foundations of

MULTICULTURAL
PSYCHOLOGY

MultiCult_TitleP.indd 1 7/21/15 3:10 PM

 foundations of
MULTICULTURAL
PSYCHOLOGY
RESEARCH TO INFORM
EFFECTIVE PRACTICE

TIMOTHY B. SMITH and JOSEPH E. TRIMBLE

American Psychological Association • Washington, DC

MultiCult_TitleP.indd 2 7/21/15 3:10 PM

Copyright © 2016 by the American Psychological Association. All rights reserved. Except
as permitted under the United States Copyright Act of 1976, no part of this publication may
be reproduced or distributed in any form or by any means, including, but not limited to, the
process of scanning and digitization, or stored in a database or retrieval system, without the
prior written permission of the publisher.

Published by To order
American Psychological Association APA Order Department
750 First Street, NE P.O. Box 92984
Washington, DC 20002 Washington, DC 20090-2984
[Link] Tel: (800) 374-2721; Direct: (202) 336-5510
Fax: (202) 336-5502; TDD/TTY: (202) 336-6123
Online: [Link]/pubs/books
E-mail: order@[Link]

In the U.K., Europe, Africa, and the Middle East, copies may be ordered from
American Psychological Association
3 Henrietta Street
Covent Garden, London
WC2E 8LU England

Typeset in Goudy by Circle Graphics, Inc., Columbia, MD

Printer: Bang Printing, Brainerd, MN

Cover Designer: Naylor Design, Washington, DC

The opinions and statements published are the responsibility of the authors, and such
opinions and statements do not necessarily represent the policies of the American
Psychological Association.

Library of Congress Cataloging-in-Publication Data

Smith, Timothy B.
Foundations of multicultural psychology : research to inform effective practice /
Timothy B. Smith and Joseph E. Trimble.
pages cm
Includes bibliographical references and index.
ISBN 978-1-4338-2057-1 — ISBN 1-4338-2057-9 1. Multiculturalism—Psychological
aspects. 2. Clinical psychology. 3. Counseling psychology. I. Trimble, Joseph E. II. Title.
HM1271.S6295 2016
305.8—dc23
2015011087

British Library Cataloguing-in-Publication Data

A CIP record is available from the British Library.

Printed in the United States of America

First Edition

[Link]
 CONTENTS

Acknowledgments...................................................................................... vii
Chapter 1. Introduction: Multiculturalism in Psychology
and Mental Health Services.............................................. 3

I. Synthesis of Multicultural Research

on Therapist Characteristics .............................................................. 19
Chapter 2. Multicultural Education/Training and Experience:
A Meta-Analysis of Surveys and Outcome Studies........ 21
Chapter 3. Therapist Multicultural Competence: A Meta-Analysis
of Client Experiences in Treatment.................................... 49

II. Synthesis of Research on the Experiences of People

of Color With Mental Health Services ............................................ 65
Chapter 4. Mental Health Service Utilization Across Race:
A Meta-Analysis of Surveys and Archival Studies......... 67

v
Chapter 5. Participation of Clients of Color in Mental
Health Services: A Meta-Analysis of Treatment
Attendance and Treatment Completion/Attrition........ 95
Chapter 6. Matching Clients With Therapists on the
Basis of Race or Ethnicity: A Meta-Analysis
of Clients’ Level of Participation in Treatment............ 115
Chapter 7. Culturally Adapted Mental Health Services:
An Updated Meta-Analysis of Client Outcomes......... 129
Chapter 8. Acculturation Level and Perceptions of Mental
Health Services Among People of Color:
A Meta-Analysis........................................................... 145

III. Synthesis of Research on the Experiences

and Well-Being of People of Color ............................................... 165
Chapter 9. The Association of Received Racism With
the Well-Being of People of Color:
A Meta-Analytic Review.............................................. 167
Chapter 10. Ethnic Identity and Well-Being of People of Color:
An Updated Meta-Analysis.......................................... 181

IV. Foundations for the Future ............................................................ 207

Chapter 11. Philosophical Considerations for the Foundation
of Multicultural Psychology.......................................... 209
Chapter 12. Firming up the Foundation for an Evidenced-Based
Multicultural Psychology.............................................. 235
Appendix: General Methods of the Meta-Analyses (Chapters 2–10)......... 249
References................................................................................................. 253
Index......................................................................................................... 297
About the Authors................................................................................... 307

vi       contents
 ACKNOWLEDGMENTS

In the late 19th century, Lone Man (isna la wican), a Lakota spiritual
leader, is thought to have said, “I have seen that in any great undertaking it
is not enough for a man to depend upon himself.” The conceptualization,
preparation, and writing of this book depended on the research conducted
by hundreds of scholars from a variety of academic disciplines whose work
we synthesized. This book also depended on the monumental efforts of
many students who searched for the research and coded the manuscripts.
Several friends and colleagues provided thoughtful guidance, criticism,
and commentary along the way. We express our profound gratitude to all
those kindred spirits and others who guided us without our awareness.
To paraphrase Lone Man, we could not have completed this undertaking
without them.
Cindy Smith and Molly Trimble shared the ebb and flow of our frustra-
tions and enthusiasm for the book from the moment we decided to embark
on our venture. Their loving spiritual encouragement and wisdom enable us
in all ways.

vii
 We acknowledge the imperfections of this book, including the long
delays between the data analyses and the publication of this volume. Our
aim to conduct multiple meta-analyses proved overly ambitious, given per-
sonal circumstances and other professional obligations. We now depend on
our colleagues to use what we have offered to continue to improve the field,
the great undertaking of infusing multiculturalism in the mental health
professions.

viii       acknowledgments

 foundations of
MULTICULTURAL
PSYCHOLOGY

MultiCult_TitleP.indd 1 7/21/15 3:10 PM

 1
INTRODUCTION:
MULTICULTURALISM IN PSYCHOLOGY
AND MENTAL HEALTH SERVICES

Recognizing that all behavior is learned and displayed in a cultural con-

text makes possible accurate assessment, meaningful understanding, and
appropriate intervention relative to that cultural context. Interpreting
behavior out of context is likely to result in misattribution.
—Paul Pedersen (2008, p. 15)

Imagine the work of a mental health professional who accepts a new

position in a close-knit community with cultural lifestyles very different from
mainstream society. The therapist was born and raised far from that commu-
nity but had been successful elsewhere. Although the therapist uses the same
approach and techniques that had previously worked well, most clients fail to
return after the first or second session. The few clients who remain in therapy
seem to understand the therapist’s intentions and respond to treatment, but
reluctantly, the therapist begins to face the fact that the approaches taken in
therapy do not align with the experiences and worldviews of most of the new
clients. The clients perceive situations in ways unanticipated by the therapist.
The clients’ explanations about emotional events seem peculiar to the thera-
pist, who realizes that trying to interpret the clients’ behavior, feelings, and
thoughts often results in misattributions. Desiring to better understand local
lifeways and thoughtways and to acquire the skills necessary to implement
that understanding, the therapist searches for evidenced-based guidelines

[Link]
Foundations of Multicultural Psychology: Research to Inform Effective Practice, by T. B. Smith and J. E. Trimble
Copyright © 2016 by the American Psychological Association. All rights reserved.

3
and resources in the professional mental health literature (G. C. N. Hall &
Yee, 2014). Where to begin?

MULTICULTURAL PSYCHOLOGY AND COUNSELING:

AN OVERVIEW

Multicultural psychology and counseling is an emerging discipline with

the potential to inform therapists of cultural considerations relevant to mental
health (Paniagua & Yamada, 2013). It is based on the premise that the ethical
provision of mental health services should include an accurate accounting of
clients’ cultural lifeways and thoughtways (Leong, Comas-Díaz, Hall, McLoyd,
& Trimble, 2014; Pedersen, 1999). As an emerging discipline, it has developed
guidelines for therapists seeking to be more effective in their work (American
Psychological Association [APA], 2003; G. C. N. Hall & Yee, 2014; Leong
et al., 2014; D. W. Sue & Sue, 2013), and it has become increasingly influential
across the mental health professions, most recently in the revised standards for
psychology graduate programs and internships (APA, 2014). Although excep-
tions persist, multicultural perspectives are becoming increasingly normative
among mental health professionals.
But to what extent are the tenets and guidelines for practice that have
arisen from multicultural perspectives based on research evidence? Psychologists
and other mental health professionals understand the benefits of using data to
inform practice and policy (APA 2005 Presidential Task Force on Evidence-
Based Practice, 2006), but to what extent has that occurred? A solid research
foundation is essential to the credibility and long-term effectiveness of multi-
cultural guidelines for practitioners.
A primary purpose of this volume is to summarize research data to inform
mental health practices relevant to client race and ethnicity, two delimited
aspects of multiculturalism. Using meta-analytic methods to summarize data
in Chapters 2 to 10, the book addresses questions that are fundamental to the
discipline. For instance, how large are racial discrepancies in mental health
service utilization and client retention, and what factors predict those racial
discrepancies? To what degree are perceptions of racism and ethnic iden-
tity associated with psychological well-being? To what extent can therapists’
training in multicultural issues and their level of multicultural competence
benefit diverse clients? These are among the key questions relevant not only
to the therapist described at the beginning of this chapter but also to every
therapist who works in a multicultural world.
Practitioners improve the effectiveness of their work when they under-
stand and apply research data (APA 2005 Presidential Task Force on Evidence-
Based Practice, 2006; G. C. N. Hall & Yee, 2014). The meta-analyses in

4       foundations of multicultural psychology

Chapters 2 through 10 of this book contain interpretations useful for prac-
titioners, students, and researchers. Practitioners and students need not be
experts in meta-analytic methods to understand the implications of the find-
ings, summarized at the end of each chapter. This book emphasizes research
findings, but that should benefit, not deter, mental health professionals seek-
ing answers. One need not be a researcher to benefit from research. The divide
between practitioners and researchers can be bridged. This book attempts to
construct a foundation for that bridge, but the reality is that research and
practice necessarily inform one another and have been doing so for decades.

Brief Historical Overview of Multiculturalism in Mental Health Services

Topics of culture, race, ethnicity, gender, religion and spirituality, sex-

ual orientation, and so forth were rarely covered in social science theories
and research until the second half of the 20th century. Mental health prac­
titioners and scholars often presumed that theories and research findings
could be applied to everyone, so they sought to establish “universal validity”
(Dawson, 1971, p. 291). Although they acknowledged that different cultures
exist around the world, most concerned themselves almost exclusively with
the majority population in their own narrow segment of the global society.
And they often reasoned that cultural influences were insufficiently strong to
merit serious consideration, let alone merit the time required to gain in-depth
familiarity and proficiency across cultures. Culture was seen as a nuance, with
the substance of theories and research presumed universal, enduring across
circumstances.
The rise of multicultural psychology and counseling in North America
came following the expansion of civil rights to historically oppressed popula-
tions and paralleled the diversification of the population in the final decades
of the last century. Mental health professionals began to realize that although
much of human experience is universal (e.g., we desire companionship and
grieve at its loss), interpretations of experience are informed by circumstances,
values, and worldviews that differ from culture to culture. “It is by no means
self-evident that a concept embodied in a theory that has its origins within a
particular culture can necessarily be operationalized into a conceptual equiva-
lent in a different culture” (Jahoda, 1979, p. 143). For instance, child rearing
is universally essential to human survival irrespective of culture, but child-
rearing practices differ dramatically from one culture to another (Whiting,
1963). Psychology that had ignored cultural differences was “guilty of suggestio
falsi [because] textbooks and articles commonly implied universality without
seeking to provide any grounds for their implicit claims” (Jahoda, 1988, p. 93).
Multiple factors influence emotional well-being and mental health, and the
field gradually began to account for those contextual variables.

introduction      5
 Inclusion of multicultural perspectives began to spread during the 1970s
when increased numbers of women and individuals from diverse backgrounds
received graduate degrees in the mental health professions and joined together
to form professional associations on multicultural issues. In 1972, for example, a
group of psychologists from different countries convened in Hong Kong to criti-
cally examine and discuss culture’s influence on the human experience (Lonner,
2000). The meeting led to the founding of the International Association for
Cross-Cultural Psychology. Two years earlier, the well-established and distin-
guished Journal of Cross-Cultural Psychology was launched (Berry, Poortinga,
Segall, & Dasen, 1992). Many other organizations with an emphasis on multi­
cultural issues also established research journals because mainstream public­ations
did not represent those considerations. In 1974 the first issue of the Journal of
Black Psychology appeared. In 1978, the White Cloud Journal of American Indian/
Alaska Native Mental Health was founded (and was renamed American Indian
and Alaska Native Mental Health Research, the Journal of the National Center in
1987). The Hispanic Journal of Behavioral Sciences and the Asian American Journal
of Psychology were first published in 1979. With publication outlets available,
opportunities for scholarship broadened.
During the 1980s and 1990s, the amount of research focusing on multi­
cultural issues increased markedly. Professional conferences such as the Winter
Roundtable at Teachers College, Columbia University, strengthened networks
and collaborations. Scholarly books began to appear with regularity. The APA
began publishing a series of annotated bibliographies to help cohere the accu­
mulated research findings. The series’ topics include African Americans (Evans
& Whitfield, 1988; Keita & Petersen, 1996), Hispanic/Latino(a) Americans
(Olmedo & Walker, 1990), Asian Americans (Leong & Whitfield, 1992), and
North American Indians (Trimble & Bagwell, 1995). By the end of the 1990s
APA’s Division 45 journal Cultural Diversity and Ethnic Minority Psychology
had appeared (previously titled Cultural Diversity and Mental Health), and
three APA divisions sponsored the first National Multicultural Conference
and Summit. It had taken several decades, but multicultural perspectives had
achieved professional recognition (D. W. Sue, Bingham, Porché-Burke, &
Vasquez, 1999).

Brief Overview of Contemporary Contexts

Infusion of multiculturalism into mental health practices, training pro-

grams, and policies is underway. Mental health professionals increasingly
understand “that all behavior is learned and displayed in a cultural context”
and that accounting for clients’ cultures “makes possible accurate assessment,
meaningful understanding, and appropriate intervention relative to that cul-
tural context” (Pedersen, 2008, p. 15). Over the past 4 decades mental health

6       foundations of multicultural psychology

services delivered to ethnic minority populations in the United States and
Canada has grown dramatically in terms of general availability as well as
in the range of care offered. This growth can be attributed to a number of
factors, notably changes in national public health policies, increasing com-
munity resources and expertise, and community demands for more compre-
hensive and culturally relevant care.
The rapid expansion of mental health services to diverse populations
has frequently preceded careful consideration of several critical components
of such care, specifically the delivery structure itself, treatment processes,
program evaluation, epidemiological data, and preventive strategies. Clarity
is lacking, and mainstream journals and professional publications persist in
inadequately addressing multicultural issues (Henrich, Heine, & Norenzayan,
2010). The relevance and applicability of general psychological knowledge
across diverse populations remain uncertain (e.g., Leong, Holliday, Trimble,
Padilla, & McCubbin, 2012; S. Sue, 1999). Multiculturalism too often remains
separated from mainstream discussions about mental health services (Wendt,
Gone, & Nagata, 2014). To better integrate multicultural considerations into
mainstream practices, government agencies are implementing more cultur-
ally sensitive mental health programs along with more accurate research and
reporting. For example, under the directive of health disparities research
national agencies have developed initiatives to promote preventive interven-
tion efforts in ethnic minority communities. These interests and the initia-
tives are positive steps and have potential for improvement of mental health
conditions among historically oppressed populations.
Looking to the future, multicultural psychology and counseling must
now establish a solid foundation of research to better meet pressing needs in a
pluralistic society. North America is increasingly culturally diverse (Statistics
Canada, 2011; U.S. Census Bureau, 2010). Individuals with ancestry from
Africa, Asia, and Central and South America, along with peoples indige-
nous to North America and the Pacific Islands, will eventually constitute the
majority of the population (U.S. Census Bureau, 2010). Accounting for cul-
tural differences can no longer be the concern of professionals chiefly working
in urban ethnic enclaves or isolated rural communities. Demographic reali-
ties signal that mental health services must account for cultural differences to
meet the needs of the majority of clients seeking services. Whereas a therapist
like the one described at the beginning of this chapter would be struck by the
cultural contrasts evident in an unfamiliar environment, therapists working
in familiar settings may only occasionally realize the realities of diversity and
take action accordingly.
The field now needs a translational pathway from practice to research
and back to training, driven by demographic realities and clients’ needs.
New priorities for research, teaching, and practice must be developed so

introduction      7
that current knowledge and new knowledge in psychology becomes relevant
and applicable across diverse contexts. Demographic changes will inevitably
move the field toward the full consideration of diversity in ways that are
inclusive and representative. How soon and with what tools?

FOUNDATIONAL MULTICULTURAL ISSUES

IN THE MENTAL HEALTH PROFESSIONS

Multicultural scholarship in the mental health professions is so broad,

encompassing global diversity in all its varieties, that it can appear fragmented
and diffuse—and thus hamper the credibility and effectiveness of the field. To
overcome this limitation, multicultural scholarship should articulate a core
set of principles and address major challenges to those principles to facilitate
genuine improvements in mental health practices. This volume provides a
partial remedy by articulating some principles and addressing their major
challenges empirically.
One key principle is that therapists must remain focused on the fun-
damental issues impacting the mental health of historically oppressed popu­
lations. Such fundamental issues pertinent to race and ethnicity include
(a) the degree of client access and involvement in mental health treatment
as a function of race or ethnicity; (b) the degree to which the experiences of
clients of color in therapy are associated with their level of acculturation and
the racial and ethnic background of the therapist; (c) the influence of cultural
experiences, particularly racism and ethnic identity, on client well-being;
and (d) the effectiveness of treatment as a function of therapist multicultural
competence, therapist training in multicultural competence, and cultural
adaptations and culture-specific approaches to treatment. Although other
critical issues merit consideration, this volume focuses on and evaluates data
relevant to these four particular topics because they are central; they address
the interaction between treatment and the cultural experiences of clients
seeking treatment.

Client Access to and Involvement in Mental Health Services

Although in an optimal world mental health services would be acces-

sible to and used by people of all backgrounds, racial discrepancies were iden-
tified by the U.S. Surgeon General in mental health service utilization (U.S.
Department of Health and Human Services, 2001). The ideal of universal
access to mental health services in many urban ethnic enclaves and in most
rural communities falls short, but to what degree are people of color systemati-
cally disadvantaged? And when people of color enter treatment, how likely

8       foundations of multicultural psychology

are they to complete it? As depicted in the scenario at the start of this chap-
ter, cultural factors unaccounted for by universalistic treatment approaches
can result in premature client discontinuation. Mental health professionals
must constructively confront racial and ethnic discrepancies in service uti-
lization and retention, if those discrepancies currently persist more than a
decade after the report of the U.S. Surgeon General. Research can ascertain
the nature and extent of racial disparities, factors contributing to the discrep-
ancies, and solutions.

Association of Client Acculturation and Therapist Race and Ethnicity

With Client Experiences in Mental Health Treatment

Despite the findings of the Human Genome Project that ethnic and
especially racial distinctions have no biological basis (Bonham, Warshauer-
Baker, & Collins, 2005), these constructs remain integral aspects of our social
fabric (Gómez & López, 2013). Racial and ethnic categories align with politi-
cal and social structures that continue to influence individuals and com-
munities. In part because of they are so integral to sociopolitical contexts,
it is often difficult to separate race and ethnicity from socioeconomic status
and experiences of migration, acculturation, and discrimination. At times,
categorical race and ethnicity may serve as a proxy for those variables (e.g.,
individuals’ ethnic self-identifications vary as a function of acculturation).
Given this complexity and the multiple problems inherent in approaches
that perpetuate ethnic gloss1 (Trimble, 1990, 1995; Trimble & Bhadra, 2013),
should scholars move beyond simplistic categories of race and ethnicity and
develop constructs that account for the reality of multivariate convergence
in these categories? Among many other factors, the answer to this question
depends on whether the individual or group experience differs substantially
in terms of acculturation style and assimilation to mainstream North American
society (which strongly overlap with race and ethnicity). In mental health
settings, does client acculturation style predict experiences and outcomes in
treatment? Alternatively, are race and ethnicity so important to clients that
the categorical race or ethnicity of the therapist affects the client’s willingness
to engage in treatment?

1Ethnic gloss is an overgeneralization stemming from simplistic labeling of ethnocultural groups, such as
Native American Indians (consisting of over 500 tribes), that ignores differences between and within
groups. An ethnic gloss presents the illusion of homogeneity where none exists and therefore may be
considered a superficial, almost vacuous, classification that further separates groups from one another
(Trimble & Bhadra, 2013).

introduction      9
Association of Racism and Ethnic Identity With Well-Being

The multicultural literature has long emphasized that therapists must

be keenly aware of clients’ cultural experiences and lifestyles (N. B. Miller,
1982). Understanding clients’ experiences with racism, for instance, would
be important for a therapist because those experiences could be relevant to
clients’ presenting problems or at the very least could exacerbate distress.
Hence, therapist efforts to ascertain not only clients’ experiences of racism
but also the degree to which those experiences affect client well-being should
inform treatment approaches otherwise ignorant of that particular distress.
Similarly, knowing a client’s strength of ethnic identification could at the
very least inform a therapist’s understanding of client self-perceptions, and if
the therapist also understood how the client’s ethnic identity was associated
with psychological coping mechanisms, emotional support from community
members, and other resources relevant to well-being, therapy would be further
strengthened. Clients’ cultural supports, resources, and sources of distress are
clearly relevant to therapy, but to what degree? To what extent are level of
ethnic identity and experiences of racism associated with individuals’ emo-
tional well-being and distress?

Therapist Multicultural Competencies, Multicultural Training,

and Cultural Adaptations and Culture-Specific Approaches
to Treatment

Therapist abilities useful for working with diverse clients have been
termed multicultural competencies, commonly broken down into components
of knowledge, skills, and awareness (Arredondo et al., 1996). Multicultural
competencies articulate ways of enhancing the therapeutic alliance and meet-
ing client needs through strategies and approaches that explicitly account for
cultural contexts. For instance, work with culturally diverse clients can be
enhanced when mental health professionals account for (a) their own cul-
tural worldview, (b) the client’s cultural worldview, (c) the interaction
between their own worldviews and those of the client, including assump-
tions related to therapy processes, and (d) the culture of the environment
in which the therapy occurs (Pedersen, Draguns, Lonner, & Trimble, 2008).
Combining these possible conditions, therapists could find themselves, in a
rather extreme case, “working with a client from another culture, on a prob-
lem relating to a third culture, in the environment of a fourth culture where
each par­ticipating culture presents its own demands” (Pedersen, Draguns, &
Lonner, 1976, p. vii). Scholars have asserted that the cultural complexities
associated with providing mental health services necessitate multicultural
competencies distinct from general therapy skills (e.g., Arredondo et al.,

10       foundations of multicultural psychology

1996), but to what extent do clients’ outcomes benefit from therapist multi-
cultural competence? To what degree do training programs facilitate therapist
multicultural competence?
An important component of multicultural competence is flexible adap-
tation to clients’ cultural experiences and worldviews, with resulting cultural
adaptations to treatment protocols, and procedures that extend beyond con-
ventional practice (G. Bernal & Domenech Rodríguez, 2012). Although it
seems obvious that treatment should account for clients’ experiences and
worldviews, a tension can arise between the benefits of aligning treatment
with individual clients and the benefits of systematic implementation of tra-
ditional forms of therapy with fidelity to the intervention model (Castro,
Barrera, & Holleran Steiker, 2010). To what degree do cultural adaptations to
traditional treatments improve client outcomes or hamper them because they
diverge from established evidence-based practices? The answer to this ques-
tion has profound implications for the future of the profession. If adapting
treatments to align with clients’ worldviews is more effective than standard-
ized approaches, then the field has only begun to account for the breadth and
depth of individuality contextualized within multiple systems. If culturally
adaptations prove to be equivalent in effectiveness to established, standard-
ized approaches, then multiculturalism can remain relegated to a secondary
consideration within the broader profession, useful in circumstances when
treatment as usual no longer seems to work, as in the scenario described at
the start of the chapter. Stated differently (and in direct opposition to the
presumed universalism characteristic of the past), to what extent must indi-
viduals’ worldviews and experiences, embedded in cultural, familial, environ-
mental, and economic circumstances, permeate mental health treatments?

BUILDING UP THE FOUNDATION FOR MULTICULTURAL

PSYCHOLOGY AND COUNSELING

Multicultural psychology and counseling is at a key juncture in terms of

its influence on the mental health professions. Following decades of incremen-
tal growth, scholarship inclusive of multicultural issues has reached a proverbial
“tipping point” (Gladwell, 2006), with the potential to now pervade all aspects
of the professions, as has been envisioned (Pedersen, 1999). An analysis of
40 years of citations cataloged in PsycINFO shows a remarkable increase in
the number of citations that reference racial and ethnic groups: About 2% of
scholarly manuscripts referred to racial or ethnic groups in the 1960s, which
doubled to about 4% in the 1980s, and doubled again to about 8% in the 2000s.
Citations making references to Africans and African Americans increased
from about 2,000 across the entire decade of the 1960s to over 5,000 in the

introduction      11
single year of 2014. References to Asians and Asian Americans increased from
about 1,000 citations across the decade of the 1960s to over 8,000 in the single
year of 2014. In 2014, over 4,000 manuscripts mentioned Hispanic/Latinos(as)
and over 500 mentioned First Nations peoples, Native American Indians, or
Alaska Natives. Overall, more than 23,000 manuscripts in PsycINFO men-
tioned concepts of race, ethnicity, or culture in 2014 alone. Whereas in previ-
ous decades scholars urged the “vigorous expansion” (M. E. Bernal & Castro,
1994, p. 797) of research on multicultural topics, the clear and present need is
now for a distillation and synthesis of this rapidly expanding literature.

Need for Literature Syntheses Using Meta-Analytic Methods

Multicultural psychology no longer lacks numbers of interested scholars;

it instead lacks a coordinated approach to scholarship, informed by data instead
of opinion. Supporters of multicultural psychology can sometimes press for-
ward, unwittingly advocating measures and approaches that go beyond the data
of what has been confirmed by research evidence. As a consequence, skeptics
of multicultural psychology continue to point to overgeneralized statements
about multicultural issues and to scattered and contradictory research findings
(e.g., O’Donohue & Benuto, 2010). This lack of clarity helps no one.
With tens of thousands of manuscripts now appearing every year on
issues relevant to race or ethnicity, a massive amount of information is avail-
able, but traditional narrative review methods would clearly be inadequate
to accurately summarize so much data. A solution is available: Meta-analysis,
the quantitative “study of studies.”
Meta-analysis . . . [is] the statistical analysis of a large collection of analysis
results from individual studies for the purpose of integrating the findings.
It connotes a rigorous alternative to the casual, narrative discussions of
research studies which typify our attempts to make sense of the rapidly
expanding research literature. (Glass, 1976, p. 3)
Meta-analysis aggregates quantitative data to provide a descriptive summary
of the results. Statistical models combine data across many individual manu-
scripts to estimate the overall strength of the effect or the relationship, the
averaged effect size. Meta-analyses have become normative in scientific jour-
nals, and scholars rely heavily on their results.
Meta-analytic methods offer multiple advantages over impressionistic
summaries of research findings. When research findings are inconsistent, which
is certainly the case in multicultural psychology (and thus only broad, ten-
tative conclusions are possible in narrative literature reviews), meta-analysis
can identify sources of variation across studies; for example, meta-analysis can
ascertain the degree to which findings differ across participant characteristics
(e.g., age, gender) and study characteristics (e.g., research design, measurement

12       foundations of multicultural psychology

used). Knowing an average effect size and, even more important, the conditions
under which effect sizes vary benefits both practitioners who use evidenced-
based practices and scholars who seek to build on current findings when design-
ing new research questions and new treatments. Whereas narrative literature
reviews provide information based on expert opinion, meta-analyses summa-
rize research data.
Reliance on aggregated data can improve the mental health professions
(APA 2005 Presidential Task Force on Evidence-Based Practice, 2006). A
failure to examine aggregated data in multicultural psychology would pose a
significant roadblock to the field, particularly if decisions about the content of
graduate training and about reimbursements for professional services become
restricted to empirically supported treatments (ESTs) and evidence-based prac-
tices (EBPs; see also G. C. N. Hall & Yee, 2014). Although rigid and narrow
decision making about ESTs and EBPs can bestow empirical data with a false
aura of truth (Slife & Williams, 1995), meta-analytic methods should inform
consequential decisions so long as data interpretation includes contextual fac-
tors. Chapters 2 through 10 in this book use meta-analytic methods to examine
the existing literature, identify gaps in present understanding, and suggest areas
for future inquiry. Chapter 11 covers issues pertinent to data interpretations.

Limitations of Meta-Analyses and This Book

Limitations characterize every scholarly undertaking. The approaches

taken in this book are necessarily qualified by limitations associated with
meta-analytic methodology and by the focus of our work, delimited to selected
topics relevant to race and ethnicity.
Meta-analyses describe research findings, and those descriptions neces-
sarily depend on the research available. When research data are limited, the
conclusions of a meta-analysis must remain tentative. And when research data
differ as a function of research design, those factors must also be considered.
Poor quality research can yield unreliable results, even in the aggregate (“gar-
bage in, garbage out”). Nevertheless, removing studies a priori from meta-
analytic reviews can restrict the generalizability of the findings: The tighter
the methodological controls within studies, the more likely those studies rep-
resent solely the conditions in which the study was undertaken. Meta-analyses
involve trade-offs between internal and external validity. In most of the chap-
ters of this book, our meta-analytic approach erred on the side of inclusion
of research manuscripts, maximizing external validity, rather than excluding
research manuscripts based on methodological considerations. However, we
also accounted for factors relevant to internal validity by analyzing differences
in findings across methods (e.g., convenience sampling vs. random selection
of participants). Thus we attempted to address both internal and external

introduction      13
validity. This approach was warranted when prior meta-analyses were unavail-
able and primary sources of variation in data were largely unknown: It was
preferable to examine all research findings and then evaluate the degree to
which findings vary as a function of methodology rather than to exclude
studies based on hypothetical variation. When previous meta-analyses had
been conducted in a given topic area (e.g., cultural adaptations of mental health
services, multicultural education and training), we restricted our analyses to
experimental and quasi-experimental research designs, with our data extrac-
tion from manuscripts accounting for differences in methods (e.g., comparison
groups using bona fide treatments vs. wait-list controls). Although we attended
to considerations relevant to both internal and external validity, ultimately
meta-analyses remain descriptive: They portray the state of current practice
in aggregate form, which does not necessarily represent the experiences of
any individual client or therapist.
Another limitation of the approach taken in this book is its delimited
focus. We address multicultural counseling competence and a few selected
topics relevant to race and ethnicity. Although a targeted focus has several
advantages, including the fact that a broader coverage of human diversity
using meta-analytic methods would be virtually impossible in a single book,
there are several disadvantages that deserve mentioning. First, emphasis on
any particular variable to the exclusion of others obscures the holistic reali-
ties of human experience. An individual can never be understood solely in
terms of race and ethnicity, no matter how important those particular sources
of identity may be to the person or to the society in which the person resides.
Second, race and ethnicity are often conceptualized in terms of discrete
categories, yet people vary substantially in terms of their experiences, attri-
butes, and degrees of identification. Variability within purportedly homoge-
neous racial and ethnic groups is many times larger and often more complex
than variability between groups (Trimble, 2007; Trimble & Dickson, 2005).
Moreover, the processes of racial and ethnic identity development can be
complex enough for individuals with a clearly defined racial and or ethnic
heritage, let alone for biracial and multiracial individuals, individuals adopted
by parents not of their same racial or ethnic heritage, and so forth. Third,
racial and ethnic categorizations perpetuate stereotypes. No research finding
presented in this volume will be completely accurate for a particular client
or therapist, so research findings specific to racial and ethnic groups can be
considered tentative possibilities for exploration. Thus, although this book
provides useful information for mental health professionals, the information
retains its benefit only to the degree that the reader uses it along with all
other sources of information available. In that sense, the content of this vol-
ume provides the reader with an opportunity to use a key professional skill:
Learn from data and improve therapeutic practices accordingly, but always

14       foundations of multicultural psychology

remain focused on the individual client’s needs and experiences. Students
and practitioners uninterested in the data tables can still benefit from the
interpretations of the findings provided at the end of each chapter. The gap
between research and practice is only as wide as our ability to bridge it.

Research Data Versus Expert Opinion

Some practitioners may question our emphasis on research and meta-

analytic findings as the foundation for effective multicultural mental health
services. Why so much insistence on evidence? A parallel from the history of
health care may prove persuasive to readers doubtful of this emphasis. Prior
to rigorous research becoming the norm in the medical profession, expert
opinion was the primary foundation for practices, yet death rates for individ­
uals in medical treatment were excessively high, even after common sources
of infections were understood and antibiotics had become available in the
20th century (Bynum & Porter, 2013). Replacing reliance on expert opin-
ion with reliance on high quality research made the difference. Systematic
lines of research identified risk factors and causal mechanisms and improved
treatment effectiveness for a broad range of health conditions, not merely
infectious diseases. A pervasive reliance on research data prevents illness and
saves lives (Watkins & Portney, 2009).
Social scientists and mental health professionals understand that all
empirical research, including medical research, is fraught with problems (Slife
& Williams, 1995). Nevertheless, the benefits of relying on research evidence
outweigh both the many disadvantages and the advantages of alternatives.
Expert opinion is no substitute for evidence. And the reality is that mental
health professionals in general (with some notable exceptions) have not sub-
jected their explanations and treatments to scrutiny as intense and as system-
atic as is necessary to clearly distinguish information from opinion.
Like related disciplines, multicultural psychology and counseling could
be accurately described as having been more reliant on expert opinion than
on data. One purpose of this volume is to shift the conversations in the men-
tal health professions toward greater inclusion of contextual factors, particu-
larly culture, race, and ethnicity. But an equally important objective is to
have those conversations become more reliant on data than opinion.

OVERVIEW OF BOOK CONTENT

Foundational questions about mental health and mental health services

across racial and ethnic groups involve multiple considerations, including
client experiences with treatment, factors that influence client well-being,

introduction      15
and therapist characteristics. This book addresses each of those broad topic
areas for an audience of practitioners, students, and researchers in mental
health professions.
The first section of the book attends to the therapist characteristics of
multicultural competence and multicultural training. Chapter 2 investigates
the degree to which therapists’ training in and experiences with multicultural
issues relates to their work with clients of color. Training in multicultural issues
is mandatory for graduate students in accredited programs, but how effective is
that training? Chapter 3 deals with the topic of therapist multicultural com-
petence. To what extent do therapists’ purported knowledge, awareness, and
skills relevant to multicultural considerations affect clients’ experiences in
treatment?
The second section of the book focuses on client experiences with treat-
ment as a function of race and ethnicity. Chapter 4 addresses the issue of utili-
zation: How large are racial and ethnic discrepancies in mental health service
utilization? This question is among the most important for mental health
providers to address. If people of color who are in need of mental health ser-
vices are not receiving them, the field needs to rectify systematic inequities.
Chapter 5 responds to a related question: How large are racial and
ethnic discrepancies in mental health treatment participation? Clients dis­
continue treatment for a variety of reasons, but the extent to which clients of
color prematurely discontinue treatment due factors related to race and eth-
nicity requires serious attention. Pursuing this same line of inquiry, Chapter 6
evaluates the extent to which clients remain in mental health services as a
function of the race of the therapist.
Chapter 7 discusses the degree to which the outcomes of clients of color
can be improved when adapting treatment to align with the client’s cultural
background. Previous research has shown that culturally adapted treatments
are more effective than treatments not explicitly accounting for client cul-
ture, and this chapter provides an updated review of that literature.
An associated issue of client level of acculturation receives attention
in Chapter 8. Clients’ attitudes about and experiences with mental health
treatments could vary substantially based on the degree of their acculturation
to North American cultural mores.
The third section of the book addresses two topics relevant to psycho-
logical well-being. Chapter 9 investigates the degree to which experiences of
racism among people of color are associated with their well-being. Chapter 10
provides an updated review on the association between the ethnic identity of
people of color and their reports of well-being.
The fourth section of the book reflects on the overall state of the field.
Diverging sharply from the preceding data-focused chapters, Chapter 11
encourages the reader to consider the underlying assumptions and popular

16       foundations of multicultural psychology

beliefs characterizing contemporary multicultural psychology and counseling
research. Purposefully distinct in tone and content, this chapter invites the
reader to take the crucial steps of asking hard questions and engaging in critical
analysis. Improvement of mental health services for multicultural populations
depends on what kinds of questions are being asked in research, how concepts
are operationalized in research, and many other factors requiring careful con-
sideration. Chapter 12 summarizes the findings of the meta-analyses reported
in Chapters 2 through 10 and provides some recommendations for the future.
Overall, this book covers several topics relevant to race and ethnic-
ity that can affect mental health and mental health treatment. The book
addresses client access to and involvement in mental health treatment, condi-
tions that affect client experiences in treatment, experiences that influence
individuals’ well-being, and factors that influence the effectiveness of treat-
ment. Although many more topics deserve consideration, the focus of this
book on research data and its invitation for critical analysis provide a founda-
tion for the work of students, scholars, and practitioners invested in the well-
being of all people.

CHAPTER SUMMARY

For decades the mental health professions have been selective in the
study and characterization of people. Most of the early research occurred in
a monocultural vacuum involving restricted classes of research participants
from Europe and North America. These populations most often studied in
social science research have been referred to as WEIRD (Western, educated,
industrialized, rich, and democratic; Henrich et al., 2010). Among other
sources, Robert Guthrie’s (1976) book, Even the Rat Was White, documented
systematic historical racial biases in psychological research and practice.
Although many racial and ethnic groups remain underrepresented in the
mental health literature, representation has increasingly been achieved (e.g.,
Case & Smith, 2000). Mental health professionals are beginning to under-
stand that multicultural considerations are central to the experiences of many
clients and that therefore these considerations should be central to their work
with these clients (Leong et al., 2014). Mental health professionals increasingly
seek information about how to better account for cultural contexts in their
work (Pedersen, 1999). They also strive to improve mental health service uti-
lization and retention rates among historically underserved populations. They
are interested in the multitude of ways in which the ethnic identity of clients
of color and their experiences with racism affect their emotional well-being.
They wonder whether cultural adaptations to existing mental health treat-
ments are justified and whether culture specific approaches are warranted.

introduction      17
They seek confirmation that multicultural education and the acquisition of
certain skills, referred to as multicultural competencies, will genuinely benefit
diverse clients seeking their services. They have many questions about the
complexity of cultural realities, and they seek answers.
This book responds to several key questions by summarizing available
research data via meta-analysis. Many books on multicultural psychology
have been published, and expert opinions on multicultural considerations
in mental health treatment have been widely circulated. But which of the
many recommendations and practices are based on evidence? A synthesis of
research findings should assist in supporting or refuting opinions, popular or
not. Although far from yielding definitive answers, research findings present
the most solid foundation on which a field of multicultural inquiry could be
built. And multicultural issues are so important to mental health practices
that no other foundation should suffice.
In sum, mental health professionals cannot fully understand the human
condition without viewing it through a lens informed by multiculturalism.
Even with the aid of this lens, the complexity of individual variations embed-
ded within multiple systems poses enough challenges to make multicultural
understanding a lifelong quest (T. B. Smith, Richards, Granley, & Obiakor,
2004). Avoiding the complexities of the human condition, including but not
limited to race and ethnicity, is not a sustainable option for purported experts
in human behavior, mental health professionals. Minimizing that complexity
or rationalizing it away through universalistic assertions has been the norm
in the past, but the harmful consequences of such minimizations become
easily apparent when working with diverse clientele whose experiences and
worldviews do not fit supposedly universal conceptualizations (D. W. Sue,
2015a). Moreover, mental health services, circumscribed and confined to
European and North American academic notions about treatment modal­
ities and well-being, remain restricted in their scope and in their potential
for ongoing refinement that could instead be expanded by the holistic, multi­
faceted conceptualizations of multiculturalism. Multiculturalism is not merely
a perspective to adopt when meeting someone perceived to be “different.”
Multiculturalism seeks to convey knowledge of factors that are part and parcel
of the human condition. The sooner mental health professionals account for
and embrace the facts of human diversity, the better they will be able to serve
the next client who seeks their services. Whether that audacious claim is
brash rhetoric (aka expert opinion) or is an invitation justified by research
evidence remains to be seen in the data.

18       foundations of multicultural psychology

 I
SYNTHESIS OF
MULTICULTURAL RESEARCH
ON THERAPIST
CHARACTERISTICS
 MULTICULTURAL EDUCATION/
2
TRAINING AND EXPERIENCE:
A META-ANALYSIS OF SURVEYS
AND OUTCOME STUDIES

Mental health professionals have a moral and ethical responsibility to

facilitate client well-being. In a pluralistic society, this responsibility includes
working in ways congruent with clients’ cultural backgrounds, commonly
referred to as multicultural counseling competencies. Descriptions of multi­
cultural competence have provided essential guidance for mental health profes-
sionals for over 40 years (D. W. Sue et al., 1982). For 3 decades, the American
Psychological Association (APA; 1986) has required accredited programs to
address multicultural issues in the curricula. APA guidelines (1994, 2003) have
clarified the importance of multicultural education for psychologists, with
other mental health professional associations having similar requirements and
guidelines. Multicultural education has become a primary strategy for improv-
ing therapists’ abilities to serve diverse clients effectively (Abreu, Gim Chung,
& Atkinson, 2000). “The critical importance of training psychologists and

A complete list of references for the studies included in this chapter’s meta-analysis is posted online.
Readers can consult it at [Link]
[Link]
Foundations of Multicultural Psychology: Research to Inform Effective Practice, by T. B. Smith and J. E. Trimble
Copyright © 2016 by the American Psychological Association. All rights reserved.

21
mental health professionals for work in an increasing multicultural society is
now unquestioned” (Ponterotto & Austin, 2005, p. 19).
The entire discipline of multicultural psychology and counseling has
been based on the assumption that multicultural awareness, knowledge, and
skills can be taught and learned (Abreu et al., 2000; Rogers & O’Bryon,
2014). Although multicultural counseling competencies overlap with gen-
eral counseling competencies (Sheu & Lent, 2007), they require specialized
instruction (Cates, Schaefle, Smaby, Maddux, & LeBeauf, 2007; Yutrzenka,
1995). Evaluating that specialized instruction can verify its effectiveness
(does it actually work?) and suggest ways for enhancing the abilities of thera-
pists to meet the needs of diverse clients, the aim of the discipline.
This chapter provides a brief overview of relevant considerations and
previous research on multicultural education, followed by a quantitative sum-
mary of research findings using meta-analytic methods. Implications of the
results of the meta-analysis conclude the chapter.

NARRATIVE REVIEW OF THE LITERATURE

The objective of multicultural education should be to enhance multi-

cultural competency (Abreu et al., 2000). To what degree is it meeting that
objective? Does multicultural education improve therapists’ abilities to effec-
tively serve diverse clients?

Relevant Theory

The tripartite model of multicultural competence (D. W. Sue et al., 1982)

provided the initial blueprint for multicultural education in psychology and
counseling. Standards subsequently derived from this model provided addi-
tional guidance (D. W. Sue, Arredondo, & McDavis, 1992b). Later opera-
tionalization of the competencies provided even more specific objectives
(Arredondo et al., 1996). Most subsequent scholarship in multicultural edu-
cation has aligned with the general paradigm of multicultural competence.
For instance, Ponterotto (1997, 1998; Ponterotto & Austin, 2005) has iden-
tified exemplary practices, proposed a model, and identified characteristics
of effective multicultural trainers and mentors, multicultural trainees, and
multicultural training environments. A cross-cultural triad training model
and related content were developed and refined by Pedersen (2000), who also
generated an extensive list of training activities and resources (2004). Brislin
and colleagues have produced many volumes on intercultural training that
can easily be applied to mental health practitioners (e.g., Landis & Brislin,
2013). Ridley and colleagues have provided a nearly comprehensive model

22       foundations of multicultural psychology

for multicultural program development (Ridley, Mendoza, & Kanitz, 1994),
specific guidelines for developing multicultural coursework (Ridley, Espelage,
& Rubinstein, 1997), and a model of general counseling competence that
explicitly integrates culture (Ridley, Mollen, & Kelly, 2011a). These models
provide valuable conceptual foundations for instruction designed to enhance
competence in general and multicultural counseling. Instructors should use
these and similar foundational resources.

Narrative Review of Previous Research

Just as there is no single “psychotherapy,” there is no single type of multi­

cultural education. The content, format, duration, intensity, and techniques
vary substantively among programs (Ponterotto & Austin, 2005). Most of
the published literature has addressed university classes in graduate programs,
although ongoing professional development for practicing clinicians is also
essential (Rogers-Sirin, 2008). The limited literature investigating multi­
cultural education provided at predoctoral internship sites indicates variabil-
ity in the extent and effectiveness of training (R. M. Lee et al., 1999), which
consists mostly of brief seminars or workshops typical for clinical settings,
with a few notable exceptions (e.g., Sevig & Etzkorn, 2001).
Although seminars and workshops for interns and clinicians can be
delimited to a specific topic (e.g., family therapy with Cuban Americans),
semester-long graduate classes devoted to multicultural psychology typically
attempt to cover a broad range of topics using a wide variety of methods
and techniques. Two separate reviews of syllabi from graduate classes across
the United States (Pieterse, Evans, Risner-Butner, Collins, & Mason, 2009;
Priester et al., 2008) found coverage of broad topics—notably race, culture,
gender, sexual orientation, age, and physical and cognitive abilities—but treat-
ment was also given to specific subgroups, such as Amish and Appalachians.
This remarkable breadth of coverage impedes depth of coverage, and schol-
ars have requested greater clarity in what content should be taught (Newell
et al., 2010). This book, hence this chapter, focuses on race and ethnic-
ity; nevertheless, guidelines regarding structure, design, instructional strate-
gies, and training philosophy are available in the literature (Ridley et al.,
1994, 1997).
The philosophical assumptions underlying multicultural education dif-
fer considerably across programs, from universal approaches that emphasize
shared humanity to culture-specific and race-based approaches (Carter &
Qureshi, 1995). Although a broad range of approaches allows flexibility for
instructors, “the lack of a unifying framework often results in confusion and
the tendency for trainers and educators to use a hodgepodge approach to
facilitating students’ multicultural competence” (Ancis & Ali, 2005, p. 95).

multicultural education/training and experience      23

 Although most instructors indicate that multicultural competencies
provide the foundation for their teaching (Reynolds, 2011), multicultural
education has typically been delimited to instruction designed to enhance
multicultural knowledge and self-awareness, failing to foster the develop-
ment of multicultural skills. A survey of 169 instructors found that the vast
majority emphasized multicultural awareness and sensitivity (93%) and
knowledge and content (82%), but only half (49%) addressed multicultural
skill development (Reynolds, 2011).
Perhaps more accurate estimates than those based on instructor self-
report come from two content analyses of syllabi collected from graduate
classes across the nation. The first of these (Priester et al., 2008) indicated
that only nine of 64 syllabi (12%) required substantive evaluations of stu-
dents’ skills, whereas 84% based evaluation on their knowledge. The authors
concluded that multicultural courses “almost completely ignore the develop-
ment of skills” (p. 29). A second review of 54 multicultural course syllabi from
across the nation (Pieterse et al., 2009) found that only 40% mentioned rel-
evant counseling interventions, with a limited 19% covering culture-specific
interventions. Evaluation of participant counseling skills occurred only 7%
of the time. Participant self-reported multicultural competence was evaluated
13% of the time. The author found it “disconcerting that multicultural courses
seem deficient in specific skill-based instruction” (Pieterse, 2009, p. 109). Thus,
the field appears to be engaged in providing content-oriented instruction, rather
than facilitating genuine shifts in abilities and worldviews (Reynolds, 1997).
Improvement is obviously needed.
Multicultural competencies appear to be used as general guidelines for
content rather than standards for evaluation of mastery. The two reviews of
syllabi from multicultural education classes consistently found assignments
indirectly assessing multicultural knowledge and self-awareness, including
journal writing (56%); a cultural self-examination paper (42%); reaction
papers to art, literature, or film (12%–34%); attendance at cultural events
(34%); a class presentation (33%); and an interview of a member of a dif-
ferent cultural group (31%; Priester et al., 2008). Infrequently used (thus
innovative) assignments fostering and evaluating skill development included
designing a culturally appropriate intervention, reviewing videotaped clinical
sessions with cross-cultural dyads, conducting role plays, presenting clinical
cases, and working with local communities or organizations to help under-
served populations (service-learning).
An unfortunate by-product of content-oriented instruction in multi-
cultural education is inadequate focus on the needs of individual students
“who differ in terms of their multicultural competence” (Ancis & Ali, 2005,
p. 95). Scholars have observed that “present multicultural training tends to

24       foundations of multicultural psychology

emphasize helping White trainees improve their counseling of clients of
color” (Chao, Wei, Good, & Flores, 2011, p. 80), finding that such training
“significantly enhanced Whites’ multicultural awareness, but did not enhance
racial/ethnic minority trainees’ awareness” (p. 72). Students from historically
oppressed groups come to a multicultural workshop or class with experiences
and abilities perhaps underutilized and underdeveloped by generic instruc-
tion (Coleman, 2006; Rooney, Flores, & Mercier, 1998). Multicultural edu-
cation can strengthen multicultural competencies independent of strength
of racial and ethnic identity (Chao & Nath, 2011), such that all participants
should be able to improve. Multicultural psychology is so vast as to require
“a lifelong journey toward the goal of increasing multicultural competence”
(Dickson, Argus-Calvo, & Tafoya, 2010, p. 262).
Although some participants resist multicultural education, openness to
multiculturalism is increasing over time: A national survey found that only
5% of instructors reported primarily negative responses among participants
(Reynolds, 2011). Nevertheless, individuals who engage in racial color blind-
ness (universalizing human experiences rather than acknowledging racial
differences) are less likely to acquire multicultural knowledge through multi­
cultural education (Chao et al., 2011). Acquiring multicultural awareness
can entail intense emotional reactions such as defensiveness, guilt, pain, and
tears (Tummala-Narra, 2009). Hence confrontation must be constantly bal-
anced with a climate conducive to emotional safety and support (Kiselica,
1998; Reynolds, 2011; Sevig & Etzkorn, 2001). As with most clinical skills,
acquiring multicultural competence requires genuine personal involvement
and, ultimately, personal stretching beyond comfort zones in gaining the
competence to enhance the therapeutic alliance with clients from every walk
of life.

Trends in Recent Research on Multicultural Education

Research on multicultural education in psychology attracted widespread

interest after APA (1986) required that multicultural issues be addressed in
accredited graduate programs. Initially, this scholarship attempted to justify
the requirement of multicultural education (Yutrzenka, 1995). However, the
focus has shifted over time: “After decades of deliberating, studying, and even
debating the role of multicultural training . . . scholars are moving from study-
ing whether trainees should take a multicultural training course to examining
the effectiveness of training” (Chao et al., 2011, p. 81).
Substantial research evidence has accumulated on multicultural educa-
tion. A meta-analytic review located 82 studies with empirical data published
before 2003 (T. B. Smith, Constantine, Dunn, Dinehart, & Montoya, 2006).

multicultural education/training and experience      25

The authors of the review identified 45 surveys of participants’ previous
exposure to multicultural education and 37 studies of the effectiveness of a
particular multicultural education program. The results of these two sets of
analyses were discussed using a hypothetical scenario of an employer desiring
to hire a therapist with multicultural competence:
[The first analysis, including 45 surveys,] was similar to an employer ask-
ing “What is the difference between an average applicant who has had
multicultural education and one who has not?” . . . In answer to the
hypothetical employer’s question, an applicant who has completed multi­
cultural education will report moderately higher multicultural competence
[Cohen’s d = .49] than an individual who has not. [The second analysis,
including 37 outcome studies,] answered a different question, similar to
an employer asking “How much did the average employee change during
the multicultural training we provided?” . . . In answer to the employer’s
second hypothetical question, the average participant completing a par-
ticular multicultural education intervention will report large increases
[Cohen’s d = .92] in multicultural competence. (T. B. Smith et al., 2006,
p. 139–140)

The cumulative research evidence has established the effectiveness of multi-

cultural education in the mental health professions.
Many research studies have appeared since 2003. Some of these recent
studies have sought to improve on the limitations of previous work, which
have been “(a) [the] small and fairly homogeneous samples, (b) the ten-
dency to examine students’ experiences in only one course, and (c) the use
of simple methods of examining teaching methods” (M. N. Coleman, 2006,
p. 180). A few studies have moved beyond simplistic evaluations to more
complex designs, such as hierarchical regression considering environmental
factors such as training environment (Dickson & Jepsen, 2007), structural
equation models testing mediation of variables between multicultural edu-
cation and multicultural competence (Spanierman, Poteat, Wang, & Oh,
2008), and other analyses of moderating and mediating factors (Chao &
Nath, 2011; Chao et al., 2011). The effectiveness of multicultural educa-
tion is an area that has continued to attract scholarship, which necessitates
periodic synthesis: “Multicultural education initiatives and the construct
of multicultural competence must undergo ongoing and rigorous scrutiny
if they are to continue receiving widespread support from mental health
professions” (T. B. Smith et al., 2006, p. 142). We therefore evaluated the
cumulative research evidence in an updated meta-analysis to verify the
effectiveness of multicultural education in mental health professions and to
estimate the degree to which study and participant characteristics moderate
the overall findings. Readers seeking a summary may skip to that section.

26       foundations of multicultural psychology

 QUANTITATIVE SYNTHESIS OF RESEARCH DATA

Our updated meta-analysis evaluated research studies conducted in

the United States or Canada that took place in professional mental health
settings (university coursework, professional conference workshop, or work-
place seminar). Each study in the meta-analysis considered a pertinent
dependent variable: therapist multicultural counseling competencies (self-
reported multicultural knowledge, awareness, and skills), therapist racial
attitudes or biases (e.g., ratings of comfort levels working with clients of a
given race), trained observers’ evaluations of therapists (e.g., ratings of thera-
pists’ congruence with the client), clients’ ratings of therapists (e.g., rating
of therapists’ empathy), client retention rates (e.g., premature dropouts
vs. treatment completers), and client outcomes in therapy (e.g., changes in
symptoms over time). We planned to ascertain any differences among these
dependent variables.
If studies simultaneously evaluated the influence of individuals’ multi-
cultural experience (i.e., number or percentage of clients of color) on those
same outcome variables, we extracted those estimates for a separate analysis.
Regarding attitudinal measures, we extracted effect sizes only for measures
specific to race or ethnicity (i.e., racial prejudice), not considering other
forms of prejudice (e.g., sexist or homophobic beliefs) because multicultural
education inconsistently addresses issues of gender, sexual orientation, and
so forth, but consistently addresses issues of race and culture (Pieterse et al.,
2009). Statistical estimates within manuscripts were converted to Cohen’s d
using meta-analytic software.
Outcomes of multicultural education delivered as part of a study were
analyzed separately from retrospective surveys that considered participants’
previous levels of multicultural training and/or experience. The effect sizes
for these two classes of studies have distinct meanings: Outcome studies
estimate the effectiveness of a specific intervention, whereas retrospective
studies evaluate participants’ level of exposure to multicultural education.
Furthermore, retrospective studies confound the issue of effectiveness with
such factors as the accuracy of participants’ memories, their interest in multi-
cultural topics, and their desire to appear experienced in multicultural issues.
Thus, we analyzed these two sets of studies separately.
We further disaggregated the data by the research design used in the
outcome studies to reflect the different meaning of effect sizes generated from
distinct research designs: (a) pre- to posttest changes for a single group of
participants (e.g., university students measured at the beginning and end of a
class) and (b) comparisons of gains across groups that differed in form of multi­
cultural education (e.g., students taking a multicultural class compared with
students taking a class without multicultural content). The data for these two

multicultural education/training and experience      27

types of research designs have distinct interpretations. Effect sizes generated
from single group pre- to posttest designs estimate the degree of change par-
ticipants experienced during the time of the multicultural education, without
considering extraneous factors. In contrast, comparative, quasi-experimental,
and experimental designs provide an estimate of the relative improvement
made by participants receiving multicultural education, while also control-
ling for several possible confounding factors. Positive values for Cohen’s d
effect sizes from these outcome studies indicated the degree of effectiveness
of the particular multicultural education program or class provided in the
study, whereas positive values for effect sizes from the retrospective studies
represented the degree to which exposure to multicultural education (or multi­
cultural experience) was associated with favorable ratings on dependent vari-
ables (i.e., multicultural counseling competence and/or racial attitudes). The
methods of the meta-analysis are reported in the Appendix of this book.

Description of Analyzed Research Literature

Our literature search yielded a total of 47 outcome studies that reported

data on 2,665 individuals’ changes as a function of multicultural education. Of
these studies, 24 evaluated pre- to posttest changes within a single group as a
function of multicultural education (i.e., differences over time for participants
completing a class), and 23 evaluated relative gains compared with a control
group using static group comparison, quasi-experimental, or experimental
designs. The 47 outcome studies typically involved a particular multicultural
education class or workshop conducted at a university or clinic studying the
people available in those settings (convenience sampling), with an average
of only 57 participants, most often involving White graduate students, with
females represented more than males (see Table 2.1). Thus, these studies
evaluated the effectiveness of actual multicultural workshops or university
classes.
We also located 68 retrospective studies that evaluated the previous
multicultural education of 9,596 total participants. These studies typically
involved one-time surveys administered to a convenience sample of about
141 participants, with the surveys measuring the number of multicultural
education courses or workshops that participants had completed, along with
a pertinent dependent variable (e.g., multicultural competence, racial atti-
tudes). Of these studies, 16 also obtained data regarding participants’ levels
of experience working with culturally diverse clients. These retrospective
studies typically involved White trainees or clinicians, with females rep-
resented more than males (Table 2.1). Only one retrospective study con-
trolled for social desirability, and none controlled for possible confounds
such as degree of commitment to multiculturalism. Thus, the results of these

28       foundations of multicultural psychology

 TABLE 2.1
Characteristics of 47 Outcome Studies of Multicultural Education and
68 Retrospective Studies of Participants’ Prior Multicultural Education
Outcome studies Retrospective studies
No. of No. of
Characteristic M studies (k) % M studies (k) %
Year of report 1996 2002
Before 1980 3 6 0 0
1980–1989 5 11 0 0
1990–1999 17 36 15 22
2000–2008 22 47 53 78
Publication status
Published 30 64 35 52
Unpublished dissertations 17 36 33 49
Sampling procedure
Convenience 46 98 47 69
Representative 1 2 21 31
(random selection)
Research design
Cross-sectional (survey) 57 84
Longitudinal 1 1
Comparison groups 3 6 1 1
Quasi-experimental/ 20 43
experimental
Pre- to posttest change, 24 51
1 group
Other (e.g., analogue, 9 14
archival)
Population sampled
Practicing clinicians 9 19 24 38
Clinical trainees 34 74 36 53
Both clinicians and trainees 3 7 6 9
Sample size 57 141
<50 27 57 10 15
50–99 15 32 17 25
100–199 3 6 28 41
200–399 2 4 12 18
400–999 0 0 1 1
>1000 0 0 0 0
Age of participantsa 32 35
Young adults (19–29 years) 22 51 16 24
Middle-aged adults 21 49 52 76
(30–55 years)
Gender (% female) 74 72
Race of participantsb (%)
African American 10 7
Asian American 5 3
Hispanic/Latino(a) American 2 4
Native American 0 0
White/European American 78 82
Other 4 4
Note. Not all variables sum to the total number of studies because of missing data; not all percentages sum
to 100 because of accumulated rounding errors.
aAlthough this category represents the average age category of participants in studies, not all participants in

the study would necessarily be in the category listed. bThe racial composition of participants across all studies
was calculated by multiplying the number of participants within studies by the percentage of participants
from each racial group and dividing that product by the total number of participants.
retrospective studies represent the degree to which individuals who self-
report their level of multicultural education (and multicultural experience)
make statements reflecting favorable dispositions. The associated effect sizes
represent correlational, not causal, estimates of the degree to which exposure
to multicultural education may relate to individuals’ perceptions about work-
ing with diverse clients.

Overall Research Findings

Outcome Studies Using Comparison Groups (Static Group Comparison,

Quasi-Experimental, or Experimental Designs)
Across 23 studies that compared groups differing in their exposure to
multicultural education, the random effects weighted effect size was d = .67
(SE = .114, 95% CI = [.44, .89], p < .0001). Thus, participants in multicultural
education tended to report moderately large gains relative to control groups
on measures. The heterogeneity of the findings was large (I2 = 75.4, 95%
CI = [63, 84]; Q(22) = 89.4, p < .0001; t2 = .21), meaning that the results were
inconsistent across studies.

Outcome Studies Measuring Pre- to Posttest Changes

for a Single Group of Participants
We found 24 studies evaluating pre- to posttest changes among partici-
pants receiving a particular multicultural education class or program. The
average random effects weighted effect size was d = .95 (SE = .154, 95% CI =
[.65, 1.25], p < .0001), indicating that on average participants reported large
improvements from pre- to posttest. The heterogeneity of the findings was
very large (I2 = 86.1, 95% CI = [81, 91]; Q(23) = 165.7, p < .00001; t2 = .44);
the results varied substantially across the 24 studies, suggesting inconsistency
in effectiveness.

Retrospective Studies
With the retrospective studies, the amount of multicultural education
(e.g., number of multicultural classes completed) was reported separately
from the amount of multicultural experience (e.g., caseload percentage of
culturally diverse clients). We therefore conducted a multivariate meta-
analysis (Becker, 2000) on those data to simultaneously account for the two
different types of effect sizes in those studies (specifying the within-study
effect size correlation to the obtained value of r = .41). The overall model
was statistically significant when the 68 studies were examined (Wald
X2 = 149.1, p < .0001). The random effects weighted average effect sizes
were d = .41 (SE = .034, 95% CI = [.34, .47]) for multicultural education and

30       foundations of multicultural psychology

d = .29 (SE = .066, 95% CI = [.16, .42]) for multicultural experience. The
difference between these two types of effect sizes did not reach statistical
significance (p > .05). Because only 16 studies contained effect sizes spe-
cific to multicultural experience, subsequent analyses were delimited to the
effect sizes pertaining to multicultural education in all 68 studies. For those
data, the heterogeneity of the findings was in the moderate range (I2 = 59.2,
95% CI = [47, 69]; Q(67) = 164.1, p < .0001), meaning that the results were
fairly inconsistent across studies.

Influence of Study and Participant Characteristics on the Findings

Study Characteristics Among the 47 Outcome Studies

Studies differed in the type of dependent variables measured (self-
reported multicultural competence, self-reported racial attitudes, observer/
client ratings), so the results had to be analyzed across these different mea-
sures. These analyses did reveal statistically significant differences in effect
size across the type of dependent variable measured (see Table 2.2). Among
the 24 studies involving pre- to posttest changes in a single group of partici-
pants, the average effect size was higher for studies using participant-rated
measures of multicultural competence (e.g., the Multicultural Competence
Inventory) than for those using measures of racial attitudes or observer/client
ratings of the therapist. Among the 23 comparison group studies, the effect
sizes were highest on observer/client ratings of therapists and much lower
on self-rated measures of multicultural competence or racial attitudes. Thus,
although participants in the 24 studies measuring pre- to posttest changes of a
single group tended to self-report strong gains over time in their multicultural
competence, the 23 studies using comparison group designs showed the effec-
tiveness of multicultural education to be more pronounced when observer/
client ratings were used.
Studies also differed in the authors’ descriptions of how they developed
the multicultural education provided in the study. Most authors explicitly
described the theoretical foundation for the multicultural education, but 20 of
the 47 outcome studies lacked that information. In the 23 comparison group
studies, multicultural education that was explicitly based on models of multi-
cultural competence or relevant theories was three times as effective as that
provided in studies containing no information about how the multicultural
education was developed (Table 2.2). The majority cited models of multi­
cultural competence (e.g., D. W. Sue et al., 1982) as the theoretical founda-
tion, with relevant models such as Paul Pedersen’s (2000) triad training model.
The kind of multicultural education provided differed among studies, so
we evaluated the results across the average duration of the education provided

multicultural education/training and experience      31

32       foundations of multicultural psychology

TABLE 2.2
Weighted Mean Effect Sizes (d ) Across Characteristics of 24 Single Group (Pre- to Posttest Changes)
and 23 Comparison Group Studies of Multicultural Education
Single group studies Comparison studies
Characteristic Qb k d+ 95% CI Qb k d+ 95% CI
Data source 0.7 6.5*
Published 17 1.03 [.68, 1.38] 13 .89 [.62, 1.15]
Unpublished 7 .76 [.24, 1.28] 10 .37 [.08, .67]
Assignment to groupsa 1.5
Convenience/self-selected 15 .57 [.31, .83]
Random assignment 8 .86 [.48, 1.25]
Population sampledb 0.4 0.1
Clinicians 5 .78 [.06, 1.50] 4 .56 [.09, 1.04]
Trainees 17 1.06 [.63, 1.51] 17 .61 [.38, .84]
Required trainingc 0.2 0.5
Required 12 1.09 [.70, 1.49] 7 .62 [.19, 1.04]
Optional 5 .95 [.37, 1.53] 13 .81 [.47, 1.15]
Theory-based training 1.3 12.9**
Yes 14 1.09 [.72, 1.47] 13 .97 [.72, 1.22]
Not reported 10 .77 [.37, 1.17] 10 .32 [.07, .57]
Dependent variabled 8.9* 7.6*
Self-ratings, competencies 16 1.18 [.87, 1.49] 18 .58 [.36, .81]
Self-ratings, racial attitudes 9 .48 [.08, .87] 4 .47 [.00, .95]
Observer/client ratings 5 .56 [.03, 1.09] 5 1.23 [.80, 1.67]
Note. d+ = random effects weighted effect size; k = number of studies; Qb = Q-value for variance between groups; CI = confidence interval.
aThis analysis contrasted studies using static group comparison and quasi-experimental designs with those using true experimental designs. bStudies includ-

ing both trainees and clinicians were excluded from these analyses. cStudies not reporting whether the training was required or not were excluded from these
analyses. dEach type of measurement occurring in studies was represented in this analysis, so it was possible for some studies to contribute more than one
effect size.
*p < .05. **p < .01.
and on whether the education was required or voluntary. We also sought
to contrast the findings from studies that assigned participants randomly to
conditions, because true experimental designs are less susceptible to threats
to internal validity. However, these analyses did not identify any statistically
significant differences in the results obtained across studies (Table 2.2).

Participant Characteristics in the 47 Outcome Studies

Studies included a wide variety of participants, so we sought to determine
whether systematic differences in findings could be attributable to several par-
ticipant characteristics: gender (operationalized as percentage of females in the
study), average age, estimated age cohort (year of study minus average age of
participants), education level, professional status (clinicians vs. trainees), and
racial diversity (operationalized as percentage of people of color in the study).
Only the latter variable reached statistical significance for the 23 comparison
group studies, with the percentage of people of color included in the sample
strongly associated with the magnitude of the effect sizes (r = .55, p = .003).
Comparison group studies with relatively more participants of color tended to
have larger effect sizes, indicating greater effectiveness of training, than those
with almost all White/European American participants.

Study Characteristics Among the 68 Retrospective Studies

Most of the studies evaluated participants’ self-reported multicultural
competence (69%), with some measuring racial attitudes (16%) or a combina-
tion of the two. The averaged results were consistent across these dependent
variables (Q = 0.4, p > .10). The type of participant sampling (convenience vs.
random) did not consistently influence the findings (Q = 0.1, p > .10), and the
magnitude of the effect sizes did not differ significantly as a function of the year
of publication of the study (r = -0.16, p > .10). Thus, these three study charac-
teristics did not moderate the overall findings of the 68 retrospective studies.

Participant Characteristics in the 68 Retrospective Studies

No differences in the findings of the retrospective studies were observed
across any of the participant characteristics analyzed: participant gender
composition (operationalized as percentage of females in the study), partici-
pants’ average age, estimated participants’ age cohort (year of study minus
average age of participants), participants’ average years of education, partici-
pants’ professional status (trainee or clinician), or participants’ racial diversity
(operationalized as percentage of people of color in the study). The associa-
tion between participants’ prior multicultural education and their current
self-reported multicultural competence and/or racial attitudes appeared to be
independent of the characteristics of participants across studies.

multicultural education/training and experience      33

Likelihood of Publication Bias Adversely Influencing the Results
Among the 47 outcome studies, possible publication bias appears to
have influenced the magnitude of the overall findings reported previously.
Although one statistical method (Duval & Tweedie, 2000) did not identify
any “missing” studies for either the 24 single group (pre- to posttest changes)
or 23 comparison group studies, visual inspection of the contour-enhanced
funnel graphs indicated a disproportionate number of studies reporting statis-
tically significant results relative to the standard errors (combined 47 studies
depicted in Figure 2.1). Moreover, Egger’s regression test (an estimate of
asymmetry of effect sizes that would indicate possible “missing” studies) was
statistically significant for both of these types of studies (p < .01), suggesting
likely publication bias. Across the 23 comparison group studies, those that
had been published had larger effect sizes than doctoral dissertations that had
remained unpublished (Table 2.2), so it is almost certain that unpublished
studies not included in the meta-analysis would have lowered the average
effect sizes reported previously.

Figure 2.1. Contour-enhanced funnel plot of effect sizes (Cohen’s d) by standard

error for 47 outcome studies of multicultural education. Normally, data portrayed in
this manner should be distributed in the shape of a pyramid, but the lack of studies
in the lower left of the other data points suggests “missing” studies, those with
small numbers of participants and nonsignificant results that may have failed to
be published or were never submitted for publication.

34       foundations of multicultural psychology

Figure 2.2. Contour-enhanced funnel plot of effect sizes (Cohen’s d) by standard
error for 68 retrospective studies of multicultural education.

The data for the 68 retrospective studies (see Figure 2.2) were more
evenly distributed around their numeric average than the 47 outcome studies
(Figure 2.1). Although the trim and fill method using random effects weighted
estimates did not identify any “missing” studies, Egger’s regression test was sta-
tistically significant (p < .05). Thus, publication bias may have mildly inflated
the results of the 68 retrospective studies, although not nearly as conspicu-
ously as in the 47 outcome studies.

DISCUSSION AND INTERPRETATION OF THE FINDINGS

Multicultural education is effective for mental health professionals.

Therapists’ and trainees’ self-reported abilities, self-reported racial attitudes,
and observer/client reported clinical performance have all been shown to
improve through multicultural education. On average, participants receiv-
ing multicultural education improved about one standard deviation (d = .95)
over their initial scores on those indicators or about two thirds of a standard
deviation (d = .67) more than people who had not received multicultural
education during the same time frame. These findings confirmed the results

multicultural education/training and experience      35

of the previous meta-analysis, which evaluated fewer studies (T. B. Smith
et al., 2006).
Separate analyses of retrospective studies estimated the cumulative
effect of multicultural education and multicultural experience. In general, the
more multicultural education individuals received, the more likely they were
to report multicultural competence and positive racial attitudes (d = .41).
Furthermore, individuals’ level of experience working with culturally diverse
clients was positively but mildly related to their self-reported multicultural
competence and racial attitudes (d = .29). These two findings were based on
correlational designs, so personal commitment to multiculturalism and other
likely confounds cannot be ruled out. Moreover, the findings were of smaller
magnitude than would be expected; other factors (not measured) must bet-
ter account for individuals’ level of multicultural competence in providing
mental health service.

Implications From the Secondary Analyses

Analyses of the influence of study and participant characteristics sug-

gest five qualifications to the foregoing overall conclusions. First and foremost
is the likelihood that the statistical estimates of the effectiveness of multi­
cultural education may be inflated due to publication bias, which occurs when
nonsignificant research findings remain unpublished or otherwise inaccessible
to researchers conducting meta-analyses (Rosenthal, 1991). The apparent
absence of some negative or nonsignificant findings likely resulted in exces-
sively high estimates of the effectiveness of multicultural education. Thus,
the overall effect size values reported earlier are liberal estimates of the effec-
tiveness of multicultural education in improving participants’ multicultural
competence.
A second qualification concerns the differences in results across outcome
measures. Participants in multicultural education tended to rate themselves
as substantially improved in multicultural competence from pre- to posttest,
but those “gains” were much lower when treatment participants were com-
pared with control groups (d = 1.18 compared with d = .58; see Table 2.2).
About one half of the apparent effectiveness of multicultural education in
improving participants’ self-reported multicultural competence is attributable
to study invalidity: failing to account for participants’ explicit expectations
for improvement to have occurred, sensitization invoked through the initial
measurement. Thus, it may be preferable to ignore the statistical estimates
generated from the 24 studies measuring pre- to posttest changes and rely
exclusively on the estimates provided by the 23 studies using control groups.
This concern relevant to study internal validity is somewhat mitigated
by the finding that the effectiveness of multicultural education was more

36       foundations of multicultural psychology

pronounced in the 23 comparison group studies when trained observers or
clients rated the study participants. Observer ratings are presumably less sus-
ceptible to bias than self-reports, assuming that the raters are blind to the par-
ticipants’ level of multicultural education (e.g., Constantine & Ladany, 2000).
So it remains impressive that trained observers or clients themselves rated
participants who had received multicultural education much higher than par-
ticipants who had not received multicultural education. Unfortunately, only
10 estimates involved observer or client ratings (Table 2.2), so these results
may be unreliable. At least the overall estimates from the 23 comparison group
studies indicate moderate effectiveness.
A third qualification of the results concerns the inconsistency in results
of the outcome studies (Figure 2.1): ranging from no improvement to sub-
stantial gains. As mentioned early in this chapter, multicultural education
varies remarkably in method, intensity, content, format, and so forth. Clearly
and obviously, multicultural education varies remarkably in its effectiveness.
A fourth qualification helps account for the high degree of variability
in effectiveness. Although 57% of the authors of outcome studies described
the theoretical and research foundations for the multicultural education that
participants received, primarily the tripartite model of multicultural compe-
tence (e.g., D. W. Sue et al., 1982), 20 reports did not include any conceptual
basis or rationale for the multicultural education provided. And the findings
from the experimental and quasi-experimental studies differed accordingly:
Studies containing details about the theoretical basis of the intervention
(i.e., multicultural competencies) were three times as effective as those not
containing this information (Table 2.2). Three times as effective! Although
the failure of authors to describe the conceptual basis of the multicultural
education provided could reflect other nuisance variables (e.g., precision in
conducting research), it seems likely that the programs not explicitly based
on the multicultural competencies or relevant models were of poorer quality
than those based on a particular model. These differences, apparently attrib-
utable to quality, cannot be overemphasized. So the previous qualification
regarding the “moderate” effectiveness of multicultural education should be
counterbalanced with a qualification about quality: Multicultural education
explicitly based on principles relevant to multicultural competence is very
effective; multicultural education not explicitly based on principles relevant
to multicultural competence is only mildly effective.
A fifth qualification of the findings concerns the racial composition of
the participants. Across the 23 comparison group studies, the effectiveness of
multicultural education was strongly associated (r = .55) with the percentage
of participants who were people of color. On average, the more people of color
participating, the more effective the multicultural education. Many scholars
had previously called attention to this possibility (e.g., Dickson et al., 2010),

multicultural education/training and experience      37

but the trend is only now being confirmed. Nevertheless, we cannot yet ascer-
tain the degree to which this finding results from one or more of the follow-
ing conditions: (a) greater racial diversity of participants allows for enhanced
learning opportunities about multiculturalism for all participants; (b) greater
interest in (and less resistance toward) multiculturalism occurs among people
of color, who improve more than Whites as a result of the education provided;
and/or (c) multicultural education is of higher quality when provided by insti-
tutions composed of greater numbers of people of color (i.e., the institutional
environment being more conducive to multicultural competence, such as
higher expectations by instructors for student growth). Whatever the specific
causality, participant racial composition does matter when comparing gains of
treatment with control groups.
Other than those five qualifications, the data did not systematically dif-
fer across multiple study and participant characteristics, a consistency worthy
of note. For instance, the effectiveness of multicultural education did not dif-
fer when it was required or voluntary, even though some would assume that
people voluntarily taking a multicultural workshop might be more motivated
to improve than people required to take a university class. The fact that the
data did not differ between voluntary and mandated settings provides indi-
rect support for the stance taken by APA and other professional organizations
to require multicultural education in accredited graduate programs. Similarly,
results did not differ when the participants were trainees or working profes-
sionals: Multicultural education equally benefitted both.

Recommendations for Practitioners

Practicing psychologists can enhance their abilities to serve diverse

clients by participating in multicultural education. For their benefit, many
professional conferences now offer specific training in multicultural issues. It
is in professionals’ best interest to seek out additional learning and experience
(Ponterotto, 1998). Practitioners should approach multicultural education
selectively, however. Some programs are clearly more effective than others
(Figure 2.1). Programs that are based on the principles of multicultural com-
petence or relevant theory offer training best suited to personal skill develop-
ment and client melioration (Rogers-Sirin, 2008). Practitioners additionally
benefit from programs with a balance between awareness, knowledge, and
skill development. Too many programs focus on awareness and knowledge
without adequate attention to skills (Pieterse et al., 2009). Ultimately, the
best type of multicultural education is a program that fills gaps in personal
skills. A self-evaluation of multicultural competence (Arredondo et al., 1996;
D. W. Sue et al., 1992a) can help identify areas in which an individual needs
improvement.

38       foundations of multicultural psychology

Considerations for Future Research

Our meta-analytic review included 115 studies containing quantitative

data, an indication that researchers have been fairly interested in evaluating
multicultural education. Although our comments in this section are based on
those quantitative studies, “we need a balance of quantitative and qualita-
tive approaches to multicultural training research” (Ponterotto, 1998, p. 64).
With that caveat stated, we offer the following questions and suggestions for
scholars to consider in future research (see Exhibit 2.1).

Real-World Applications
Meta-analyses evaluate “effect sizes,” but the interpretation of an effect
size is not straightforward. Improvements reported in this chapter in terms
of standard deviation units (Cohen’s d) are not at all the same as individu-
als actually achieving adequate multicultural competence. Consider the fol-
lowing example. A national survey evaluating multicultural competence
(Vereen, Hill, & McNeal, 2008) obtained mediocre average subscale scores
of about 55 (possible range of 20 to 80) and standard deviations of about 6 on
the Multicultural Awareness/Knowledge/Skills Survey (D’Andrea, Daniels,
& Heck, 1991). Assuming the average effect size from studies using com-
parison group designs (d = .67), these scores would represent an absolute
gain (relative to controls) of one point on the rating scale (e.g., improving
from agree to strongly agree) on four of 20 items of each subscale, equating to
an average 7% improvement on the subscale scores (relative to controls). Is
that change meaningful? Do these results indicate that participants are truly
more effective as a result of multicultural education? In terms of the scaling
used in this particular example, the participants from the national survey
would still rate themselves between disagree and agree in their multicultural
competence, even after the hypothetical gains. Do we remain satisfied with

EXHIBIT 2.1
Recommendations for Researchers
• Examine practical/clinical significance: clients’ experiences, improvement, and
retention as a function of therapist multicultural education, experience, and
competence.
• Use research designs that will answer research questions with minimal likelihood
of bias.
• Incorporate concepts and resources from related disciplines.
• Recruit as many participants as possible to minimize sampling error.
• Control for or examine the influence of responses attributable to social desirability.
• Develop measures focused on skills and behaviors associated with multicultural
competence.

multicultural education/training and experience      39

mediocre abilities because they are better than incompetence? What level of
abilities sufficiently enhances clients’ experiences in therapy? Future research
will benefit from considering issues of practical significance and real-world
applications, not merely issues of statistical “improvement.”

Client Perspectives
We doubt that anyone seeking treatment for a medical condition would
be impressed by a research study in which medical students rated their own
improvement on their ability to perform that treatment. A prospective medical
patient would want to know whether current and potential physicians who had
received training specific to that treatment were able to perform it effectively to
improve patient outcomes. The same principle applies to mental health prac-
tices: “The ultimate purpose of counseling competence is therapeutic change.
All other aims and activities of psychologists and counselors . . . are subordinate
to this raison d’être” (Mollen, Kelly, & Ridley, 2011, p. 919–920). However,
the vast majority of studies identified in this meta-analytic review used par-
ticipants’ self-reports on variables indirectly relevant to therapy. We identified
only two outcome studies and one retrospective study in which clients pro-
vided the ratings. The field does not need more research measuring therapists’
self-reported changes; it desperately needs research on clients’ experiences,
improvement, and retention.

Clinical Settings
If researchers attended more to clients’ experiences, a related gap in the
current literature would be corrected: assessment of multicultural education
provided in the workplace. Although some research has evaluated clinic-
based programs and work environments, the primary emphasis has been on
graduate school programs.

Evaluations Versus Retrospective Surveys

Surveys can be used for a wide variety of research purposes. For example,
surveys about current opinions or practices provide needed information about
specific topics, such as what percentage of programs evaluate participants’
multicultural skills: 12% according to Priester et al. (2008) and 7% accord-
ing to Pieterse (2009). Similarly, retrospective surveys can help in estimating
general relationships of variables when experimental or quasi-experimental
research designs are impossible.
Our review located 68 surveys that attempted to evaluate therapists’ and/
or trainees’ multicultural education and experience retrospectively (e.g., corre-
lating the number of multicultural classes taken with self-reported multi­cultural

40       foundations of multicultural psychology

competence or racial attitudes). These kinds of surveys cannot accurately
evaluate questions regarding effectiveness, efficacy, usefulness, and so forth, and
they are characterized by all kinds of threats to internal validity, such as par-
ticipant self-selection bias (i.e., likelihood of people interested in multicultural
issues choosing to complete the survey). If conducted at all, such surveys should
control for confounding factors, such as preexisting multicultural interest and
motivation and topic social desirability (Castillo, Brossart, Reyes, Conoley, &
Phoummarath, 2007; Constantine & Ladany, 2000). After summarizing the
results of 47 outcome studies, we anticipate little if any benefit from conduct-
ing a 69th retrospective survey. Retrospective surveys are easy to conduct, so
dissertation committees will be tempted to approve such projects. We urge
constructiveness over convenience.

Seeking Out Best Practices From Related Disciplines

Any organization tends to seek solutions internally, and this practice
certainly characterizes the literature that we reviewed; most authors cited
exclusively other psychologists and/or counselor educators. Yet, it can be use-
ful to learn from the successes (and failures) of other disciplines (Baca et al.,
2007). For instance, the profession of teacher education and the National
Association for Multicultural Education have a rich literature on the same
issues and problems so frequently cited in the mental health professions. The
fields of social work and international business have multiple and varied
resources that can help individuals work effectively across cultures. Cross-
fertilization of ideas may improve similar abilities among mental health ser-
vice providers.

Procedures for Correcting Publication Bias

Negative and nonsignificant research findings are typically inaccessible
(not submitted for publication or presentation), so we encourage research-
ers to archive such findings online (e.g., submitting results to databases such
as ERIC). Another way to diminish the likelihood of publication bias is for
researchers to substantially increase the number of study participants, which
would both reduce sampling errors and increase the likelihood that the results
will achieve statistical significance, even if effect size remains minimal.

Recommendations for Instructors and Program Directors

Multicultural education is required for every professionally accredited

graduate program. Hundreds of instructors, thousands of students, and mil-
lions of clients stand to benefit from improvements in multicultural education.

multicultural education/training and experience      41

 EXHIBIT 2.2
Recommendations for Instructors and Clinical Directors
• Provide specific, concrete, tangible illustrations of multicultural competencies.
• Demonstrate the overlap between multicultural and general counseling
competencies.
• Focus instruction on and provide models of multicultural competencies.
• Integrate multicultural issues across the curriculum, particularly in practica.
• Maintain high expectations for multicultural competence.
• Identify and meet individual students’ needs.
• Promote mastery of skills through practice.
• Evaluate student skills regularly.
• Evaluate the multicultural education provided.
• Foster effective learning environments.
• Implement published guidelines:
• Teaching counseling based on competence (Ridley, Mollen, & Kelly, 2011b)
• Multicultural program development model (Ridley, Mendoza, & Kanitz, 1994)
• Multicultural course development model (Ridley, Espelage, & Rubinstein, 1997)
• Multicultural competency model of training (Ponterotto, 1997, 1998)
• Characteristics of effective instructors, students, and programs (Ponterotto, 1998)

Several recommendations should help promote that improvement (see

Exhibit 2.2).

Keep the Focus of Instruction on Multicultural Competence

The objective of multicultural education is multicultural competence
(Abreu et al., 2000). Anything that does not promote multicultural com-
petence can be dropped, no matter how interesting the facts, engaging
the activities, enlightening the self-scrutiny, and so forth. Knowing about
a cultural group becomes valuable in therapy only when that knowledge
facilitates worldview congruence with a particular client. The objective for
gaining cultural knowledge is to be able to apply it in a therapeutic relation-
ship (e.g., Ridley, Mollen, & Kelly, 2011b). As the data show (Figure 2.1),
entire groups of people may fail to improve after months in a multicultural
psychology class.
Every activity, every bit of information provided in multicultural educa-
tion should enhance participants’ motivation and abilities to provide effective
therapy (e.g., Mollen et al., 2011). Although some individuals may experi-
ence discouragement when they become more aware of their incompetence
(and thus rate themselves low at posttest), the realistic appraisal provided by
effective instruction contributes to skill enhancement. Activities commonly
used to promote self-awareness (such as cultural autobiographies) should be
made explicitly relevant to therapy or replaced with activities that promote
skills simultaneously with self-awareness, such as those suggested by Pedersen

42       foundations of multicultural psychology

(2000). We hope that within a few years, research reviews of multicultural
course syllabi will show a balance of all aspects of multicultural competence,
which will necessitate a substantial and sustained increase in skill develop-
ment (Pieterse, 2009).

Integration of Multicultural Issues Across the Curriculum, Particularly Practica

For decades scholars have called for integration of multicultural issues
across the curricula of professional graduate programs (Abreu et al., 2000;
Ponterotto, 1998; Vereen et al., 2008). Research confirms that such inte-
gration fosters a cultural learning environment that predicts multicultural
competence (Dickson & Jepsen, 2007); students in programs with strong
support for multiculturalism gain most from multicultural education (M. N.
Coleman, 2006). The literature emphasizes particularly the benefits of inte-
grating practicum supervision with a multicultural focus (Vereen et al., 2008).
Explicit integration with clinical practice is necessary if multiculturalism is
to be taken seriously, learned, and applied (Abreu et al., 2000; Pieterse et al.,
2009). The topic of multicultural supervision is receiving increased attention
in the literature (e.g., Constantine, 2001; Lassiter, Napolitano, Culbreth, &
Ng, 2008; Ober, Granello, & Henfield, 2009), providing guidance to graduate
programs, internship sites, and mental health clinics. Programs should also
facilitate opportunities for trainees to work with underserved populations
(Toporek & Vaughn, 2010; Vereen et al., 2008).

Model the Processes Essential to Multicultural Competence

People learn by observing. If participants observe how an instructor
maintains an open learning environment (López et al., 1989), they will
understand a foundational principle of multiculturalism: Mutual benefits
come through sharing power (Toporek & Vaughn, 2010). If they observe
how an instructor facilitates mutual understanding during inevitable par-
ticipant value conflicts, they will have learned an essential clinical skill
seldom witnessed in the real world: genuine dialogue. Participants do not
need self-analysis or motivational gimmicks as much as they need to observe
multiculturalism applied. If instructors respect differences among the par-
ticipants, the resulting emotional safety will facilitate participants’ growth
(M. N. Coleman, 2006; Reynolds, 2011). Participants learn how to work
with personal vulnerabilities as the instructor invites genuine engagement
by modeling genuine attitudes and behavior.

Maintain High Expectations for Multicultural Competence

People rise to expectations. If therapists and trainees realize that cli-
ents expect their therapist to understand their worldview, make them feel

multicultural education/training and experience      43

comfortable, and promote well-being in ways congruent with their values,
these professionals may focus on improvement more than if the training is
perceived as just a workshop for credit or a class for a grade.

Identify and Meet Individual Participants’ Specific Needs

Participants in a multicultural education program are diverse in interests
and experiences; a generic training program will not meet everyone’s needs
(Chao et al., 2011; Dickson et al., 2010; Ridley et al., 2011a). As suggested
by the finding reported previously that greater racial diversity among partici-
pants is associated with enhanced effectiveness of multicultural education,
differences in experiences of White/European Americans and people of color
may require attention from instructors of racially mixed groups (Rooney et al.,
1998) as they attempt to make the experience meaningful for people at all
levels of multicultural competence. One useful procedure involves four steps:
(a) require participants to self-evaluate their own limitations and competen-
cies (which vary with the individual), (b) bolster motivation to improve on
the identified areas of needed growth, (c) show examples of replacement
behaviors and worldviews, and (d) provide opportunities to practice and
solidify the newly adopted procedures and perspectives.
The instructor teaches principles underlying multicultural competen-
cies and then provides examples of how those principles apply to particular
cases. Important principles include promoting recovery from experiences of
cultural incongruence, maintaining a multicultural perspective, verifying cul-
tural meaning, practicing flexibility in meeting client needs, and so forth. The
participants remain responsible for applying those principles. Participants
benefit from setting specific self-motivating goals and being accountable to
report their progress. The efforts of any instructor are multiplied when par-
ticipants actively self-instruct—instituting for themselves the individualiza-
tion necessary to improve despite different experiences and abilities.

Support Participants in Mastering Challenges That Engage Them

on Many Levels
Workshops and classes are time-limited. After preliminary instruction and
modeling, participants may tend to learn best by engaging in learning activities
that provide opportunities to practice the principles described. Certain activi-
ties and experiences, such as conducting an interview focused on issues of race
or culture, can simultaneously facilitate awareness of self, knowledge of others,
and skills for counseling. “Identifying which activity engages students on many
levels (e.g., cognitive, affective, and behavioral) seems crucial to promoting
the development of professionals who can effectively work with diverse popu-
lations” (Ancis & Ali, 2005, p. 96–97). As instructors help students analyze

44       foundations of multicultural psychology

complex clinical vignettes, consult with cultural informants, reevaluate their
own cultural assumptions reflected in role plays or simulations, practice skills
demonstrated on videotapes, and so forth, they both enhance participants’
self-confidence and develop essential skills (Pieterse, 2009).

Conduct Multiple Performance Evaluations

Many scholars have encouraged relevant performance evaluations: “In
the future more programs will hold students to a competence standard in
classes that are designed to teach them about racial-cultural issues, especially
since the cultural competence standards have been in existence for more than
[30] years” (Carter, 2003, p. 30). Learners absorb as much as they are required
to demonstrate. If participants in multicultural education know they will
soon need to demonstrate a particular skill, they tend to pay attention. For
instance, an instructor of a 3-hour workshop could invite culturally diverse
surrogates to engage participants in role plays at the end of the session, pro-
viding feedback on participants’ application of principles. Or at the end of
each clinical placement, supervisors could observe videotaped sessions of
their students, using a rubric of multicultural competencies to provide formal
evaluations (Fouad & Arredondo, 2007; Sevig & Etzkorn, 2001). Clinicians
could be told that their employers will be tracking client retention data across
race. Performance improves when people remain accountable.

Perform Internal Evaluations of the Multicultural Education Provided

Improvement requires self-scrutiny. Program directors, clinical super-
visors, and instructors can regularly evaluate the quality of the multicultural
education they provide, using external metrics and guidelines created for
multiculturally competent programs (Fouad & Arredondo, 2007; Ponterotto
& Austin, 2005; Utsey, Grange, & Allyne, 2006). They may conduct evalu-
ations of the multicultural competence of instructors (Spanierman et al.,
2011), use narrative descriptions of exemplary programs as models (Fouad,
2006; Ponterotto & Austin, 2005), and consult faculty members from
those exemplary programs. Also essential is ongoing personal professional
development—training the trainer (Newell et al., 2010; Ponterotto, 1998;
Toporek & Vaughn, 2010). The whole enterprise of multicultural education
assumes that instructors understand the literature of multicultural psychol-
ogy. An effective curriculum necessitates this familiarity. Programs should
never have to scramble to find an adjunct instructor for a multicultural class.
Not every faculty member has the required preparation to teach a class on
forensic psychology; similarly, not every instructor has the genuine experi-
ence with multicultural psychology that should be prerequisite to teaching
it (see characteristics listed by Ponterotto, 1998).

multicultural education/training and experience      45

Apply the General Factors of Effective Psychotherapy
to the Learning Environment
Limited research has evaluated the effectiveness of different methods,
formats, and approaches to multicultural education (Ancis & Ali, 2005).
Although it may be tempting to compare the effectiveness of those approaches,
findings from psychotherapy research emphasize the importance of general
factors over specific techniques. Thus, it may be wiser to transfer and pro-
mote those same general factors in multicultural education (e.g., enhancing
the alliance of instructors with students, raising student expectations for self-
improvement) than to scrutinize the nuances of unproven methods and tech-
niques (Priester et al., 2008).

Incorporate Recommendations Provided in the Literature

Recommendations for improving multicultural education regularly
appear in professional publications. Excellent guidelines have been provided
by Ponterotto (1997, 1998; Ponterotto & Austin, 2005) and Ridley et al. (1994,
1997), among others. Beyond information in those foundational sources, recent
specific recommendations for improvement include the following:
77 Address international perspectives and global psychology
(Ægisdóttir & Gerstein, 2010).
77 Cover indigenous healing practices and perspectives on mental
health (Bojuwoye & Sodi, 2010).
77 Facilitate “service learning” by having participants practice
essential multicultural skills while meeting needs of local com-
munities (Tomlinson-Clarke & Clarke, 2010), collaborating
with families and community organizations (Newell et al., 2010).
77 Use technology and Internet resources to enhance students’
learning (Ancis, 2003).
77 Provide instruction regarding spiritual and religious diversity
(Crook-Lyon et al., 2012).
77 Ensure that students understand the principles of contextual-
ization and the intersections of diversity, rather than teaching
one content area at a time as if the distinct topics existed in
isolation (Pieterse et al., 2009).
77 Provide direct and indirect supervision of participants’ activities
designed to promote multicultural competence (Newell et al.,
2010; Vereen et al., 2008).
Instructors and program directors are intensely busy, but time spent in the
literature will not only prevent serious mistakes but also promote maximal
benefit from the time already invested in providing multicultural education.

46       foundations of multicultural psychology

 CONCLUSION

Mental health professionals would benefit from receiving instruction

in multicultural issues as part of their graduate degree requirements and
across their careers through professional development programs and work-
shops. Although multicultural education results in moderate improvements
in self-ratings, its quality is highly variable. The most effective training
programs focus on participants’ acquisition of multicultural competencies.
Openly talking about historically oppressed groups and gaining insight into
personal feelings about those groups is important for the development of a
therapist (López et al., 1989), but dialogue and insight alone do not neces-
sarily enhance therapists’ abilities to work effectively with clients from those
groups. Education, in the true sense of the word, is functional. It is past time
for the field to focus more on skill development (Pieterse, 2009); it is time to
demonstrate that skill acquisition results in improved client outcomes.
To increase the focus on skill development, multicultural issues must be
integrated across other aspects of professional training, particularly clinical
supervision (e.g., Ober et al., 2009). This need for systematic integration of
multicultural issues has been repeatedly emphasized (Dickson et al., 2010;
Fouad & Arredondo, 2007; Toporek & Vaughn, 2010). Acquiring and main-
taining multicultural competence in mental health services is more likely
when the institution or clinic supports it (Dickson & Jepsen, 2007; Utsey
et al., 2006). Sporadic workshops and a single university class are better than
nothing at all, but genuine multicultural education entails “a lifelong jour-
ney toward the goal of increasing multicultural competence” (Dickson et al.,
2010, p. 262).
This chapter has provided specific recommendations for improving multi-
cultural education for practitioners, program directors, researchers, and
instructors. We have much to improve. A question to guide the actions of all
parties was suggested by Rogers-Sirin (2008): “How does training relate to
improvements in [client] use of services, client satisfaction, client retention?”
(p. 318). If we collectively redesign and then implement multicultural educa-
tion with that question as our primary consideration, it seems likely that the
statistical estimates reported in this chapter will appear small by comparison
with what we will achieve in the future.

multicultural education/training and experience      47

 3
THERAPIST MULTICULTURAL
COMPETENCE: A META-ANALYSIS
OF CLIENT EXPERIENCES
IN TREATMENT

Mental health professionals have an ethical responsibility to facilitate

effective psychotherapeutic interventions for all clients; this necessarily
entails accounting for and being sensitive to human diversity (Arredondo &
Toporek, 2004; S. Sue, 1998). Although mental health professionals would
not intentionally mistreat clients from diverse multi­cultural backgrounds,
inadequate cultural knowledge or awareness may result in unintentional
harm to the client (Pope-Davis, Liu, Toporek, & Brittan-Powell, 2001;
S. Sue, 1988). For example, when a clinician misunderstands a client’s
cultural worldviews, lifestyles, and experiences, the mental health needs
of the client may remain unrecognized and unmet (S. Sue & Zane, 1987).
To prevent this type of situation, scholars have suggested that therapists

Alberto Soto and Derek Griner of Brigham Young University contributed to the writing of this chapter.
A complete list of references for the studies included in this chapter’s meta-analysis is posted online.
Readers can consult it at [Link]
[Link]
Foundations of Multicultural Psychology: Research to Inform Effective Practice, by T. B. Smith and J. E. Trimble
Copyright © 2016 by the American Psychological Association. All rights reserved.

49
develop multicultural competencies (Arredondo et al., 1996; Arredondo
& Tovar-Blank, 2014) that will enable them to adjust their practices to
meet the needs of culturally diverse populations (American Psychological
Association, 2009).
The broad concept of multicultural counseling competence (MCC) has
been operationalized in terms of multicultural counseling competencies (MCCs),
which are most often described as therapists’ awareness, knowledge, and skills
in working with diverse clients (Constantine, 2002; D. W. Sue, Arredondo,
& McDavis, 1992a; D. W. Sue et al., 1982). The potential benefits to diverse
clients of having therapists who are aware, knowledgeable, and skillful in
handling multicultural issues may be so obvious that perhaps some scholars
immersed in multicultural psychology have taken them for granted. Despite
wide acknowledgement of MCCs and attempts to evaluate them across
more than three decades, scholars in multicultural psychology actually have
limited data about how these competencies affect clients’ perspectives and
experiences in therapy (D’Andrea & Heckman, 2008; Huey, Tilley, Jones,
& Smith, 2014).
Some of the difficulties associated with evaluating MCCs may be due
in part to the methods traditionally used to measure these constructs (e.g.,
Constantine & Ladany, 2001; Hoyt, Warbasse, & Chu, 2006). Self-report
measures have long been used to determine clinicians’ multicultural case
conceptualization abilities, but self-report measures overestimate abilities
compared with expert ratings (Cartwright, Daniels, & Zhang, 2008). The
inherent confounds associated with self-evaluation raise serious questions
about the validity of self-report measures of MCCs (e.g., overconfidence,
social desirability). For example, Constantine and Ladany (2001) found a
positive correlation between MCC self-report measures and a general index
of social desirability: After the researchers had controlled for social desir-
ability, none of the self-report scales were significantly related to a clinician’s
MCC conceptualization ability. Thus, many research questions should be
asked about therapists MCC (Ridley & Shaw-Ridley, 2011; Worthington
& Dillon, 2011). Most notably, to what degree do MCCs increase the effec-
tiveness of therapeutic services provided to culturally diverse clients, reduce
clients’ premature discontinuation of therapy, and enhance clients’ experi-
ences in therapy?
It would not be an exaggeration to say that the MCC construct is
the most fundamental concept in applied multicultural psychology, which is
based on the premise that therapeutic services received by culturally diverse
clients are more effective when therapists skillfully attend to specific cul-
tural variables (Arredondo et al., 1996; Arredondo & Tovar-Blank, 2014;
S. Sue, 1998, 2003). Without the concept of MCC, multicultural psychol-
ogy informs mental health treatment but does not prescribe specific actions.

50       foundations of multicultural psychology

 REVIEW OF THE LITERATURE

Relevant Theory

The seminal paper written for The Counseling Psychologist by D. W. Sue

and his colleagues (1982) established the foundational tripartite conceptual-
ization of MCC as comprising therapists’ cultural knowledge, awareness, and
skills. The authors also presented a detailed listing of specific MCCs, later
operationalized by Arredondo and colleagues (1996). Subsequent scholar-
ship regarding MCC has continued to build and expand on the foundational
tripartite model, with occasional calls for more objective data that include
clients’ perceptions of the multicultural competency of their therapist
(Constantine, 2002; Imel et al., 2011; Pope-Davis et al., 2001; Worthington
& Dillon, 2011). Although elaboration on the various conceptualizations
and multiple lists of MCCs published in the literature is beyond the scope
of this chapter, we describe briefly the components of the tripartite model.

Awareness
Therapists who are keenly aware of their own cultural values, beliefs,
and worldviews should be able to more accurately discern and interpret the
cultural values, beliefs, and worldviews of their clients than those who are not.
Therapists lacking cultural awareness risk misunderstanding clients’ actions
and comments, perhaps assuming that their clients’ values are the same as their
own. Or they may fail to account for their own implicit biases (A. D. Katz &
Hoyt, 2014). For instance, a middle-class male therapist who fails to account
for his own insecurity over his socioeconomic status may falsely conclude
that because his affluent female client was “born into money,” she does not
appreciate the value of work. This preconceived notion could subtly (or per-
haps not so subtly) affect the approach of the therapist and the relationship
between the therapist and client. The following are brief examples of multi-
cultural awareness (Arredondo et al., 1996):
77 awareness of one’s own limits in multicultural competency;
77 awareness of how one’s own background shapes personal values,
assumptions, and biases; and
77 awareness of how different methods of treatment (including
theoretical orientations and their sociopolitical contexts) can
affect work with people from culturally diverse backgrounds.

Knowledge
Therapists with cultural knowledge can ascertain both differences and
similarities across various domains, such as race and ethnicity, gender, sexual

therapist multicultural competence      51

orientation, and religion. They are able to put into context and accurately
interpret the meaning of the actions and perceptions of others. For instance,
a psychologist familiar with the traditional Latin American value of familismo
(“family first”) may welcome a client’s request that family be included in the
therapy process, rather than interpreting this request as a form of enmesh-
ment. Examples of multicultural knowledge include the following (Arredondo
et al., 1996):
77 knowledge of the impact that culture and history have had on
psychological theory, inquiry methods, and professional practice;
77 knowledge of specific contexts of oppression, discrimination,
and prejudice that many culturally diverse clients have encoun-
tered and experienced; and
77 knowledge of cultural attitudes about mental health and mental
health services.

Skills
Multiculturally skilled therapists apply their awareness and knowl-
edge to engage effectively with others and to use culturally appropriate
strategies in therapy (Arredondo et al., 1996). They avoid overgeneralizing
or over-individualizing treatment (S. Sue, 1998) by accounting for cultural
contexts as they meet the needs of the client. For instance, a therapist who
learns of an African immigrant’s strong beliefs regarding gender roles (knowl-
edge) that differ from those of the therapist (awareness) will appropriately
seek common ground with the client in other areas to strengthen the thera-
peutic alliance (skills) before exploring how gender roles relate to the cli-
ent’s presenting concern. The following are examples of multicultural skills
(Arredondo et al., 1996):
77 ability to look beyond color, culture, religion, sexual orientation,
accent, and so forth, and see individuals in a holistic way, thus
viewing the client in the context of his or her historical, socio-
political, and economic background;
77 initiative to seek out educational and consultative experiences
to increase the therapist’s own effectiveness in working across
cultural differences; and
77 ability to modify assessment and treatment methods to better
match the needs of multicultural clientele.
MCCs have strongly influenced formal guidelines adopted by the
American Psychological Association (2003) and other professional organi-
zations. Applied psychology practice needs specific guidelines that stipulate

52       foundations of multicultural psychology

what therapists must do to improve the effectiveness of their work (Kaslow
et al., 2007), including their work with diverse clientele. Some research has
shown that the abilities of therapists to understand and attend to cultural
variables in treatment influence therapeutic processes (Imel et al., 2011).
Evaluating the degree to which therapists’ MCCs affect treatment processes
and outcomes will ultimately improve the delivery of services to a broad
range of clients. Therapists who develop MCCs increase their effective-
ness in working with culturally diverse clients. Therapists who do not may
cause harm to clients or provide therapy that is culturally incongruent and
therefore less effective (D. W. Sue & Sue, 2013).

Narrative Review of Previous Research

During the 3 decades since D. W. Sue and colleagues (1982) published

their conceptualization of MCC, scholars have examined both conceptual and
empirical components of multicultural competence (e.g., Arredondo, Rosen,
Rice, Perez, & Tovar-Gamero, 2005; Drinane, Owen, Adelson, & Rodolfa,
2014; Dunn, Smith, & Montoya, 2006; Owen, Leach, Wampold, & Rodolfa,
2011a). Unfortunately, empirical research focused specifically on the valid-
ity of MCC (Pope-Davis, Coleman, Liu, & Toporek, 2003) has remained
limited, with little attention to its effect on client outcomes (D’Andrea &
Heckman, 2008).
Several researchers have conducted content analyses to identify themes
in the MCC literature (Arredondo et al., 2005; Ponterotto, Fuertes, & Chen,
2000; Pope-Davis et al., 2001). Ponterotto and his colleagues (2000) conducted
a comprehensive review of MCC models, considering both therapists’ acquisi-
tion of MCCs and clients’ ratings of their therapists’ MCCs. They concluded
that although indirect support for MCCs was available, outcome research with
actual clients was needed. They also recommended that MCC research move
away from self-report measures toward evaluation of the abilities of therapists
to conceptualize cases from multicultural perspectives.
In a content analysis of research spanning 20 years, Worthington,
Soth-McNett, and Moreno (2007) examined empirically based studies of
MCCs published between 1986 and 2005. Across the 81 studies reviewed,
56 (72.7%) reported findings from descriptive field surveys and 19 (24.7%)
used some form of analogue research. The vast majority involved therapists’
self-reported levels of MCCs, with only nine studies (11%) using an assess-
ment of MCCs completed by someone other than the therapist. Of the studies
that included clients’ ratings of therapists’ MCCs, the clear majority (82%)
involved volunteers rather than actual clients, which raises questions about
the generalizability to clinical settings.

therapist multicultural competence      53

 The dearth of client outcome research evaluating therapists’ MCC
has been openly lamented (D’Andrea & Heckman, 2008). Critics have
questioned the utility of the construct, arguing that therapists who claim
to have obtained such competence may gain a false sense of effectiveness
(Weinrach & Thomas, 2002, 2004). Other scholars, however, have contested
these criticisms while still acknowledging the need for increased empirical
research of MCC (Arredondo & Toporek, 2004; Arredondo & Tovar-Blank,
2014; H. L. K. Coleman, 2004; Owen, Leach, Wampold, & Rodolfa, 2011b).
Increased research evaluation of MCC is clearly needed (McCutcheon &
Imel, 2009; Worthington et al., 2007).
The historical trend to discuss rather than investigate MCC is begin-
ning to reverse (Worthington et al., 2007). For example, several studies have
examined the correlation of clients’ ratings of their therapists’ MCCs and
their satisfaction with psychotherapy (Constantine, 2002, 2007; Fuertes et al.,
2006; Owen et al., 2011a; Owen, Tao, Leach, & Rodolfa, 2011). This increas-
ing attention to research indicates that the field will benefit from a current
systematic evaluation of recent research findings to support widespread imple-
mentation of the MCCs (Pope-Davis et al., 2003; Worthington et al., 2007).
We therefore conducted a meta-analysis to determine the degree to which
therapists’ multicultural competence is associated with clients’ experiences in
therapy, including their perceptions of the therapist, their participation levels
in treatment, and their clinical outcomes. The following section describes the
data, and a subsequent section summarizes the findings.

QUANTITATIVE SYNTHESIS OF RESEARCH DATA

Our meta-analysis evaluated studies with a quantitative measure of ther-

apist multicultural competence that was statistically associated with at least
one quantitative measure of client experiences in mental health services with
that therapist in three broad categories: (a) client perceptions of the therapist
and treatment (e.g., client evaluations of therapists’ skills, client satisfaction
with the treatment received), (b) client level of participation in treatment
(e.g., premature termination vs. completion of treatment), and (c) and client
clinical outcomes (e.g., symptom reduction). In most cases, the manuscripts
reported associations in terms of Pearson’s r, with other types of statistics
converted to that metric. Positive correlations indicated positive client expe-
riences in therapy associated with higher levels of therapist multicultural com-
petence, and negative correlations indicated positive client experiences in
therapy associated with lower levels of therapist multicultural competence.
Additional information regarding the methods of the meta-analysis is pre-
sented in the Appendix of this book.

54       foundations of multicultural psychology

Description of the Existing Research Literature

We identified a total of 16 studies that reported data on 2,025 clients’

experiences in mental health treatments as a function of their therapists’
level of multicultural competence. Ten studies (67%) involved clients receiv-
ing individual psychotherapy. One study involved an outdoor adventure pro-
gram for at-risk youth, another consisted of volunteer clients who attended
five sessions of counseling in an experimental design, and four studies evalu-
ated clients participating in several modalities (individual and group mental
health treatments). The previously mentioned three aspects of client treat-
ment were considered: (a) client perceptions of therapists and treatments,
(b) client participation in treatment, and (c) client outcomes (i.e., symptom
reduction). Two studies evaluated both the participation of the client and
the client perceptions of the therapist; those distinct areas were evaluated
separately in our analyses.
Table 3.1 contains breakdowns of several study and participant charac-
teristics. The majority of studies have appeared since the year 2000, includ-
ing several unpublished doctoral dissertations. All studies used convenience
samples, with more than half involving university study clients. Studies aver-
aged 127 clients, with only two studies having sample sizes greater than 200.
Most often clients were either African Americans or Hispanic/Latino(a)
Americans, with limited research investigating other racial or ethnic groups.
Only one study involved clients with relatively low levels of acculturation to
Western society (Li & Kim, 2004); this was also the only one that explicitly
adapted the treatment to client culture. The typical study involved cross-
sectional (correlational) data, with only four studies evaluating changes in
client symptoms over time.

Overall Research Findings

Across 10 studies that evaluated client perceptions of therapists, the

random effects weighted correlation with therapist multicultural competence
was r = 0.50 (95% CI = [.31, .65], p < .0001). Across four studies that evalu-
ated the level of client participation in treatment, the value was r = 0.26
(95% CI = [.05, .44], p = .02). Finally, across four studies that evaluated client
outcomes, the value was r = 0.16 (95% CI = [.03, .28], p = .01). The differ-
ences among these three types of outcomes reached statistical significance
(Q = 7.6, p = .02).
The heterogeneity of the findings was very high for studies evaluating
client perceptions of therapists (I2 = 86.3, 95% CI = [77, 92]; Q(9) = 65.7,
p < .001) and for those evaluating client participation in treatment (I2 = 80.9,
95% CI = [50, 93]; Q(3) = 15.7, p = .001). Heterogeneity was moderately high

therapist multicultural competence      55

 TABLE 3.1
Characteristics of 16 Studies of the Association Between
Therapist Multicultural Competence and Client Experiences
in Mental Health Treatments
No. of
Characteristic M studies (k) %
Year of report 2005
Before 1990 0 0
1990–1999 2 12
2000–2011 14 88
Publication status
Published 9 56
Unpublished 7 44
Sampling procedure used
Convenience 16 100
Representative (random selection) 0 0
Research design
Cross-sectional 12 75
Longitudinal 4 25
Population sampled
Outpatient mental health clients 5 31
University/college student clients 9 56
At-risk group members 2 13
Sample size 126.6
<50 5 31
50–99 3 19
100–199 6 38
200–399 1 6
400–999 1 6
Age of participantsa 28.2
Children (<13 years) 1 6
Adolescents (13–18 years) 2 13
Young adults (19–29 years) 5 31
Middle-aged adults (30–55 years) 6 37
Senior adults (>55 years) 0 0
Not reported 2 13
Gender of participants (% female) 64.1
Race of clientsb (%)
African American 46
Asian American 9
Hispanic/Latino(a) American 33
Native American Indian 1
Other ethnic minority 4
aAverage age category of participants in studies. Not all participants in the study would necessarily be in the

category listed. bThe racial composition of clients across all studies, calculated by multiplying the number of
clients in studies by the percentage of clients from each racial group and dividing that product by the total
number of clients across all studies. Three studies included White/European American participants, whose
data were accounted for in statistical models.

56       foundations of multicultural psychology

for the four studies evaluating client outcomes (I2 = 67.3, 95% CI = [5, 89];
Q(3) = 9.2, p = .03). In the 16 research studies identified, the observed corre-
lations prior to aggregation ranged from -0.25 to 0.83. Inconsistent findings
characterized this meta-analysis, with the association between therapist multi-
cultural competence and client experiences in treatment so varied across stud-
ies that it was difficult to interpret the averages reported earlier. We therefore
sought explanations for the variability we observed.

Factors Influencing the Association

As shown in Figure 3.1, the effect sizes in all types of studies were very
unevenly distributed. Specifically, studies tended either to cluster around
r = 0 (indicative of no effect) or to be statistically significant (located beyond
the shaded regions to the right in Figure 3.1), without studies filling in the
space between those two extremes. We therefore examined the data for

Figure 3.1. Contour-enhanced funnel plot of effect sizes (Pearson r) by standard

error. This graph depicts the correlation coefficients between therapists’ multicultural
competence and clients’ experiences in therapy as a function of the number of partici-
pants in the study (operationalized as standard error). The results are highly scattered.
Particularly, the six studies in the region of statistical nonsignificance (white back-
ground) were disconnected from the other studies, which were all statistically signifi-
cant at p < .01. This discrepancy was found to be attributable to study characteristics.

therapist multicultural competence      57

systematic differences that could account for this unusual distribution.
We found that the 11 studies with the largest effect sizes all involved the
same method of evaluating client perceptions of therapist multicultural
competence by using the same measure, the Cross-Cultural Counseling
Inventory—Revised (CCCI–R; LaFromboise, Coleman, & Hernandez,
1991). Thus, the larger effect sizes obtained in these studies likely reflected
shared rater variance (clients completing all measures) and/or shared mea-
surement variance (the content of CCCI–R items having similar meaning
to the content of items on the measures of clients’ perceptions of thera-
pists; see Drinane et al., 2014). The 11 effect sizes from studies measur-
ing client perceptions of therapists’ multicultural competence averaged
r = 0.54, whereas the seven effect sizes from studies using a measure of multi-
cultural competence completed by the therapist averaged r = .05. The differ-
ence reached statistical significance in a random effects weighted regression
model that controlled for the type of outcome evaluated and explained a
remarkable 71% of the variance in effect sizes (Q = 36.3, p < .00001). Thus,
this model accounted for the disparate findings shown in Figure 3.1. All of
the nonsignificant studies in the center of the graph involved therapist self-
reported multicultural competence, and all but one of the statistically sig-
nificant findings (to the right of the shaded lines) involved client ratings of
therapist multicultural competence.
Thus, shared rater variance explained the findings obtained within
studies, even after accounting for the three different types of outcomes mea-
sured (client perceptions of therapists, client participation in treatment, and
client clinical outcomes), with the added possible confound of shared mea-
surement variance (similar content between the CCCI–R and clients’ ratings
of other therapists’ attributes; see Drinane et al., 2014) likely influencing the
findings of the studies measuring client perceptions of therapists. Across those
studies of client perceptions of therapists, the random effects weighted correla-
tion was r = .64 in seven studies in which ratings of therapists’ multicultural
competence were completed by clients, compared with r = -.02 in three studies
in which the therapists rated their own multicultural competence.
We also examined the unusual distribution of effect sizes (Figure 3.1)
for the possibility of publication bias influencing the results. Publication bias
occurs when studies with insignificant or even negative effect sizes are not
located in a literature search (typically because those studies remain unpub-
lished). In the case of this meta-analysis, the notable gap between statistically
significant studies (on the right) and nonsignificant studies (in the center)
suggested several “missing” studies (i.e., studies conducted but unpublished
with values that would fill in the missing spaces in the existing distribution).
Nevertheless, when we controlled for the type of outcome evaluated by studies
located in this meta-analysis, published studies yielded results of about the same

58       foundations of multicultural psychology

magnitude as unpublished studies (p = .17 when testing for differences), sug-
gesting that the results of the meta-analysis were not attributable to publication
status. Furthermore, one statistical method (Duval & Tweedie, 2000) failed to
identify any “missing” studies when conducted on the overall data and applied
separately for each of the three types of study outcomes, the findings of which
also contradict the hypothesis that publication bias influenced the reported
data. Thus, we concluded that the unusual distribution of data (Figure 3.1) was
accounted for in the regression model reported previously (type of outcome and
source of data) and was not attributable to publication bias.

DISCUSSION AND INTERPRETATION OF THE FINDINGS

Overall Findings

We located only 16 studies that considered how therapist MCCs related

to the experiences of culturally diverse clients in therapy, with only four of
those studies evaluating client outcomes in treatment. Those small numbers
restrict our degree of confidence in interpreting the findings of the meta-
analysis; also, the high variability in the findings is problematic. Thus, we
presently have a very unreliable estimate of the association between therapist
MCC and client experiences in therapy, given the limited data available.
Nevertheless, even preliminary interpretations of the available data are pref-
erable to unsupported rhetoric.
Overall, MCCs are positively associated with clients’ experiences in
therapy. However, the findings clearly depend on the type of dependent vari-
able evaluated. Clients’ perceptions of therapists’ characteristics (expertness,
warmth, trustworthiness, etc.) and clients’ satisfaction with the therapist were
strongly related (r = .50) to clients’ perceptions of the therapists’ multicultural
competence. Clients’ level of participation in treatment (i.e., premature termi-
nation vs. completion) was moderately positively associated with clients’ per-
ceptions of therapists’ multicultural competence (r = .26), but clients’ outcomes
in treatment were only mildly associated with ratings of therapists’ multi­cultural
competence (r = .16).
The pattern of these findings is understandable, given that therapist attri-
butes typically explain a much smaller percentage of variance in client clinical
outcomes than do other factors (Norcross & Lambert, 2011). If client outcomes
typically vary about 12% on the basis of therapeutic alliance and 7% on the
basis of therapist attributes, the total amount of variance in client outcomes
possibly attributable to MCCs is already quite small. The average correlation
of r = .16 obtained across four studies suggests that about 2.6% of the vari-
ance in outcomes for culturally diverse clients could possibly be attributable

therapist multicultural competence      59

to therapist MCCs. That value corresponds with about 13.7% of the variance
possibly attributable to therapists and the therapeutic alliance (7 + 12 = 19%
and 2.6/19 = 13.7%, assuming an unlikely situation of no overlap with other
therapist attributes). Thus, MCCs do positively influence client outcomes to
a small degree, along with many other relevant factors. Given the small per-
centage of the variance in client outcomes explained by therapist competence
in general (Webb, DeRubeis, & Barber, 2010), therapist MCCs deserve con-
sideration. Nevertheless, the small number of studies and the wide variability
in the findings prohibit any conclusive interpretation of the data.

Other Factors Influencing the Findings

Secondary analyses confirmed the presence of a major confound in

research on MCCs: the source of the evaluations of therapist MCCs. When
therapists rated their own MCCs, those evaluations consistently explained
less than 1% of the variance in the therapy experiences of diverse clientele,
irrespective of the outcome measured. When culturally diverse clients rated
their therapists’ level of multicultural competence using the CCCI–R, how-
ever, those ratings explained 61% of the variance in clients’ perceptions of
the therapist, 44% of the variance in clients’ participation in treatment, and
16% of the variance in clients’ clinical outcomes (i.e., symptom reduction).
That latter value of 16% accounts for the vast majority of variance possibly
attributable to therapist attributes and the therapeutic relationship (16/19 =
84%, still assuming the unlikely scenario of MCCs being independent of other
therapist attributes).
Therapist MCCs may not matter much if we rely on therapist judgment,
but therapist MCCs may be critical from the perspective of the client. As
already stated, these findings are based on few studies, but the contrast is so
remarkable as to deserve substantial attention in the future: The method for
evaluating therapist multicultural competence clearly influences the data. In
fact, the method for evaluating therapist MCCs is more consequential than
any other factor in this line of research.

Considerations for Future Research

For decades scholars have been calling for increased research on thera-
pists’ MCC, yet few studies have been conducted from the perspectives of
clients. Much of the early scholarship on MCC was necessarily conceptual,
and subsequent scholarship was appropriately concerned with measurement
and psychometric considerations. After these earlier phases, dozens of cor-
relational studies appeared. Despite the slowly increasing variety in approaches,
researchers have persistently avoided working with clinical data.

60       foundations of multicultural psychology

 We offer an analogy for consideration. If a health care breakthrough
were proposed, widely advocated, and required as knowledge for professional
licensure, would it be relegated to fewer than five clinical trials during the
following 30 years? This would never happen in the practice of health care,
but it has happened with the construct of multicultural competence in the
profession of psychology.
Simply stated, fundamental topics deserve the highest priority, not only
to firmly establish their importance but also to promote and support progress
in the field. Despite the innate interest of other topics in multicultural psy-
chology or the popularity of certain lines of inquiry, multicultural psychology
applied to counseling and psychotherapy has little reason for existing without
the concept of therapist multicultural competence. A stronger research foun-
dation must be established to support the weight MCC carries in the field.
The highest priority for scholars in applied multicultural psychology
must be to thoroughly evaluate therapist multicultural competence in terms of
its real world benefits. Therapist multicultural competence was and continues
to be a breakthrough idea in mental health treatment, but it is also a profes-
sional skill set, not merely popular jargon used to maintain an image of cosmo-
politan therapy. The MCCs must be operationalized with sufficient specificity
to measure their influence—not merely to document their importance (now
affirmed for over 30 years) but also to systematically improve their application
among clinicians who work with people in need of professional assistance. It is
essential to better promote applied research that evaluates the outcomes that
matter most: clients’ experiences.
The meta-analytic review also suggested that study quality requires
attention. Experimental designs should be used, randomly assigning partici-
pants to therapists with clearly demonstrable MCCs or to “treatment as usual.”
This line of research will also benefit from larger samples and random selec-
tion of participants (D’Andrea & Heckman, 2008), optimally from mental
health clinics rather than from the overrepresented college student clientele.
As study quality improves, the publication rate of research on MCCs should
also improve beyond the current estimate of 56%.
Given the discrepancies in the findings of research studies across the
MCCs evaluation sources (therapist self-report vs. client report), researchers
must separate out their findings accordingly. Data reported by clients tend to
yield very high correlations with clients’ experiences in therapy (on average
r = .64), but therapists’ self-reported data on their MCC are not related at
all to clients’ experiences (on average r = -.02). For instance, in one study
that measured both therapist and client ratings of therapist multicultural
competence (Fuertes et al., 2006), the MCCs ratings completed by clients
correlated r = .65 with measures of clients’ perceptions of therapists, but the
self-reported MCC ratings completed by therapists only correlated r = .06

therapist multicultural competence      61

with the measures of clients’ perceptions. In the same study, the correlation
between the MCCs ratings by clients and the MCCs ratings by therapists
themselves averaged r = -.03, indicating no meaningful relationship. Such
data should inspire a wave of new inquiry into the operationalization and
evaluation of MCCs.
Specifically, future research will need to sort out at least two likely prob-
lems in this line of inquiry. First, which MCCs do clients distinguish from
general counseling skills (Drinane et al., 2014)? As currently operationalized,
MCCs may be difficult for clients to distinguish from other therapeutic alli-
ance constructs such as connectedness, empathy, and so forth. Although many
MCCs differ conceptually from general counseling competence, in reality they
overlap; therapists who demonstrate MCCs would likely demonstrate general
counseling competencies as well. Thus, some research might involve trainees
whose general counseling skills and MCCs involve a broader range than those
of more experienced clinicians. Measures of MCCs involving observation and
client ratings must be carefully evaluated for content and revised for major
overlap with general counseling competencies. Detailed operationalization
of specific behaviors and microskills relevant to MCC may help make MCCs
measurement less abstract and thus less likely to correlate so highly with other
constructs rated by the client.
Second, the field should consider the degree to which self-rated MCCs
evaluations are valid. Therapists’ evaluations of their own abilities may involve
abstractions and generalities difficult for a conscientious self-evaluator to rate
with precision. Previous research has already shown that self-ratings of MCCs
differ markedly from expert evaluations of MCC (Cartwright et al., 2008), but
as undesirable as that may be, the results of the present meta-analysis depict
an even more dire problem: Therapists’ self-ratings of MCCs are not at all
related to clients’ experiences in treatment. A future meta-analysis specific
to MCCs measurement could inform this controversy, but the issue is highly
complex. Some seasoned therapists may insist that they have much more to
learn about multicultural issues (and thus score themselves only moderately
high on measures of MCCs), yet other therapists with relatively limited expe-
rience with diverse clientele would insist they are adequately informed to treat
a broad range of clients equitably (and thus score moderately high on self-
reported MCCs). Scholars should undertake the important work of improving
evaluations of therapist MCC (Ridley & Shaw-Ridley, 2011; Worthington &
Dillon, 2011), perhaps starting with evaluations of client outcomes across race
of therapist and race of client (Owen et al., 2011a, 2011b).
Until adequate data suggest consensus on best practices, we recommend
considering existing measures involving either therapist self-ratings or client
ratings as problematic in clinical research, albeit for very different reasons. Past
research is suspect unless procedures were in place to address the two likely

62       foundations of multicultural psychology

confounds just described. Of course the alternative of observational mea-
surement entails a different set of problems, certainly in terms of the human
resources required. Yet until MCCs measurements are validated against clearly
defined practices in therapy that account for client experiences, the field will
lack the data needed to improve clinical practice with multicultural clientele.
As a temporary measure until such definitions and measurements receive sup-
port, researchers can administer multiple measures of therapist MCCs (i.e.,
completed by clients and by trained observers; see also the approach recom-
mended by Worthington & Dillon, 2011) or evaluate client outcomes by race
within and across therapists (Owen et al., 2011a, 2011b). Using only one
type of traditional measurement of MCCs (particularly therapist self-reports)
in research with clients without accounting for likely confounds will yield
misleading results.

Suggestions for Practitioners

All individuals, therapists included, are subject to a variety of pre-

conceived biases and personal limitations. We may underestimate, or more
likely, overestimate our competencies and skills, which could possibly be
damaging to clients if our own certainty prevents us from accurately under-
standing client characteristics and needs. In the present meta-analysis the
essential message for practitioners is to continue to be humble, acknowledge
our limitations, and seek to learn from every single client. A culturally com-
petent therapist is a continual student, reexamining beliefs and enhancing
skills, benefitting from ongoing clinical supervision and peer feedback (e.g.,
Soheilian, Inman, Klinger, Isenberg, & Kulp, 2014). Clinicians also gain valu-
able insights from clients. It would therefore be wise to solicit and respond to
explicit client feedback (both verbal and data-based; Lambert, 2010), rather
than rely on our own impressions of multicultural competence. In fact, the
contrast between client- and self-reported data suggests that therapists should
continually work toward increasing multicultural competencies rather than
assuming competence can be fully achieved. The knowledge required to under-
stand how cultural worldviews interact with aspects of the human experience is
highly complex, so of course all therapists, regardless of experience, will benefit
from systematic learning over time—perhaps a lifetime.
Overall, practitioners should continue to have faith in the construct of
therapist MCC (Owen et al., 2011b), even though several problems require
attention (Ridley & Shaw-Ridley, 2011; Worthington & Dillon, 2011).
Although this chapter has highlighted major deficiencies in the research,
the results of the meta-analysis in no way detract from the centrality or the
utility of the underlying construct of MCC. MCC enhances the effectiveness
of therapy. In fact, when relying on clients’ evaluations (arguably the most

therapist multicultural competence      63

important perspective to consider), the data showed that therapist MCCs
are associated with clients’ perceptions of the therapist. Clients are much
more likely to complete treatment and experience positive outcomes when
therapists demonstrate MCCs.

CONCLUSION

The available data affirm that therapist multicultural competence is favor-

ably associated with the experiences of culturally diverse clients in treatment.
The data are imprecise, necessitating caution in deciding how much can be
extrapolated to therapeutic practice, but the data obtained from clients (rather
than therapist self-reports) develop a striking picture. Diverse clients tend to
see therapist multicultural competence as highly related to, yet distinct from,
other positive counselor attributes. In addition, culturally diverse clients are
moderately more likely to prematurely discontinue treatment when their thera-
pists do not demonstrate multicultural competence. Client outcomes improve
when their therapists are able to competently attend to and value the varying
experiences of culturally diverse clients.
Our review located only 16 studies that considered the association of
therapist MCC with client experiences in therapy. The research on this topic
has been increasing in recent years; thus, future reviews will be needed as
the field continues to advance. We emphasize that although the limitation of
available data should not be twisted to undercut the utility of therapist multi-
cultural competence, our review represents a direct challenge to researchers to
focus more attention on the MCC construct. Other variables that have been
fiercely calling for research attention must be temporarily put on hold until
researchers have squarely dealt with the most fundamental variable in applied
multicultural psychology: the multicultural competencies of therapists.

64       foundations of multicultural psychology

 II
SYNTHESIS OF RESEARCH ON
THE EXPERIENCES OF PEOPLE
OF COLOR WITH MENTAL
HEALTH SERVICES
 4
MENTAL HEALTH SERVICE
UTILIZATION ACROSS RACE:
A META-ANALYSIS OF SURVEYS
AND ARCHIVAL STUDIES

This chapter reviews data regarding racial discrepancies in mental health

service utilization (see also Snowden & Yamada, 2005). Historically, racial dis-
crimination has restricted access to mental health services by people of color
in North America (Richards, 2012). In previous centuries, inadequate access
to quality mental health services was simply one of countless forms of systemic
discrimination. Over many decades, as individuals, then social organizations,
then finally the government gradually recognized and then slowly enforced civil
rights, the racial desegregation of civic institutions, then health institutions,
and then finally mental health institutions became increasingly normative.
Attention was given to equity. Clinic doors opened to multicultural America.
Now, many decades after the civil rights movement, we have to ask
whether we have achieved racial equity in mental health service utilization,
with all racial groups accessing mental health services at the same rate as

A complete list of references for the studies included in this chapter’s meta-analysis is posted online.
Readers can consult it at [Link]
[Link]
Foundations of Multicultural Psychology: Research to Inform Effective Practice, by T. B. Smith and J. E. Trimble
Copyright © 2016 by the American Psychological Association. All rights reserved.

67
White/European Americans. If utilization rates are now equal across race,
multicultural psychology has overcome a major consequence of systemic racial
discrimination. However, if contemporary mental health services continue to
be underutilized by persons from groups that have been historically oppressed,
we of the present generation must work with policymakers to remove the
entrenched racial inequities (McGuire & Miranda, 2008; Meyer & Takeuchi,
2014; Valentine, DeAngelo, Alegría, & Cook, 2014).
Following a review of relevant theory and an overview of relevant
research, this chapter describes the results of a meta-analysis of studies exam-
ining mental health service utilization across race. Implications of the results
of the meta-analysis for practitioners and researchers are presented. The term
utilization in this chapter refers to the receipt of mental health services; it does
not include referrals for services or retention and dropout during services.

NARRATIVE REVIEW OF THE LITERATURE

Relevant Theory

Why do some individuals in need of professional mental health services

not obtain them? Scholars commonly provide at least three types of explana-
tion (Meyer & Takeuchi, 2014). First, individual and social factors include
perceived susceptibility to future distress, external and social cues to take
action, and extent to which professional services are perceived to be benefi-
cial (Rosenstock, 1966). Second, poverty and related socioeconomic factors
decrease access and utilization (Dressler, Oths, & Gravlee, 2005; Lo, Cheng,
& Howell, 2014). Third, system and delivery factors include the number of
clinicians available in the client’s area, location of and distance to services,
length of services, prerequisite conditions for services, and nature of services
(Andersen & Newman, 1973; Cook, Doksum, Chen, Carle, & Alegría, 2013).
These three classes of factors interact with race to exacerbate racial discrepan-
cies (e.g., Snowden & Yamada, 2005). Specifics are provided in Exhibit 4.1
and in the sections that follow.

Individual and Social Factors

Cultures conceptualize well-being and mental illness differently
(Kleinman, Eisenberg, & Good, 1978). For instance, in some Asian cul-
tures only chronic or violent conditions may be considered to be mental ill-
ness requiring professional treatment, whereas anxiety and depression may
be seen as problematic conditions requiring informal, not formal, inter-
vention (S. Chen, Sullivan, Lu, & Shibusawa, 2003). Some people per-
ceive emotional difficulties to be caused by imbalances in forces of nature

68       foundations of multicultural psychology

 EXHIBIT 4.1
Common Explanations for Racial Discrepancies
in Mental Health Service Utilization
Individual/social factors
• Perceptions of mental health and self-care/desire to obtain professional mental
health treatment
• Conceptualization of mental illness and its causes
• Cultural values and social norms regarding treatment for mental illness
• Unfamiliarity with professional services/options
• Use of informal/family networks (rather than professionals) to cope with mental illness
• Negative stigma about seeking professional treatment for mental illness
• Mistrust of mental health services
• Possible racial differences in levels of need for mental health services
• Individual/social characteristics that systematically differ across racial groups
(age, education, English language fluency, religious beliefs, acculturation to
Western society)
Socioeconomic factors
• Incidence and prognosis of mental illness associated with poverty
• Social/family instability associated with poverty
• Lack of access to quality professional services associated with poverty
• Lack of insurance coverage for mental illness (until only recently)
• Legal or undocumented immigrant status
Systems/delivery factors
• Lack of mental health services located in areas where people of color reside
• Poorer quality of mental health services located in areas where people of color reside
• Greater likelihood of involuntary mental health treatments for people of color
• Bias/racism in assessment and service delivery

or supernatural causes (Alvidrez, 1999) irrelevant to Western methods of

professional treatment. Alternatively, mental health may be conceptual-
ized in terms of physical health, such that individuals seek treatment from
physicians rather than mental health specialists (e.g., Neighbors et al.,
2007). Still other cultural beliefs suggest that emotional difficulties result
from a lack of motivation or willpower (Alvidrez, 1999) and that individuals
should be strong enough to handle difficulties alone (Alegría et al., 2002;
Cachelin & Striegel-Moore, 2006). Thus, we cannot assume that individuals
with cultural origins outside of Europe would seek out Western professional
mental health services (Bosworth et al., 2000). They may not perceive such
services to be helpful, or they may be unfamiliar with professional service
options and availability (Broman, 1987; Cachelin & Striegel-Moore, 2006).
Even when people of color perceive professional services to be helpful, they
may still underutilize those services (Diala et al., 2000).
When individuals from non-Western cultures need mental health treat-
ment, they may rely on familiar informal networks for assistance (Golding &
Wells, 1990). In particular, they may feel that mental health issues are best

mental health service utilization across race      69

addressed within the family (Alvidrez, 1999). Individuals with strong family
support experience several benefits that promote well-being (Berkman, Glass,
Brissette, & Seeman, 2000), and some cultures implicitly or explicitly discour-
age taking personal difficulties outside the family (M. Yeh et al., 2005). Thus,
some cultures prefer to seek mental health assistance from sources other than
psychotherapy (Broman, 1987).
Many of these individual and social considerations interact with nega-
tive stigmata associated with professional mental health services (Alvidrez,
1999; Corrigan, 2004). People of color who are contemplating using mental
health services may experience guilt or shame that prevents them from
entering therapy (Cachelin & Striegel-Moore, 2006), but treatment deci-
sions are not so simple (Alvidrez, Snowden, & Patel, 2010). The common
assumption that negative stigmata about mental illness and use of mental
health treatment may affect people of color more intensely than Whites
may or may not be justified, with only limited research directly confirming
that hypothesis (Conner, Koeske, & Brown, 2009).
Related to, yet distinct from, negative stigma is outright mistrust of pro-
fessional mental health services. The psychology literature has long recognized
mistrust stemming from past racial oppression and associated mistreatment of
people of color by institutions historically serving the needs of Whites (e.g.,
Triandis, 1976). For instance, several years ago Blacks considering mental
health services feared being involuntarily hospitalized (Sussman, Robins, &
Earls, 1987). Contemporary mistrust of professional mental health services
may still influence differences across race (Henderson et al., 2014).
Racial differences in utilization rates could also reflect differences in
mental illness rates (Lo et al., 2014). Racial groups may differ in their level
of psychological distress (e.g., Broman, 1987), and obviously “need for
treatment . . . is a strong and consistent predictor of use of outpatient men-
tal health services” (Swartz et al., 1998, p. 141). Despite the unequivocal
impact of need for treatment on service utilization, most community-based
studies of racial differences conducted in previous decades did not control
for this factor. More recent research has begun to adjust for factors such as
symptom severity and duration, an important step toward understanding
the nature of racial differences in mental health service utilization.
A variety of other individual and social factors thought to influence uti-
lization rates across race have been mentioned in the literature. These include
age and gender (e.g., Neighbors et al., 2007; Swartz et al., 1998), but variables
more likely to differ across race include English language fluency (Alegría et al.,
2002), religious beliefs (Alvidrez, 1999), level of acculturation to Western soci-
ety (e.g., Cachelin, Veisel, Barzegarnazari, & Striegel-Moore, 2000), and level
of education (Cachelin & Striegel-Moore, 2006). Of course, level of education
is also highly related to socioeconomic status, the next area to be considered.

70       foundations of multicultural psychology

Socioeconomic Factors
Socioeconomic differences persist across racial groups (U.S. Census
Bureau, 2011) and confound racial differences in mental health service utili-
zation. “Generally speaking, socioeconomic status not only affects the onset,
development, and prognosis of mental illnesses, but also has an effect on help
seeking and service receiving” (S. Chen et al., 2003, p. 27).
Poverty is clearly associated with multiple risk factors for mental ill-
ness, including social and family instability, which could contribute to under­­
utilization of mental health services. “Neighborhoods with high concentrations
of poor people . . . tend to have high rates of unemployment, homelessness,
crime, and substance abuse. There is high residential turnover and little oppor-
tunity for the development of . . . community services.” (Chow, Jaffee, &
Snowden, 2003, p. 792). Given these multiple considerations, the emphasis
on socioeconomic factors in the research literature has become so strong that
some authors have suggested that reported findings of no racial differences in
mental health service utilization are due to failure to control for socioeconomic
differences (Bosworth et al., 2000).
Related to socioeconomic status is mental health insurance coverage.
Until recent decades low-income individuals and families not qualifying
for public assistance (e.g., Medicaid) typically lacked mental health cov-
erage either because they could not afford insurance or because mental
health treatments were excluded from coverage in low-cost health insur-
ance plans. Some scholars have even described the relationship between
insurance coverage and mental health service utilization as causal: “Lack
of insurance may cause underrepresentation in services” (Garland et al.,
2000, p. 135).
Despite the strong emphasis on socioeconomic factors and insurance cov-
erage shown by the recent literature, research data have been mixed. Although
some research findings substantiate commonly held assumptions about the
strong influence of socioeconomic factors explaining the racial discrepan-
cies in utilization, the results of several studies were unaffected by socio-
economic variables (e.g., Neighbors et al., 2007; Padgett, Patrick, Burns,
& Schlesinger, 1994). One author concluded that among people of color
“research indicates that availability of health insurance does not neces-
sarily promote greater contact with mental health services” (Cachelin &
Striegel-Moore, 2006, p. 159), which suggests that racial differences in
utilization are not solely economic. Discrepancies cannot be explained by
any single set of variables. Socioeconomic, intrapersonal, interpersonal, and
racial–cultural variables interact. For example, immigration status is associ-
ated not only with cultural beliefs but also with socioeconomic status, access
to mental health insurance, and familiarity with systems.

mental health service utilization across race      71

Systems and Delivery Factors
Scholars repeatedly mention restricted access to mental health services
as an explanation for racial discrepancies in utilization (e.g., Alegría et al.,
2002). Many mental health clinics are located in suburbs that are predomi-
nantly White; fewer are located in urban areas with higher concentrations of
people of color. Underutilization also occurs in rural areas where few clinics
are available (e.g., Angold et al., 2002). Thus, likelihood of mental health
service utilization is a function of density of mental health service providers
(Cook et al., 2013). “Availability and accessibility of mental health services
are clearly factors in ethnic differences in utilization” (Alvidrez, 1999, p. 515).
Mental health clinics in inner-city and rural areas tend to be public
institutions, which are likely to provide different kinds of services than those
in suburban facilities, which are commonly private institutions. Systematic
differences in quality of care, range of services, office procedures, and general
accessibility could account for racial differences in utilization rates (e.g., Cook
et al., 2013).
Quality of care may also differ systematically across client race. For
instance, traditional mental health services are less effective with clients
of color than culturally adapted services (see Chapter 7, this volume), so
people of color receiving traditional mental health services are at a compar-
ative disadvantage, whereas those receiving culturally congruent services
may be more likely to refer others to treatment and thus perpetuate positive
cycles of service utilization within a community.
Differential treatment of clients across race can be indirectly evaluated
by such variables as involuntary versus voluntary utilization. Research has
shown that people of color are more likely than Whites to use emergency
mental health services (which are not necessarily voluntary) and that people
in impoverished areas tend to receive services that are “more coercive and less
volitional” (Chow et al., 2003, p. 796). Macro-level dynamics that maintain
racism in professional mental health services include denial of or selective
attention to race, marginalization of the experiences of people of color and
normalization of the experiences of Whites, and minimization of discrepancies
across race (Thompson & Neville, 1999). To the extent that those dynamics
continue in contemporary mental health practices, discrepancies in utilization
rates across race remain possible.

Narrative Review of Previous Research

In the middle of the previous century community mental health services

were promoted as a means to overcome common barriers to mental health
treatment. However, the lofty ambitions of early community psychology
remained unachieved (Sarason, 1974). In the 1970s, research documentation

72       foundations of multicultural psychology

of continued racial discrepancies in the utilization of community mental health
services, despite egalitarian ideals, brought national attention to the issue of
race in psychology (S. Sue, 1977). The evidence failed to confirm aspirations.
If utilization discrepancies across race continued, even after years of effort, the
entire system warranted reconsideration. Clearly, additional effort was needed
to achieve equity, but increasing evidence showed that the effort would have to
come from an entirely different direction. Revising the existing mental health
system would not suffice. New perspectives would be needed—multicultural
perspectives. The ground had been cleared for the rise of multicultural scholar-
ship in the 1980s.
Awareness of racial discrepancies in mental health utilization became
widespread in 1999 when the first report of the U.S. Surgeon General specific
to mental health highlighted those inequities. The topic received even more
attention through a follow-up government report, Mental Health: Culture, Race,
and Ethnicity in 2001. Among other points, the report strongly encouraged sev-
eral means for increasing equal access to services and reducing racial barriers to
services, such as those listed in Exhibit 4.1. A recent special section of American
Psychologist provided updated perspectives (López, Barrio, Kopelowicz, & Vega,
2012; Snowden, 2012; S. Sue, Cheng, Saad, & Chu, 2012). Of all the topics in
multicultural psychology, utilization discrepancies have received the broadest
public attention.
Notwithstanding the attention given to racial discrepancies, it is plau-
sible that racial discrepancies have diminished over time. Mental health ser-
vices have become more commonly accepted in the general population and
mental health providers increasingly receive training in multicultural compe-
tence, such that the results of contemporary research may differ from findings
of earlier studies.
It is also possible that differences in utilization attributed to race may
be more attributable to socioeconomic factors (or other factors). To address
the interactions of the multiple explanations for utilization discrepancies
(Exhibit 4.1) with race, researchers increasingly use statistical models that
include factors such as differences in socioeconomic status and symptom
severity (Alegría et al., 2002; McGuire, Alegría, Cook, Wells, & Zaslavsky,
2006; Snowden & Yamada, 2005).
Among other contributions to the literature, a meta-analysis of research
findings would be able to identify whether research findings have changed over
time and would help to clarify whether race or other factors account for the uti-
lization discrepancies. Our objective in conducting a meta-analysis was to con-
trast the reported number of actual clients with estimates of the total number
of individuals available in the area (non-clients) across race to (a) determine
the relative degree to which people of color in the United States and Canada
utilize mental health services and (b) estimate the degree to which study and

mental health service utilization across race      73

participant characteristics moderated any racial discrepancies in mental health
service utilization. That is, we sought to determine the size of the racial discrep-
ancies and to identify factors that could account for them.
The methods and results of the meta-analysis are described in the follow-
ing section. Individuals uninterested in the details of meta-analytic methodol-
ogy may find the details cumbersome and the tables difficult to interpret, but a
subsequent section summarizes the findings (see “Discussion and Interpretation
of the Findings”).

QUANTITATIVE SYNTHESIS OF RESEARCH DATA

To estimate the degree to which people of color have utilized mental

health services, we carried out a meta-analysis of research conducted in the
United States or Canada that evaluated individuals’ use of mental health
services across race. Data typically came from two sources: (a) studies of indi-
viduals’ current and past mental health utilization broken down by race (i.e.,
a survey of whether participants had ever used mental health services, with
responses to that question analyzed by race) and (b) utilization figures in
clinics compared with estimates of the available population by race (i.e., the
racial percentages represented by clients at a clinic compared with census
estimates of the racial composition of the city and vicinity in which the clinic
is located). We necessarily excluded studies containing only information
about clients’ racial composition that could not be contrasted with a pool of
potential clients (individuals who should have had access to the same clinic)
because the relevant effect size statistic requires that baseline information.
Almost all studies reported data from White/European Americans and
Canadians, so we used that population as the contrast group when calculating
effect sizes because no other racial group was consistently represented in the
literature. Having made that determination based on the need for a consistent
comparison group,1 we excluded the few studies that did not evaluate that group.
After gathering manuscripts, we reviewed data reported in studies to
identify truly independent estimates. Two manuscripts included the results
of more than one completely distinct study and database, which we treated as
distinct studies when coding effect sizes. We also found multiple instances of
duplicated findings involving the same database, such as the Epidemiological

1We oppose the assumption that Whites are a normative reference group. The experiences of other racial
groups do not need to be compared with Whites to be understood or valued. Our decision to contrast
the utilization rates of people of color with Whites in this chapter was based on our preference to include
as many studies in our analyses as possible, and no other racial group was represented as consistently in
the literature. Moreover, having a consistent contrast group allowed for contrasts involving every group.
Our strategy therefore maximized the number of studies we could include in the meta-analysis while also
enabling comparisons to be made involving every group.

74       foundations of multicultural psychology

Catchment Area Study and county mental health databases in California.
Inclusion of multiple manuscripts with the same participants would have
violated the assumption of statistical independence; therefore, we retained
only manuscripts that did not overlap substantially with the data in other
manuscripts (<20% of the same participants or over 7 years between esti-
mates with the same participants). When the data in manuscripts did overlap
substantially, we retained the manuscript with the largest number of partici-
pants; when the number of participants was identical across manuscripts, we
retained only the manuscript that statistically controlled for participant men-
tal health status. In all, we eliminated 26 manuscripts located in our literature
search because of duplication of data publication.
Different rates of use of mental health services across race were com-
monly reported in the metric of an odds ratio. An odds ratio of 0.50 would
indicate that people of color were 50% less likely than Whites to utilize
mental health services, given their representation in the population. Odds
ratio values cannot be meaningfully combined using simple averages: An
odds ratio of 2.0 (meaning twice as likely, a 2:1 ratio) and an odds ratio of
0.50 (meaning half as likely, a 1:2 ratio) average to 1.0 (meaning equally
likely, a 1:1 ratio). To account for that difficulty and thus enable analyses,
we transformed the data to the natural log of the odds ratio.
Statistical estimates derived from other metrics were converted to log
odds ratios using meta-analytic software and subsequently converted back to
odds ratios to enable interpretation. If studies contained both unadjusted and
adjusted effect sizes, we coded those that had been adjusted for possible con-
founds (e.g., mental health status, socioeconomic status) as long as the adjusted
values pertained to the entire sample, not to a restricted subsample. We tracked
whether estimates were based on unadjusted or adjusted data in order to sub-
sequently analyze possible differences in those types of effect sizes. The general
methods of the meta-analysis are reported in the Appendix to this book.

Description of the Existing Research Literature

We located 130 independent studies that met inclusion criteria. Typically,

this research has involved one of two approaches: (a) obtaining responses on
surveys about individuals’ prior mental health service use and subsequently
analyzing the data by race, or (b) reviewing archived records of mental health
agencies and comparing the percentages of actual clients by race with cen-
sus estimates of the available clientele. The effect sizes therefore represent
the degree to which individuals of different races report having used mental
health services or having actually used services relative to comparable indi-
viduals who would have been potential clients. In 60% of the studies these
comparisons were made with general community members, but several studies

mental health service utilization across race      75

conducted comparisons with groups at risk of mental health challenges (e.g.,
medical patients) or groups known to have symptoms of mental illness (e.g.,
screened for clinical conditions; see Table 4.1).
The total number of clients identified in studies exceeded 4,771,472.
(Ten studies not included in this total reported only the total number of
participants, not the number of clients.) Large databases containing over
1,000 mental health clients (i.e., county or state records or nationwide sur-
veys) were used in 40% of the studies. The extremely large average number
of clients per study reported in Table 4.1 was based on a positively skewed
distribution; the median number of clients was 603. The large number of
participants in many studies resulted in a cluster of studies with small stan-
dard error values (those at the top of Figure 4.1).
African American clients were evaluated in 94 studies, Hispanic/
Latino(a) American clients in 66 studies, and Asian Americans in 31 studies.
Only 13 studies contained estimates specific to Native American Indians, so
those results were grouped with 35 studies that did not specify the race of the
“ethnic minority” participants or contained estimates for “other” ethnic groups.
Overall, the data appeared to be characterized by strong external validity.
Studies contained large numbers of participants, data were collected from
every major population center and many rural areas in North America, and
multiple age groups were represented.

Overall Research Findings

We examined racial differences in mental health service utilization

by conducting a multivariate meta-analysis (accounting for the observed
value of r = 0.42 for within-study correlations). The overall model reached
statistical significance (Wald X2(4) = 77.9, p < .0001), with the results
by race reported in Table 4.2. African Americans, Asian Americans, and
Hispanic/Latino(a) Americans all statistically differed from White/European
Americans in their use of mental health services (p < .001), but participants
from unspecified/other racial groups did not (p = .10). Odds ratios can be
interpreted in terms of likelihood, with an odds ratio of 0.67 equating to a
33% difference in likelihood of service utilization (1.0 - 0.67 = 0.33), which
value J. Cohen (1988) characterized as a “small” effect size (a 60% difference
in likelihood being a “moderate” effect size according to J. Cohen). Thus,
African Americans were only 21% (1.0 - 0.79 = 0.21) less likely (meaning
slightly less likely) than White/European Americans to use mental health
services; Asian Americans were 51% less likely, and Hispanic/Latino(a)
Americans were 25% less likely than Whites to use mental health services.
The confidence intervals in Table 4.2 can be used to compare findings
among all groups. Asian Americans were less likely to use mental health

76       foundations of multicultural psychology

 TABLE 4.1
Characteristics of 130 Studies of Utilization of Mental Health
Services Across Race
No. of
Characteristic M studies (k) %
Year of report 1996
Before 1980 12 9
1980–1989 23 18
1990–1999 32 25
2000–2008 63 49
Publication status
Published 112 86
Unpublished dissertations 18 14
Research design
Cross-sectional 67 51
Longitudinal 12 9
Archival 50 39
Mixed (more than one of the above) 1 1
Population sampled
General community members (former clients) 87 67
Clinical populations (currently in treatment) 43 33
Comparison group
General community members (non-clients) 78 60
At-risk group members 23 18
Diagnosable populations (in need of treatment) 29 22
Sample size (number of clients) 40,087
<50 8 7
50–99 11 9
102–199 18 15
200–399 11 9
400–999 23 19
>1000 48 40
Age of participantsa 30.7
Children (<13 years) 17 15
Adolescents (13–18 years) 12 11
Young Adults (19–29 years) 19 17
Middle-aged Adults (30–55 years) 58 52
Senior Adults (>56 years) 6 5
Gender of participants (% female) 55.8
Race of clientsb (%)
African American 5
Asian American <1
Hispanic/Latino(a) American 3
Native American Indian <1
Unspecified/other ethnic minority 12
White/European American 76
Note. Not all variables sum to the total number of studies because of missing data.
aAverage age category of participants in the studies is given; however, not all participants in the study would

necessarily be in the category listed. bThe racial composition of mental health clients across all studies was
calculated by multiplying the number of clients within studies by the percentage of clients from each racial
group and dividing that product by the total number of clients.

mental health service utilization across race      77

Figure 4.1. Contour-enhanced funnel plot of effect sizes (natural log odds ratios) by
standard error for 130 studies of utilization of mental health services across race.
This graph shows the distribution of effect sizes as a function of the number of
participants in the study (operationalized as standard error). In this case, the overall
average is slightly to the left of 0, indicating a small racial discrepancy in utilization of
mental health services, but the results are scattered, with that variability weakening
the interpretability of the numerical average.

TABLE 4.2
Weighted Mean Effect Sizes (Odds Ratios) Across Participant Race
Group k OR+ SE 95% CI
African Americans 94 .79a .064 [.69, .89]
Asian Americans 31 .49 .099 [.41, .60]
Hispanic/Latino(a) Americans 66 .75a .051 [.68, .83]
Unspecified/other ethnic minorities 48 .86a .092 [.72, 1.03]
All groups combined 130 .74 .027 [.70, .78]
Note. Effect sizes were derived from contrasts of mental health utilization rates with those of White/European
Americans; nevertheless, all groups can be contrasted within the table. CI = confidence interval; k = number
of studies; OR+ = random effects weighted effect size (odds ratio), a value of 1.0 indicating no difference, a
1:1 ratio of utilization; SE = standard error.
aStatistically significantly different from Asian Americans (p ≤ .0001).

78       foundations of multicultural psychology

services than any other group, but groups did not differ statistically from one
another. Specifically, Asian Americans were 37% less likely to use mental
health services than African Americans, 34% less likely to use mental health
services than Hispanic/Latino(a) Americans, and 43% less likely to use men-
tal health services than unspecified/other people of color. We emphasize that
although these numbers may appear impressive, they are of small magnitude
according to J. Cohen’s (1988) interpretation guidelines.
Given difficulties in interpreting odds ratios, we found it useful to pres-
ent the data in terms of averaged within-study utilization rates. Across studies,
Whites within samples tended to use mental health services 24% of the time,
African Americans 19% of the time, Hispanic/Latino(a) Americans 17% of
the time, and Asian Americans 12% of the time. Overall, two trends were
clearly shown in the data: White/European Americans tended to use mental
health services most often, and Asian Americans used mental health services
least often.
Across all 130 studies, the random effects weighted effect size was
OR = 0.739 (95% CI = [0.70, 0.78], p < .0001). The heterogeneity of the find-
ings was extremely large (I2 = 99.7, 95% CI = [99.6, 99.7]; Q(129) = 43,324.1,
p < .00001), meaning that the results were very inconsistent across studies. This
variability of the overall findings was so large that we decided to conduct all
subsequent analyses within racial groups. Race-specific analyses would reduce
the amount of variability and be easier to interpret than conglomerate results.

Likelihood of Publication Bias Adversely Influencing the Results

Publication bias occurs when the data obtained in a meta-analysis fail

to represent the entire universe of studies due to the increased probability
of nonsignificant results remaining unpublished (and therefore less acces-
sible for meta-analytic reviews). As seen in Figure 4.1, the data in this meta-
analysis appeared somewhat imbalanced toward the left line, which indicates
statistical significance; the distribution of the data was dense on the left side
but sparse on the right, which is the area of nonsignificance. This distribution
normally suggests that some nonsignificant studies were missing in the meta-
analysis. Both Egger’s regression test and an alternative to that test recom-
mended for odds ratio data (Peters, Sutton, Jones, Abrams, & Rushton, 2006)
reached statistical significance (p < .01), which suggested possible publication
bias. One statistical method (Duval & Tweedie, 2000) estimated 26 “missing”
studies, and when these were accounted for, the resulting overall effect size was
OR = 0.85 (95% CI = [0.81, 0.88]), a value closer to 1.0 (no difference) than
the overall finding of OR = 0.74 reported earlier. It thus seemed likely that the
effect sizes reported in Table 4.2 overestimate discrepancies in mental health
service utilization across race.

mental health service utilization across race      79

 Before reaching the conclusion of overestimated discrepancies, however,
we sought to rule out an alternative explanation for the suspicious distribution
of the data in Figure 4.1. In coding the data we had observed that most of the
studies at the top of Figure 4.1 had involved census records and other large
databases (e.g., health department records), so it was possible that the results
of the studies at the top of Figure 4.1 differed from the others as a function of
methodology. We evaluated that possibility as part of a comprehensive evalu-
ation of study characteristics, as discussed in the following section.

Influence of Study and Participant Characteristics on the Findings

Study Characteristics
We examined differences in effect sizes as a function of research design,
statistical adjustment of data, type of comparison group, and year of data
collection. Statistically significant differences were found across all study char-
acteristics for every group except Asian Americans (see Tables 4.3 and 4.4).
Three findings were consistent across African American, Hispanic/Latino(a)
American, and unspecified/other people of color. First, studies surveying par-
ticipants regarding past mental health utilization yielded much larger effect
sizes than studies using archival designs involving institutional databases
(which yielded no discrepancy in mental health service utilization across
racial groups except Asian Americans). Second, studies using institutional
databases or surveys found utilization rates of people of color significantly
lower than those of Whites, but studies using census estimates (of poten-
tially available clients) found utilization rates equivalent to or higher than
those of Whites. Third, statistical estimates that controlled for potential
confounds (e.g., principally mental health status as an indicator of need for
services but also socioeconomic status, age, gender, insurance status) were asso-
ciated with larger effect sizes than those based on unadjusted values.
Because the 130 studies had been conducted over many decades, it was
essential to also ascertain whether racial discrepancies had changed over time.
As suggested previously, we considered it possible that racial discrepancies in
mental health service utilization have decreased over decades. However, initial
analyses indicated the opposite: Studies in recent years demonstrated greater
racial inequities in mental health service utilization than studies conducted in
previous decades (r = -0.27, p < .0001). This finding was so surprising that we
doubted its validity. When coding manuscripts, we had found that study quality
had improved over time, with many of the recent studies statistically adjust-
ing for possible confounds (uncommon among earlier studies) such that meth-
odological differences could possibly account for the results opposite of those
expected. We therefore conducted random effects weighted regression models
to evaluate the simultaneous relationship among all study characteristics.

80       foundations of multicultural psychology

 TABLE 4.3
Weighted Mean Effect Sizes (Odds Ratios) for African Americans and
Asian Americans Across Study and Participant Characteristics
African Americans Asian Americans
Characteristic Qb k OR+ 95% CI Qb k OR+ 95% CI
Study characteristics
Data source 0.6 2.0
Published 82 .73 [.67, .79] 25 .48 [.41, .56]
Unpublished 12 .80 [.64, .99] 6 .62 [.45, .85]
dissertations
Research design 37.8*** 1.6
Cross-sectional survey 61 .61 [.56, .67] 14 .45 [.36, .56]
Archival 32 .99 [.87, 1.11] 17 .54 [.45, .66]
Comparison data sourcea 119.8*** 0.9
Census/population 16 1.46 [1.27, 1.67] 10 .46 [.36, .58]
Institution database 13 .63 [.54, .74] 7 .52 [.39, .71]
Sample surveyed 65 .61 [.57, .67] 14 .53 [.43, .66]
Statistical controlsb 21.3*** 1.7
Unadjusted effect size 65 .82 [.76, .89] 25 .52 [.44, .60]
Adjusted effect size 25 .56 [.49, .65] 5 .39 [.27, .57]
Participant
characteristics
Client populationc 33.6*** 2.1
Former clients 66 .64 [.58, .70] 19 .55 [.46, .65]
(retrospective)
Clients at a clinic 28 1.00 [.88, 1.13] 12 .45 [.36, .55]
Comparison groupd 18.3*** 0.6
Community members 53 .86 [.79, .95] 24 .51 [.44, .60]
Clinical populations 22 .58 [.50, .68] 6 .45 [.33, .61]
Acculturation levele 5.3*
Low 6 .41 [.34, .51]
High 6 .59 [.47, .74]
Payment method 27.6*** 0.2
Not specified 27 .59 [.49, .70] 9 .48 [.36, .63]
Public 24 1.09 [.91, 1.30] 10 .51 [.40, .64]
Private and mixed 43 .65 [.57, .75] 12 .52 [.41, .65]
public/private
Note. CI = confidence interval; k = number of studies; OR+ = random effects weighted odds ratio; Qb = Q-value for variance
between groups.
aThis variable overlapped with study research design. Studies either compared clinic records with census data or institutional

databases (archival designs) or compared survey responses of utilizers with nonutilizers. bThis variable contrasted effect
sizes computed from raw data with those statistically adjusted for participants’ mental health status and other variables. The
few studies that adjusted for variables unrelated to mental health status were omitted to make the contrast as clean as possible.
cThis variable contrasted studies evaluating retrospective recall of mental health service utilization with actual client records

maintained by a mental health agency. dThis variable contrasted studies in which the non-client comparisons were general
community samples or individuals identified as being in need of mental health services. eStudies not containing adequate
information relevant to acculturation were excluded. This variable was calculated only for Asian Americans and Hispanic/
Latino(a) Americans because those are the groups most likely to be recent immigrants to North America.
*p < .05. **p < .01. ***p < .001.

mental health service utilization across race      81

 TABLE 4.4
Weighted Mean Effect Sizes (Odds Ratios) for Hispanic/Latino(a) Americans and
Unspecified/Other Ethnic Minorities Across Study and Participant Characteristics
Hispanic/Latino(a) Americans Unspecified/other
Characteristic Qb k OR+ 95% CI Qb k OR+ 95% CI
Study characteristics
Data source 12.1*** 1.6
Published 58 .68 [.63, .73] 44 .90 [.85, .96]
Unpublished 13 .92 [.79, 1.08] 4 1.03 [.85, 1.25]
dissertations
Research design 10.1** 65.1***
Cross-sectional survey 40 .66 [.61, .71] 27 .69 [.63, .75]
Archival 26 .79 [.73, .86] 21 1.12 [1.04, 1.21]
Comparison data sourcea 17.5*** 129.6***
Census/population 16 .88 [.78, .99] 16 1.33 [1.23, 1.45]
Institution database 10 .69 [.59, .81] 5 .67 [.57, .78]
Sample surveyed 40 .65 [.59, .70] 27 .71 [.65, .77]
Statistical controlsb 6.5* 25.5***
Unadjusted effect size 47 .75 [.70, .81] 33 1.00 [.94, 1.07]
Adjusted effect size 15 .62 [.54, .71] 12 .68 [.60, .78]
Participant
characteristics
Population sampledc 18.8*** 0.9
Former clients 43 .64 [.59, .69] 31 .87 [.79, .95]
(retrospective)
Clients at a clinic 23 .85 [.77, .94] 17 .94 [.83, 1.05]
Comparison groupd 3.7 2.9
Community members 43 .75 [.70, .81] 27 1.00 [.91, 1.09]
Clinical populations 15 .65 [.57, .74] 13 .87 [.73, 1.00]
Level of acculturatione 0.1
Low 11 .84 [.63, 1.11]
High 5 .81 [.55, 1.19]
Method of payment 12.5** 24.4***
Not specified 18 .61 [.52, .71] 13 .71 [.60, .85]
Public 21 .83 [.74, .93] 12 1.24 [1.05, 1.47]
Private and mixed 27 .68 [.62, .75] 23 .81 [.71, .91]
public/private
Note.╇CI = confidence interval; k = number of studies; OR+ = random effects weighted odds ratio; Qb = Q-value for variance
between groups.
aThis variable overlapped with study research design. Studies either compared clinic records with census data or institutional

databases (archival designs) or compared survey responses of utilizers with nonutilizers. bThis variable contrasted effect sizes
computed from raw data with those statistically adjusted for participants’ mental health status and other variables. The few studies
that adjusted for variables unrelated to mental health status were omitted to make the contrast as clean as possible. cThis variable
contrasted studies evaluating retrospective recall of mental health service utilization with actual client records maintained by a
mental health agency. dThis variable contrasted studies in which the non-client comparisons were general community samples or
individuals identified as being in need of mental health services. eStudies not containing adequate information relevant to accul-
turation were excluded. This variable was calculated only for Asian Americans and Hispanic/Latino(a) Americans because those
are the groups most likely to be recent immigrants to North America.
*p < .05. **p < .01. ***p < .001.

82 ╇╇╇╇╇ foundations of multicultural psychology

 The regression models by race included all study characteristics, includ-
ing the recommended estimate of publication bias (Peters et al., 2006). Thus,
the models would differentiate not only between the effects of time and study
quality but also between study quality and possible publication bias, as men-
tioned in the previous section.
The five variables in the model explained 44.2% of the variance in effect
sizes with data from African American participants (see Table 4.5). Two pre-
dictors reached statistical significance: research design and type of contrast.
Specifically, studies using archival data tended to have less discrepancy across
race than studies in which data had been collected using surveys (standard-
ized beta = .16), and studies that compared the racial composition of clients
with local census estimates showed much higher rates of mental health service

TABLE 4.5
Random Effects Regression Weights for Study Characteristics
Associated With Effect Sizes
Group/variable R2 SE p b
African Americans (k = 93) 44.2***
Year of data collection .005 .63 -.03
Archival researcha .040 .02 .16
Census comparisonb .104 <.001 .53
Statistical controlsc .082 .38 -.06
Publication bias estimated 26.3 .63 -.03
Asian Americans (k = 31) 17.2
Year of data collection .013 .91 .03
Archival researcha .142 .46 .20
Census comparisonb .254 .11 -.37
Statistical controlsc .296 .37 -.19
Publication bias estimated 167 .17 -.30
Hispanic/Latino(a) Americans (k = 66) 10.8***
Year of data collection .004 .04 .14
Archival researcha .031 .26 .08
Census comparisonb .071 <.001 .26
Statistical controlsc .064 .10 -.10
Publication bias estimated 24.3 .26 -.07
Unspecified/Other (k = 48) 24.7***
Year of data collection .004 .07 .13
Archival researcha .045 .03 .15
Census comparisonb .092 <.001 .47
Statistical controlsc .087 .99 .01
Publication bias estimated 30.6 .50 -.03
Note. k = number of studies; SE = standard error.
aThis variable contrasted archival designs with cross-sectional surveys. bThis variable contrasted studies

that used relevant census estimates as the comparison for non-clients with studies that used non-clients
identified in surveys or databases. cThis variable contrasted studies that statistically controlled for potential
confounds (primarily mental health status but also variables such as age, gender, socioeconomic status, and
insurance coverage) with studies that used unadjusted data. dEstimate of publication bias recommended by
Peters and colleagues (2006) for data in OR format.
***p < .001.

mental health service utilization across race      83

utilization among African Americans than the studies that used surveys or
institutional databases (standardized beta = .53). The other variables in the
model failed to reach statistical significance. Thus, the trend for the findings
of studies to have changed over time was accounted for by study methodology.
Similarly, publication bias seemed an unlikely threat to the findings when
considered simultaneously with study methodology.
The model with data from Asian Americans failed to reach statistical
significance (Table 4.5). The variables entered in the model for that racial
group did not account for systematic differences in the results. Nevertheless,
the results of the model with data from unspecified/other ethnic minorities
were very similar to the results obtained with African Americans (Table 4.5).
The model with data from Hispanic/Latino(a) Americans was also similar,
except that year of data collection did achieve statistical significance. Notably,
the direction of the association had reversed from that reported for the bivari-
ate correlation; whereas the association had been negative, it was positive
(the expected direction) in the regression model. After controlling for study
characteristics, we found racial discrepancies somewhat less likely to occur in
recent years than in previous years (standardized beta = .14).
In the other models, year of data collection failed to reach statistical
significance; thus, after accounting for other factors, racial discrepancies in
mental health service utilization appear to have changed very little over the
past 3 decades. The estimate of possible publication bias did not reach statis-
tical significance in any model, indicating that the univariate data and the
unusual pattern of data in Figure 4.1 can be accounted for by study method-
ological characteristics. We therefore concluded that publication bias had
not adversely affected the overall findings.

Participant Characteristics
Studies involved participants with a variety of backgrounds and char-
acteristics. We therefore sought to determine whether results systematically
differed as a function of participant age, approximate age cohort (year of the
data collection minus the average age of participants), sample gender com-
position (operationalized as percentage female), sample racial heterogeneity
(estimated by the percentage of White participants involved in the study),
type of client (former client retrospectively recalling service utilization or cur-
rent client in a specific clinic), type of comparison group member (community
member or individual identified as needing mental health services), partici-
pant level of acculturation (estimated for Asian Americans and Hispanic/
Latino(a) Americans on the basis of descriptions of the sample within stud-
ies), and client form of payment for the mental health services. The categori-
cal variables were analyzed using random effects weighted analyses of variance,
and the continuous variables were analyzed using random effects weighted

84       foundations of multicultural psychology

correlations, with statistically significant variables included in subsequent ran-
dom effects weighted regression models, also conducted by race.
Participants’ age, estimated age cohort, and gender were not statistically
significantly associated with effect size magnitude for any of the four racial
groups analyzed. However, the racial heterogeneity of the samples was associ-
ated with effect sizes in the analyses involving African Americans (r = .21,
p = .002), Asian Americans (r = .48, p < .0001), unspecified/other people of
color (r = .37, p < .0001), but not Hispanic/Latino(a) Americans (r = .09,
p = .22). Thus, for the three former groups the studies conducted in settings
with relatively fewer White participants were more likely to yield rates of
mental health service utilization discrepant from that of Whites. Stated differ-
ently, people of color used mental health services relatively less often when the
total research sample had greater representation of people of color. Given that
some community mental health centers specifically serve communities with
large percentages of people of color, this unexpected finding should receive
particular scrutiny in the future.
Tables 4.3 and 4.4 portray the results of the random effects weighted
analyses of variance with the categorical variables. The first analysis evalu-
ated whether any differences were found when the clients participating in
studies had recalled past mental health utilization or were identified with a
particular clinic at the time of the study. Among studies involving African
American and Hispanic/Latino(a) American participants, individuals pro-
viding retrospective information were relatively less likely to report using
mental health services than individuals who had access to a particular clinic.
Thus, people from those groups tended to report using mental health ser-
vices more frequently when they were actually using the services than when
asked generically about their past experiences with services.
A second analysis involved the contrast groups (non-clients), which
consisted either of general community samples (ostensibly with low rates of
mental illness) or of individuals identified as being in need of mental health
treatment (those in the clinical range on a screening instrument). In the data
from African American participants, statistically larger racial discrepancies in
service utilization were found when contrasting utilization rates with those of
clinical populations than when contrasting them with rates of general commu-
nity members. Thus, African Americans who needed mental health services
were at greater risk of underutilizing those services (Lo et al., 2014). This find-
ing did not occur with the three other racial groups evaluated.
We considered it possible that level of acculturation to Western society
could be related to mental health service utilization (e.g., Cachelin et al.,
2000). This information was only available in studies of Asian Americans and
Hispanic/Latino(a) Americans, and what was found was limited. Analyses
indicated that the difference between participants of high versus low levels of

mental health service utilization across race      85

acculturation reached statistical significance only in the data involving Asian
American participants. Asian Americans low in acculturation to Western
society were less likely to use mental health services than those who appeared
to be highly acculturated.
As noted earlier in the review of relevant theory, previous researchers
have strongly suggested that utilization discrepancies across race have been
a function of socioeconomic status, particularly insurance status and method
of payment for services (e.g., S. Chen et al., 2003; Garland et al., 2000).
Although we could not directly evaluate socioeconomic status of participants
because studies commonly omitted relevant information, we did analyze dif-
ferences in the type of payment arrangement for clients in the studies. Very
few studies explicitly limited data collection to clinics or practices that did
not accept public payment for services, so we combined those few studies
involving such clinics with the more common category involving clinics that
accepted either public or private payment. That contrast revealed statistically
significant differences across all groups except Asian Americans. Participants
who received public mental health services were relatively more likely to use
those services than participants in studies that had failed to report any infor-
mation regarding payment or those in studies of clinics that accepted mixed
private and public payment or solely private payment. Public mental health
services negated racial discrepancies for all groups except Asian Americans.
Recognizing that participant characteristics could interact with one
another, we conducted a random effects weighted regression model for each
racial group to ascertain which participants remained statistically significant
in the presence of one another. The results of these four models are presented
in Table 4.6.
In the data involving African American participants, all four of the vari-
ables that had been statistically significant at the univariate level remained
statistically significant in the regression model. Thus, African Americans were
less likely to use mental health services in studies when there were relatively
fewer White participants, when individuals retrospectively recalled mental
health service use, when participants not using mental health services needed
those services, and when clients attended anything other than exclusively
public agencies.
In the regression model with Asian American participants, the only pre-
dictor variable to remain statistically significant was the percentage of White
participants in the study. Asian Americans had relatively higher utilization
rates in studies with greater percentages of White participants. The other vari-
ables in the model did not reach statistical significance, although the magni-
tude of the standardized coefficients was moderate in some cases.
The regression model involving Hispanic/Latino(a) American partici-
pants explained only 11% of the variance in effect sizes, and the only predictor

86       foundations of multicultural psychology

 TABLE 4.6
Random Effects Regression Weights for Participant Characteristics
Associated With Effect Sizes
Group/Variable R2 SE p b
African Americans (k = 92) 32.8***
% of White participantsa .002 .006 .21
Clients at a clinicb .104 .01 .22
Public payment of servicesc .108 <.001 .34
Clinical population comparisond .105 .007 -.21
Asian Americans (k = 31) 16.0***
% of White participantsa .004 .002 .44
Clients at a clinicb .154 .30 -.15
Public payment of servicesc .152 .88 .02
Clinical population comparisond .186 .29 -.15
Low acculturatione .180 .10 -.23
Hispanic/Latino(a) Americans (k = 64) 11.0***
% of White participantsa .002 .29 .07
Clients at a clinicb .079 .004 .23
Public payment of servicesc .082 .14 .12
Clinical population comparisond .083 .63 -.03
Low acculturatione .100 .60 .04
Unspecified/Other (k = 46) 21.4***
% of White participantsa .285 <.001 .36
Clients at a clinicb .124 .70 -.03
Public payment of servicesc .131 <.001 .32
Clinical population comparisond .139 .78 .02
Note. k = number of studies; SE = standard error.
aStudies not reporting this information were excluded from the analysis. bThis variable contrasted studies evalu-

ating retrospective recall of mental health service utilization with actual client records maintained by a mental
health agency. cThis variable contrasted studies in which payment for mental health services was provided by
public or other means with studies not reporting method of payment and reporting private or mixed methods
of payment. dThis variable contrasted studies in which the non-client comparisons were general community
samples or individuals identified as being in need of mental health services. eThis variable was calculated only
for Asian Americans and Hispanic/Latino(a) Americans; it contrasted studies estimated to involved participants
with low levels of acculturation to Western society (recent immigrants) with all other studies.
***p < .001.

variable to reach statistical significance was the type of clients within studies.
Studies in which former clients retrospectively recalled their service utiliza-
tion found lower relative mental health use compared with studies in which
participants were clients at a particular clinic involved in the study.
The two predictor variables in the regression model involving unspecified/
other people of color that reached statistical significance were the percent-
age of White clients in the study and the type of payment made for services.
Studies with greater percentages of White clients and with public payment
tended to find greater relative utilization of mental health services for the
unspecified/other people of color in the study. Thus, these variables remained
statistically significant, even when considered simultaneously with the other
two variables in the model.

mental health service utilization across race      87

 DISCUSSION AND INTERPRETATION OF THE FINDINGS

People in North America use mental health services at rates that differ by
race. Asian Americans are the least likely of any group evaluated to use mental
health services. The differences observed among racial groups (Table 4.2) fall
in the range of “small” magnitude, according to J. Cohen’s (1988) benchmark
value, equivalent to OR = 0.67 (a 33% reduced likelihood of service utilization),
but we here provide some contextualization to J. Cohen’s generic guidelines for
interpretation.

Issues of Interpretation

Admittedly, odds ratios can be difficult to interpret.2 Illustrating the find-

ings in terms of human lives can facilitate interpretation of the odds ratios in
Table 4.2.
Consider a hypothetical research study of 1,000 participants exactly
representative of the 2010 U.S. census populations by race (i.e., 122 African
American participants to correspond with the census estimate of 12.2% of
the total population), with an average within-study mental health service
utilization rate of 24% among White participants (the average utilization rate
found across all studies in the meta-analysis), and with the same statistical
estimates as those presented in Table 4.2. Given those parameters, the hypo-
thetical study would have identified 19 people of color (1.9% of the study’s
total participants and 5.2% of the participants of color) who would have
failed to receive mental health services had those services been equitable with
White/European Americans. Of those 19, there would have been five African
Americans, eight Hispanic/Latino(a) Americans, five Asian Americans, and
one “other” ethnic minority individual.
Thus, the overall percentage of inequity is very small (with the racial
inequity negatively affecting 19 of about 363 people of color). However, that
interpretation would be no consolation to the 19 untreated individuals and
their family members. Moreover, when multiplying the findings of the pres-
ent meta-analysis by the total U.S. population, the figures would suggest that
tens of thousands of people of color may fail to receive needed mental health
services if those services remain inequitable across race.

2For individuals familiar with the common statistic R2, the overall differences in race accounted for
0.7% of the variance in the relative likelihood of mental health service utilization across the 130 studies.
The greatest contrast observed in the meta-analysis occurred between Whites and Asian Americans,
a difference which explained about 3.7% of the variance in the results of 31 studies with Asian
American participants.

88       foundations of multicultural psychology

Implications From the Secondary Analyses

The overall findings differed across several study and participant char-
acteristics. These differences help illuminate the meaning of the overall esti-
mates. Five major themes emerged from the secondary analyses, detailed next.

Contrasts Involving Census Estimates

Initially, publication bias was identified as a threat to the interpretation
of the overall results of the meta-analysis, but the results of subsequent analy-
ses suggested that studies using different research methods yielded different
results—and those different methods accounted for the non­symmetric data
distribution (Figure 4.1). Specifically, studies involving census estimates of
the population typically included very large numbers of people, so the effect
sizes extracted from those studies tended to have very small standard error
values (resulting in the cluster of studies at the top of Figure 4.1). Except
for studies involving Asian Americans, when researchers contrasted client
utilization rates by race with relevant census estimates of the population’s
racial composition, relatively greater mental health service utilization was
observed among people of color. In fact, studies involving African Americans
and unspecified/other people of color that used census estimates actually found
higher rates of mental health service utilization for these groups than for
Whites, a finding completely opposite from the direction of the overall results.
Two possible explanations could account for this finding. Censuses of
ethnic neighborhoods may have underrepresented people of color; how-
ever, the small degree of error in census data would make this explanation
untenable. A more plausible explanation stems from the fact that studies
using census estimates had to have at least one mental health clinic in the
vicinity (that is why the researchers conducted the study in that location),
whereas general surveys likely involved some participants who had no easy
access to a mental health facility: Proximity of mental health services could
possibly account for the differences in the findings. This same explanation
could account for the differences observed between cross-sectional surveys and
archival research, the latter design necessarily entailing clinic accessibility.

Racial Heterogeneity of Clients

Across studies of all racial groups except Hispanic/Latino(a) Americans,
the greater the percentage of White clients, the greater the relative utilization
rates shown for clients of color. This finding is not attributable to a statisti-
cal artifact (odds ratios are based on the relative utilization rates and are not
affected by the percentages of clients within racial groups). However, this
finding could be attributable to either clinic accessibility and/or individuals’

mental health service utilization across race      89

level of acculturation. If mental health services are disproportionately more
available in neighborhoods with predominantly White/European Americans,
people of color living elsewhere may have relatively decreased access to clin-
ics (e.g., Cook et al., 2013). Another congruent possibility is that people of
color who attend a clinic with a large percentage of White clients could be
more acculturated to White culture than those who attend a clinic with lower
percentages of Whites.

Retrospective Reporting of Mental Health Utilization

In studies with African Americans and Hispanic/Latino(a) Americans,
less frequent mental health utilization rates were found in retrospective reports
by participants (i.e., responding to a survey) compared with data derived from
actual clients at a clinic. This finding could be attributable to inaccurate
reporting by survey participants, who may have felt disinclined to report past
mental health service use because of an internalized negative stigma about
seeking these services, despite apparent confidentiality. Whatever the cause of
inaccuracy, the data from actual clinic records would likely be more depend-
able than estimates obtained from participant retrospective recall.

Public Payment of Services

In studies with African Americans and unspecified/other people of color,
public payment of mental health services was associated with utilization rates
equitable with those of Whites. Thus, in those instances, racial inequities
appeared to be related to economic factors. Public payment of services did not
significantly affect the utilization rates of Asian American or Hispanic/Latino(a)
American clients, which suggests that for those populations cultural views or
other factors such as immigration or legal status are more relevant to mental
health service utilization than economic matters. Overall, these findings deserve
future attention. For instance, there is a need to rule out the possibility of sam-
pling bias, given that individuals who apply for public assistance may differ from
those who could benefit from public assistance but do not apply (e.g., clients with
chronic mental health conditions and minimal support networks).

Mental Health Status of the Comparison Group

Although most studies used the general population as the comparison
group of non-clients, some studies involved participants who had been identi-
fied by clinical screening instruments or recent clinical records as in need of
mental health services. Comparisons using the general population sought to
answer the question “To what degree do individuals of different races avail
themselves of mental health services?” Comparisons involving at-risk or diag-
nosable populations asked, “To what degree do individuals of different races

90       foundations of multicultural psychology

use mental health services that are needed?” The answer to these questions
proved to be similar for all populations except for African Americans, for
whom mental health service utilization was lower when the individuals were
in need of services. If African Americans needing services are less likely to
receive them, the finding is troubling. Negative stigmata or racial biases may
or may not explain the finding (Alvidrez et al., 2010), but the precise causes
must be identified and corrected.

Considerations for Future Research

For several decades scholars have been aware of multiple factors plausibly
resulting in utilization discrepancies across race (Exhibit 4.1; McGuire et al.,
2006; Snowden & Yamada, 2005). Of course, researchers could not possibly
evaluate or control for all of these factors simultaneously in any given research
study. Thus, in most studies we located, scholars (particularly those who found
minimal differences across race) hedged their interpretations with the reality
that they did not account for all possible factors—and they listed unmeasured
variables that could have possibly influenced the results (see Exhibit 4.1). A
typical discussion section might read, “We did not find large racial discrep-
ancies, but this finding could be accounted for by X, Y, or Z that we did not
measure.” This understandable qualification in data interpretation, charac-
teristic of many research reports, has led to an undesirable trend: repetition of
the same plausible confounds and explanations (Exhibit 4.1) without strong
confidence in the overall results and without much evidence regarding the
relative contribution of the various explanations for the discrepancies.
In that context, the present meta-analysis contributed substantially to
the literature by evaluating findings across studies that controlled for a variety
of factors, particularly socioeconomic status. Meta-analytic aggregation of
data involving different sets of statistical controls provides less methodologi-
cal control than would be desirable, but it provides an estimate of a hypo-
thetical general condition with multiple variables controlled within many
studies. And the findings indicated that the combined data remained of low
magnitude, whether or not the studies controlled for potential confounds.
Those findings contrast with the general tone of discussions in the literature
about utilization discrepancies across race: Small racial discrepancies are not
necessarily qualified by multiple confounds. Some unknown deficiency of
the meta-analysis may yet qualify the findings, but the meta-analytic results
are highly unlikely to have been influenced by any single confound, let alone
some improbable combination of variables that happened to diminish the
magnitude of the findings across 130 studies, making the racial discrepancies
appear less strong than they are in reality. Thus, we can be fairly certain that
the small differences in utilization across race were racial differences.

mental health service utilization across race      91

 The meta-analysis was also a useful step forward for researchers seeking
to identify the relative contributions of some of the causal and confounding
factors mentioned in the literature (Exhibit 4.1). Among the factors evalu-
ated consistently within studies, results sometimes differed as a function of the
five conditions detailed in the previous section: contrasts involving census
estimates, racial heterogeneity of clients, retrospective reporting of mental
health utilization, public payment of services, and mental health status of the
comparison group. In some cases, these factors remained explanatory even in
the presence of one another. Thus, future research can continue to investi-
gate these particular variables with confidence. However, the data were not
moderated by gender, age, or age cohort, so those particular variables should
not require sustained attention when considering utilization discrepancies
across race.
Indirectly, the data showing differences between studies using census
estimates and those using surveys seemed to indicate that proximity and access
to mental health services may account for racial discrepancies. Underlying dif-
ferences in access to care across race remain problematic (Cook et al., 2013).
We therefore encourage research evaluating proximity to mental health
services through geospatial analysis (e.g., de Smith, Goodchild, & Longley,
2007), a method perfectly suited to address that important topic.
A major limitation to the research findings would have occurred if the
underlying rates or severity of mental illness differed across race (e.g., Lo
et al., 2014). However, most of the recent research studies controlled for
symptom incidence or severity. Systematic differences in the rates or the
severity of mental illness across race may occur but seem unlikely explana-
tions for underutilization of mental health services by people of color (e.g.,
Alegría et al., 2002).
Many scholars have emphasized the influence of socioeconomic factors
on utilization discrepancies across race. Our finding that public payment of
services minimized racial discrepancies involving African American partici-
pants and unspecified/other people of color certainly lends support to that
prevalent conclusion (Thomas & Snowden, 2001). Nevertheless, this finding
did not characterize the data from Asian American and Hispanic/Latino(a)
American clients. The fact that in some cases
socioeconomic factors cannot explain ethnic differences in use points to a
need to go beyond conventional wisdom that liberalizing health benefits
will be sufficient to address the health and mental health needs of ethnic
minority groups. Thus changes in health policies designed to reduce eco-
nomic barriers to care are necessary but not sufficient to close the gap in
unmet need for these groups. (Padgett et al., 1994, p. 358)
Cultural explanations cannot be overridden by socioeconomic explanations;
both must be considered simultaneously (see Exhibit 4.1).

92       foundations of multicultural psychology

Suggestions for Practitioners and Agencies

The tenets of multicultural psychology can inform practitioners seeking

to increase mental health service utilization among people of color. A key
principle of multicultural psychology is to make the services provided congru-
ent with the cultural worldviews and experiences of the clients. When clients
perceive services to be helpful, they are more likely to utilize them and rec-
ommend them to others; when they have negative experiences in treatment,
the opposite occurs: “Underuse of mental health services . . . [may not be due]
to intrinsic negative attitudes, but to problems in the health services delivery
systems that have a negative impact on those attitudes following the use of
mental health services” (Diala et al., 2000, p. 462). Practitioners should under-
stand clients’ cultural worldviews, including the stigma attached to receiving
mental health services, and they should work in ways that alleviate rather
than exacerbate incongruence between therapist and client expectations.
Despite the importance of therapist multicultural competence during
therapy, therapist expertise alone does not attract new clients in large num-
bers. Community awareness initiatives and outreach programs may be necessary
if people of color are to seek mental health services when needed (Neighbors
et al., 2007), particularly among Asian Americans, who are least likely to utilize
mental health services. “Better outreach and public education of mental health
care workers are essential to improve [client] access” (Chow et al., 2003, p. 796).
Practitioners can also influence local referral networks. In most communi-
ties, a variety of agencies (health, legal, employment, youth, religious, commu-
nity development, etc.) tend to work frequently with culturally diverse clientele.
Regular communication and consultation with those agencies would likely
increase referrals. Moreover, because people of color tend to be more likely to
use emergency services than outpatient clinics (Chow et al., 2003), practitioners
providing outpatient services can maintain close ties with emergency service
professionals to enhance efforts aimed at prevention and long-term recovery. In
addition, practitioners can invite current clients and community leaders sympa-
thetic to professional mental health services to pass along information to people
they believe would consider those services. These and similar methods for work-
ing with local referral networks should greatly enhance utilization by individuals
otherwise unlikely to consider mental health care.
Practitioners should also keep in mind the multiple barriers to treatment
(Exhibit 4.1) and recognize that financial considerations do not necessarily
trump cultural considerations. “Changes in health policies designed to reduce
economic barriers to care are necessary but not sufficient to close the gap in
unmet need for [people of color]” (Padgett et al., 1994, p. 358). Practitioners
can consider reduced or sliding fees, but ultimately, they will need to find out
and address multiple concerns of the potential clients in their locale. The

mental health service utilization across race      93

concerns about mental health services among Native American Indian mili-
tary veterans residing in Sioux Falls, South Dakota, likely diverge from those
of African American high school students in Asheville, North Carolina, and
from those of immigrant families in Calexico, California. Practitioners can
identify the issues and concerns of local clientele by consulting with com-
munity members or conducting a focus group at a public health center. With
that essential local information, they can find local solutions to facilitate
individuals’ engagement in treatment when needed.

CONCLUSION

In North America, individuals of Asian descent are the least likely to

use professional mental health services, but other people of color are still less
likely to use mental health services than Whites. Nevertheless, the magnitude
of inequity is lower than has been commonly assumed, and racial discrepan-
cies decreased over time for all groups except Asian Americans in studies with
census data. Truly, “little progress has been made to eliminate the disparities in
mental health service access for Asian American populations” (S. Sue et al.,
2012, p. 540).
We felt grateful that the racial inequities were smaller than we had imag-
ined them to be, but like most people receiving news too good to be true, we
could not help questioning the data. In the end, we could not dismiss the
facts that the data came from many research studies, most of them by recog-
nized advocates of multicultural psychology, with participants totaling in the
millions, conducted in rural and urban settings across multiple geographic
regions. If a meta-analysis has been conducted in multicultural psychology
that could claim stronger external validity, we have not yet found it. And the
meta-analytic data indicate that race accounts for small differences in mental
health service utilization, except in the case of Asian Americans.
Multicultural psychology attracted national attention when research
documented racial discrepancies in mental health service utilization (e.g.,
S. Sue, 1977). We are grateful that the inequities were shown to be smaller
than long presumed. We are keenly aware that inequities persist. And we rec-
ognize that access to services is not the same as access to high quality services.
We join others in urging policy makers and mental health providers to take
additional steps necessary to eradicate all inequity and to promote high qual-
ity services for all populations (López et al., 2012; McGuire & Miranda, 2008;
Meyer & Takeuchi, 2014; Snowden, 2012; S. Sue et al., 2012; Valentine
et al., 2014). We have offered specific recommendations that can increase
accessibility of mental health services to all cultural and racial groups.

94       foundations of multicultural psychology

 5
PARTICIPATION OF CLIENTS OF
COLOR IN MENTAL HEALTH SERVICES:
A META-ANALYSIS OF TREATMENT
ATTENDANCE AND TREATMENT
COMPLETION/ATTRITION

After an individual has decided to utilize mental health services and has
entered treatment, multiple concerns can arise with every session: Is ther-
apy as helpful as anticipated? How much trust, effort, emotional discomfort,
money, and time away from work and family should be invested and for how
long? What if an employer or an acquaintance finds out? Other consider-
ations include, but are not limited to, preexisting level of distress and moti-
vation for change, external and environmental contingencies, social norms
and peer support, interactions with the therapist, and expectations for and
beliefs about therapy. A client’s decision to enter treatment is an extremely
important step, but it is among the first in a complex process.
The complexity increases for clients of color, whose experiences and
worldviews may not be understood or supported by their community or thera-
pist. Clinicians must be skilled in adapting traditional therapeutic methods to

A complete list of references for the studies included in this chapter’s meta-analysis is posted online.
Readers can consult it at [Link]
[Link]
Foundations of Multicultural Psychology: Research to Inform Effective Practice, by T. B. Smith and J. E. Trimble
Copyright © 2016 by the American Psychological Association. All rights reserved.

95
meet the needs of diverse clientele (e.g., a recent immigrant from Bangladesh
or a first-year college student who is Hopi). Cultural factors clearly influence
not only a client’s perceptions and experiences in therapy but also the thera-
pist’s perceptions and skills in working with diverse clients (e.g., Hayes, Owen,
& Bieschke, 2015). In addition to the multiple retention factors common
to most clients’ experience, continuation in treatment for clients of color
also depends on a therapist’s multicultural competence (Gaztambide, 2012;
Langer, 1999). Multiculturally competent therapists may be more likely to rec-
ognize the concerns of culturally diverse clients and address those concerns in
a sensitive manner that will encourage continued participation in treatment.
For decades researchers have reported that clients of color discontinue
mental health services more frequently and earlier in treatment than White/
European Americans (e.g., S. Sue, 1977). However, the magnitude of that
discrepancy has remained unspecified. Mental health providers advocating
for the needs of clients of color would be in a better position to help serve
those needs if they had accurate data regarding the extent of inequities. How
great are racial and ethnic differences in treatment participation? Are those
differences affected if participation is considered in terms of attendance
(number of sessions) versus treatment completion? Do socioeconomic differ-
ences account for participation differences across racial and ethnic groups?
The answers to these and similar questions would facilitate the work of both
scholars and therapists concerned with providing culturally competent men-
tal health services.

REVIEW OF THE LITERATURE

Improved client functioning and well-being are among the principal

goals of mental health services, but achieving them is contingent on client
participation in treatment. When rates of client participation and discon-
tinuation differ across race or ethnicity, the most logical explanation seems to
be that the services inadequately accounted for cultural values, expectations,
and worldviews (e.g., failing to understand that a Japanese American client
might prefer directive counseling methods to open-ended exploration). The
field of multicultural psychology seeks to help practitioners meet diverse cli-
ents’ mental health needs through participation in treatment that is appro-
priately aligned with those needs (La Roche & Lustig, 2013).

Relevant Theory

Clients’ therapy experiences vary not only according to their cir-

cumstances but also in terms of their expectations, so social psychological

96       foundations of multicultural psychology

theories regarding expectations provide a framework to conceptualize cli-
ent participation (e.g., Swift & Callahan, 2011). At the most basic level,
unmet expectations lead to discontinuation of treatment, but the associ-
ations are more complex: Expectations affect every aspect of treatment,
including the roles that clients take in therapy, their interpretations of the
recommendations provided in therapy, and the perceived racial, ethnic, and
gender compatibility between the client and therapist. Client expectations
are related to their attitudes and subjective norms, which are components
of the theory of reasoned action (Ajzen & Fishbein, 1977). Seen through
that theoretical lens, client participation in treatment is a function of both
beliefs/attitudes about therapy and interpretations about what constitutes
normative behavior. So a client whose culture values self-sufficiency over
interactivity may be suspicious of treatments that magnify misalignments
with the therapist.

Narrative Review of Previous Research

In practice, client participation in therapy can be conceptualized several

ways. The broadest categories are client treatment attendance (number of
sessions completed, percentage of appointments kept, and number of days in
treatment) and client treatment completion/attrition (also referred to as reten-
tion, dropout, or premature termination/discontinuation; Alegría et al., 2008).
The latter variable, completion/attrition, can be operationalized in at least
five ways: (a) completion of a prespecified number or percentage of sessions,
(b) completion of a treatment protocol or manual, (c) unilateral termination
(client missing a scheduled appointment without subsequently rescheduling
or contacting the therapist), (d) discontinuation prior to clinically significant
or reliable improvement in presenting symptoms, and (e) therapist judgment
of whether discontinuance of therapy was warranted (see Swift & Greenberg,
2012). The latter method, therapist judgment, accounts for contexts known
to the therapist and not otherwise measurable, but this method results in the
largest rates of attrition compared with the others (Swift & Greenberg, 2012),
although it does not accurately correspond with client symptom reduction
(Swift, Callahan, & Levine, 2009). This inaccuracy may result from thera-
pists’ self-serving biases that can prevent them from accurately evaluating cli-
ent termination (Murdock, Edwards, & Murdock, 2010). Of the five methods,
the fourth (symptom reduction trajectory), which would logically be the most
clinically useful indicator, has accumulated research support (e.g., Swift et al.,
2009), but few research studies evaluate client mental health symptoms fre-
quently enough to calculate symptom trajectories. Overall, the research lit-
erature is inconsistent in how completion/attrition is operationalized (Kazdin,
1996; Swift & Greenberg, 2012).

participation of clients of color      97

 Irrespective of how client treatment completion/attrition is operation-
alized, research in this area deserves high priority status among clinicians and
researchers. Clients frequently discontinue treatment prior to adequate symp-
tom reduction and/or without therapist foreknowledge. Widely cited esti-
mates of premature or unilateral termination have ranged from 20% to 60%,
with a previous meta-analysis estimating the number at 47% (Wierzbicki &
Pekarik, 1993) and a recent meta-analysis specific to adult clients yielding an
estimate of 20% (Swift & Greenberg, 2012), although client attrition may
be higher in clinics not explicitly tracking client progress. Unfortunately, the
modal number of therapy sessions most commonly reported in the literature
is one (e.g., Connolly Gibbons et al., 2011).
The fact that so many clients discontinue treatment is extremely prob-
lematic. Clients who discontinue treatment prior to symptom improvement
may experience minimal or no benefit, which occurs much more often than
therapists realize (Lambert, 2010). Some experience symptom exacerbation.
Client attrition also affects the quality of care provided to others in clinical set-
tings; missed appointments reduce the availability of services for other clients
and increase the length of waiting lists, sometimes resulting in negative client
perceptions about the agency (Barrett, Chua, Crits-Christoph, Gibbons, &
Thompson, 2008). As financial losses to the agency may result, “the problem
of attrition is particularly acute in agencies that provide mental health services
to those who are economically disadvantaged” (Barrett et al., 2008, p. 248).

Commonly Cited Reasons for Treatment Completion/Attrition

Some of the reasons clients of color choose to discontinue therapy prior to
adequate improvement are outside the control of client or therapist. However,
many impediments to clients’ active consistent engagement in treatment are
psychological and social and thus potentially amenable to change in the thera-
peutic relationship (e.g., reframing, empowering; see Alegría et al., 2008). Two
commonly referenced factors amenable to therapist intervention are (a) client
expectations about the nature and duration of treatment and (b) the therapeu-
tic alliance (Piselli, Halgin, & MacEwan, 2011; Reis & Brown, 1999). Clients
entering therapy may expect rapid improvement but have little prior experi-
ence using therapy techniques. When therapists clarify treatment expectations
with clients (e.g., Swift & Callahan, 2011) and establish strong relationships
with their clients (Sharf, Primavera, & Diener, 2010), particularly culturally
congruent relationships (La Roche & Lustig, 2013), clients are much less likely
to discontinue treatment before symptom improvement.
Although client expectations and the client–therapist alliance have
received much attention in the literature, previous research has identified
multiple risk factors for premature discontinuation of treatment in the gen-
eral population (e.g., Kazdin, 1996; Reis & Brown, 1999). Exhibit 5.1 lists the

98       foundations of multicultural psychology

 EXHIBIT 5.1
General Factors Associated With Client Premature/Unilateral
Discontinuation of Mental Health Treatment

Client factors supported by research findings

• Perceived need for services and level of psychological functioning and symptom
severity, including comorbidity and physical illness
• Outcome and role expectations, including the length and purpose of therapy
• Dissatisfaction with therapist; strength of therapeutic alliance
• Socioeconomic factors, including level of education, time constraints, and
residential stability
• Parent or caregiver support and influence for child and adolescent clients
• Age (older clients being slightly more likely than younger clients to remain in
therapy)
• Life circumstances (e.g., crises, family needs, workplace demands, documentation
status)

Other factors supported by research findings

• Setting (settings being equivalent with exception of university-based clinics, which
have higher attrition rates)
• Time limits (lower attrition in time-constrained therapy)
• Operationalization of completion/attrition (higher attrition observed when based on
therapist judgment alone)
• Experience level of therapist (trainees having higher rates of client attrition than
experienced therapists)
• Availability of equivalent services elsewhere; physical distance to services
• Changes in treatment procedures during treatment
• Session availability; length of waiting list

risk factors most commonly mentioned in the literature and indicates the fac-
tors that are likely to interact. For instance, symptom severity and treatment
expectations are associated with socioeconomic factors, which are associated
with physical distance to viable treatment and the quality of treatment, and
so on. Even when a single factor may explain why a particular client discon-
tinued early, consideration of multiple risk factors is preferable to focusing on
single factors when looking at aggregate data. The inadequacy of evaluating
a single risk factor in isolation from others becomes readily apparent when
considering the research on differences in treatment participation across race
or ethnicity.

Racial and Ethnic Discrepancies

Racial and ethnic discrepancies in client treatment attendance and treat-
ment completion/attrition have been widely reported in the research literature.
A 1993 meta-analysis (Wierzbicki & Pekarik) concluded that ethnic minor-
ity clients were more likely to discontinue treatment than White clients, and

participation of clients of color      99

subsequent narrative literature reviews reached the same conclusion (Barrett
et al., 2008; Reis & Brown, 1999). The professional literature commonly asserts
racial and ethnic discrepancies, often referring to them in passing as if they
are common knowledge.
However, a recent meta-analysis found small racial and ethnic differ-
ences in the rate of premature discontinuation across 11 studies reporting
data by race (not statistically significant at p = .06) and found no correlation
between the percentage of clients of color and the rate of premature cli-
ent discontinuation in 243 studies (Swift & Greenberg, 2012). These recent
findings suggest the possibility that racial and ethnic discrepancies in client
treatment completion/attendance have decreased over time. Such a decrease
may suggest improved multicultural competence among mental health pro-
fessionals, increased exposure to and understanding of mental health treat-
ments among clients of color, or influence from a factor not yet identified.
Unfortunately, previous research has failed to identify specific predictors of
racial and ethnic discrepancies in client treatment participation. Authors
commonly list race and ethnicity among several factors that have been found
to be associated with participation (Table 5.1), but few authors address the
interactions of race and ethnicity with those other factors—and even fewer
examine underlying causes for the discrepancies across race and ethnicity.
Sometimes cultural explanations are given for racial and ethnic differ-
ences in treatment participation. For instance, clients from some cultures prefer
to avoid confrontation or assume passive roles with clinicians (Alegría et al.,
2008) and may discontinue treatment rather than openly raise concerns. Lack
of familiarity with mental health treatments or communication difficulties may
account for lower participation among some clients of color, particularly recent
immigrants. Other authors emphasize that client participation depends on the
therapist’s multicultural competence (e.g., Langer, 1999). Research confirms
that client unilateral termination varies across therapist characteristics, with
some therapists having greater retention of clients of color than others (Owen,
Imel, Adelson, & Rodolfa, 2012). Thus, therapists should facilitate attunement
and alliance across cultural differences (Gaztambide, 2012).
Although cultural explanations for client completion/attrition are cer-
tainly reasonable, research has yet to confirm the associated assumptions.
Evidence of potential causes for racial discrepancies and research designs that
establish causality are presently limited. The vast majority of research has
been either descriptive (means and percentages) or correlational:
Research on attrition from treatment is frequently less well executed than
would be desirable. The literature reflects an extensive yet mainly atheo-
retical search for correlates of attrition, which thus far has provided little
firm direction for understanding why clients leave treatment. (P. M. Harris,
1998, p. 302)

100       foundations of multicultural psychology

 Our literature search located no studies of racial and ethnic discrepan-
cies that involved causal modeling or even such obviously appropriate statisti-
cal methods as survival analysis (Corning & Malofeeva, 2004). We did locate
a handful of studies involving path analysis or fixed/random effects models, but
in our review we found that most researchers simply collected demographic
data at intake and then evaluated subsequent differences in attendance or
completion/attrition, without accounting for treatment processes, symptom
levels, cultural values and expectations, or other likely explanations for racial
and ethnic differences.
Despite the methodological limitations of the available research on
racial and ethnic discrepancies in client treatment participation, extensive
research has been completed using descriptive and correlational designs. This
evidence can and should be synthesized to inform future inquiry. As far as
we can determine, our summary in this chapter of the extant findings is the
first specific to racial and ethnic differences; previous meta-analyses have
treated a contrast of White/European Americans with people of color as one
of several variables, without specifically searching for manuscripts contrast-
ing findings across race or ethnicity (Swift & Greenberg, 2012; Wierzbicki
& Pekarik, 1993). We undertook this meta-analysis to generate estimates of
racial and ethnic differences in client treatment attendance and client treat-
ment completion/attrition and to estimate the degree to which study and
participant characteristics moderated the findings. A summary of the findings
can be found in the next section (Discussion and Interpretation of Findings).

QUANTITATIVE SYNTHESIS OF RESEARCH DATA

We conducted a meta-analysis of the available research literature. To

evaluate individuals’ degree of participation (attendance or completion/
attrition or both) in mental health services across race or ethnicity, we had
to calculate an effect size that was conceptually consistent across studies by
comparison with a single racial group. Because almost all studies we located
reported data for White/European American clients, we selected that group
as the contrast.1 We used Cohen’s d as the effect size metric, with positive
values indicating greater participation rates (higher attendance or comple-
tion, lower attrition) than the contrast group and negative values indicating

1We oppose the assumption that Whites are a normative reference group; the experiences of other racial
and ethnic groups do not need to be compared with Whites to be understood or valued. Our decision to
contrast the participation rates of people of color with Whites in this chapter was based on our desire to
include as many studies in our analyses as possible, and no other racial or ethnic group was represented
as consistently in the literature. Moreover, having a consistent contrast group allowed for contrasts
between every group.

participation of clients of color      101

lower participation rates than those of the contrast group. General methods
of the meta-analysis are reported in the Appendix to this book.

Description of the Existing Research Literature

We located 67 studies that compared client participation (attendance

and/or completion/attrition) in a mental health treatment by race. As may
be seen in Table 5.1, the majority of studies involved archival data, retrieving
information regarding client attendance or completion/attrition, and then
comparing those data across racial groups. With so many studies involving
archival data, the number of participants tended to be large (overall median
of 423 clients per study with 275,037 total clients). Most studies were con-
ducted in outpatient treatment facilities (e.g., community mental health
clinics), but some involved treatment programs for youth or college counsel-
ing centers. Only five evaluated inpatient/residential treatments.
Adolescents and senior adult clients were underrepresented. African
Americans tended to be evaluated most frequently, but Asian Americans and
Hispanic/Latino(a) Americans were also frequently studied. Only three studies
reported data specific to Native American Indian clients, and 14 studies col-
lapsed participants of color into a catchall “ethnic minority” group.
Research studies operationalized client participation both as treatment
attendance (number of sessions attended, return to treatment after an ini-
tial session, and number of days in treatment) and as treatment completion/
attrition (finishing a prescribed protocol or number of sessions or being judged
by a therapist as either having completed necessary treatment or having pre-
maturely discontinued treatment), with 12 studies using both methods of evalu-
ation. The effect sizes represent simple comparisons of these data across race,
with no causality inferred.
The 48 effect sizes based on client attendance data averaged d = -0.06,
whereas the 31 effect sizes based on client completion/attrition data averaged
d = -0.15; this difference reached statistical significance (Q = 4.3, p < .05).
Evaluation of client participation based on measures of treatment completion/
attrition consistently resulted in greater discrepancies with White/European
American clients than that based on measures of client attendance. We there-
fore report analyses of data for client attendance and for client completion/
attrition separately.

Findings Across Measures of Client Attendance

Differences in client attendance (e.g., number of sessions completed)

by race were examined through a multivariate meta-analysis, in which effect
sizes specific to each racial group were examined simultaneously (accounting

102       foundations of multicultural psychology

 TABLE 5.1
Characteristics of 67 Studies of the Participation of Clients
in Mental Health Services Across Race
No. of
Characteristic M studies (k) %
Year of report 1993
Before 1980 7 10
1980–1989 17 26
1990–1999 25 37
2000–2008 18 27
Publication status
Published 41 61
Unpublished dissertations 26 39
Research design
Archival 43 64
Comparison groups 16 24
Longitudinal 8 12
Clinical setting
Outpatient treatment 38 57
Day treatment program 7 10
College counseling center 7 10
Inpatient/residential treatment 5 8
Mixed settings 10 15
Sample size 4,105
<50 2 3
50–99 7 11
100–199 13 19
200–399 9 14
400–999 13 19
>1,000 23 34
Age of participantsa 30.3
Children (<13 years) 10 17
Adolescents (13–18 years) 4 7
Young adults (19–29 years) 6 11
Middle-aged adults (30–55 years) 36 63
Senior adults (>56 years) 1 2
Gender of participants (% female) 48.5
Race of participantsb (%)
White/European American 53
African American 22
Asian American 10
Hispanic/Latino(a) American 14
Native American Indian 0
Other 1
Note. Not all variables sum to the total number of studies because of missing data.
aAverage age category of participants within studies, though not all participants within the study would

necessarily be in the category listed. bThe racial composition of participants across all studies was calculated
by multiplying the number of participants within studies by the percentage of participants from each racial
group and dividing that product by the total number of participants.

participation of clients of color      103

for the observed value of r = .00 for the results of different racial groups within
studies). The overall model reached statistical significance (Wald X2(4) =
43.2, p < .0001); the results by race are reported in Table 5.2. Asian American
clients had higher average attendance rates than all other groups, includ-
ing White/European Americans. African American clients had lower aver-
age attendance rates than all other groups. Although most of the differences
between groups reached statistical significance, they were all of very small
magnitude. Across all 48 studies examining client attendance, the random
effects weighted effect size was d = -.06 (SE = .011, 95% CI = [-.08, -.04],
p < .0001). The heterogeneity of the findings was moderate (I2 = 64.3, 95%
CI = [51, 74]; Q(47) = 131.6, p < .00001), meaning that the results tended to
be inconsistent across studies (see Figure 5.1).

Effects of Study and Participant Characteristics

Studies involved a wide variety of participants and procedures, so we
evaluated differences across participant age and gender (percentage of females
in the study), year of study publication, estimated age cohort (year of study
publication minus average age of participants), and racial diversity of the set-
ting (percentage of people of color in the study). None of these correlations
reached statistical significance. Thus, the variables coded were unrelated to

TABLE 5.2
Weighted Mean Effect Sizes (Cohen’s d) Across Participant Race
Group k d SE 95 % CI
Client attendance discrepancies
African Americans 33 –.11a,b .02 [-.15, -.07]
Asian Americans 14 .08b,c .03 [.02, .13]
Hispanic/Latino(a) Americans 30 –.06a .01 [-.09, -.04]
Unspecified/other ethnic minorities 7 –.07a,c .05 [-.17, .03]
All groups combined 48 –.06a,c .01 [-.08, -.04]
Client completion/attrition discrepancies
African Americans 17 –.23a,b .05 [-.33, -.13]
Asian Americans 8 –.13c .09 [-.31, .05]
Hispanic/Latino(a) Americans 15 –.04c .04 [-.12, .05]
Unspecified/other ethnic minorities 12 –.15b .07 [-.29, -.02]
All groups combined 31 –.15b .04 [-.24, -.06]
Note. Effect sizes were derived from contrasts of participation data with White/European Americans;
nevertheless, all groups can be contrasted within the table. CI = confidence interval; d = random effects
weighted effect size; k = number of studies; SE = standard error.
aStatistically significantly different from Asian Americans (p ≤ .05). bStatistically significantly different from

Hispanic/Latino(a) Americans (p ≤ .05). cStatistically significantly different from African Americans (p ≤ .05).

104       foundations of multicultural psychology

Figure 5.1. Contour-enhanced funnel plot of effect sizes (Cohen’s d) by standard
error for 48 studies of client treatment attendance across racial groups. This graph
shows the distribution of effect sizes as a function of the number of participants in
the study (operationalized as standard error). The studies at the top of the graph are
those with many participants (and small standard error values), which studies yielded
results that clustered very slightly to the left of zero (evenly distributed around the
overall average, d = -0.06). By contrast, inconsistent results were obtained when
studies involved few participants, as seen by the scattered data points toward the
middle and bottom of the graph. This kind of pyramid-shaped distribution is expected
in meta-analyses.

racial discrepancies in client attendance, although the small number of studies

restricted the statistical power of these analyses, which diminished the likeli-
hood of obtaining statistically significant results.
Given that 14 of the 48 effect sizes were adjusted for potential con-
founds, including client socioeconomic status (in 10 of the 48 studies) and
client mental health status or severity of symptoms (in 13 of the 48 studies),
it was important to ascertain whether the results differed when they were
adjusted versus unadjusted. Nevertheless, the average effect size values were
essentially the same (d = -0.05 for adjusted values and d = -0.07 for unad-
justed values, p > .05). Thus, the results across studies did not consistently
differ with the inclusion of covariates.

participation of clients of color      105

Influence of Publication Bias
When research manuscripts with nonsignificant results remain unpub-
lished, they are less likely to be located in a literature search and thus bias the
findings of a meta-analysis. Publication bias appeared to have influenced the
overall findings reported in the previous section. Although the data in Figure 5.1
were fairly symmetric, Egger’s regression test (an estimate of asymmetry of effect
sizes) was statistically significant (p < .001). In addition, one statistical method
(Duval & Tweedie, 2000) identified six “missing” studies in the distribution.
However, when those hypothetical data were accounted for, the resulting effect
size was d = -0.05 (95% CI = [-0.08, -.03]), essentially the same as the overall
value of d = -0.06 reported earlier (Table 5.2). Thus, although publication bias
may have occurred, its effect on the results appears to have been minimal.

Findings Across Measures of Client Treatment Completion/Attrition

We next examined racial differences in client treatment completion and

attrition patterns (e.g., either finishing a prescribed protocol or prematurely
discontinuing treatment) through multivariate meta-analysis, with effect
sizes specific to each racial and ethnic group examined simultaneously. The
overall model reached statistical significance (Wald X2(4) = 27.8, p < .0001);
results by race are reported in Table 5.1. African American clients had the
lowest levels of treatment completion (highest levels of attrition), followed
by unspecified/other ethnic minorities and Asian American clients. Hispanic/
Latino(a) American clients had almost the same average completion/attrition
rates as White/European American clients.
Across all 31 studies examining client completion and attrition pat-
terns, the random effects weighted effect size was d = -.15 (SE = .044, 95%
CI = [-.24, -.06], p < .0001). The heterogeneity of the findings was very
large (I2 = 85.4, 95% CI = [80, 89]; Q(30) = 205.8, p < .0001), meaning that
the results were highly inconsistent across studies (see Figure 5.2).

Influence of Study and Participant Characteristics

As in previous analyses, we correlated effect sizes with participant age
and gender composition, year of study publication, estimated age cohort, and
racial diversity of setting. The random effects weighted correlations of these
variables with study effect sizes did not reach statistical significance. Thus,
the racial discrepancies in treatment completion and attrition patterns were
unrelated to the characteristics analyzed, but the small number of studies
restricted the statistical power of the analyses.
The minimal differences between the effect sizes of studies using sta-
tistical controls (d = -0.17) versus not using them (d = -0.14) did not reach

106       foundations of multicultural psychology

Figure 5.2. Contour-enhanced funnel plot of effect sizes (Cohen’s d) by standard
error for 31 studies of client treatment completion/attrition across racial groups. In this
case the overall average is also slightly to the left of 0, but the results are widely scat-
tered, with high variability even among studies with a larger number of participants
(and lower standard errors) depicted at the top of the graph. This inconsistent
distribution limits the interpretability of the overall average.

statistical significance in a random effects weighted analysis of variance. Effect

sizes remained of similar magnitude whether or not the data had been statisti-
cally adjusted for potential confounds.

Effects of Publication Bias

The data in Figure 5.2 were asymmetric, shifted toward negative effect
size values (indicating less participation of clients of color). This unbalanced
distribution suggested possible missing studies with data indicating greater
participation of clients of color. Egger’s regression test, an estimate of asym-
metry of effect sizes, was statistically significant (p < .001), but one statistical
method (Duval & Tweedie, 2000) failed to identify any “missing” studies in
the distribution. This contradictory information seemed to have occurred
because the study with the smallest standard error (at the top of Figure 5.2)
fell on the right side of the distribution, the side where missing data would
have been expected. To evaluate the possible impact on the findings if a few

participation of clients of color      107

studies were actually missing, we temporarily inserted six hypothetical values
where gaps appeared on the right side of the distribution. This procedure
resulted in an overall mean of d = -0.09, a value that was small but still statis-
tically different from 0 (p < .01). Thus, although publication bias was possible
in the data, the results appear to have been minimally affected.

DISCUSSION AND INTERPRETATION OF FINDINGS

Notable differences were found between estimates based on metrics of

attendance (number of sessions completed) and metrics based on treatment
completion/attrition (as evaluated by the therapist or by a set number or
percentage of sessions); thus, separate analyses were conducted with those
distinct classes of data. The differences observed between racial and ethnic
groups in these analyses ranged from almost nonexistent to small, according
to commonly used guidelines for interpreting Cohen’s d (J. Cohen, 1988).
On average, racial and ethnic groups participated similarly in treatment.
However, results may be enhanced by considering the meaning of the effect
size (Cohen’s d) in real-world terms: number of sessions completed (for studies
reporting attendance data) and percentages of clients completing treatment
(for studies reporting completion/attrition data).
To conceptualize the meaning of Cohen’s d in terms of therapy atten-
dance, we suggest a hypothetical scenario in which an average of 15 therapy
sessions are expected, with substantial variability in the actual number of ses-
sions completed (SD = 9). Given those parameters and the Cohen’s d values
for attendance data found in Table 5.2, African American clients complete on
average 14 sessions, one less than expected (transforming d = -0.11). Using
those same parameters and attendance data transformed from Cohen’s d val-
ues in Table 5.2, Asian Americans would attend an average of 15.7 sessions,
and Hispanic/Latino(a) Americans would attend an average of 14.5 sessions.
Typically, 14 or more total sessions are required for the majority of clients to
demonstrate clinically significant change, and the number of total sessions is
very weakly correlated with client outcome (Lambert, 2007), so missing on
average of one session of therapy would not likely have a meaningful impact
on therapy effectiveness. Average racial and ethnic differences in therapy
attendance appear to have minimal real-world consequences unless very few
sessions are provided.
The effect size Cohen’s d takes on a different meaning when trans-
formed into the metric of percentages of clients completing treatment. Across
41 studies that reported absolute completion rates, about 61% of all clients
completed treatment. Using that baseline estimate of 61%, the racial dis-
crepancy corresponding with J. Cohen’s (1988) description of a “small” effect

108       foundations of multicultural psychology

size (d = 0.20) would be about 10%, which would mean that one out of every
10 clients discontinued therapy on the basis of reasons apparently related to
race and ethnicity. To reach the value for a “medium” effect size (d = 0.50), the
racial discrepancy would have to be about 24%: one out of every four clients
discontinuing therapy on the basis of reasons apparently related to race. In
the real world, a racial discrepancy of 24% would be unconscionable. Mental
health professions would come under scathing attack and legal action—well
deserved—if racial and ethnic inequities were that obvious. Even the “small”
discrepancy of 10% seems truly problematic when multiplied by the number
of clients seeking mental health services. Thus, interpretations of the data in
terms of the real-world metric of percentages of treatment completion appear
preferable to the standard guidelines given for values of Cohen’s d.
Using the observed average baseline participation rate of 61%, trans-
formation of the treatment “criterion and completion” data for African
Americans in Table 5.2 (d = -0.23) indicates that African American clients
remained in treatment on average 11.5% less often than White/European
American clients. Using those estimates, for every five African American cli-
ents who completed therapy, an additional individual should have completed
therapy if race had not been a factor in treatment participation. Calculated
differently, for every nine African American clients who walked into a clinic,
one would have failed to complete treatment for reasons associated with race.
Asian Americans, who we found in previous analyses to be more likely to
attend therapy than White/European Americans, are nevertheless 6.5% less
likely to complete treatment (transformation of d = -0.13 in Table 5.2). One
out of every 15 Asian Americans who walk into a clinic would fail to com-
plete treatment for reasons related to race. Hispanic/Latino(a) American cli-
ents completed treatment 2% less often than White/European Americans but
9.5% more often than African Americans. For every 50 Hispanic/Latino(a)
American clients who walked into a clinic, one would have failed to com-
plete treatment for reasons related to race or ethnicity. Whereas one out of
50 clients is extremely small in real-world terms, one out of nine clients is a
consequential number, a true racial discrepancy deserving serious attention.
By comparison, one out of 15 clients seems much less egregious but certainly
far from optimal.
Nonetheless, the statistical estimates of racial and ethnic discrepancies
in client participation may be mildly inflated due to publication bias, which
occurs when nonsignificant or counterintuitive research findings remain
unpublished or otherwise inaccessible to researchers conducting meta-analyses
(Rosenthal, 1991). Thus, the effect size values just interpreted (and reported
in Table 5.2) may represent liberal estimates of racial discrepancies.
The analyses also indicated that averaged participant characteristics of
age, age cohort, and gender were unrelated to the degree of racial and ethnic

participation of clients of color      109

discrepancies in client participation. And the results across studies were
essentially the same whether or not participants’ characteristics (commonly
including those variables but also including mental health functioning and
socioeconomic status) were statistically controlled. These findings suggest
that commonly measured demographic variables are on average unrelated
to racial and ethnic discrepancies in client attendance or client treatment
completion/attrition.

Recommendations for Future Research

Researchers must consistently evaluate mental health treatments to

ensure that clients are receiving the best services available. On the basis of
research specific to client participation in therapy, the results of this meta-
analysis strongly suggest that the dependent variables of client (a) attendance
and (b) completion/attrition patterns must be kept completely distinct.
“Attendance and retention differ both conceptually and empirically” (Alegría
et al., 2008, p. 248). Unless very few sessions are planned, measures of client
attendance are less consequential than measures of client treatment comple-
tion. Further studies of client attendance rates may not contribute much to
the body of literature already accumulated; future research should focus on
treatment completion/attrition and the other recommendations listed in
Exhibit 5.2.
To identify why African American, Asian American, and unspecified
clients of color are at risk of therapy attrition, scholars should conduct both

EXHIBIT 5.2
Recommendations for Researchers for Improving the Study
of Racial Discrepancies in Treatment Completion
• Statistically control for preexisting symptom severity and length of treatment.
• Use optimal statistical methods: survival analysis, including proportional hazards
regression.
• Use proximal cultural characteristics, not race only, in models that account for
multiple likely predictors of treatment completion (see Exhibit 5.1).
• Optimally evaluate client termination in terms of symptom reduction or reduction
trajectory. Examine clinical significance of outcomes.
• Contact clients who discontinued treatment to ascertain their reasons, rather than
assume that discontinuation was unwarranted; also deepen qualitative inquiry into
client experiences of termination.
• Report absolute and adjusted rates of treatment completion/attrition.
• Evaluate possible differences across treatment settings that are currently under-
represented in the literature, particularly inpatient and residential (controlling for
whether treatment was voluntary).
• Evaluate the alignment of the treatment provided with client expectations and
worldviews.

110       foundations of multicultural psychology

qualitative inquiry and psychotherapy process research. Scholars should also
make efforts to locate former clients who discontinued therapy and learn about
those clients’ experiences in therapy and their rationale for discontinuing it.
Too little of this kind of scholarship presently exists (Knox et al., 2011).
Qualitative and quantitative differences in the reasons for client discon-
tinuation of treatment can inform refinement of dependent variables, which
include operationalization of treatment completion/attrition. Therapists’
evaluations of client dispositions may be problematic (e.g., Murdock et al.,
2010; Swift et al., 2009). Other commonly used methods, such as predetermined
cutoff values for number of sessions completed and dichotomous conceptual-
izations of treatment completion, fail to account for underlying reasons for
discontinuation. For instance, in one study “clients who were better adjusted
initially tended to withdraw from treatment; clients receiving particularly
severe diagnoses tended either to finish or to drop out” (Snowden, Storey, &
Clancy, 1989, p. 116). In another study, clients who successfully completed
treatment averaged 40 sessions, whereas those who discontinued “early” in
treatment averaged 10 sessions, and those who discontinued “late” in treat-
ment averaged 47 sessions (Talebi, 2006). If a client attended 47 sessions and
discontinued because her or his symptoms remained unimproved, certainly
researchers should not categorize that client alongside another client who
failed to schedule a return appointment following intake. We concur with
Kazdin’s observation (1996):
The variables that are selected for investigation to predict dropping out
often appear to be “variables of convenience,” that is, measures that are
readily available on clinic intake forms. . . . It is [more] useful to concep-
tualize the process of dropping out and then to place these (and other
variables) in the context of that conceptualization. (p. 138)
We believe that the evaluation of client completion and attrition patterns
should include clients’ symptom trajectories (Swift et al., 2009) and underly-
ing motives (Knox et al., 2011). Using precise dependent variables informed
by clients’ reasons for treatment discontinuation and their rate of symptom
improvement should enhance the interpretability of future research findings.
Finally, scholars conducting future research on client participation
should identify and control for probable confounds, not merely demographic
variables. Only 19% of the 67 studies in the meta-analysis statistically controlled
for client socioeconomic variables, yet economic constraints undoubtedly influ-
ence treatment participation rates. Furthermore, the most likely confound,
client symptom severity, was controlled in only 24% of the studies. Although
our analyses with this limited number of studies did not find evidence that
change in statistical controls altered averaged results across studies, results
did change within studies when statistical controls were used (e.g., Snowden

participation of clients of color      111

et al., 1989). Logically, it seems essential to account for likely confounds
when attempting to examine differences across race and ethnicity: for exam-
ple, to account for the overall length of time in treatment, not merely mea-
sure the percentage of expected sessions completed. How can researchers
speak with any confidence about racial and ethnic differences if we fail to
systematically evaluate likely confounds? Future research must control for
preexisting symptom severity and should optimally use statistical methods
such as survival analysis (Corning & Malofeeva, 2004) and proportional
hazards regression (M. H. Katz & Hauck, 1993), rather than simply report-
ing percentages and means. “The phenomenon of attrition from treatment
requires more rigorous attention” (P. M. Harris, 1998, p. 293) and more
rigorous methodology to distinguish what differences are attributable to race
and ethnicity.

Suggestions for Practitioners

Previous research has shown that therapists can facilitate completion

of therapy for clients of color by implementing culturally competent prac-
tices (e.g., Alegría et al., 2008). The multicultural psychology literature
contains a myriad of suggestions for appropriately engaging diverse clients
(e.g., Gaztambide, 2012; Ponterotto, Casas, Suzuki, & Alexander, 2010;
T. B. Smith, 2010). Practitioners can also keep in mind that several macro-
level factors influence client continuation in treatment. For instance, most
of the factors thought to adversely affect mental health service utilization
(see Chapter 4, this volume, Exhibit 4.1) are likely to adversely affect client
continuation in treatment. If a clinic is located in a suburban area consisting
of predominantly White/European Americans, clients of color might find
it more difficult to repeatedly travel to the clinic. Although the publicly
available data do not identify which specific factors are most influential, all
levels of explanation for discrepancies (Exhibit 5.1) can be considered by
practitioners to identify the particular factors most likely to affect racial and
ethnic discrepancies at their particular clinic and to change procedures and
practices to better facilitate client engagement in and completion of needed
mental health services.
In adopting practices that will increase client engagement in therapy (see
Exhibit 5.3), practitioners must also take a broad perspective on the reasons for
client premature discontinuation. Our meta-analysis indicated that an average
of 61% of clients completed therapy across 41 of 67 studies reporting those
data. The fact that so many clients prematurely discontinue treatment (Swift
& Greenberg, 2012; Wierzbicki & Pekarik, 1993) is the proverbial “elephant
in the room” that mental health professionals typically minimize or avoid. The
discontinuance average of four out of 10 clients shown by our analyses is far

112       foundations of multicultural psychology

 EXHIBIT 5.3
General Recommendations for Practitioners
in Increasing Client Treatment Completion
• Help clients clearly/precisely articulate their reasons for desiring treatment.
• Establish agreement regarding the focus of treatment; verify that the content and
process of therapy align with clients’ values and goals, including readiness for
change.
• Empower clients by facilitating opportunities to ask questions, seek information,
and involve themselves in treatment decisions.
• Strengthen the therapeutic alliance, and take immediate corrective action when
client–therapist misalignments occur.
• Explicitly solicit client expectations regarding the duration of treatment, then align
treatment to match client expectations and/or provide clients with specific reasons
why treatment of a different (longer) duration would be beneficial, and finally revisit
client expectations.
• Openly discuss the pros and cons of therapy from the client’s perspective.
• Identify and correct misalignments in understanding.
• Involve client support networks/family in treatment to the extent helpful.
• Facilitate client imagery of systematic improvement over time.
• Limit treatment to an attainable duration (rather than leave treatment open-ended).

too many failures, even before accounting for racial and ethnic discrepancies.
Eliminating racial and ethnic discrepancies in treatment completion should
occur alongside efforts to raise overall percentages of clients who complete
treatment. Thus, the findings from this meta-analysis align with ample extant
data that “suggest the need for providers and service managers to give high
priority to keeping patients in treatment longer” (Lambert, 2007, p. 4).
Of the common recommendations for increasing client engagement in
and completion of therapy (Exhibit 5.3), research strongly supports working
with client expectations about the nature and duration of treatment and the
roles of the client in therapy (e.g., Swift & Callahan, 2011). Although some
authors describe didactic methods of client induction and orientation that
implicitly maintain power imbalances favoring the clinician, a method spe-
cifically designed to empower multicultural clients focuses on activation—
facilitating clients in formulating questions, seeking information, and managing
their own mental health care decisions and processes (Alegría et al., 2008).
This activation approach also strengthens the quality of client–therapist inter-
action, a critical factor in client completion of treatment (Sharf et al., 2010).

CONCLUSION

In our perusal of the relevant research literature, we identified two dis-

tinct qualitative tones taken by authors when referring to racial and ethnic
discrepancies in mental health treatment participation. Some authors openly

participation of clients of color      113

lamented the discrepancies, voicing alarm that current mental health prac-
tices are extremely problematic for clients of color. Many others referred to
the racial and ethnic discrepancies in passing, as one of several variables
associated with treatment participation, without considering under­lying
dynamics—as if nothing could be done to rectify the situation. Neither
of these tones accurately captures the available data. Our analyses of the
research data raise concern about racial and ethnic differences, but we
found those differences much less severe than we had supposed. Treatment
attendance differs minimally across race and ethnicity, but on average treat-
ment completion and attrition patterns and differences have adverse effects
on African American clients, with a smaller but still notable impact on
Asian American clients. Our interpretation of the completion and attrition
data demonstrated that even small statistical differences have real-world
consequences.
We believe that it is now time for scholars and practitioners to rectify
these differences through identifying and addressing the underlying causes.
For instance, although most therapists oppose racist beliefs, therapists vary
in their effectiveness in cross-cultural therapy (Hayes et al., 2015). It is not
difficult to imagine that the behavior of one out of nine therapists might
influence the one out of nine African American clients (or the one out of
15 Asian American clients) who discontinue therapy for reasons apparently
related to race. This particular interpretation exceeds the bounds of our data;
we did not evaluate therapist racism, and although therapists account for
some of client treatment completion and attrition patterns (Owen et al.,
2012; Piselli et al., 2011), many other concomitant factors also contribute to
racial discrepancies (Exhibit 5.1).
Racial discrepancies in client treatment completion may occur rela-
tively infrequently, but the fact that racial discrepancies exist at all deserves
remediation, not resignation. If a client discontinues therapy for reasons
apparently related to her or his race or ethnicity, the profession should take
steps to remove those reasons.

114       foundations of multicultural psychology

 6
MATCHING CLIENTS WITH
THERAPISTS ON THE BASIS
OF RACE OR ETHNICITY:
A META-ANALYSIS OF CLIENTS’
LEVEL OF PARTICIPATION
IN TREATMENT

Mental illness occurs across all races and ethnicities, but not all races
and ethnicities receive equivalent mental health treatment. People of color
in North America may receive an inferior quality of care (Alegría et al.,
2008; Gone & Trimble, 2012; Shin et al., 2005; S. Sue, 1988), and they tend
to report greater dissatisfaction with mental health services than White/
European Americans (Garroutte, Kunovich, Jacobsen, & Goldberg, 2004;
S. Sue & Zane, 1987). To help improve the quality of care provided to people
of color, scholars have emphasized the need to adapt treatments culturally
and to establish cultural congruence between clients and therapists (Griner
& Smith, 2006; La Roche & Lustig, 2013; S. Sue, 1998). Although not the
same as adapting treatments, matching clients with therapists according

Racquel R. Cabral provided the foundational work for this chapter, including data collection and cleaning.
Derek Griner and Alberto Soto of Brigham Young University contributed to the writing of this chapter.
A complete list of references for the studies included in this chapter’s meta-analysis is posted online.
Readers can consult it at [Link]
[Link]
Foundations of Multicultural Psychology: Research to Inform Effective Practice, by T. B. Smith and J. E. Trimble
Copyright © 2016 by the American Psychological Association. All rights reserved.

115
to race or ethnicity has received consistent attention over the past several
decades as one way to foster cultural congruence (Presnell, Harris, & Scogin,
2012; S. Sue, Fujino, Hu, Takeuchi, & Zane, 1991).
This matching approach assumes that therapeutic outcomes may
improve because clients and therapists share similar worldviews and values
(Kelly & Strupp, 1992). In addition to a preference for a therapist who shares
their racial or ethnic background, some clients may experience a stronger
therapeutic alliance and greater levels of participation in treatment when
matched on race or ethnicity (S. Sue, 1977; S. Sue et al., 1991). Shared
race and ethnicity may also attenuate factors that often preclude clients of
color from seeking professional mental health services—for example, mis-
trust of cultural differences, unfavorable views of mental health profession-
als (Whaley, 2001), lack of culturally congruent services, and institutional
barriers (D. W. Sue et al., 2007; S. Sue, 2003). Cultural barriers may be so
prominent that some clients of color may feel a sense of relief when they see
a mental health professional who phenotypically appears to share their racial
or ethnic background.
Shared race or ethnicity, however, cannot be presumed to align with the
more precise, proximal variables of shared values and worldviews (Helms,
Jernigan, & Mascher, 2005). People are remarkably different within as well
as across cultures (see Trimble & Bhadra, 2013). For instance, Hispanic/
Latino(a) Americans include individuals from dozens of ethnic groups, but
even if working with only one delimited group, such as the Runakuna of
Peru, a wide variety of attributes and worldviews should still be expected.
More­over, perceived similarity based on visible attributes, such as race,
can mask actual differences in values and worldviews. So matching mental
health clients with therapists of the same race or ethnicity only accounts for
some aspects of human experience, albeit important ones. It should there-
fore come as no surprise that the effects of matching clients with therapists
of the same race or ethnicity are mixed and likely to vary on the basis of the
type of evaluation conducted (e.g., preferences, dropout, outcome; Cabral
& Smith, 2011; H. L. K. Coleman, Wampold, & Casali, 1995; Flaskerud &
Liu, 1990).
A question that has not yet been resolved in the research literature is
the degree to which clients participate in treatment (i.e., attendance and
completion/attrition) as a function of the racial or ethnic match with the
therapist. Research has found that clients prefer therapists of their own race
and perceive therapists of their own race somewhat more positively than
other therapists (Cabral & Smith, 2011). Therefore, clients may be more
likely to remain in treatment when they are matched with therapists of their
same race or ethnicity.

116       foundations of multicultural psychology

 REVIEW OF THE LITERATURE

Relevant Theory

The idea that client engagement in treatment could be improved by

matching clients and therapists according to race or ethnicity is loosely based
on social psychology. Oversimplified, several theories indicate that individu-
als tend to have greater affiliation with people perceived to be similar to
themselves (e.g., D. E. Byrne, 1971). Social scientists refer to greater prefer-
ence for similarity as homophily (McPherson, Smith-Lovin, & Cook, 2001),
which is frequently observed in behavioral tendencies. Our choices tend to
be homogenous. Race and ethnicity influence the strongest divides in our
personal environments, with age, religion, education, occupation, and gen-
der following in roughly that order (McPherson et al., 2001).
On the basis of such principles, clients and therapists of the same race
or ethnicity should experience a stronger working relationship than client–
therapist dyads with dissimilar backgrounds (e.g., Simons, Berkowitz, & Moyer,
1970). Nevertheless, such similarity could become problematic if differences are
overlooked. And differences in perspective can be beneficial when they facili-
tate insight, create dialogue, and promote change. The degree to which racial
or ethnic matching actually benefits clients, therefore, deserves investigation.

Narrative Review of Previous Research

Research on the effects of matching clients with therapists on the basis

of race or ethnicity has accumulated over many decades (Cabral & Smith,
2011; H. L. K. Coleman et al., 1995; Flaskerud & Liu, 1990; Shin et al., 2005).
This research has addressed a variety of relevant variables, including (a) client
preferences for therapist race or ethnicity, (b) client perceptions of therapists
across race or ethnicity, and (c) client outcomes as a function of therapist
race or ethnicity. Because negative biases can adversely affect people of color
and these experiences can be difficult to discuss (D. W. Sue, 2015; D. W. Sue
et al., 2007), having a therapist of one’s own race or ethnicity can decrease dis-
comfort. Previous research has suggested that Hispanic/Latino(a) Americans
and African Americans generally prefer that arrangement (Flicker, Waldron,
Turner, Brody, & Hops, 2008), although there is great variability in research
findings, and generalization of findings can be problematic (Dumas, Moreland,
Gitter, Pearl, & Nordstrom, 2008). Overall, when given the option, culturally
diverse individuals tend to have a moderately strong preference for thera-
pists from their own racial or ethnic background, with a recent meta-analysis
reporting a d = .63 across 52 studies (Cabral & Smith, 2011).

matching clients with therapists      117

 Client preferences for therapists of their own race or ethnicity may lead
to the assumption that clients could also perceive those therapists more posi-
tively than therapists from other cultural backgrounds, expecting them to
be more competent in addressing cultural issues (Knox, Burkard, Johnson,
Suzuki, & Ponterotto, 2003). Overall, research supports the notion that
clients view therapists of their own race or ethnicity somewhat more posi-
tively than those of other races, with a reported d = .32 across 81 studies
(Cabral & Smith, 2011).
Given the relevance of racial or ethnic matching to client preferences
and perceptions of therapists, to what degree does matching improve client
outcomes? Overall, research has found little evidence to support the hypoth-
esis that therapeutic outcomes improve as a result of racial or ethnic matching
(Maramba & Nagayama Hall, 2002; Presnell et al., 2012; Shin et al., 2005),
although in some instances improvements have been noted (Ruglass et al.,
2014). Across 53 studies, therapy outcomes were only slightly more positive
(d = .09) for clients who were matched with therapists on race or ethnicity
(Cabral & Smith, 2011), although the effect was more notable among African
American clients (d = .19). Ethnic or racial matching does not necessarily
improve client–therapist communication (Zane & Ku, 2014), and generally
client outcomes are not greatly affected by the race or ethnicity of the therapist.
The finding that client preferences and perceptions relevant to ther-
apist match do not necessarily translate into improved outcomes requires
explanation. One possible explanation is that clients with higher levels of
discomfort related to racial or ethnic differences may be less likely to initi-
ate and attend therapy if unmatched with a therapist of their own race or
ethnicity. In addition, clients may be more likely to drop out of treatment if
they feel misunderstood by a therapist of a different race or ethnicity. Thus,
the results of the client outcome research could be confounded by client
attrition: Clients who are uncomfortable with unmatched conditions could
be less likely to complete the research, such that research findings without
those participants would be biased in the direction of no effect. Examining
client levels of participation in therapy as a function of racial or ethnic match
is therefore essential (Cabral & Smith, 2011).
One meta-analysis specific to African American clients reported a very
minor effect of racial match on dropout and number of sessions attended
(Shin et al., 2005), but some research indicates that ethnic or racial match
decreases premature treatment dropout (Ibaraki & Hall, 2014). The field will
now benefit from investigation of differential rates of client participation
in therapy across other groups using updated research reports available in
the literature. The objectives of our meta-analytic review were to generate
estimates of differences in client participation in mental health treatments
across racial or ethnic match with therapists and to estimate the degree to

118       foundations of multicultural psychology

which study and participant characteristics moderate the overall findings. A
summary of the findings is presented in the subsequent section.

QUANTITATIVE SYNTHESIS OF RESEARCH DATA

We conducted a meta-analysis of the research literature in the United

States and Canada that evaluated individuals’ degree of participation (atten-
dance or completion/attrition) in mental health services as a function of
whether the client and therapist were of the same race or ethnicity. Statistical
estimates in the studies were converted to Cohen’s d. Positive values for effect
sizes indicated greater participation rates (higher attendance or completion,
lower attrition) among the clients who were matched with a therapist of
their same race or ethnicity, and negative values indicated lower participa-
tion rates for the matched clients. General methods of the meta-analysis are
described in the Appendix to this book.

Description of the Existing Research Literature

We located 53 studies that reported comparisons of client participation

(measured by session attendance and/or treatment completion/attrition) as
a function of client racial or ethnic match with the therapist. As shown in
Table 6.1, studies typically selected participants among those conveniently
available and rarely randomized clients to matched versus unmatched condi-
tions. Data tended to either be culled from clinic databases or gathered across
clients available in a clinic, with differences between matched and unmatched
conditions described rather than experimentally tested. Although only one
study involved senior adults, clients of a variety of ages were represented in the
literature. The fact that 56% of studies involved middle-aged adults reflected
the fact that data were most often collected in community mental health clin-
ics. Treatment involved outpatient mental health services of various types and
modalities.
Native American Indians and Alaska Natives were not examined in
any of the studies we found, but other racial groups were represented. Clients
of a single race (e.g., African American clients with either African American
therapists or therapists of another race) were examined in 24 studies, cli-
ents of multiple races with therapists of a single race (e.g., White/European
American therapists with clients of color contrasted with White/European
American clients) were involved with 15 studies, and a combination of thera-
pist and client races evaluated in a binary matrix, matched versus unmatched,
was the subject of 14 studies. Overall, the 53 studies contained data from
62,434 individual clients, with a median number of 250 clients per study.

matching clients with therapists      119

 TABLE 6.1
Characteristics of 53 Studies of the Participation of Clients in Mental
Health Services as a Function of Racial or Ethnic Match With the Therapist
No. of
Characteristic M studies (k) %
Year of report 1997
Before 1980 2 4
1980–1989 5 9
1990–1999 28 53
2000–2013 18 34
Publication status
Published 32 60
Unpublished dissertations 21 40
Research design
Archival 22 41
Comparison groups 30 57
Retrospective survey 1 2
Participant selection
Random 7 13
Convenience 46 87
Assignment to therapist
Random 6 11
Convenience 47 89
Treatment modality
Individual therapy 32 61
Group or family therapy 5 9
Mixed, individual and group/family 16 30
Sample size 1,178
<50 3 6
50–99 11 21
103–199 10 19
200–399 8 15
400–999 6 11
> 1,000 15 28
Participant agea 31.0
Children (< 13 years) 8 15
Adolescents (13–18 years) 6 11
Young adults (19–29 years) 6 11
Middle-aged adults (30–55 years) 30 56
Senior adults (> 56 years) 1 2
Participant gender (% female) 57.3
Participantb race (%)
African American 21
Asian American 29
Hispanic/Latino(a) American 19
Native American Indian 0
White/European American 30
Other 0
Note. Not all variables sum to the total number of studies because of missing data.
aAverage age category of participants in studies, although not all study participants would necessarily be

in the category listed. Three studies did not report information about client age. bThe racial composition of
participants across all studies was calculated by multiplying the number of participants in studies by the
percentage of participants from each racial group and dividing that product by the total number of participants.

120       foundations of multicultural psychology

 Research studies operationalized client participation both as (a) treat-
ment attendance (number of sessions attended, return for additional treatment
after an initial session, and number of days in treatment) and (b) treatment
completion/attrition (completion of a proscribed protocol or certain number
of sessions or judged by a therapist as having completed necessary treatment
vs. premature treatment discontinuation). The effect sizes represent simple
comparisons of these data across conditions of client–therapist match based
on race or ethnicity.

Research Findings Across Racial Groups

Racial and ethnic differences in client attendance were first examined

through a multivariate meta-analysis in which effect sizes specific to each
racial or ethnic group were examined simultaneously (accounting for the
observed value of r = .07 for within-study correlation of effect sizes). The
overall model reached statistical significance, Wald X2 (5) = 85.7, p < .0001,
with the results by race as reported in Table 6.2. Asian American clients were
moderately more likely to continue participating in mental health treatment
when they were matched with a therapist of their own race, and they were
much more likely than any other group to have increased participation in
mental health treatment when matched with a therapist of their own race.
Hispanic/Latino(a) Americans and African Americans were somewhat more
likely to participate when they were matched, but the effect of matching on
client participation was minimal for unspecified/other ethnic minorities. The
participation level of White/European American clients was influenced very
minimally by the therapist’s race or ethnicity.
Across all 53 studies examining client attendance, the random effects
weighted effect size was d = .22, a value that obscures the differences found
across client race. The heterogeneity of the overall findings was very high,

TABLE 6.2
Weighted Mean Effect Sizes (Cohen’s d ) Across Participant Race
k d SE 95% CI

African Americans 16 .19a .08 [.04, .34]

Asian Americans 16 .46b .07 [.31, .60]
Hispanic/Latino(a) Americans 15 .22a, c .06 [.10, .33]
Unspecified/other ethnic minorities 22 .13a .02 [.08, .18]
White/European Americans 8 .09 .06 [-.02, .20]
All groups combined 53 .22a, c .03 [.16, .28]
Note. CI = confidence interval; k = number of studies; d = random effects weighted effect size; SE = standard
error.
aStatistically significantly different from Asian Americans (p < .001). bStatistically significantly different from all

other groups (p < .001). cStatistically significantly different from White/European Americans (p ≤ .05).

matching clients with therapists      121

Figure 6.1. Contour-enhanced funnel plot of effect sizes (Cohen’s d ) by standard
error for 53 studies of client participation in treatment as a function of racial or ethnic
match with the therapist. In this graph the overall average is to the right of 0, but the
results are widely scattered, with high variability even among studies with a large
number of participants (and lower standard errors) depicted at the top of the graph.
This inconsistent distribution restricts the interpretability of the overall average.
Analyses were conducted by client race to reduce variability.

I2 = 87.7, 95% CI = [85, 90], Q(52) = 428.9, p < .0001, meaning that the overall
results were very inconsistent across studies (see Figure 6.1).

Study and Participant Characteristics Influencing the Results

Studies involved a wide variety of participants and procedures, so we
sought to determine if systematic differences in findings could be attribut-
able to participant or study characteristics. Specifically, we conducted ran-
dom effects weighted correlations with continuous level variables and random
effects weighted analyses of variance with categorical level variables and with
the study effect size as the dependent variable. Overall, no differences were
found across client gender (operationalized as percent female), average age,
education level, or socioeconomic status, and no differences were found across
type of research design, including whether participants were or were not
randomly selected and/or randomly assigned to matched versus unmatched
conditions.

122       foundations of multicultural psychology

 We observed a strong decline over time in the benefits of ethnic and
racial matching on client participation in treatment, with the year of study
publication correlating -.48 (p < .001) with effect sizes. Studies published
before 1990 averaged d = .42, studies published in the 1990s averaged d = .23,
and studies published since the year 2000 averaged d = .14.
The benefits of ethnic or racial matching were stronger when client par-
ticipation was operationalized in terms of treatment completion (d = .27) rather
than treatment attendance (d = .15, p = .03). The 33 studies that matched
on the basis of race tended to show smaller benefits from the match than
did the 20 studies that operationalized the match by ethnicity (e.g., Mexican
American clients matched with Mexican American therapists; d = .17 and .28,
respectively, p = .03). The 27 studies reporting the race of therapists showed
a negative correlation (r = -.39, p = .03) between the percentage of White/
European American therapists in the study and the effect size; thus, studies
with predominantly White/European American therapists tended to find less
benefit from racial or ethnic matching on the degree of client participation
in treatment.
The five studies that explicitly matched clients who preferred a language
other than English with a therapist who spoke the preferred language showed
statistically significantly fewer benefits from a match based on race or ethnicity
(d = .08) compared with the results of 18 studies with unknown procedures in
which bilingual therapy or therapy in the preferred language may or may not
have been provided to clients who spoke English as a second language (d = .31,
p < .05). When therapy was provided in English to English speakers, the effect
of racial or ethnic matching was d = .18, a value between those of the other
two groups. These findings suggested that the effect of racial or ethnic matching
was influenced by language matching for clients who preferred a language other
than English. Specifically, the benefit of matching for clients who preferred to
speak a language other than English was apparently more attributable to the
match based on language than to the match based on race or ethnicity. When
a therapist neither spoke the clients’ preferred language nor shared the clients’
race or ethnicity, the effect of racial or ethnic matching was almost four times
as large as when language matching was provided.
Overall, the effect of racial or ethnic matching on client participation in
treatment was moderated by language match, ethnic versus racial matching,
percentage of White/European American therapists in the study, treatment
completion versus treatment attendance, and year of publication. These
several apparent moderating variables may have interacted with one another.
We therefore sought to ascertain the degree to which these several variables
affected effect sizes in the presence of one another by conducting random
effects weighted regression models. Because the eight effect sizes specific to
White/European Americans had previously been shown to be unrelated to

matching clients with therapists      123

client participation (Table 6.2), we excluded those data in the regression
models to avoid confounding race with language matching and ethnic versus
racial matching. Thus, the dependent variable for this analysis was the aver-
aged effect size across only clients of color.
In the first model, which used all 53 studies, 38.9% of the variance in
effect sizes was explained (p < .0001) by the combination of ethnic versus
racial matching, treatment completion versus treatment attendance, language
matching, and year of publication. At the univariate level, the two variables of
explicit non-English language matching and year of publication reached statis-
tical significance (standardized betas = -.27 and -.37, respectively, p < .05). In
a second model, which considered the 27 studies that reported the racial com-
position of the therapists, the percentage of White/European American thera-
pists in the study was entered along with the same four variables entered in the
first model. This second model explained 46.4% of the variance in effect sizes
(p = .02), again with only explicit non-English language matching (standard-
ized beta = -.45, p = .02) and year of publication (standardized beta = -.48,
p = .01) reaching statistical significance. Thus, although the combination of
variables explained a substantial amount of effect size variance, apparent trends
over time and explicit language matching among clients who preferred a lan-
guage other than English were the variables that most explained study findings.

Likelihood of Publication Bias Adversely Influencing the Results

Meta-analytic results can be adversely affected by publication bias, given
that unpublished studies are less likely to be located. In this meta-analysis,
32 published studies yielded stronger average effect sizes than 21 unpublished
studies (d = .27 and .14, respectively, p = .03). Nevertheless, the inclusion of
so many unpublished studies mitigated the possibility of publication bias. The
overall data were fairly balanced on either side of the mean (see Figure 6.1),
Egger’s regression test (an estimate of asymmetry of effect sizes) did not
reach statistical significance (p > .05), and one statistical method (Duval &
Tweedie, 2000) did not identify any “missing” studies. These three analyses
suggested that publication bias did not influence the overall findings.

DISCUSSION AND INTERPRETATION OF THE FINDINGS

Data from 53 research studies indicated that clients participated in men-

tal health treatment at slightly higher rates and prematurely discontinued
treatment less often when they had a therapist of their same race or ethnicity.
However, the results differed across client race, with Asian American clients
benefitting substantially when they had a therapist of their own race (d = .46),

124       foundations of multicultural psychology

particularly a therapist of their own ethnicity (e.g., Japanese American client
with a Japanese American therapist). In contrast, the treatment participation
of White/European American clients differed to a small degree (d = .09) when
they were matched or unmatched with the therapist.
On average, client age, education level, socioeconomic status, and gen-
der did not affect findings across studies: Ethnic or racial matching yielded
similar data irrespective of those attributes. Research findings have, however,
changed substantially over time. The year in which a study was published was
strongly related to the observed effect size (d = -.48), such that the impact of
ethnic or racial matching on client participation levels has decreased mark-
edly in recent years. Matching a client with a therapist of her or his same race
would have been an important treatment consideration in the 1970s, but
more recently that consideration would be one of several relevant to increas-
ing client participation and decreasing premature termination.

Considerations for Future Research

The results of any meta-analysis are qualified by the quality, method-

ology, and research design of the studies included in the analysis (Cooper,
Hedges, & Valentine, 2009). In this meta-analysis the majority of studies
(87%) did not randomly select participants; thus, the available data lacked
evidence of external validity. More problematic was that only 11% of the
studies involved random assignment of clients to matched versus unmatched
conditions. This flaw greatly restricts our confidence in the internal validity
of the studies. Multiple plausible confounds could have adversely affected the
research findings, including the fact that clients with a therapist of their same
race or ethnicity may have preferred that arrangement. Such methodologi-
cal confounds are so problematic that we see no value in conducting further
research on this topic that does not involve randomization to matched versus
unmatched conditions.
Prior research had indicated that potential clients tend to prefer a thera-
pist of their own race or ethnicity but that this match improves therapeutic
outcomes only to a small degree (Cabral & Smith, 2011). According to the
data analyzed in this meta-analysis, that apparent discrepancy between strong
client preferences and weak outcome effects cannot be primarily attributed to
differential participant attrition across matched versus unmatched conditions.
Client attrition is problematic and requires attention, but it is not strongly
affected by therapist race or ethnicity except for Asian American clients.
The specific factors that influence Asian American clients’ participa-
tion in treatment deserve additional research attention (e.g., S. Sue, Cheng,
Saad, & Chu, 2012). Among other explanations, the differences across cli-
ent race likely reflect underlying differences in worldviews. Psychotherapy

matching clients with therapists      125

originated in Europe and North America and was therefore initially designed
for White/European American clients (Pedersen, 2000). Asian Americans
who engage in mental health treatments appear to benefit when Asian
American therapists understand and account for their cultural experiences
and worldviews that are pertinent to mental health and symptom change.
Research to confirm this assertion that acculturation to Western methods of
treatment accounts for the observed differences across race would be helpful,
particularly because research has not yet examined groups such as Native
American Indians and Alaska Natives.
Researchers can also attend more to issues of language. Because of
numerous logistical difficulties, people with limited English proficiency can
be excluded from research studies (Acevedo, Reyes, Annett, & López, 2003).
In our analysis, only five studies (9%) explicitly matched clients who pre-
ferred a language other than English with a therapist who spoke that lan-
guage. Data analyses suggested that the effects of racial or ethnic matching
were primarily accounted for by language matching for clients who spoke
English as a second language. The field would benefit from further investiga-
tion of language issues and from taking seriously the benefits of conducting
treatment in the clients’ preferred language if other than English.
Researchers must keep in mind that although the variables of race and eth-
nicity are important, they are broad descriptions that tell us little about how a
client and therapist may actually interact in therapy. Variables that may increase
understanding of client–therapist interactions include the therapist’s multi-
cultural competence and worldview congruence with the client. Those more
specific variables have not received nearly as much attention in the research lit-
erature, yet likely influence client participation in therapy more than therapist
race or ethnicity. Clients and therapists who share the same race or ethnicity
do not necessarily share similar worldviews (Ibaraki & Hall, 2014). Although
matching may be important for some clients, not all will benefit from this prac-
tice (S. Sue, 1988). As S. Sue and Zane (1987) suggested several decades ago,
researchers should shift their focus from distal to proximal variables.

Recommendations for Practitioners

Hypothetically, matching clients with therapists of their same race or

ethnicity should foster feelings of being understood, improve the therapeutic
relationship, and increase a client’s level of participation in treatment. Clients
of color, particularly African American clients, tend to prefer this arrange-
ment (Cabral & Smith, 2011). Therapists working with clients from races or
ethnicities other than their own may therefore have some concern about how
the client will engage in treatment. Most clients seeking therapy are more
interested in factors other than therapist race, notably decreasing emotional

126       foundations of multicultural psychology

pain and working through difficulties. Although perceived similarities with
the therapist can provide a feeling of immediate connection, genuine respect
must eventually be earned by the therapist, irrespective of race.
In particular, therapists of color can be confident that racial differences
typically do not affect White/European American clients’ participation in
treatment. On average White/European American clients’ choice to remain
in treatment is essentially unaffected by whether the therapist shares their
racial or ethnic background.
Moreover, matching clients of color with therapists of their same race
or ethnicity is often unfeasible for a variety of reasons, including the fact
that there are fewer available therapists of color (American Psychological
Association, 2005). Even if it were possible to consistently match clients and
therapists according to race or ethnicity, doing so might have unintended
negative consequences (Alladin, 2002). For example, well-intentioned ther-
apists may inadvertently fail to acknowledge other meaningful differences
(e.g., level of acculturation) or dismiss the fact that within-group differences
can be larger than between-group differences. Such oversights could curtail
the effectiveness of psychotherapy and inadvertently contribute to the client
being misunderstood.
Nevertheless, the data from this meta-analysis do suggest that some-
thing systematic needs to be done to better retain Asian American clients
who have a therapist from another background. When Chinese American
therapists are available to work with Chinese American clients, for example,
such an arrangement would likely increase client participation. However,
other practices could also increase client retention, such as incorporating
culturally sensitive adaptations to therapy and providing therapists with
ongoing multicultural training. Therapists working with Asian American
clients especially have to proactively explore and implement practices to
improve engagement and retention.
Practitioners may increase client participation and retention by provid-
ing therapy in clients’ preferred language when other than English. Clients
for whom English is a second language can find this a barrier to receiving high
quality services (Acevedo et al., 2003). American Psychological Association
guidelines (2003) suggest that mental health professionals interact in the lan-
guage requested by the client or provide a professional language interpreter.

CONCLUSION

When clients of color are matched with a therapist of their same race or
ethnicity, they tend to remain in treatment somewhat longer and drop out less
frequently than when they have a therapist from a different racial or ethnic

matching clients with therapists      127

background. The benefits were particularly notable for Asian American cli-
ents. However, the research findings were more robust in past decades than
at present, and findings vary substantially across studies. Many factors other
than therapist race or ethnicity affect client levels of participation in mental
health treatment.
At present, the research literature is of insufficient quality to justify more
solid conclusions. Nevertheless, given the declining influence of racial or eth-
nic matching on client participation rates, we do not believe that continuing
this line of inquiry will yield as much benefit to treatment participation among
clients of color as would other strategies. These may include working to increase
the number of bilingual therapists and therapists of color and enhancing the
multicultural competence of therapists from all racial and ethnic backgrounds.
In effect, few clients who discontinue therapy would identify the therapist’s
race or ethnicity as the deciding factor. More likely, they would speak of mis-
understandings, insufficient emotional support, diverging views, incompatible
opinions about what should be done, and so forth. Those are the key issues for
therapists and researchers to address. It is time to identify precisely the reasons
why clients prematurely discontinue therapy—and implement solutions.

128       foundations of multicultural psychology

 7
CULTURALLY ADAPTED
MENTAL HEALTH SERVICES:
AN UPDATED META-ANALYSIS
OF CLIENT OUTCOMES

When clients from diverse backgrounds seek professional mental health

services, several concerns may arise: Will this therapist respect my beliefs and
values? Will this therapist promote mainstream (White/European American)
worldviews and solutions? Can I truly be myself in therapy? Clients may
not verbalize these specific concerns, but these are among many that may
adversely affect client outcomes in traditional therapy (S. Sue & Zane, 1987;
S. Sue et al., 2006). Thus, clinicians should ask this critical question: To what
extent does the therapy I provide align with the cultural beliefs, values, and
goals of this client?
Although mental health treatments have proliferated in recent decades,
most treatments are based on theories that reflect European and European

Alberto Soto and Derek Griner, both of Brigham Young University, contributed to the writing of this
chapter.
A complete list of references for the studies included in this chapter’s meta-analysis is posted online.
Readers can consult it at [Link]
[Link]
Foundations of Multicultural Psychology: Research to Inform Effective Practice, by T.â•–B. Smith and J.â•–E. Trimble
Copyright © 2016 by the American Psychological Association. All rights reserved.

129

14068-07_CH07_2ndPgs.indd 129 7/28/15 9:07 PM

 American ideals and values (e.g., psychodynamic, client-centered, cognitive–
behavioral). As such, these traditional mental health treatments have fre-
quently ignored or minimized contextual variables such as culture, gender,
race and ethnicity, socioeconomic status, and religious and spiritual values
(G.â•–C.â•–N. Hall, 2001; Ibaraki & Hall, 2014; Ponterotto, Casas, Suzuki, &
Alexander, 2010). Rather than assume that any particular treatment will
meet the needs of every client, therapists must consider these contextual and
environmental variables to effectively align the treatment with the needs and
experiences of the client (Cardemil, 2010b; Norcross & Wampold, 2011).
Cultural adaptations should be made to traditional treatments to better
meet the needs of a diverse clientele (Castro, Barrera, & Holleran Steiker,
2010; Gone & Trimble, 2012; T.â•–B. Smith, 2010). For instance, because
Western values such as assertive individuation may contradict psychological
well-being as conceptualized by clients from collectivistic cultures, therapists
working with these populations may find that explicitly addressing the clients’
social cohesion may prove more effective than the individualistic focus of
many mental health treatments. Thus, a therapist working with Hispanic/
Latino(a) American clients could consider family involvement in treatment
(Falicov, 2009; Hurwich-Reiss, Rindlaub, Wadsworth, & Markman, 2014).
Alternatively, a therapist who generally practices a strictly behavioral orien-
tation to treatment might benefit from considering the cultural beliefs of a
client who may think that ancestral spirits cause depression, even if most of
the interventions remain behavioral in nature. A therapist who has discerned
that an Alaska Native client favors traditional worldviews could incorporate
cultural imagery and metaphors. Culturally competent treatments involve
adaptations to methods of delivery, content, and conceptualizations, such
that the client becomes more likely to engage in and successfully complete
the treatment provided (S. Sue, Zane, Nagayama Hall, & Berger, 2009; Zane
& Ku, 2014).
Psychologists have an ethical obligation to provide the most effective
service available to their clients (Trimble, Scharrón-del-Río, & Hill, 2012;
Vasquez, 2012). Much of the therapy literature focuses attention on evidence-
based practices (EBPs; Kazdin, 2008), but EBPs, like most standardized treat-
ment approaches, do not account for the needs of culturally diverse clients
(La Roche & Christopher, 2008). Treatments provided should be based on evi-
dence, but therapists must remember that no single EBP will work with every
client. They must attend to cultural factors that influence therapy (Cardemil,
2010a; T.â•–B. Smith, 2010; S. Sue, 2003; Trimble et al., 2012) and realize that
cultural adaptations to EBPs are justified (Castro et€al., 2010). Therapists who
assume that an EBP is equally effective across all cultures take a universalistic
approach to treatment, giving deference to the particular EBP over the reali-
ties of a particular client. These therapists may inappropriately apply the EBP

130 ╇╇╇╇╇ foundations of multicultural psychology

14068-07_CH07_2ndPgs.indd 130 7/28/15 9:07 PM

 by setting goals and embracing values that coincide with their own culture but
not the client’s culture (Comas-Díaz, 2006; Falicov, 2009). The field of multiÂ�
cultural psychology should help to replace culturally insensitive treatments
(even those that may be evidence based) with culturally appropriate services.
Although cultural adaptations to mental health treatments are benefi-
cial, they must be done systematically, using empirical data to support imple-
mentation (Barrera, Castro, Strycker, & Toobert, 2013; Domenech Rodríguez
& Bernal, 2012; Huey, Tilley, Jones, & Smith, 2014). Multicultural psychol-
ogy cannot assume that particular culturally adapted treatments are superior
to traditional treatments without evidence and without simultaneous reliance
on empirical data to refine the cultural adaptations made (G. Bernal, Jiménez-
Chafey, & Domenech Rodríguez, 2009; Cardemil, 2010b; Castro et al., 2010;
Huey et al., 2014). Cultural adaptations should be subject to the same intense
scrutiny necessary for any high-risk activity involving individuals’ well-being.
Research on the effectiveness of cultural adaptations will help mental health
professionals understand how best to apply the abstract principles of multi-
cultural psychology to real world practices, with the goal of improving the
outcomes of diverse clients over outcomes of treatment as usual.

REVIEW OF THE LITERATURE

Relevant Conceptual Issues

Although psychotherapy is relatively modern, mental health treatments

are ancient. Cultures across the world have long practiced healing rituals or
provided worldviews to enhance coping with psychological distress and mental
illness (Calabrese, 2008; McCubbin & Marsella, 2009). Significant improve-
ments in mental health treatments over the past century should not be mis-
interpreted to mean that all other conceptualizations and practices relevant
to mental health are invalid or archaic. Psychologists have no monopoly on
effective mental health treatments.
Furthermore, although many individuals in North America have come
to see psychotherapy as a socially acceptable method for treating mental
health concerns, individuals from other cultural origins may not feel the
same way. Many individuals from diverse cultural backgrounds consider
Western forms of psychotherapeutic interventions to be strange, invalidat-
ing, and intrusive (C.â•–C.â•–I. Hall, 1997; Jackson, Schmutzer, Wenzel, & Tyler,
2006). Although psychotherapy is effective (Campbell, Norcross, Vasquez,
& Kaslow, 2013; Lambert, 2007), racial and ethnic minorities may encounter
barriers such as difficulties discussing the negative consequences of racism
with White therapists (D.â•–W. Sue, 2015) and may therefore have outcomes

culturally adapted mental health services╇╇╇╇╇ 131

14068-07_CH07_2ndPgs.indd 131 7/28/15 9:07 PM

 less positive than expected (S. Sue, 1998). Some scholars have therefore
advocated that psychologists in North America attend to indigenous concep-
tualizations and practices, with methods specific to local contexts increasingly
being developed (U. Kim, Yang, & Hwang, 2006; Ramos & Alegría, 2014). In
situations where indigenous approaches to psychology are insufficiently under-
stood or are impractical to inform mental health treatments, mental health
practitioners can use existing EBPs, adapting them to align with clients’ cultural
values and worldviews (Barrera et al., 2013).
Several decades ago, Stanley Sue (1977) specifically recommended that
therapists culturally adapt treatment at the client–therapist level as well as at
a system level. His original suggestions have become a catalyst for subsequent
culturally sensitive treatments. These suggestions included (a) conducting
therapy in the client’s preferred language, (b) matching clients and therapists
according to race or ethnicity, (c) developing mental health clinics that cater
to specific racial or ethnic groups, and (d) providing alternative methods for
mental health services delivery (G. Bernal & Flores-Ortiz, 1982; Flaskerud,
1986; Miranda et al., 2005; S. Sue, 1977). The first suggestion, language
matching, was an obvious improvement over the common English-only ser-
vices, but simply translating content into a client’s preferred language would
not correct underlying cultural differences in conceptualizations of mental
health or methods for enhancing well-being. The second recommendation,
racial matching, has proven effective in delimited circumstances (Cabral
& Smith, 2011; see also Chapter 6, this volume), such as those involving
African Americans and immigrant populations. The third recommendation
has not been widely heeded, because few contemporary clinics provide spe-
cialized services for a given racial or ethnic group. Over time, the fourth
recommendation, providing alternative methods and cultural modifications
to existing methods, has received more emphasis than the other three (e.g.,
S.â•–W. Chen & Davenport, 2005; T.â•–B. Smith, 2010; S. Sue et al., 2009).
Cultural adaptations can be made according to the concept of dynamic
sizing (S. Sue, 1998), accounting for commonalities in culturally diverse groups
while also respecting and attending to individual differences (La Roche &
Lustig, 2010). For example, a Native American Indian client who embraces
indigenous spirituality may benefit from a spiritual healing ritual integrated into
therapy (Calabrese, 2008; Trimble, King, Morse, & Thomas, 2014). However,
a Chinese American client who was raised and educated in the United States
may prefer standard cognitive behavioral therapy (CBT) techniques to a
version of CBT adapted to account for traditional Chinese worldviews and
values (S.â•–W. Chen & Davenport, 2005). Thus, therapists should be mindful
to align treatment with individual clients’ cultural worldviews (La Roche &
Lustig, 2013), rather than assume that standard (nonadapted) treatment is suf-
ficient (Cardemil, 2010a) or implement cultural adaptations without verifying

132 ╇╇╇╇╇ foundations of multicultural psychology

14068-07_CH07_2ndPgs.indd 132 7/28/15 9:07 PM

 whether they actually match the client’s worldviews, including acculturation
level (S. Sue, 2010). Therapists should tailor the treatment to the individual cli-
ent (Norcross & Wampold, 2011), not assume generalization without evidence
for generalization (Cardemil, 2010b; S. Sue, 1999).
Given the essential culturally specific expertise needed for such adapta-
tions to treatment, clinicians may feel overwhelmed, particularly when consid-
ering the many possible ways that treatment could be modified to better align
with cultural factors. To help specify the types of cultural adaptations that
should be most effective in mental health treatments, scholars have developed
conceptual frameworks to help guide clinicians.
An ecological validity model (G. Bernal, Bonilla, & Bellido, 1995)
addressed eight specific cultural domains that clinicians should consider
when working with culturally diverse clients:
77 language adaptions—extending beyond interpretation and trans-
lation to include terminology and methods for communicating
that are appropriate to the client’s particular background (e.g.,
for Mexican Americans, the phrase échale ganas rather than
“hang in there”);
77 persons—the attention that should be directed to the client–
therapist relationship, particularly cultural similarities and differ-
ences (Asnaani & Hofmann, 2012);
77 metaphors—the symbols, folklore, and concepts shared by a
cultural group of individuals (Parra Cardona et al., 2012);
77 content—culture-specific beliefs and practices (e.g., religious
and spiritual beliefs, cultural history and traditions) that can
be infused in therapeutic interventions (McCabe, Yeh, Lau, &
Argote, 2012);
77 concepts—ways in which presenting concerns are conceptual-
ized in a culturally congruent manner (e.g., accepting somatic
conceptualizations of depression vs. depression conceptualized
only in terms of emotional distress; exploring religious or spiri-
tual concepts to explain suffering and healing; S.â•–W. Chen &
Davenport, 2005);
77 goals for treatment, which should align with the cultural world-
view of the client (e.g., collectivism vs. individualism; see Diaz-
Martinez, Interian, & Waters, 2010);
77 methods of treatment, which should be adapted for the individual
needs of the client, congruent with his or her worldview (e.g.,
family-centered therapy vs. individual therapy); and
77 context—the broader issues (e.g., acculturative stress, racial
microaggressions, immigration, poverty) that culturally diverse

culturally adapted mental health services╇╇╇╇╇ 133

14068-07_CH07_2ndPgs.indd 133 7/28/15 9:07 PM

 clients may commonly face (Hinton, Rivera, Hofmann, Barlow,
& Otto, 2012; La Roche, D’Angelo, Gualdron, & Leavell, 2006).
Therapists should attend to multiple dimensions such as these when cul-
turally adapting mental health treatments. Additional considerations and
frameworks for practice also provide guidance but cannot be detailed in this
chapter (Barrera & Castro, 2006; Hwang, 2006, 2009; Lau, 2006; Leong, 2011;
Whitbeck, 2006).
Although research-based models provide a framework for addressing cul-
tural factors that may be ignored by traditional forms of therapy (Domenech
Rodríguez & Bernal, 2012; Huey et al., 2014), clinicians should be receptive
to client feedback about what has and has not worked for them (Lambert,
2010). Researchers and clinicians may find that appropriate adaptations
can be assessed through focus groups with individuals from the relevant
community. A therapist might also consult community leaders, religious
and spiritual leaders, scholars, and client family members for insight into
the appropriateness of proposed adaptations for a specific cultural group,
rather than attempting to address several culturally diverse groups with the
same intervention (T.â•–B. Smith, Rodríguez, & Bernal, 2011). The balance
of appropriate etic versus emic adaptations is difficult, but it can be attained
by thoughtful and purposeful consultation with clients and examination of
the relevant literature, which includes several books on the topic of cul-
tural adaptations (G. Bernal & Rodríguez, 2012; La Roche, 2012; C.â•–J. Yeh,
Parham, Gallardo, & Trimble, 2011).

Narrative Review of the Research Literature

Several years after guidelines for cultural adaptations of treatments

were publicized, culturally congruent services remained rare (S. Sue et al.,
2006). Nevertheless, the topic has received increasing professional interest
(G. Bernal et al., 2009; Cardemil, 2010b; Huey et al., 2014; Ramos & Alegría,
2014). Meta-analyses have shown that cultural adaptations of treatments are,
in fact, effective (Benish, Quintana, & Wampold, 2011; Chowdhary et al.,
2014; T.â•–B. Smith et al., 2011; van Loon, van Schaik, Dekker, & Beekman,
2013). The first of these reviews (Griner & Smith, 2006) found that the aver-
age effect sizes for quasi-experimental and experimental designs were d = .42
and d = .40, respectively. A subsequent update to that meta-analysis (T.â•–B.
Smith et al., 2011) found that when the results were adjusted for apparent
publication bias, the average effect size was d = .27. A review of culturally
adapted treatments for youth reported an average effect size of d = .22 when
the control group received treatment as usual (Huey & Polo, 2008). Another

134 ╇╇╇╇╇ foundations of multicultural psychology

14068-07_CH07_2ndPgs.indd 134 7/28/15 9:07 PM

 meta-analysis (Benish et al., 2011) reported an average effect size of d = .32
when comparing culturally adapted treatments with bona-fide treatments. A
recent review, limited to nine studies of outpatient treatment for depression
and anxiety (van Loon et al., 2013), found substantially greater effectiveness
(d = 1.06). A review of 20 studies specific to the treatment of depression also
found a large average effect size (d = .72; Chowdhary et al., 2014). Another
recent review has summarized much of that work (Huey et al., 2014).
Overall, these meta-analyses and reviews have documented benefits of
culturally adapted treatments when working with culturally diverse clients.
Researchers and clinicians are, however, continually finding new and inno-
vative ways to attend to cultural factors in therapy. The increasing number
of research studies of culturally adapted treatments means that the literature
must be continuously evaluated so practitioners can benefit from the latest
findings and trends. Thus, we sought to update our previous meta-analysis
of culturally adapted interventions by searching out additional data. After
presenting the details of the meta-analysis, a summary of the findings is pre-
sented in a subsequent section.

QUANTITATIVE SYNTHESIS OF RESEARCH DATA

Our updated meta-analysis included studies identified in previous meta-

analyses (Benish et al., 2011; T.â•–B. Smith et al., 2011) that were conducted
in the United States or Canada; evaluated clients’ experiences in mental
health services that were adapted on the basis of culture, race, or ethnicity;
and involved a control group using a quasi-experimental or experimental
research design. We included any type of control group, and we recorded
the nature of the control group because we were planning to compare those
studies involving equivalent mental health services not culturally adapted
(e.g., treatment as usual) with no treatment (e.g., clients on a waiting list
receiving no services). We excluded studies with more than 10% White/
European American clients or control group members because the cultural
adaptations were presumed to enhance the experiences of people of color,
such that the treatments would have different meaning for White/European
American clients.
Statistical estimates within studies were converted to Cohen’s d using
meta-analytic software. Positive values for effect sizes indicated improved
client outcomes (decreased symptoms, higher treatment completion rates)
over the control group, whereas negative values indicated worse client outcomes
compared with the control group. General methods of the meta-analysis are
reported in the Appendix of this book.

culturally adapted mental health services╇╇╇╇╇ 135

14068-07_CH07_2ndPgs.indd 135 7/28/15 9:07 PM

 Description of the Existing Research Literature

We located 108 studies that evaluated a culturally adapted mental

health treatment. However, we analyzed data from only 79 studies because
17 involved a single group of participants tracked over time (evaluating pre-
to posttest changes without a control group), eight involved a substantial
number (>10%) of White/European American participants, two involved
correlational designs, one reported client treatment participation but no
client outcome information, and one involved extended follow-up data to
a previously published study already included in the review. As shown in
Table€7.1, the 79 studies were conducted primarily during the past 2 decades,
with most of the studies located being published. Subjects tended to be either
children and adolescents or middle-aged adults, reflecting a trend in the
literature for treatments to be either prevention-oriented (treating “at-risk”
groups) or clinically oriented (examining groups in community mental health
clinics). Asian Americans and Hispanic/Latino(a) Americans were the two
groups most commonly evaluated in studies. By contrast, African Americans
and Native American Indians were evaluated in only 14% and 7% of the
studies, respectively.
Overall, the manuscripts contained data from 12,014 individual clients,
but the median number of clients was 60, a small number of participants
given that about half were in control group conditions. The types of cultural
adaptations varied substantially across studies. Some studies clearly followed
existing guidelines available in the professional literature (e.g., G. Bernal
et€al., 1995), but others had produced thin descriptions and/or had delimited
adaptations to a few aspects of culture:
77 76% included explicit mention of cultural content and values
in treatment,
77 70% provided treatment in the client’s preferred language when
other than English,
77 60% matched clients with therapists of similar ethnic and racial
backgrounds,
77 53% addressed clients’ contextual issues (e.g., experiences of
racism, employment),
77 48% used metaphors from client cultures,
77 47% modified the methods of delivering therapy on the basis of
cultural considerations,
77 46% indicated that they had developed the cultural adapta-
tions through consultation with individuals from the culture,
77 43% adhered to the client’s conceptualization of the presenting
problem,

136 ╇╇╇╇╇ foundations of multicultural psychology

14068-07_CH07_2ndPgs.indd 136 7/28/15 9:07 PM

 TABLE 7.1
Characteristics of 79 Studies of Culturally Adapted Treatments
No. of
Characteristic M studies (k) %
Year of report 2000
Before 1980 0 0
1980–1989 8 10
1990–1999 26 33
2000–2012 45 57
Publication status
Published 63 80
Unpublished dissertations 16 20
Research design
Quasi-experimental 19 24
Experimental 60 76
Sample type
Community members 9 11
At-risk clients 41 52
Clinical clients 29 37
Sample size 152
<50 31 39
50–99 30 38
100–199 9 11
200–399 4 5
400–999 2 3
>1,000 3 4
Age of participantsa 25.5
Children (<13 years) 22 28
Adolescents (13–18 years) 20 25
Young adults (19–29 years) 4 5
Middle-aged adults (30–55 years) 29 37
Senior adults (>56 years) 4 5
Gender of participants (% female) 58.4
Race of participantsb (%)
African American 14
Asian American 44
Hispanic/Latino(a) American 35
Native American Indian 7
Other <1
Note.╇ Not all variables sum to the total number of studies because of missing data.
aAverage age category of participants within studies, not all participants necessarily listed. bThe racial

composition of participants across all studies, calculated by multiplying the number of participants within
studies by the percentage of participants from each racial group and dividing that product by the total
number of participants.

culturally adapted mental health services╇╇╇╇╇ 137

14068-07_CH07_2ndPgs.indd 137 7/28/15 9:07 PM

 77 42% based their cultural adaptations on published research or
theoretical models,
77 21% reported that mental health staff had received training in
the cultural adaptations,
77 18% modified the wording of outcome instrumentation to be
culturally appropriate, and
77 11% explicitly solicited culturally congruent outcome goals
from the client.
On average, studies reported 3.9 out of the eight components of G. Bernal’s
model (G. Bernal et al., 1995). Only 14 studies (18%) involved more than
five of the eight components. Thus, on the whole, the studies made attempts
to culturally adapt treatment without adhering closely to best practices.
The typical study involved an experimental design in which clients
were randomly assigned to either a culturally adapted mental health treat-
ment or “treatment as usual.” In those studies, the effect size represents an
estimate of the effectiveness of culturally adapted treatments. For the 24%
of studies using quasi-experimental designs (nonrandom group composition)
and the 47% of studies using no-treatment controls (clients on a waiting list),
the magnitude of the effect size estimates would be influenced by factors other
than the culturally adapted nature of the intervention; these studies thus
warranted separate analyses. We also considered it essential to distinguish the
results from studies using mental health treatments with clinical populations
from results of studies involving prevention-oriented interventions for at-risk
populations, because the nature of the services provided would necessarily
differ, and client outcomes and rate of change would also likely differ.

Overall Findings by Research Design, Control Group Type,

and Treatment Type

Across all 79 studies examining a culturally adapted mental health treat-

ment, the random effects weighted effect size was d = 0.47 (SE = .043, 95%
CI = [0.39, 0.57], p < .0001). The heterogeneity of the findings was high
(I2 = 72.0, 95% CI = [65, 78]; Q(78) = 278.9, p < .0001), meaning that the results
tended to be very inconsistent across studies (see Figure 7.1). In the 60 studies
in which participants were randomly assigned to treatment conditions (true
experimental designs), average results were statistically significantly more effec-
tive than in the 19 studies using nonrandom assignment of clients to treatment
conditions (quasi-experimental designs; d = 0.55 vs. d = 0.28, p = .004). This
finding was unexpected because experimental designs that remove some plau-
sible confounds (e.g., impact of self-selection on treatment effectiveness) typi-
cally result in more conservative effect sizes than designs in which confounds

138 ╇╇╇╇╇ foundations of multicultural psychology

14068-07_CH07_2ndPgs.indd 138 7/28/15 9:07 PM

 Figure 7.1.╇ Contour-enhanced funnel plot of effect sizes (Cohen’s d) by standard
error for 79 studies of culturally adapted treatments. This graph shows the dis-
tribution of effect sizes as a function of the number of participants in the study
(operationalized as standard error). The studies at the top of the graph are those
with many participants (and small standard error values), studies that yielded
results in the range of d = 0.0–0.5. The fewer the participants, the less consistent
the results. Moreover, in the middle and bottom of the graph are few studies with
nonsignificant results (note the absence of dots in the white area relative to the
top of the graph and to the right of the graph). This distribution strongly suggests
publication bias in the available literature; the overall average effect estimate
should therefore be adjusted closer to zero (d = .31) to account for apparently
“missing” nonsignificant findings.

are uncontrolled. Also contrary to expectations, no meaningful difference was

evident between the findings of the 37 studies comparing outcomes of treat-
ment groups to those of no-treatment control groups (i.e., clients on a waiting
list) and the findings of the 42 studies comparing outcomes in the experimen-
tal group with outcomes of clients receiving a bona fide treatment (treatment
as usual; d = 0.49 vs. d = 0.46, p > .10).
Comparison of treatments for clinical populations versus prevention-
oriented programs for at-risk populations yielded similar results (d = 0.47 vs.
d = 0.52, p > .10). When we restricted analyses of those two types of studies
to experimental designs with comparison groups receiving a comparable (but
nonadapted) intervention, 12 treatments for clinical populations yielded an

culturally adapted mental health services╇╇╇╇╇ 139

14068-07_CH07_2ndPgs.indd 139 7/28/15 9:07 PM

 effect size of d = 0.56, and 15 prevention-oriented programs for at-risk popula-
tions yielded an average effect size of d = 0.59.

Likelihood of Publication Bias Adversely Influencing the Results

When a researcher obtains results contrary to expectation (i.e., null find-
ings), those results are less likely to be published and therefore more difficult to
locate in a literature search and meta-analysis. This so-called publication bias
can shift meta-analytic data toward the hypothesis favored by scholars, because
nonsignificant results were unrepresented. In this meta-analytic review, pub-
lication bias apparently did influence the overall findings we reported. The
data in Figure 7.1 were asymmetric, with a notable dearth of studies with few
participants that did not achieve statistically significant results, indicating
that studies with negative or null results had not been located in our literature
search. Egger’s regression test (an estimate of effect size asymmetry) was statisti-
cally significant (p < .0001), indicative of publication bias. In addition, one sta-
tistical method (Duval & Tweedie, 2000) identified 22 “missing” studies in the
distribution. When those hypothetically missing data were accounted for, the
resulting omnibus effect size was reduced to d = 0.31 (95% CI = [0.22, 0.40]).
Evidence of publication bias was also found in the restricted sample of studies
using an experimental design with a comparable treatment for the control
group (i.e., culturally adapted treatment vs. “treatment as usual”). Thus, the
results presented in the previous section represent liberal estimates, and the
influence of publication bias will need to be accounted for when interpreting
the results and in subsequent analyses.

Study and Participant Characteristics Influencing the Results

Study and participant characteristics had been found to moderate the
results of a previous meta-analysis (T.â•–B. Smith et al., 2011). Specifically, the
effectiveness of culturally adapted mental health treatments had been found
to be greater among (a) adult client populations older than 35 to 40 years,
(b) racially homogeneous samples of clients (with homogeneity being one
indicator of specificity of cultural adaptions), (c) Asian American clients, and
(d) studies involving multiple cultural adaptations (with more cultural adap-
tations producing more effective treatments). We sought to ascertain whether
these variables would remain predictors of treatment effectiveness when the
influence of publication bias and experimental versus quasi-experimental
research design were considered. A meta-regression including these six vari-
ables explained 33.1% of the variance in effect sizes (p < .0001), with the
results depicted in Table 7.2. All variables contributed at least 1% of variance
to the model, as indicated by beta weights above .10, but two variables—the
percentage of Asian American participants and the racial homogeneity of

140 ╇╇╇╇╇ foundations of multicultural psychology

14068-07_CH07_2ndPgs.indd 140 7/28/15 9:07 PM

 TABLE 7.2
Random Effects Regression Weights for Study
Characteristics Associated With Effect Sizes
Variable R2 b p
33.1***
Estimate of publication biasa .25 .006
Sum of cultural adaptationsb .25 .005
Random assignment to treatment type .27 .005
Average client age .26 .006
Percentage of Asian American clients .14 .16
Client racial homogeneity within studiesc .11 .23
Note.╇ aInverse of the number of participants in the study (Peters, Sutton, Jones, Abrams, & Rushton, 2006).
b
Sum of the eight indicators of the ecological model (G. Bernal, Bonilla, & Bellido, 1995). cContrast of studies
in which all participants were of the same race with studies in which participants’ race varied, *** p < .001.

the client participants—failed to reach statistical significance in the presence

of the other variables (k = 79). The other four variables contributed equiva-
lently to the model, each explaining between 6% and 7% of the variance in
effect sizes. Culturally adapted mental health treatments continued to appear
to be most effective for adult populations over age 40, likely interacting with
acculturation level. And treatments involving multiple cultural adaptations
were more effective than those with only a few types of cultural adaptations:
The more closely a treatment aligned with recommended practices (e.g.,
G. Bernal et al., 1995), the more effective the treatment.
A separate meta-regression was conducted to examine whether any par-
ticular cultural adaptation was more predictive of positive client outcomes
than any other. Indicators of the eight components of the ecological validity
model (G. Bernal et al., 1995) explained 16.3% of the variance in effect sizes
(p = .02). The two types of cultural adaptations that remained statistically
significant in the presence of the others were (a) explicitly basing treatment
on the client’s goals, informed by cultural values (standardized beta = 0.25,
p = .03) and (b) providing treatment in the clients’ preferred language (stan-
dardized beta = 0.21, p = .04). Both of these adaptations had proven effective
in a previous meta-analysis (Griner & Smith, 2006).

DISCUSSION AND INTERPRETATION OF THE FINDINGS

Compared with treatment as usual, culturally adapted treatments result

in better outcomes for clients of color. Nevertheless, the underlying find-
ings are highly variable (Figure 7.1), such that some culturally adapted
treatments are clearly preferable to others. In fact, some culturally adapted

culturally adapted mental health services╇╇╇╇╇ 141

14068-07_CH07_2ndPgs.indd 141 7/28/15 9:07 PM

 treatments are about as effective as or worse than nonadapted treatment
as usual (see also Huey et al., 2014). The distinguishing features of studies
with findings of effective versus ineffective cultural adaptations were
random assignment of clients to treatment conditions, treatment of adult
clients above age 40 (likely conflated with level of acculturation), and more
comprehensive cultural adaptations. Treatments were particularly effective
when based on clients’ goals, informed by cultural contexts, and provided
in clients’ preferred language. However, we observed a troubling trend
for publication bias. Studies with null findings are apparently remaining
unpublished. After accounting for publication bias, the magnitude of the
overall results for studies using true experimental designs to compare cultur-
ally adapted treatments to bona fide treatments was d = .31.

Considerations for Future Research

Researchers must continually evaluate psychotherapeutic interven-

tions to ensure that clients are receiving the best services available. Because
several meta-analytic reviews have confirmed that cultural adaptations do
result in better client outcomes than bona fide treatments, researchers can ask
additional questions: What makes culturally adapted treatments more effec-
tive than traditional practices? Why are some culturally adapted treatments
very effective but others not much better than control group conditions (see
Figure€7.1)? In such examinations, scholars can explicitly evaluate the postu-
lates of relevant conceptual models (Barrera & Castro, 2006; G. Bernal et€al.,
1995; Hwang, 2006, 2009; Lau, 2006; Leong, 2011; Whitbeck, 2006). We
know that cultural adaptations work, but we now need more specific informa-
tion about the underlying mechanisms and processes (T.â•–B. Smith, 2010).
We need greater attention to cultural adaptations to treatments involv-
ing Native American Indian clients, who are currently underrepresented in
research studies. Additional studies involving African American clients would
also be useful. The field already has enough studies with relatively small num�
bers of participants, given that the median value was only 60 clients across
79 studies. With the variability in research findings among studies using rela-
tively few participants (note range of effect sizes across bottom two thirds
of Figure 7.1), large multi-site research projects would be more useful. We
encourage scholars in the field to collaborate on large-scale projects rather
than work on separate small ones.
We also invite scholars as well as journal reviewers and editors to con-
sider the issue of publication bias. Studies with statistically significant findings
appear to be published more frequently than studies with nonsignificant results.
Authors may be reluctant to submit manuscripts with findings that contradict
the data evaluated in previous meta-analyses, and editors may be reluctant

142 ╇╇╇╇╇ foundations of multicultural psychology

14068-07_CH07_2ndPgs.indd 142 7/28/15 9:07 PM

 to publish them. This reluctance creates a problem: Not all cultural adapta-
tions work (note the many points in the center area of nonsignificant results in
Figure 7.1), and we do not benefit from science when failures remain hidden.

Suggestions for Practitioners

In our review of the literature, we located several models and conceptual

frameworks that can be used by clinicians to culturally adapt mental health
treatments (Barrera & Castro, 2006; G. Bernal et al., 1995; Hwang, 2006, 2009;
Lau, 2006; Leong, 2011; Whitbeck, 2006). Clinicians should follow these
models (Castro et al., 2010), but we found few instances of clinical practice that
were explicitly based on them. On average, the treatments in the 79 studies we
examined used 3.9 out of eight aspects of G. Bernal’s model (G. Bernal et al.,
1995), with only 14% using more than five. Clinical practices reported in the
literature are beneficial but fall short of the ideal. There are, no doubt, practi-
cal reasons why few clinicians consider the conceptual frameworks available
in the literature and why only about half of the recommended dimensions
of adaptation are implemented. Identification of those reasons can facilitate
removal of barriers and eventual implementation of best practices. The more
cultural adaptations a therapist makes to a treatment, the more effective that
treatment is likely to be with clients of color. We urge practitioners to imple-
ment professional recommendations for culturally adapting mental health
treatments.
Among the practices clinicians use to culturally adapt treatment, align-
ing treatment goals with the cultural worldviews and values of each client
currently seems to be the most effective. This strategy yielded the strongest
client outcomes across the few studies that implemented it. Therapists should
give particular heed to working toward client-generated goals that incorpo-
rate cultural contexts. This process will help the clinician gain insight into the
cultural worldview of the client and will also promote an egalitarian relation-
ship, empowering the client.
The meta-analytic data also support providing treatment in the client’s
preferred language, consistent with existing professional guidelines:
Psychologists interact in the language requested by the client and, if this
is not feasible, make an appropriate referral. . . . If this is not possible, psy-
chologists offer the client a translator with cultural knowledge and an
appropriate professional background. When no translator is available, then
a trained paraprofessional from the client’s culture is used as a translator/
culture broker. (American Psychological Association, 1993, p. 47)
We hope that practitioners will make services available to clients who prefer
a language other than English by using language interpreters or by actively
using referral networks to find professionals with the necessary language skills.

culturally adapted mental health services╇╇╇╇╇ 143

14068-07_CH07_2ndPgs.indd 143 7/28/15 9:07 PM

 CONCLUSION

Culturally diverse clients benefit from efforts to align mental health treat-
ments with their cultural values and worldviews. However, culturally adapted
treatments vary in their effectiveness; some are very effective, whereas others
are only slightly better, or even less effective, than nonadapted treatment as
usual (Huey et al., 2014). Effective treatments tend to involve comprehensive
cultural adaptations, as clinicians align their work with the cultural values and
worldviews of the clients.

144 ╇╇╇╇╇ foundations of multicultural psychology

14068-07_CH07_2ndPgs.indd 144 7/28/15 9:07 PM

 8
ACCULTURATION LEVEL AND
PERCEPTIONS OF MENTAL HEALTH
SERVICES AMONG PEOPLE OF COLOR:
A META-ANALYSIS

A number of behavioral and social scientists take the position that: Social
change is accompanied by the intensification of social and cultural sources
of psychological conflict, by new stresses and new adaptation require-
ments in new milieus, and by the loss of the stabilizing effect of old cul-
tural patterns.
—Ari Kiev (1972, p. 9)

As psychiatrist Ari Kiev has pointed out, social and cultural change
intensifies the psychological conflicts typically brought on by the need to
adapt to new and unfamiliar circumstances. Vast societal and organizational
changes in the Western world over the past 400 years have produced corre-
sponding changes in both indigenous and immigrant populations. For indige-
nous aboriginal populations, those changes were invariably imposed through
colonization, legislation, war, and disease. In contrast, immigrants sought
change. Indeed, many individuals from a wide variety of nations immigrated
to the Western Hemisphere to find new lifestyles and opportunities. Change
of any type has clear ramifications for psychology.
Changes require coping strategies. Sudden social changes are likely to
be especially disruptive for individuals. A sudden change in cultural tradi-
tions produces acculturative stress (Berry, 1980), increasing the likelihood

A complete list of references for the studies included in this chapter’s meta-analysis is posted online.
Readers can consult it at [Link]
[Link]
Foundations of Multicultural Psychology: Research to Inform Effective Practice, by T. B. Smith and J. E. Trimble
Copyright © 2016 by the American Psychological Association. All rights reserved.

145
of destructive coping mechanisms such as substance abuse at the individual
level and massive disruption of cultural values and norms at the group level.
Among the most obvious and most frequently reported acculturation
consequences are societal disintegration and personal crisis (Berry & Kim,
1986). As long-held cultural worldviews and norms disappear, individuals
may report feeling “lost” amidst the change. At the group level, previous
patterns of authority, civility, and welfare are absorbed. At the individual
level, hostility, uncertainty, and depression may occur; some immigrants may
even backtrack as they struggle with the challenges of acculturation. All
together these changes comprise the negative side of acculturation, which
can adversely affect family relationships, peer relationships, occupation, and
physical health (Haasen, Demiralay, & Reimer, 2008; Mui & Kang, 2006;
Steffen, Smith, Larson, & Butler, 2006; Vinokurov, Trickett, & Birman,
2002). Acculturation processes affect both interpersonal and intrapersonal
functioning.
The challenges for the mental health profession lie in sensitively under-
standing the complexities of acculturation and social change, especially
among people of color who encounter racism and sometimes stark cultural
differences. This learning process for the clinician involves understanding
both the stress of acculturation and the stress individuals may feel from their
view of mental health and mental health treatment. Conceptualizations of
mental health differ across cultures, and non-Western cultures have prac-
tices and beliefs about effective cures that differ from psychotherapy. Thus,
understanding the degree to which the acculturation of clients of color
affects their perceptions of and experiences in mental health services in
North America is essential for practitioners working with those clients (e.g.,
Leong, Kim, & Gupta, 2011).

NARRATIVE REVIEW OF THE LITERATURE

The principal goal of mental health services is to improve client func-

tioning and well-being. Achieving this goal is contingent on several fac-
tors related to the client’s ethnic and racial background. One essential factor
concerns the client’s ability to adapt to sociocultural change, including the
extent to which they (a) are capable of coping with the demands associ-
ated with the acculturative process and (b) understand and utilize Western
methods of mental health treatment. Some basic background information
contributes bases for understanding this point.
Immigrant populations are increasing. In the United States’ last decade
some 700,000 immigrants became naturalized citizens. According to news col-
umnist John Cookson (2012),

146       foundations of multicultural psychology

 In the 1990s, the annual average was 500,000, and in the 1980s it was
200,000 . . . yet as a share of the total population, this is a change from
0.1% of the population becoming naturalized citizens each year in the
1980s to 0.2% now. (p. 1)
In the past few years most of the naturalized citizens have been immigrants
from Mexico, followed in prevalence by those from the Philippines, India, and
China; over half of them live in California, Florida, and New York (Cookson,
2012). Data are inconsistent on mental health clinic visits by the immigrant
populations, but there is a general expectation that some of the new immi-
grants will seek out mental health services: “A 2003 survey supported by the
National Institutes of Health found that a sixth of new legal immigrants . . .
became depressed during the [acculturation] process” (Cookson, 2012, p. 1).
Such diverse clientele will require culturally appropriate services from mental
health professionals.
As immigrant populations have increased, so has interest in accultura-
tion as a psychosocial cultural construct; 93% of publications in PsycINFO
on “acculturation and mental health” occurred since 1990. Acculturation has
emerged as a leading variable in mental health research (Birman, 2011; Birman
& Simon, 2014; Heath, Neimeyer, & Pedersen, 1988). The acculturation pro-
cess, particularly its association with perceptions of and experiences in men-
tal health services, should be a major concern for mental health practitioners
who provide services for immigrant populations and others experiencing rapid
sociocultural change.

Relevant Theory

Acculturation had been defined as “culture change that is initiated by

the conjunction of two or more autonomous cultural systems, [including] the
selective adaptation of value systems [and] the processes of integration and
differentiation” (Social Science Research Council, 1954, p. 974). The most
significant concepts in this 1954 definition were represented by the words
change and adaptation. Subsequent research and exploration of the two pro-
cesses generated different views of the acculturation construct. When con-
sidering such views, acculturation should be differentiated from enculturation,
which is “the process by which a person learns the requirements of the culture
by which he or she is surrounded, and acquires values and behaviors that are
appropriate or necessary in that culture” (Grusec & Hastings, 2007, p. 547).
To these early formulations (which are concerned mainly with cultural
phenomena) has been added a psychological component focusing on changes
that individuals undergo during the acculturation of their group, referred
to as psychological acculturation by Graves (1967). This construct refers to

acculturation level and perceptions      147

the dynamic process of transformation for a particular individual coming
into contact with a new culture. With the variety of personal, community,
and societal factors that shape individual immigrants’ cultural experiences,
psychological acculturation refers to more than the mere passage of time in
a new country or generational status (Birman & Simon, 2014; Schwartz,
Pantin, Sullivan, Prado, & Szapocznik, 2006).
The more traditional definition implies that a cultural group moves
from a native or tradition-oriented state through a transitional stage and pro-
gresses eventually to an “elite acculturated” stage (Spindler & Spindler, 1967).
According to this notion, cultural changes proceed away from one’s own cul-
tural lifeway in a linear manner to culminate in the full and complete inter-
nalization of another culture’s lifeway. Contemporary social researchers have
difficulty with the traditional view, claiming that acculturation is neither a
linear process nor an achievable end, especially if the process occurs during the
initial contact and change period.
Increasing evidence suggests that acculturation is a multifaceted phe-
nomenon. Moderating variables, individual preferences, and the desire for
ethnic affiliation must be factored into the process (B.â•–S.â•–K. Kim & Abreu,
2001; M.â•–J. Miller, 2007; Tsai, Ying, & Lee, 2000; Zea, Asner-Self, Birman,
& Buki, 2003). Thus, contemporary researchers have adopted bidirectional
and multidimensional perspectives. Bidirectional perspectives view accultur-
ation as a process in which elements of both one’s own and the donor culture
are retained and internalized (e.g., LaFromboise, Coleman, & Gerton, 1993;
Mendoza, 1984).
Instead of attempting to isolate an individual on an index that approaches
full assimilation, one must consider the possibility that many available options
depend on the situation. Mendoza (1984) suggested that an acculturating indi-
vidual may reject religious practices, assimilate dress customs, and integrate
food preferences and selective holiday celebrations; one’s acculturative status,
therefore, is best understood from a composite of indices rather than from an
aggregated summative index. Trimble (1988) advocated a similar view, empha-
sizing the intricate recursive relationship among person variables, situational
characteristics, and acculturative patterns; his model emphasized the potency
of contextual and situational variables in determining behavioral, perceptual,
and cognitive appraisals. The context in which acculturation occurs contrib-
utes to the process (B.â•–S.â•–K. Kim & Abreu, 2001; B.â•–S.â•–K. Kim, Atkinson, &
Umemoto, 2001; R.â•–M. Lee, Yoon, & Liu-Tom, 2006). Along with indicating
that acculturation involves seven dimensions (behavioral, affective, cognition,
personality, identity, attitudes, and stress), Berry (1980, 1997; Berry & Sam,
1996) has pointed out that contextual factors such as politics and econom-
ics can significantly influence adaptation (S. Sue & Okazaki, 2009; Tran &
Birman, 2010). Cabassa (2003) noted that the acculturative process depends

148 ╇╇╇╇╇ foundations of multicultural psychology

on one’s preimmigration history, the initiating means and reasons for immigra-
tion, and the postimmigration context.
From these definitions, characteristics of the process can be identified.
First, contact or interaction between cultures must be continuous and first
hand, excluding short-term accidental contact and single cultural practices dif-
fused over long distances. Second, change occurs in the cultural or psychologi-
cal phenomena among the people in contact. Thus, a distinction is apparent
between acculturation as a process and as a state: In the process, dynamic activ-
ity occurs during and after contact, and the result is relatively stable, though
ongoing change may continue.
Mental health workers must recognize that, like group-level phenom-
ena, individual acculturation does not cohere as a neat predictable package.
Groups and individuals vary in their response to and participation in accul-
turative influences; some domains of culture and behavior may be altered
without changes in other domains. For example, attitudes toward the value
of technology may change without parallel changes in associated beliefs and
behaviors. Thus, the process of acculturation is uneven, not uniform in its
cultural and psychological effects.
Much of the acculturation literature tends to emphasize negative and
unhealthy adjustment outcomes. A few researchers, however, point out that
groups that acculturate can adapt successfully to new environments. Individuals
in pluralistic societies develop attitudes about the society as a whole as well
as ways of relating to individuals and groups (Berry, Trimble, & Olmedo, 1986).
Similarly, attitudes of those in the acculturation process largely determine
orientations and perceptions of one’s own group and relationships with other
groups.
The most cited acculturation theory is the work of Canadian cross-cultural
psychologist John Berry. According to Berry (1994, 1995, 1997), individuals
can hold acculturation attitudes toward any of the following dimensions: assim-
ilation, integration, separation, and marginalization. A review of the definitions
of adjustment, adaptation, acculturation, assimilation, and effectiveness, especially
as they relate to intercultural effectiveness, revealed several skills and traits that
contribute to successful or effective adaptation, including ability to communi-
cate, ability to establish and maintain relationships, orientation toward knowl-
edge, capability in linguistics, disposition toward flexibility, realistic view of the
target culture, and willingness to develop cultural empathy (Hannigan, 1990).
Literature review findings also suggest that factors such as dependent anxiety,
perfectionism, ethnocentrism, rigidity, narrow-mindedness, and self-centered
role behaviors can contribute to negative adjustments. Mendenhall and Oddou
(1985) also pointed out that expatriate acculturation is a multidimensional
process that includes dimensions of self-orientation, other-orientation, per-
ception, and cultural toughness. Such factors as relationship development,

acculturation level and perceptions      149

communicative openness, and stress reduction activities can promote effective
adaptation. Groups and individuals can anticipate the effects of the accultura-
tive process if culture contact is imminent.
In a related acculturation theory, Olmedo (1979) proposed that accul-
turation research use a “full-measurement model,” which provides a way to
investigate relationships among multidimensional sets of quantitatively defined
variables. The model utilizes a set of acculturation scales or factors like those of
a cognitive or personality test. This framework allows researchers to determine
(a) the interdependence structure of acculturation variables, (b) the interdepen-
dence structure of cognitive or personality test variables, and (c) the structure
of relationships between the two sets of variables. Thus, research may explore
fully the possible relationships among variables, such as perceptions of and
experiences in psychotherapy.
Acculturation theories have their critics and skeptics. Some contend
that the construct is inadequately defined, the measures are sometimes lim-
ited to specific populations, and the scales may be measuring experiences that
are confounded by multiple life situations (Birman & Simon, 2014; Hunt,
Schneider, & Comer, 2004; Rudmin, 2003, 2009; Rudmin & Ahmadzadeh,
2001; Schwartz, Unger, Zamboanga, & Szapocznik, 2010). Considering the
prevalence of debates and contentiousness of some critics in the field, Escobar
and Vega (2000) suggested,
Until such time as we have clarification of these important matters,
we recommend suspending judgments about the necessity of includ-
ing acculturation measures in peer reviewed research, or presupposing
the meaning and value of acculturation measures in the absence of an
explicit theoretical rationale for their inclusion. (p. 739)
Hence, it is time for a synthesis of research findings on the degree to which
clients’ level of acculturation is associated with their perceptions of and expe-
riences in mental health services.

Summary of Previous Research Findings

Although most contemporary researchers emphasize that accultura-

tive processes are multifaceted, bidirectional, and situational, inspection of
a number of acculturation measures reveals unidimensional linear evalua-
tions: Most measures yield a single score that identifies one’s position on a
nativist–traditionalist–assimilationist dimension. Moreover, some measures
contain as few as three items, whereas a few others contain as many as 43, a
discrepancy that can raise serious validity questions (Rudmin, 2003, 2009).
Acculturation scale content typically includes one theme or a combi-
nation of three themes: natality, behavioral predisposition, and subjective

150       foundations of multicultural psychology

preferences. All three themes can contain items often found in measures of
ethnic identification. Some researchers have identified an ethnic identity
factor embedded in their acculturation scales, and researchers continue to
use items in acculturation scales that also are found in ethnic identity scales
(Trimble, 2005, 2007). Such issues are but a few of the emerging concerns
about measuring acculturation.
In studying and assessing the acculturative process, cross-cultural psycho­
logists prefer to emphasize an individual’s experiences. As a consequence,
the bulk of the studies flowing from this orientation measure psychological
acculturation and attempt to isolate an individual’s cultural orientation
on a bipolar linear continuum. For example, Triandis, Kashima, Shimada,
and Villareal (1986) used a participatory measure of acculturation to assess
the existence of cultural differences in a sample of Hispanic and non-
Hispanic navy recruits. They created a single acculturation score by summing
four items: length of residence in the United States, media acculturation
(e.g., preference for television, radio, and movies), number of non-Hispanic
coworkers, and number of non-Hispanic friends and romantic partners.
Padilla (1980) proposed an acculturation framework that embraces ele-
ments of the contact-participation dimension and one’s perceived loyalty to
one’s own culture. This framework includes 11 dimensions (e.g., language
preferences, name preference for children) to measure loyalty and 15 dimen-
sions to identify cultural awareness. Padilla (1980) concluded that “cultural
awareness is the more general component,” and “ethnic loyalty is the more
tenuous” (p. 65). This model formed the basis of a multifactorial acculturation
model developed by Richman, Gaviria, Flaherty, Biz, and Wintrob (1987)
containing five dimensions: language, customs, ethnic identity, sociability,
and discrimination. Many other multidimensional frameworks and measures
have subsequently appeared (e.g., Cuéllar, Arnold, & Maldonado, 1995;
Suinn, Khoo, & Ahuna, 1995; Zea et al., 2003).
A number of factors can influence the degree to which immigrants
adapt and adjust to the new host cultural environment (Birman & Simon,
2014). One of the interesting questions arising from the literature is the
degree to which one’s own heritage and the host culture’s general heri-
tage contribute to psychological adaptation and adjustment. The research
results are mixed, depending to some extent on the ethnic background of
the immigrant. The degree to which immigrants identify with their cul-
tural and ethnic background can partially influence adaptation, especially
among Hispanic Americans (Rivera, 2007; Rodríguez, Mira, Paez, & Myers,
2007; Torres, 2010). In a few studies researchers found that for some Asian
Americans adjustment was not difficult because they believed they were
more assimilated than others (Hwang & Myers, 2007; Oh, Koeske, & Sales,
2002; C. J. Yeh, 2003).

acculturation level and perceptions      151

 Recently Eunju Yoon et al. (2013) published a meta-analysis of the rela-
tionship among acculturation, enculturation, and mental health. Examining
data from 325 studies, the researchers found that acculturation was positively
associated with mental health outcomes such as self-esteem, life satisfaction,
depression, distress, anxiety, and so forth. In addition, they found that the
use of language and certain behaviors related positively to mental health, as
did the extent to which individuals identified with their cultural background.
The authors concluded that the most favorable acculturation strategy was to
strive for integration.
One of the major findings from a review of the literature suggested that
most researchers rely heavily on correlating acculturation scales. This is a
problem. Specific analyses must be conducted to investigate different dimen-
sions of acculturation and account for possible curvilinear relationships
(Birman & Simon, 2014; Rudmin, 2003). Authors should explicitly report
and emphasize subscale as well as total score findings. Optimally, authors will
identify classes of participants (e.g., cluster analysis) and report additional
differences accordingly. Moreover, measurement should evaluate different
components of the acculturative process rather than drawing limited conclu-
sions from one or two factors. The different components seem likely to relate
differently to mental health treatment, with cultural knowledge and attitudes
mattering less than cultural values and cognitive styles. Finally, most accul-
turation research consists of surveys. Comparison designs that evaluate differ-
ences across contexts are minimal. Yet we know that acculturation contexts
matter.
To summarize, change is an inevitable human and environmental con-
dition, and acculturation is one form of sociocultural change that has clear
implications for mental health, including perceptions of and experiences in
mental health services. Originally identified and conceptualized by anthro-
pologists, acculturation is recognized as central to the research agendas of
psychologists, psychiatrists, sociologists, social workers, and educators. Most
researchers in psychology have attempted to attribute the results of socio-
cultural change to the acculturative process, blaming negative adaptation
and adjustment on acculturation as if it has a direct effect on mental health
outcomes. Such attributions, however well meaning, have confounded the
research process and muddled the field of inquiry. Yet there is no doubt that
when two or more cultural groups come into direct contact, conflicting expec-
tations regarding what constitutes “normal” mental health and its treatment
are inevitable. How immigrants and people of color perceive and experi-
ence mental health services in North America continues to be an impor-
tant and significant research question. The objectives of this meta-analytic
review were to estimate the degree to which clients’ level of acculturation has
been associated with individuals’ perceptions of and experiences with mental

152       foundations of multicultural psychology

health services and to estimate the degree to which study and participant
characteristics have moderated those findings.

META-ANALYSIS OF THE RESEARCH DATA

We reviewed U.S. and Canadian studies that included a quantitative

measure of individuals’ level of acculturation that was statistically associ-
ated with at least one quantitative measure of an aspect of mental health
services (utilization, retention, treatment attitudes, clinical outcomes). We
included studies measuring perceptions of mental health services (e.g., likeli-
hood to seek mental health services, perceived usefulness of mental health
services, evaluations of the helpfulness of mental health therapists), but we
coded studies of individuals’ perceptions separately from studies of actual cli-
ents. We transformed statistics in manuscripts into the common metric of
Pearson’s r, with positive values indicating that greater acculturation was
associated with more favorable experiences in or perceptions about mental
health services. The Appendix of this book contains additional information
about methods of the meta-analysis.

Description of Existing Research Literature

Our literature search yielded 107 studies reporting data on 23,173 indi-
viduals’ perceptions of or experiences with mental health services as a func-
tion of their level of acculturation while residing in the United States or
Canada. As shown in Table 8.1, the topic has been consistently investigated
over the past 2 decades, but few studies have involved actual mental health
clients (k = 15). The modal study involved cross-sectional, correlational data
obtained by convenience from nonclinical samples of young adults. Children,
adolescents, and senior adults were rarely evaluated. Relatively few studies
involved methodologically rigorous designs (i.e., randomly selected individu-
als who have experienced therapy in contrast to matched controls who have
not). Researchers have typically administered a survey of people’s perceptions
of mental health services along with a measure of acculturation. Thus, the
overall results of this meta-analysis primarily represent nonclient attitudes
and perceptions about mental health services.
Notably, only three of the 107 studies involved multiple racial groups;
in contrast to most of the other topics covered in this volume, data collection
for acculturation tended to be limited to participants of a single racial group,
primarily to individuals with Asian or Hispanic/Latino(a) ancestry. Few
studies addressed African Americans, Native American Indians, Polynesian
Americans, or other groups; specifically, the construct of acculturation has

acculturation level and perceptions      153

 TABLE 8.1
Characteristics of 107 Studies of the Association Between Acculturation
and Perceptions of and/or Experiences With Mental Health Services
No. of
Characteristic M studies (k) %
Year of report
Before 1980 6 6
1980–1989 8 7
1990–1999 46 43
2000–2007 47 44
Publication status
Published 73 68
Unpublished 34 32
Sampling procedure
Convenience 96 90
Representative (random selection) 11 10
Research design
Cross-sectional 67 63
Longitudinal 2 2
Comparison groups 18 17
Other (e.g., analogue, archival) 20 18
Sample population
General community members 87 81
At-risk group members 5 5
Clinical populations (in treatment) 15 14
Sample size 171
<50 9 8
50–99 28 26
100–199 37 35
200–399 22 21
400–999 7 6
>1,000 4 4
Participant agesa 27.7
Children (<13 years) 2 2
Adolescents (13–18 years) 3 3
Young adults (19–29 years) 53 52
Middle-aged adults (30–55 years) 43 42
Senior adults (>56 years) 1 1
Participant gender of (% female) 59.1
Participant raceb (%)
African American 2
Asian American 65
Hispanic/Latino(a) American 28
Native American Indian/Alaska Native 1
Other 3
Note. Not all variables sum to the total number of studies because of missing data.
aAverage age category of participants within studies; not all participants within the study would necessarily be in

the category listed. bThe racial composition of participants across all studies was calculated by multiplying the
number of participants within studies by the percentage of participants from each racial group and dividing that
product by the total number of participants.

154       foundations of multicultural psychology

been most often applied by researchers to two racial groups with a recent his-
tory of large-scale immigration to the United States and Canada. In contem-
porary research, the construct of acculturation has not been widely applied
to racial groups indigenous to North America or historically enslaved. This
issue will be covered later in more detail. Overall, research data described
in the remainder of this chapter should be qualified as applying primarily to
Asian and Hispanic/Latino(a) Americans.

Overall Research Findings

Research findings included a high degree of variability. Across the

107 research studies identified, the observed correlations between indi-
viduals’ level of acculturation and their perceptions of or experiences with
mental health services ranged from -0.45 to 0.88; research studies reported
anywhere from a strong inverse relationship to a very strong positive rela-
tionship. In essence, the results were “all over the map” (see Figure 8.1).
On average, across all types of measures of perceptions of or experiences
with mental health services, the random effects weighted correlation with

Figure 8.1. Contour-enhanced funnel plot of effect sizes (Pearson r) by standard error.
Effect sizes ranged from strongly negative to strongly positive, with no consistent
pattern. Although the overall average approximated zero, the results were too variable
to be interpreted in the aggregate.

acculturation level and perceptions      155

acculturation was .08 (SE = .017, 95% CI = [.05, .11], p < .0001). The hetero-
geneity of the findings (I2 = 82.1, 95% CI = [79, 85], Q(106) = 592.7, p < .0001)
rendered problematic the exclusive reliance on a numerical average. Clearly,
multiple factors influenced the association of acculturation with perceptions
of and experiences in therapy. Acculturation is sometimes related to more
favorable attitudes about mental health services and other times (although
less frequently) to less favorable attitudes. Extreme values on both sides occur
but are washed out in the averages. Interpreting the meaning of the data
therefore benefitted from additional analytic steps.

Factors Influencing the Results

Study Characteristics
Studies evaluated different aspects of individuals’ perceptions of or
experiences in therapy, requiring that results be analyzed across the type
of dependent variable used. The three broad types of variables were client
utilization/retention in therapy, individuals’ perceptions of therapy or the
therapist, and client symptom reduction as a function of therapy. Specifically,
24 studies measured client utilization/retention, 86 measured individuals’
perceptions of therapy or therapists, and eight measured client outcomes.1
We conducted a multivariate meta-analysis of these data, using an estimate
of r = .60 for the correlation of effect sizes within studies (to account for
statistical interdependence). The random effects weighted effect sizes were
low for client utilization/retention, r = .09 (SE = .025, 95% CI = [.04, .13]);
for individuals’ perceptions of therapy or therapists, r = .09 (SE = .023, 95%
CI = [.04, .13]); and for client outcomes, r = -.01 (SE = .062, 95% CI = [-.13,
.11]). These three types of data did not differ from one another (p > .10).
To evaluate whether our initial three categories of dependent variables
were too broad, we grouped the data into five specific categories. However,
no differences were found across the more specific categories of evaluation/
outcome used across studies (Q = 6.2, p = .19; see Table 8.2). Studies of mental
health service utilization or retention yielded similarly small results to studies
of individuals’ perceptions/expectations about mental health services. On
average, acculturation was shown to be only minimally related to both actual
involvement in therapy and perceptions about therapy—an important find-
ing in our analyses, given its implications (see also Chapter 5, this volume).
When perceptions about therapy were broken down by the name
of the instrument used, the results remained in the same minimal range

1Of the 107 total studies, two measured both client outcomes and client retention/utilization, two
measured both client outcomes and client perceptions, and six measured both client perceptions and
client retention/utilization.

156       foundations of multicultural psychology

 TABLE 8.2
Weighted Mean Correlations Across Levels of Several
Moderator Variables in Studies of Perceptions of or Experiences
With Mental Health Services as a Function of Acculturation
Variable Qb p k r+ 95% CI
Data source 2.3 .13
Published 34 .04 [-.01, .10]
Unpublished 73 .10 [.06, .14]
Sampling procedure 0.1 .84
Convenience 96 .08 [.04, .12]
Representative (random selection) 11 .07 [-.03, .17]
Research design 3.1 .37
Cross-sectional 67 .09 [.05, .13]
Comparison groups 17 .06 [-.02, .15]
Analogue 13 .00 [-.10, .10]
Others 10 .11 [-.01, .22]
Sample population 0.1 .98
General community members 87 .08 [.04, .12]
At-risk groups or clinical populations 20 .08 [.00, .16]
Participant gender a 1.5 .23
Female 17 .17 [.07, .27]
Male 9 .07 [-.08, .20]
Participant raceb 8.4 .04
African American 4 .08 [-.12, .28]
Asian American 59 .12 [.07, .16]
Hispanic/Latino(a) American 35 .01 [-.05, .07]
Native American Indian/Alaska Native 2 .02 [-.25, .28]
Acculturation level 8.7 .03
Low 16 .17 [.09, .25]
Moderate 47 .08 [.03, .13]
High 11 -.02 [-.12, .08]
Insufficient information 33 .07 [.01, .13]
Acculturation type 14.4 .003
Proxy variable (i.e., language use, time) 10 .02 [-.08, .12]
Ethnic-specific acculturation measure 74 .06 [.02, .10]
General acculturation measure 12 .25 [.15, .34]
Miscellaneous (more than one above) 11 .12 [.02, .22]
Evaluation type c 6.2 .19
Utilization of mental health services 18 .10 [.01, .18]
Retention in mental health services 6 .07 [-.07, .21]
Attitudes/expectations about services 81 .09 [.05, .13]
Client perceptions of therapists 9 -.04 [-.16, .08]
Client outcomes in therapy 8 -.02 [-.16, .12]
Measurec 8.8 .07
ATSPPHS 46 .09 [.03, .14]
CERS 10 -.10 [-.23, .04]
CCCI 5 -.04 [-.22, .15]
CRF 4 .13 [-.07, .33]
Help-seeking 33 .10 [.03, .16]
Note. CI = confidence interval; k = number of studies; Q b = Q-value for variance between groups; r+ = random
effects weighted correlation; ATSPPHS = Attitude Toward Seeking Professional Psychological Help Scale
(E. H. Fischer & Farina, 1995); CERS = Counselor Effectiveness Rating Scale (L. K. Jones, 1974); CCCI =
Cross-Cultural Counseling Inventory (LaFromboise, Coleman, & Hernandez, 1991); CRF = Counselor Rating
Form (Atkinson & Wampold, 1982).
aThis analysis involved only studies with exclusively male or female participants. bThis analysis involved

studies with participants of a single racial group. cThis analysis involved all effect sizes within studies that
contained distinct outcome measures, such that studies could contribute more than one effect size to the
analysis but only one effect size per category.
(Q = 8.8, p = .07; see Table 8.2). Studies using the Counselor Effectiveness
Rating Scale (L. K. Jones, 1974) tended to have negative effect sizes in con-
trast to the positive effect sizes of the Attitude Toward Seeking Professional
Psychological Help Scale (E. H. Fischer & Farina, 1995) and general measures
of help seeking for mental health. These findings may not be reliable, given
the small number of studies using several of the instruments.
Differences were found across the way in which researchers operational-
ized acculturation (Q = 14.4, p = .003; see Table 8.2). Studies using general
measures of acculturation to Western society (which included a variety of
indicators) yielded much higher correlations than all other types of accul-
turation indicators.
The overall results obtained across all 107 studies were found to be con-
sistent over time; the correlation between the year of study and its effect size
was a trivial r = -0.03. As shown in Table 8.2, the results did not differ across
the type of research design used or across the type of participant sampling used.

Participant Characteristics
Studies included a wide variety of participants, so we sought to determine
whether systematic differences in findings could be attributable to participant
characteristics: gender, age, age cohort, race, level of acculturation, aspect of
acculturation reported, and clinical status.
The overall findings were not affected by participant gender. The cor-
relation between studies’ effect sizes and their percentage of female partici-
pants was very small (r = 0.05, p > .10), and there were trivial differences
between studies using exclusively male or female participants (see Table 8.2).
However, the average age of participants did moderate the results (r = 0.25,
p = .004); studies with older participants tended to have higher effect sizes.
Acculturation was more salient to middle-aged and senior adults’ perceptions
of or experiences with mental health services than to those of younger popula-
tions. This finding proved to be independent of age cohort: In a simultaneous
random effects weighted regression, the association of effect size with average
age remained moderate (b = 0.22), but estimated average year of participant
birth (year of study minus average age of participants) did not at b = -0.03.
Minor differences were found across participants’ race. Studies investi-
gating Asian Americans had a somewhat higher average effect size (although
still small) compared with studies investigating other racial groups (Q = 8.4,
p = .04; see Table 8.2).
Participant level of acculturation moderated the overall results (Q = 8.7,
p = .03; see Table 8.2). Studies with participants experiencing relatively low
levels of acculturation (i.e., recent immigrants) had an average effect size
more than double that of studies using more acculturated participants. Thus,

158       foundations of multicultural psychology

lower acculturation levels affect perceptions of and experiences in therapy
more than higher levels.
Overall results of the studies did not differ as a function of the clinical sta-
tus of participants (Table 8.2). The average results across individuals receiving
treatment and those receiving no treatment were virtually identical.
We next sought to determine which of the statistically significant parti�
cipant characteristic variables remained statistically significant in the pres-
ence of the others. We conducted a random effects weighted metaregression
to predict effect size with those variables found to be statistically significant in
the univariate analyses. The model reached statistical significance (R2 = .14,
Q = 19.6, p = .0006).
The only variable that remained statistically significant in the model
was participant level of acculturation (b = -.20, p = .03). To facilitate inter-
pretation of the data, variables entered into both models were centered on
their means. The B value for the variable labeled constant (see Table 8.3) rep-
resents the average effect size one would expect to find if all variables included
in the model had the expected mean value. This value corresponds with the
overall effect size reported earlier (r = .08). Interpretation of the B€weights
(Table 8.3, Column 1) reveals the extent to which each study characteristic
would be expected to influence the observed overall effect size when con-
trolling for the presence of the other variables. Thus, when researchers use
samples of participants with low levels of acculturation, the expected effect
size would be r = .159 (.079 + .08), and when they use samples of participants
with high levels of acculturation, the expected effect size would be r = -.001
(.079 - .08). When people of color have acculturated to Western society,
their level of acculturation is irrelevant to their perceptions of and experi-
ences in therapy. However, when people of color have not acculturated to
Western society, their level of acculturation is modestly associated with their
average perceptions of or experiences in therapy.

TABLE 8.3
Random Effects Regression Weights for Study
Characteristics Associated With Effect Sizes
Variable B SE p b
R 2 = .14 (k = 102)
Constant .079 .018 .0006
Participants’ average age .003 .002 .10 .15
Asian American participantsa .042 .057 .46 .10
Hispanic/Latino(a) American participantsa -.066 .061 .28 -.15
Participants’ level of acculturationb -.080 .037 .03 -.20
Note.╇ SE = standard error. aSamples of participants exclusively from this racial group contrasted with samples
of participants from all other racial groups. bHigher levels of acculturation have higher values.

acculturation level and perceptions╇╇╇╇╇ 159

Possible Influence of Publication Bias
Most studies (73 or 68%) in this meta-analysis were doctoral disserta-
tions. It seemed improbable that the remaining 34 published manuscripts
could have substantively affected the overall results in either direction. The
average effect size obtained across published versus unpublished dissertations
did not differ (Table 8.2).
As shown in Figure 8.1, effect sizes were widely distributed around the
average value of r = .08. There were no obviously “missing” corners in the
distribution that would suggest studies excluded unintentionally from analy-
ses. One statistical method (Duval & Tweedie, 2000) did not identify any
“missing” studies, and Egger’s regression test was nonsignificant. Therefore,
we concluded that publication bias was not a threat to the results of this
meta-analysis.

DISCUSSION AND INTERPRETATION OF THE FINDINGS

What Do the Overall Results Suggest?

In terms of broad research findings, participant level of acculturation

is inconsistently associated with perceptions of and experiences with mental
health services. The observed correlations ranged from -0.45 to 0.88, suggest-
ing a completely unreliable relationship. However, this meta-analysis identified
several explanations for the observed inconsistencies.
The primary explanation for the overall variability is participant accul-
turation level. Low acculturation is more strongly associated with attitudes
about and experiences with mental health services than moderate and high
levels of acculturation. Because only 16 of the 107 research studies evaluated
only participants with low levels of acculturation, those data were averaged
out, such that the overall association with acculturation appeared to be mini-
mal. However, acculturation is somewhat relevant to mental health services
when it is low (r = .17), not half so much when it is moderate (r = .08), and
not relevant at all when it is high (r = -.02).
Even among populations with low levels of acculturation, the average
effect size of r = .17 is not exceptionally strong. Mental health services pro-
vided for clients of color may or may not be inconsistent with their tradition-
ally held views about psychological disorders and their melioration. Averaged
results reveal little about the underlying nature of the mental health treat-
ments and the participants’ experiences with them. Nevertheless, the find-
ings of this meta-analysis suggested that acculturation level is more relevant
to client utilization/retention and client perceptions of therapists than it is to

160       foundations of multicultural psychology

client treatment outcome. More to the point, clients with low acculturation
levels who engage in a mental health treatment and remain in that treat-
ment tend to improve about the same as clients with higher levels (see also
Chapter 5, this volume).
Other explanations for the variability in the overall research findings
include three factors related to level of acculturation. First, the observed
effect sizes varied across participant age. Studies with older participants
tended to have higher effect sizes. However, this effect did not remain once
the researchers accounted for the level of acculturation; older populations
are less likely to be highly acculturated than younger groups. Second, the
findings suggested a trend for acculturation to be more relevant to percep-
tions and experiences with mental health services by Asian Americans than
by Hispanic/Latino(a) Americans. This finding could be attributable to dif-
ferences in Western and Eastern worldviews, which have distinct cultural
explanations for and stigma about mental illness, disclosure of personal prob-
lems, and so forth (Leong et al., 2011). Third, the way in which acculturation
was measured influenced the data, with studies using general acculturation
measures yielding higher correlations than all other acculturation indica-
tors. This particular finding is both easy and difficult to explain. On one
hand, a psychometrically robust instrument directly measuring the intended
construct of acculturation should be more predictive than proxy variables
such as time of residence in the United States, English language proficiency,
and so forth. Indirect measures of acculturation typically function less effec-
tively than direct measures. On the other hand, it is difficult to explain why
general measures of acculturation would be more predictive of mental health
attitudes than measures of acculturation that were designed for a specific
ethnic group. This finding may be due to systematic differences in measure-
ment or to a plausible curvilinear association that remained unaccounted for
in correlational studies.

Considerations for Future Research

The findings of the meta-analysis present several challenges for research.

Overall, the quality of research has to improve. The fact that two thirds of
the studies on this topic are unpublished dissertations reflects the widespread
use of surveys and correlational analyses that frankly yield limited informa-
tion. At the very least, researchers using cross-sectional designs should have
attended more to curvilinear relationships, rather than limiting themselves
almost exclusively to linear associations. We now have conclusive evidence
that research findings vary as a function of participant acculturation level, so
future research must account for this relationship. And the focus of inquiry
should shift to individuals with low levels of acculturation.

acculturation level and perceptions      161

 Optimally, researchers interested in mental health issues would have
examined clinical populations, yet only 14% did. The field of multicultural
psychology does not need any more surveys about attitudes in the general pop-
ulation that involve a few hundred participants and therefore lack external
validity. The field does need many more studies of actual client experiences.
Those studies can include qualitative investigations and case studies to illus-
trate the multiple ways in which acculturation level relates to mental health
services. We also recommend that future research examine causal pathways:
for example, evaluating individuals entering a medical clinic or social agency
and then contrasting the individuals who do and do not follow through with
a subsequent referral for mental health treatment. Until researchers system-
atically investigate causal pathways and collect data in clinical settings, we
will lack useful information about acculturation and mental health services.
Most research studies included people of Asian and Hispanic/Latino(a)
descent because those populations are often influenced by relocation and
immigration experiences. Native American Indians were the least repre-
sented ethnic population in the sample, yet many tribal members continue to
experience assimilation and adaptation difficulties and struggles. The struggles
continue because many want to retain and maintain their historical lifestyles
yet have to struggle with laws and conditions imposed by federal and state
government regulations, ongoing historical oppression and discrimination,
and the imposing lifestyles of the dominant culture. The same can be said
about Polynesian Americans and Arab Americans. There is a pressing need
for research with those populations.
Researchers should also consider increasing the number of research par-
ticipants. About 60% of the study samples involved fewer than 200 partici-
pants, but the research findings were more consistent in studies with more
than 300 participants. Using larger samples will also allow researchers to
better examine factors that account for inconsistent findings: for example,
differences that have been found across age; type of measure used and pos-
sible item bias or culturally uneven item translations; length of time in the
host country; and variable personal, social, and economic experiences with
relocation and contact with mainstream society.
Most critical, however, scholars have to ask more specific research ques-
tions and use methods and measures that will yield specific answers. The field
will not benefit from additional scattered studies (Figure 8.1) that involve
broad questions using generic measures. More specificity is needed with out-
come variables and with ways measures of acculturation are operationalized,
analyzed, and interpreted. If a research question concerns experiences in men-
tal health treatment, then measuring generic acculturation status is totally
insufficient; researchers must specifically focus on individuals’ acculturation
to Western worldviews about mental health and mental health treatments.

162       foundations of multicultural psychology

That someone is acculturated to the U.S. education system (adolescents) or
workplace (adults) cannot be assumed to indicate that the individual perceives
mental health and well-being in ways congruent (or incongruent) with main-
stream mental health service providers. Researchers should measure precisely
what needs to be known. And what mental health professionals need to know
is the specific cultural attitudes about mental health, mental health treatments,
and circumstances affecting well-being.

Suggestions for Practitioners

A therapist cannot assume that a client with low levels of acculturation

will understand or appreciate the mental health treatment being offered. A
mental health provider must seek to understand the cultural contexts and
unique cultural characteristics of each client, extending his or her efforts
beyond what is typical. Salzman (2001) recommended,
[Therapists must] respect culture as a necessary psychological defense
and design interventions accordingly; promote interventions empha-
sizing meaning construction at the community level and support the
collective (community) and individual construction of meaning that
sustains adaptive action; support and assist individuals and communities
in the identification of standards and values within the cultural world-
view they identify with that promote adaptive action in current realities;
and support and assist communities in cultural recovery. (pp. 189–190)
The findings of this meta-analysis suggest that therapists and clinics should
attend to client utilization/retention. If a clinic is presently not serving clients
with low levels of acculturation, focusing on this population might present a
significant opportunity; outreach and recruitment efforts will increase clien-
tele for the clinic and meet a critical need for those individuals.
This meta-analysis also suggests that acculturation is a multifaceted
phenomenon. Viewing acculturation as a univariate construct may result
in an inconsistent prediction of how a client will respond in treatment. A
clinician may find it optimal to discuss directly with clients their experiences
and expectations about mental health treatments, rather than assuming any-
thing. Clients unfamiliar with North American mental health treatments
and psychotherapy may (a) hold excessively high expectations (e.g., of an
instant cure), which could result in unmet expectations and dissatisfac-
tion or (b) mistrust the methods, either failing to engage or misrepresent-
ing their level of engagement to the therapist. Acculturation that has a
minimal impact on a client’s perceptions about treatment may still be highly
relevant to the presenting symptoms, patterns of coping, and contextual life
circumstances (Yoon et al., 2013).

acculturation level and perceptions      163

 CONCLUSION

Low acculturation, often found among immigrants and older adults, is

relevant to perceptions of and experiences in mental health services. Among
potential clients who are moderately to highly acculturated, the association
varies too much to predict attitudes about mental health services. Future
research using higher quality methods and more specific research questions
and measures will be necessary before other conclusions can be reached.
Much work remains for scientists to understand the processes of accul­
turation, its relationship to social and sociocultural change, and the experi-
ences of multicultural populations with mental health services. Our findings
raise a few pointed questions for consideration. Which change-related con-
structs best describe acculturation processes relevant to mental health treat-
ment utilization and outcomes (e.g., psychological acculturation, acculturative
stress, sociocultural change, cultural borrowing/fusion)? Which cultural beliefs
are most relevant to mental health and mental health services? Precisely when
and how do North American mental health treatments align and not align
with cultural worldviews of clients of color? Considering the contents of this
chapter and others in this volume, relying on simplistic descriptions and mea-
surements of acculturation and social change constructs would be insufficient
and shortsighted. The challenges are significant, but so are the opportunities
for improvement.

164       foundations of multicultural psychology

 III
SYNTHESIS OF RESEARCH
ON THE EXPERIENCES
AND WELL-BEING OF
PEOPLE OF COLOR
 9
THE ASSOCIATION OF RECEIVED
RACISM WITH THE WELL-BEING
OF PEOPLE OF COLOR:
A META-ANALYTIC REVIEW

A chapter on racism in the Handbook of Multicultural Psychology begins

as follows:
This chapter should not be here. In a more perfect world, a handbook
celebrating the influence of and importance of multiculturalism within
psychology would be complete without a specific analysis of prejudice
and racism. The world’s imperfection, however, makes this chapter an
unfortunate necessity that can inform and improve us. (Czopp, Mark, &
Walzer, 2014, p. 361)
Although the nature of racism has changed over time (Yoo & Pituc,
2013), it is engrained in many spheres of activity (D. W. Sue, 2005, 2013,
2015). Thus, this chapter, like that of Czopp et al. (2014), is critical because

Dr. Hokule’a Conklin of Brigham Young University contributed to the writing of this chapter.
A complete list of references for the studies included in this chapter’s meta-analysis is posted online.
Readers can consult it at [Link]
[Link]
Foundations of Multicultural Psychology: Research to Inform Effective Practice, by T. B. Smith and J. E. Trimble
Copyright © 2016 by the American Psychological Association. All rights reserved.

167
we need to more thoroughly understand the influence of racism and prejudice
on mental health and well-being.
Racism and prejudice have always included both ideological and insti-
tutional dimensions (Neville, Spanierman, & Lewis, 2012) affecting multiple
aspects of life, including employment, education, health care, and housing
(National Research Council, 2004; U.S. Department of Health and Human
Services, Office of the Surgeon General, 2001). Considerable evidence has
revealed that multiple inequities attributable to racism have negative health
consequences for people of color (Paradies, 2006). Racism clearly affects indi-
viduals’ mental health and emotional well-being (Carter, 2007) and thus
should be a concern for mental health professionals.
Scholars have increasingly investigated the psychological consequences
of racism. The resulting data counter the false supposition that racism no
longer poses a significant mental health risk for people of color (Constantine
& Sue, 2006). Such data should embolden mental health professionals to
oppose racism in all of its various forms (D. W. Sue, 2005, 2013).
This chapter summarizes research specific to the association between
experiences of racism and psychological well-being among people of color
in the United States and Canada. Mental health professionals who attend
to the adverse effects of racism can bolster the well-being of the individuals,
groups, and communities they serve.

REVIEW OF THE LITERATURE

Relevant Theory

Racism and its psychological consequences can be conceptualized in

many ways. Several decades ago many theories focused on psychological pro-
cesses. For instance, social comparison theory (Festinger, 1954) emphasized
how self-evaluations are generated by contrasts with others. Individuals and
groups whose race is judged negatively by others can internalize those nega-
tive evaluations (i.e., internalized racism). Although the basic components
of Festinger’s (1954) theory remain intact, significant contributions and
advancements have extended to such variables as self-enhancement, self-
evaluation maintenance, and closure avoidance, as well as various aspects of
self and other attributions (Garcia, Tor, & Gonzalez, 2006; Suls, Martin, &
Wheeler, 2002; Tesser & Campbell, 1982).
Other theories, such as relative deprivation theory (Walker & Smith,
2001), have addressed how self-evaluations differ as a function of inequitable
distribution of power and resources. Intraindividual processes, such as resil-
iency factors, have also been addressed by the transactional model of stress
and coping (Lazarus & Folkman, 1984), which posited that abilities to cope

168       foundations of multicultural psychology

with stressors, such as racism, mediate an individual’s appraisal of the racism
and corresponding distress and responses to it. Nevertheless, scholars have
noted that models emphasizing psychological interpretations of racism fail
to account for other factors: “Research and theory on stress and coping . . .
has fallen short of comprehensively capturing experiences and characteris-
tics that emerge from person–environment transactions involving race and
culture” (Harrell, 2000, p. 44).
In the past, psychological theories that emphasized the internalization
of social experiences were clearly distinguished from sociological theories
that emphasized systems and structures (i.e., inequitable access to resources).
Contemporary conceptualizations attend to both and are termed the psycho-
social, or “the interface between socially structured arrangements and intra-
individual processes” (Schnittker & McLeod, 2005, p. 77). For example,
a biopsycho­social model (R. Clark, Anderson, Clark, & Williams, 1999)
explained differences in health outcomes by allowing for differences in indi-
viduals’ perceptions, coping responses, environments, and level of exposure
to racism.
Prominent among contemporary models, critical race theory (Delgado
& Stefancic, 2012) posits that all structures that contribute to racial stratifica-
tion place racial minority individuals at greater risk of various mental health
problems (Brown, 2003). Five tenets of this theory are (a) racial stratification
is ubiquitous, (b) racism is complex and difficult to remedy, (c) race is a social
invention, (d) a phenomenological understanding of oppression is appropri-
ate, and (e) critical race theorists should try to bring about social justice.
The emphasis of this theory on dynamics of power and privilege has made it
widely influential in multicultural psychology. Building on that foundation,
Neville et al.’s (2012) expanded psychosocial model of racism illustrated how
social structures and ideologies intersect and maintain racism over time.
Contemporary models thus emphasize a variety of mechanisms through
which racism affects mental health, with some factors receiving substantial
research attention. Steele’s (2010) landmark work on stereotype threat has
attracted the interest of the general public. Overall, the mental health profes-
sions are better positioned than ever to expand research investigations into
how racism affects psychological well-being and to identify the degree to
which other variables mediate and moderate that relationship (Harrell, 2000;
Howe, Heim, & O’Connor, 2014).

Narrative Review of Previous Research

The increased attention to issues of racism in the recent research lit-

erature has been notable. One estimate indicated that from 1960 to 1970
only 115 studies of experiences of racism by racial or ethnic minority group

the association of received racism      169

members were published, whereas at least 4,669 studies on the topic appeared
between 2000 and 2010 (Yoo & Pituc, 2013).
The growing corpus of research has already settled on one conclusion:
The association of racism with well-being is complex. “Discrimination is, by
its very nature, a subjective experience. By implication, reports of discrimina-
tion are subjective and depend on a complex process involving the perception,
recall, and reporting of past life experiences” (Schnittker & McLeod, 2005,
p. 90). Such complexity creates challenges for measurement and methodology.
For obvious ethical reasons, the scientific study of racism should not involve
experimental manipulation of racist encounters (A. R. Fischer & Shaw, 1999).
The majority of research published in this area has been correlational and cross-
sectional. These kinds of studies cannot isolate the effects of racism and do not
easily distinguish among alternative explanations for decreased well-being.
Reliance on correlational research involving perception data has raised
some questions about bidirectionality of effects. “Although many have argued
that experiencing racial discrimination has deleterious consequences for mental
health, it is also plausible that individuals experiencing lower levels of mental
health are more inclined to interpret ambiguous events as being racially moti-
vated” (Sellers & Shelton, 2003, p. 1081).
Individuals who are harmed by racism can acquire a protective sensitiv-
ity (Comas-Díaz & Jacobsen, 2001), which can be better understood in terms
of race-based traumatic stress injury (Carter, 2007). Mental health profes-
sionals do not blame an individual who was raped for developing a protective
sensitivity; similarly they should not speak of racism in ways that shift blame
to people who are disadvantaged by it.
Although expressions of racism in North America are considerably
less overt than in earlier decades, racism remains pervasive and problematic
(D. W. Sue, 2015; Zárate, Quezada, Schenberger, & Lupo, 2014). Racial
hierarchies and racial microaggressions (D. W. Sue et al., 2007) can result in
anger, exhaustion, withdrawal, anxiety, and diminished self-esteem, among
other adverse mental health symptoms, collectively described as racial battle
fatigue (W. A. Smith, Allen, & Danley, 2007). It seems obvious that discrimi-
nation of any kind would have consequences that are primarily harmful, even
though some consequences, such as increased resilience and willingness to
combat injustice on behalf of others, could also accrue (D. W. Sue, 2005). At
least four published literature reviews concluded that perceptions of racism
are associated with decreased physical and psychological well-being (Carter,
2007; Paradies, 2006; D. R. Williams & Mohammed, 2009; D. R. Williams,
Neighbors, & Jackson, 2003).
Four prior meta-analyses of the literature also found fairly consistent
results. Across 23 studies specific to Asians and Asian Americans, the cor-
relation between racial discrimination and mental health variables was r = .23

170       foundations of multicultural psychology

(D. L. Lee & Ahn, 2011). A separate meta-analysis of 66 studies of Black
American adults reported an average correlation of r = .20 between perceived
racism and mental health variables (Pieterse, Todd, Neville, & Carter, 2012).
In a broader meta-analysis (Pascoe & Smart Richman, 2009), the average
effect size was r = .16 across 105 studies after accounting for publication bias.
In the fourth and most comprehensive meta-analysis (Schmitt, Branscombe,
Postmes, & Garcia, 2014), racism correlated -.21 with well-being across
211 studies using correlational designs. The fourth meta-analysis also evalu-
ated 11 experimental studies of racism, in which the average difference in well-
being attributable to the conditions of racism was d = .11, a very modest value
relative to the correlational data.
Across the four meta-analyses, different variables were found to moder-
ate the overall association between psychological well-being and perceptions
of racism. In the meta-analysis that examined several possible moderating
variables (Pieterse et al., 2012), only the type of well-being measured mod-
erated the results. In another (Pascoe & Smart Richman, 2009) the type of
measurement of well-being did not moderate the results. The third (D. L. Lee
& Ahn, 2011), which identified significant associations with resources and
coping among Asian and Asian American populations, found some differ-
ences between studies involving participants from several ethnic backgrounds
and studies involving only individuals with ancestry from China, India, or
Vietnam. In the most comprehensive meta-analysis (Schmitt et al., 2014),
the results from disadvantaged populations were of greater magnitude than
those of relatively advantaged populations. Effect sizes obtained from African
Americans were lower than those of other racial groups. No differences were
observed across participant age, but measures of self-esteem yielded effect sizes
that were about half the magnitude (r = -.13) of those involving measures
of psychological distress (r = -.25). Pervasive experiences of discrimination
proved more problematic than single events, as would be expected. Notably,
the authors also found evidence for causality, through the results of longitudi-
nal studies and studies in which the discrimination was manipulated.
Presently, we have not confirmed whether the association between rac-
ism and well-being truly differs across several study characteristics—such as
the sampling procedures used, the research design implemented, the method
for measuring racism, or the time period in which the study was conducted—
with results possibly changing over several decades. We also cannot discern
whether the results differ as a function of participant gender, clinical status,
level of acculturation, or socioeconomic status.
To address these important considerations, we present the results of a
meta-analysis that we had conducted prior to the publication of the four already
described. Our meta-analysis involved moderation analyses not reported in the
other four, but we analyzed only 81 studies. This lower number of studies was

the association of received racism      171

due in part to our use of different screening procedures: (a) restricting inclu-
sion to populations in the United States and Canada; (b) excluding measures
of perceived racism that included the construct of distress (which would arti-
ficially inflate the magnitude of the association between perceived racism and
psychological well-being); and (c) including only studies with measures of
individuals’ personal experiences, excluding generic measures of people’s per-
ceptions of racism in society. We recently learned about a fifth meta-analysis,
not yet published, being conducted by Robert Carter (mentioned in Carter &
Pieterse, 2013). We regret our delay in publishing our analyses because of per-
sonal circumstances and the amount of time required to compile all chapters
in this book, but we are pleased that data synthesis is occurring with increased
frequency in multicultural psychology; the field will benefit from increased
attention to data. After presenting the data, we summarize the findings in a
subsequent section.

QUANTITATIVE SYNTHESIS OF RESEARCH DATA

In this section we describe our meta-analysis of the literature evaluating

the association between racism and well-being among people of color living in
the United States or Canada. We evaluated studies with a quantitative measure
of individuals’ experiences with or perceptions of racism (using the terms
racism or racial/ethnic prejudice, discrimination, or oppression) that was statisti-
cally associated with at least one quantitative measure of a component of
personal well-being (i.e., mental health, self-esteem). However, to avoid con-
ceptual overlap we excluded studies using measures of racism that had embed-
ded the construct of emotional distress in its items or scaling (e.g., evaluations
of how stressful the individuals felt as a result of the racism they encountered).
This procedure was essential to distinguish the occurrence of racist events from
emotional reactions related to well-being. Statistical estimates in manuscripts
were all converted to Pearson’s r using meta-analytic software, with negative
values indicating that more encounters with racism were associated with lower
levels of well-being and positive values indicating that more encounters with
racism were associated with higher levels of well-being. The general methods
of this meta-analysis are reported in the Appendix to this book.

Description of the Existing Research Literature

We analyzed 81 studies reporting data on a total of 44,158 individuals’

psychological well-being associated with their self-reported experiences with or
perceptions of racism. Only three studies involved actual mental health clients
(see Table 9.1). All age groups except adults over age 55 were represented.

172       foundations of multicultural psychology

 TABLE 9.1
Characteristics of 81 Studies of the Association Between
Received Racism and Well-Being
No. of
Characteristic M studies (k) %
Year of report 2003
Before 1980 0 0
1980–1989 1 1
1990–1999 13 16
2000–2010 67 83
Publication status
Published 69 85
Unpublished 12 15
Research design
Cross-sectional survey 66 82
Longitudinal survey 13 16
Archival 1 1
Comparison groups 1 1
Sample type
Community members 51 62
University students 24 29
Clinical clients 3 4
Mixed sample (more than one of the above) 4 5
Sample size 545
<50 1 1
50–99 8 10
100–199 30 37
200–399 25 31
400–999 7 9
>1,000 10 12
Participant age a 25.9
Children (<13 years) 4 5
Adolescents (13–18 years) 21 27
Young adults (19–29 years) 26 34
Middle-aged adults (30–55 years) 26 34
Senior adults (>56 years) 0 0
Participant gender (% female) 61.5
Participant raceb (%)
African American 49
Asian American 17
Hispanic/Latino(a) American 25
Native American 2
Other/combined groups 7
Note. Not all variables sum to the total number of studies because of missing data.
aAverage age category of participants in studies is given, though not all participants in the study would

necessarily be in the category listed. bThe racial composition of participants across all studies was calculated
by multiplying the number of participants within studies by the percentage of participants from each racial
group and dividing that product by the total number of participants.

the association of received racism      173

Half of the research studies were specific to African Americans, but no studies
were specific to Polynesian Americans, American Indians, or Alaskan Natives.
The typical study involved cross-sectional correlational data obtained
from convenience samples of participants who completed a questionnaire
containing measures of experiences with or perceptions of racism received
and one or more measures of psychological well-being. Thus, the overall
results provided an estimate of the correlation expected from a survey about
racism and psychological well-being administered to people of color.

Overall Research Findings

Across all types of measures of psychological well-being, the random

effects weighted correlation with all types of measures of racism was -.183
(SE = .003, 95% CI = [-.21, -.16], p < .0001). The heterogeneity of the find-
ings was very large (I2 = 84.1, 95% CI = [81, 87], Q(80) = 503.4, p < .001).
Across the 81 studies, the observed correlations between indicators of indi-
viduals’ well-being and their perceptions of racism ranged from -0.46 to 0.09,
with about two thirds being between -.30 and -.10. On average, reports of
racism were modestly associated with well-being, but the data were so incon-
sistent as to caution against reliance on the overall average. We examined
several factors that could possibly explain the variability of the findings.

Factors Influencing the Results

Possible Influence of Publication Bias

As shown in Figure 9.1, effect sizes were scattered in a circular pattern
around the average value of r = -.18, whereas typically such data take the shape
of an elongated pyramid. This scattered distribution could possibly have indi-
cated “missing” studies due to publication bias. Egger’s regression test (a method
to evaluate data symmetry) did not indicate publication bias, but one statistical
method for estimating publication bias (Duval & Tweedie, 2000) identified
12 “missing” studies from the distribution. When those 12 studies were
accounted for (“filled”), the newly computed effect size of r = -.16 remained
statistically significant (p < .01, 95% CI = [-.18, -.13]) and similar to the
previously reported overall average. Therefore, we concluded that even though
about a dozen studies had likely been missing from the meta-analysis, correc-
tion for their absence changed the results to a very small degree, such that
publication bias was only a minor threat to interpretation of the results.

Study Characteristics
Meta-analyses can detect differences in effect size across research
study characteristics. We considered the influence of year of study publication,

174       foundations of multicultural psychology

Figure 9.1. Contour-enhanced funnel plot of effect sizes (Pearson’s r) by standard
error for 81 studies of the association of perceived racism with well-being. The
results were very inconsistent, irrespective of sample size. Moreover, there were
fewer than expected studies with nonsignificant results (fewer dots in the white area
adjacent to the left shaded portion of the graph) because the data should typically
fall in the form of a pyramid, not the scatter depicted here. This distribution suggested
that some studies may have been missing from the meta-analysis, with the overall
average effect estimate adjusted to r = -.16.

research design, statistical controls, type of measure of racism, and type of

well-being measured.
The results obtained across all 81 studies were found to be relatively
stable over time (the correlation between the year of a study and its effect
size was r = -.13, p = .13). Nevertheless, the majority of studies have occurred
in the very recent past, and substantive social changes typically require more
time for effects to be observed.
The results differed (p = .01) across the type of participant sampling,
with eight large-scale surveys of participants randomly selected from the pop-
ulation having lower averaged results (r = -.11) than 67 studies using conve-
nience sampling (r = -.19) and six studies involving random selection from a
local setting or sampling with greater than 60% nonparticipation (r = -.23).
Correlation values obtained from 66 cross-sectional studies averaged
r = -.18, which was essentially the same as the averaged value of r = -.19

the association of received racism      175

obtained from 13 longitudinal studies. Moreover, the 12 studies that statistically
controlled for possible confounds (age, education, gender, race, and socio­
economic status) obtained similar averaged results (r = -.17) to the 69 studies
that did not control for any other variable (r = -.19).
No differences were observed across the way racism was measured: The
averaged results did not vary when racism was operationalized in terms of
perceptions, incidence or frequency recall, or combined measurement meth-
ods. However, statistically significant differences were observed across the
type of well-being that was measured. On average, the association of racism
with adverse mental health symptoms (e.g., depression, anxiety) was notably
stronger than with measures of positive well-being (e.g., self-esteem, life sat-
isfaction; r = -.22 vs. -.12, p = .01).

Participant Characteristics
Characteristics of research participants varied widely across research
studies, and we analyzed the possible influence of participants’ average
age, estimated birth year (for possible cohort effects), gender composition
(% female), type of population (community members, students, clinical
samples), estimated level of acculturation, estimated socioeconomic status,
and race.
The results did not differ across race (p > .10). The findings of 32 studies
specific to African Americans were similar (r = -.18) to the findings of 16 studies
specific to Asian Americans (r = -.20) and to the findings of 16 studies specific
to Hispanic/Latino(a) Americans (r = -.16); the percentages of participants
from each racial group were unrelated to the study’s results. Furthermore, no
significant differences were found across participant age, estimated birth
year, gender, education level, acculturation level, or socioeconomic status.
Only three studies investigated clinical populations, which averaged r = -.25,
and the averaged results from community members and university students
averaged r = -.175.

DISCUSSION AND INTERPRETATION OF THE FINDINGS

Overall Findings

Abundant data has confirmed that experiences of racism are negatively

associated with individuals’ self-reported well-being (D. L. Lee & Ahn, 2011;
Pascoe & Smart Richman, 2009; Pieterse et al., 2012; Schmitt et al., 2014).
Our overall results were slightly less robust than those reported in the larger

176       foundations of multicultural psychology

meta-analysis by Schmitt and colleagues (2014), and after adjusting for pos-
sible publication bias (r = -.16) our results were identical to those of Pascoe
and Smart Richman (2009), who also adjusted for publication bias. To inter-
pret, a correlation value of that magnitude is equivalent to the association
between socioeconomic status and well-being (Pinquart & Sörensen, 2000)
and also to the overall association between personality variables and well-
being (DeNeve & Cooper, 1998). The association is not nearly as large as
that between physical health and well-being (r = .32; Okun, Stock, Haring,
& Witter, 1984), but it remains consequential. Nevertheless, the research
findings are highly variable, so we cannot accurately predict how any particu-
lar group, let alone any single individual, will interpret and respond to racism.

Secondary Analyses

The primary contributions of this particular meta-analysis involve the

investigation of potential moderating variables either unexamined in previ-
ous meta-analyses or inconsistently identified across meta-analyses. One of
our notable findings involved an analysis of how well-being was measured in
studies. Although Pascoe and Smart Richman (2009) did not find statisti-
cally significant differences across the form of well-being measurement used
in studies, our results were essentially identical to those of the meta-analysis
of Schmitt et al. (2014) and Pieterse et al. (2012), who both found that the
association between perceived racism and positive aspects of well-being (e.g.,
self-esteem, quality of life) was much lower than the association when mea-
surement involved negative well-being indicators (e.g., distress). We believe
that these data are sufficiently robust to support the assumption that negative
aspects of well-being are indeed more strongly associated with perceptions of
racism than are positive aspects. We conjecture that this relationship exists
because racism is a distressing series of events that overlap experientially with
states of distress more than with overall life satisfaction or well-being.
Our meta-analysis confirmed no differences across participant race, gen-
der, age, and socioeconomic status; the association between racism and well-
being is independent of these variables. However, results differed between
large-scale surveys with strong external validity and surveys with high rates of
participant attrition. This finding suggests that the results of surveys with high
participant attrition are suspect, due to possible selection bias: Individuals
acutely affected by racism may be more likely to participate in the research
than individuals with less distress ascribed to racism. Thus, even though the
size of the correlation between perceived racism and well-being has remained
fairly consistent across four meta-analyses, that association is inflated by the
inclusion of studies with high rates of participant attrition.

the association of received racism      177

Considerations for Future Research

Our findings suggest that research quality does matter. When the vast
majority of research is correlational, a design highly susceptible to bias, efforts
to reduce bias become even more essential. Specifically, if the people partici-
pating are more invested in the topic of racism (i.e., have had more negative
experiences with racism) than nonparticipants, why should we trust the results?
We also found that relatively few studies statistically controlled for possible
confounds. Although those few studies with statistical controls yielded similar
results to those using unadjusted data, in the future researchers should account
for obvious sources of bias, and journal editors are encouraged to reject manu-
scripts with sources in which bias remains uncontrolled, particularly high rates
of nonparticipation.
Measurement precision typically enhances the consistency of results.
Unfortunately, the measures of racism in studies we examined were inconsis-
tent in how participants were instructed (or not instructed) about the mean-
ing of racism and what time period was evaluated (lifetime incidence vs.
recent time periods). We cannot help but wonder whether the large hetero­
geneity of the findings was attributable to measurement issues such as cul-
tural measurement equivalence and item bias (Trimble & Vaughn, 2013).
We found that cross-sectional studies yielded essentially the same results as
longitudinal studies (r = -.18 and -.19, respectively), indicating that the
relationship between racism and well-being did not vary due to being mea-
sured concurrently or at a later time, perhaps because most measures were
already retrospective, measuring encounters with racism distanced in time.
We understand the convenience of using retrospective methods; however,
the field now has more than enough studies examining racism in the undeter-
mined past, a measure easily confounded by present mood and other factors
affecting recall accuracy. Optimally, research should seek out participants
who have recently experienced racism, provided representative samples can
be obtained.
We were surprised to find few studies of mental health clients. Clinical
research could be helpful in finding explanations for the differences in
results concerning positive versus negative well-being. People vulnerable
to mental illness may also be more likely to be targets of individual acts of
racism. Or people may cope with racism less effectively because of their
existing level of distress. In addition, some clients may present racist and
discriminatory experience as their main concern yet feel as though the ther-
apist does not truly understand the deeply damaging psychological effects
they experience.
Psychological distress is correlated more strongly with racist events
than general life satisfaction. Future research should explicitly confirm that

178       foundations of multicultural psychology

this difference between positive and negative well-being is not solely attrib-
utable to preexisting levels of distress (Allen, Lewis, & Johnson-Jennings,
2015), although we sincerely doubt that preexisting levels of distress com-
pletely account for the severity of the racism perceived.
We found that almost half (49%) of the literature was specific to African
Americans, yet the data suggest minimal differences across race, with the
average effect size being smaller among African Americans relative to Asian
Americans (Schmitt et al., 2014). Race-specific research can be valuable
for many reasons, but such research could better consider the experiences of
other groups, particularly Arab Americans, Polynesian Americans, Native
American Indians, and Alaskan Natives. We also see little value in further
comparisons among groups, as some researchers have a tendency to view the
groups as more homogeneous than they are and thus gloss over deep cultural
variations within groups (Trimble & Bhadra, 2013).
We encourage researchers to examine potential mediating variables,
guided by the work of D. L. Lee and Ahn (2011). These authors found evi-
dence that coping strategies, social support, cultural identity, and personal
strengths influence the association between perceptions of racism and well-
being. Investigating perceptions of racism in isolation, without concurrent
consideration of mediating variables, does not seem particularly effective in
moving the field forward.

Suggestions for Practitioners

If perceptions of racism are associated with well-being to about the same

extent as personality or socioeconomic variables, therapists should consider
clients’ experiences of racism in the same ways they would account for per-
sonality and socioeconomic status. They cannot afford to ignore such experi-
ences. Although individual clients vary widely in their experiences of racism
and their reactions to it, similar variability characterizes personality attributes
(DeNeve & Cooper, 1998), which are already seriously considered by many
mental health professionals.
Given the differences of association between negative and positive mea-
sures of well-being, therapists should particularly attune to and interpret inci-
dents of racism among populations in greatest distress. We assume but cannot
yet confirm (given the correlational nature of the data) that high levels of
distress make racism particularly problematic, both in terms of psychological
impact and in terms of reactive coping. We also assume that the therapist
can better assist clients by framing discussions of racism in terms of resilience
and resistance, neither implicitly blaming individuals for their reactions nor
implicitly assuming client passivity in the face of inequitable and dehuman-
izing situations.

the association of received racism      179

 Therapists should not have to worry about whether a client describes
racism in terms of frequency or severity; no differences were found in the data
across measures of perceptions of severity, incidence or frequency, or a com-
bination. Thus, therapists can consider any method of client self-report; the
associated negative consequences for well-being would be similar.
Finally, therapists should not assume that an individual of any particu-
lar background is more or less vulnerable to racism. It affects all groups. Its
impact is equivalently adverse and equally unpredictable.

CONCLUSION

The findings of the present meta-analysis confirm and extend the results
of other recent meta-analytic reviews: Racism is associated with diminished
psychological well-being across racial and ethnic groups. In the lives of people
of color, racism is as consequential to well-being as socioeconomic conditions
(Pinquart & Sörensen, 2000). Mental health professionals should attend to
clients’ experiences interacting across social groups.
The data we reviewed reminded us of an anecdote shared by a colleague
who studies issues of racism. Media coverage of that person’s research gener-
ated a substantial amount of general public response. One correspondent, who
had apparently experienced racism for decades, congratulated the researcher
but also commented that it seemed bizarre that the scientific community
would only now be validating what was so obvious—painfully obvious. Mental
health professionals and researchers can catch up with the pace of reality.
We conclude with a recommendation for the profession that exceeds the
bounds of the data we reviewed. We believe that mental health practices and
research must be inspired by an objective even more consequential than vali-
dating past and current trauma and injury. Our primary objective must become
to prevent racism (D. W. Sue, 2005, 2013; Zárate et al., 2014). No one would
be satisfied with medical explanations of HIV or cancer that merely described
individual suffering and its correlates. Mental health professionals attend to
suffering, but that focus may have detracted from the work essential to pre-
vention. Mental health professionals are well positioned to take a leading
role in scientific efforts to understand racism in order to prevent it.

180       foundations of multicultural psychology

 10
ETHNIC IDENTITY AND
WELL-BEING OF PEOPLE OF COLOR:
AN UPDATED META-ANALYSIS

For there are nearly as many ways in which such identities, fleeting or
enduring, sweeping or intimate, cosmopolitan or closed-in, amiable or
bloody-minded, are put together as there are materials with which to put
them together and reasons for doing so. . . . [A]nswers people sometimes
give to the question, whether self-asked or asked by others, as to who
(or, perhaps, more exactly, what) they are, simply do not form an orderly
structure.
—Clifford Geertz (2000, p. 225)

Culturally sensitive and multiculturally competent mental health pro-

fessionals often ask about a client’s ethnic and racial identity. Responses to
such queries can strengthen the therapist’s understanding of the worldviews
relevant to the client’s presenting concerns, make explicit the client’s expec-
tations for the treatment process, and enhance the client’s comfort with
the therapist, among other factors. Yet, as cultural anthropologist Clifford
Geertz (2000) suggested, discussions and evaluations of ethnic identity are
complicated because human beings have multiple intertwined identities
that interact and affect one another in ways that are not fully understood.
Unfortunately, answers to the question of ethnic identity “simply do not form
an orderly structure” (Geertz, 2000, p. 225).
Most of the research on ethnic and racial identity has been limited
to the abstraction of race and ethnicity at a social and psychological level

A complete list of references for the studies included in this chapter’s meta-analysis is posted online.
Readers can consult it at [Link]
[Link]
Foundations of Multicultural Psychology: Research to Inform Effective Practice, by T. B. Smith and J. E. Trimble
Copyright © 2016 by the American Psychological Association. All rights reserved.

181
of analysis (Amiot, de la Sablonnière, Terry, & Smith, 2007; Gong, 2007;
Negy, Shreve, Jensen, & Uddin, 2003; Yip, Douglass, & Sellers, 2014). Other
dimensions of an individual’s identity, such as multiple identities, situational
and contextual influences, and role expectations and enactments, are given
less attention in the psychological literature. The widespread interest in the
topic of ethnic identity suggests that we must stop and take stock of the field’s
progress and its future directions.
The growth in the field, with its attendant methodological and proce-
dural problems and theoretical debates, points to the need for inquiry into the
conceptualization and measurement of these seemingly elusive constructs.
Accordingly, this chapter provides an overview of the construct of ethnic
identity, particularly its relationship to psychological well-being, along with
a brief discussion of ethnic identity measures. An updated meta-analysis
summarizes research findings on the association between measures of ethnic
identity and personal well-being. Although there is considerable literature on
racial identity, those articles were not included in our analysis because of the
differences between many racial identity and ethnic identity scales, including
the ways the former are typically operationalized and measured.

NARRATIVE REVIEW OF THE LITERATURE

Origins of Ethnic Identity Constructs

Typically the term identity expresses the notion of sameness, likeness, and
oneness. More precisely, identity means “the sameness of a person or thing at all
times in all circumstances; the condition or fact that a person or thing is itself
and not something else” (Simpson & Weiner, 1989, p. 620). Moynihan (1993)
argued that identity is “a process located in the core of the individual and yet,
also, in . . . his communal culture” (p. 64). It is a powerful conceptualization
that strongly influences one’s personality, sense of belonging and sameness,
and quality of life.
To extend their understanding of identity, most social and psychological
theorists must contend with the concept of self. And to approach an under-
standing of self-concept, one is obliged to provide plausible, if not substantial,
explanations for the following domains: physical traits and characteristics,
personal experiences past and present, social affiliations and integration, and
personal values and behaviors, along with messages received from other peo-
ple about those several issues (Cirese, 1985). Explanations for these domains
consume volumes.
Although the bulk of the social and behavioral studies circumscribed by
the ethnic variables concentrate on ethnic minority populations, generally

182       foundations of multicultural psychology

the literature on ethnicity is actually far more inclusive. Social science inter-
est in ethnicity, especially as a potential explanatory variable, began in the
late 1940s (Cross, 1991; P. A. Katz & Taylor, 1988). Interest was fueled by at
least two fundamental concerns: (a) the contentious notion of America as a
melting pot of different nationalities, ethnic groups, and religious affiliations;
and (b) the continuous concerns about pluralism and integration. The era
“witnessed an outbreak of what might be called ‘ethnic fever’” in which “the
nation’s racial and ethnic minorities sought to rediscover their waning eth-
nicity and to reaffirm their ties to the cultural past” (Steinberg, 1981, p. 3).
Although ethnicity and race are often used interchangeably, schol-
ars have offered various distinct definitions for both constructs. The term
race has multiple meanings and therefore is difficult to define. For example,
“Ethnicity is often used as a euphemism for race,” maintained Janet Helms
(1994), “as well as for other sociocultural affiliations (such as religious and
linguistic groups).” She continued, “Thus it might better be defined as social
identity based on the culture of one’s ancestors’ national or tribal group as
modified by the demands of the CULTURE in which one’s group currently
resides” (p. 293). To add to the complexity of the construct, Helms (1994)
also suggested that “race has three types of definitions: (1) quasi-biological,
(2) sociopolitical-historical, and (3) cultural. Each type may have relevance
for how race becomes one of an individual’s collective identities” (p. 297).
Helms (2007) firmly maintained that “racial identity theories do not
suppose that racial groups in the United States are biologically distinct
but rather suppose that they have endured different conditions of domina-
tion or oppression” (p. 181). Thus, for Helms, racial identity refers to the
“psychological mechanisms that people develop to function effectively in a
society where some people enjoy social and political advantage because of
their ancestors’ (presumed) physical appearances, but others suffer disadvan-
tage and lower status for the same reasons” (Trimble, Helms, & Root, 2002,
pp. 249–250; see Helms, 1994). As an alternative to race, Helms (1994) rec-
ommended using the term sociorace to acknowledge “the fact that typically
the only criteria used to assign people to racial groups in this country are
socially defined and arbitrary” (p. 147).
Research on ethnicity appears to dominate the social science literature;
however as long as racism exists, the term race must continue to be used to draw
attention to the racist experiences of millions of people who are constantly
subjected to it. To merely classify these experiences with the terms prejudice
or discrimination obfuscates the painful sting of racism; hence, to directly and
forcefully confront racism, race must be kept at the forefront of our vocabulary
when discussing intergroup and interpersonal relations (J. M. Jones, 2003).
Phinney (1990) noted “widely discrepant definitions and measures of
ethnic identity, which makes generalizations and comparisons across studies

ethnic identity and well-being      183

difficult and ambiguous” (p. 500). Phinney (2003) developed the most widely
used definition of the construct: “Ethnic identity is a dynamic, multidimen-
sional construct that refers to one’s identity, or sense of self as a member of an
ethnic group” (p. 63). From her perspective, one claims an identity within the
context of a subgroup; this subgroup claims a common ancestry and shares at
least a similar culture, religion, language, kinship, or place of origin (Phinney
& Ong, 2007). Phinney (2003) added, “Ethnic identity is not a fixed categori-
zation, but rather is a fluid and dynamic understanding of self and ethnic back-
ground. Ethnic identity is constructed and modified as individuals become
aware of their ethnicity, within the large (sociocultural) setting” (p. 63). At
another level the term identity is almost synonymous with the term ethnicity,
prompting some sociologists, such as Herbert Gans (2003), to suggest that
identity is no longer a useful term.

Brief Overview of Relevant Theory

Social and psychological interest in ethnic and racial identity has resulted
in a copious increase in journal articles and books on the subject. A few skep-
tics doubt that ethnicity is a benign topic; some refer to it as the “new ethnic-
ity” because it is viewed as divisive, inegalitarian, and racist (Morgan, 1981).
On occasion, the mention of ethnicity and identity, especially in academic
circles, sparks discussion claiming that without segregation, ethnicity would
not survive. Discussion can and often does turn to assertions that Americans
tend to exaggerate the existence and beneficence of ethnicity (Yinger, 1986),
with such phrases as “imagined ethnicity” and “pseudo-ethnicity” referring to
those who foist some ethnic factor to justify an action. Similarly, concerning
research on ethnic factors, Gordon (1978) asserted that “students of ethnic-
ity run the risk of finding ethnic practices where they are not, of ascribing an
ethnic social and cultural order where they do not in fact influence the person”
(p. 151). Consequently, critics argue over some fanciful line to separate ethnic
from nonethnic influences. The argument poses some questions: When can
behavior, personality, values, attitudes, and so forth, be attributed to ethnic
and racial factors? If an ethnic or racial attribution is not discernible, what
sociocultural and psychological influence can account for the phenomenon?
Several anthropologists, historians, psychologists, and sociologists have
written extensively on ethnic and racial identity (see reviews by M. E. Bernal
& Knight, 1993; Carter, 1996; Cross, 1991; H. W. Harris, Blue, & Griffith,
1995; Helms, 1990, 1994; Sellers & Shelton, 2003; T. B. Smith & Silva, 2011;
Steinberg, 1981; R. H. Thompson, 1989; Trimble, 2005; Trimble et al., 2002;
van den Berghe, 1981; Yip et al., 2014). Theoretical positions vary, includ-
ing some lodged in individual experiences and some formed from a socio-
biological perspective. Barth’s (1969) position represents the former: the

184       foundations of multicultural psychology

native’s worldview defines relationships, boundaries, lifestyles, and thought-
ways. The sociobiological perspective, most fervently represented by Pierre
van den Berghe (1981), is that “ethnic and racial sentiments are extensions
of kinship sentiments” (p. 18) and that “descent . . . is the central feature of
ethnicity” (p. 27). To support his argument, van den Berghe asserted that
“there exists a general predisposition, in our species as in many others, to
react favorably toward other organisms to the extent that those organisms are
biologically related to the actor” (p. 19).
A review of the various treatises written about ethnicity leads one to an
inevitable conclusion of its complexity (Helms, 2007; Phinney & Ong, 2007;
Ponterotto & Park-Taylor, 2007; Trimble, 2007; Trimble & Dickson, 2005; Yip
et al., 2014). In its broadest form, ethnicity refers to “any differentiation based
on nationality, race, religion, or language” (Greeley, 1974, p. 187). Typically,
ethnic identity is an affiliative construct by which an individual is viewed by
self and others as belonging to a particular group. An individual can choose
to associate with a group, especially if other choices are available (i.e., having
mixed ethnic or racial heritage). Affiliation can be influenced by racial, natal,
symbolic, and cultural factors (Cheung, 1993). Racial factors include physi-
ognomic and physical characteristics; natal factors refer to “homeland” (or
ancestral) origins of individuals, their parents, and kin; and symbolic factors
include those that typify or exemplify an ethnic group (e.g., holidays, foods,
clothing, artifacts). Symbolic ethnic identity usually implies that individuals
choose their identity; however, to some extent, the cultural elements of the
ethnic or racial group have a modest influence on their behavior (Kivisto
& Nefzger, 1993). Cultural factors, which involve the specific lifeways and
thoughtways of an ethnic group, are probably the most difficult to assess and
measure (see Cheung, 1993, for more details). In conceptualizing ethnic iden-
tity, the totality of racial, natal, symbolic, and cultural factors must be consid-
ered to achieve a full and complete understanding of the construct. In the next
section a few notable ethnic and racial identity scales are summarized to illus-
trate the range of approaches researchers may use to measure the constructs.

Measurement of Ethnic Identity

In 1990 Jean Phinney summarized the existing ethnic identity literature

for adolescents and adults, emphasizing primarily measurement and concep-
tualization. She noted that “there is no widely agreed on definition of ethnic
identity” and “the definitions that were given reflected quite different under-
standings or emphasis regarding what is meant by ethnic identity” (p. 500).
Undoubtedly, social and behavioral scientists believe they have a general
sense for the ethnic construct; some indeed are rather firm about their posi-
tions (van den Berghe, 1981; Weinreich, 1986; Weinreich & Saunderson,

ethnic identity and well-being      185

2003). Identity as a psychological construct is also the subject of considerable
debate; however, the addition of ethnic has cast the debate and subsequent
hodgepodge of opinion into another domain. In fact, about a quarter of the
studies reviewed by Phinney were not built on a theoretical framework.
Several conceptual approaches to ethnic identity emphasize an indi-
vidual level of analysis linking notions of identity formation and develop-
ment to self-concept. Much of the work in this area relies on Henri Tajfel’s
(1982) theory of social identity, which maintains that social identity strongly
influences self-perception and thus should be the central locus of evaluation.
When ethnicity forms the nexus of an in-group, self-identity will be influ-
enced accordingly. One’s distinctive ethnic characteristics, however, can be
restrictive, as one may reject “externally based evaluations of the in-group”
and therefore “may establish [one’s] own standards and repudiate those of the
dominant out-group” (M. E. Bernal, Saenz, & Knight, 1991, p. 135). Other
responses are possible; individuals might withdraw or choose to dissociate
with the referent, thereby adding psychological complications for themselves.
Tajfel’s social identity theory has generated considerable influence on ethnic
identity research. Some prefer to carry out the work under the ethnic self-
identification rubric (see Helms, 1994, 2007; Phinney, 1990, 1992; Phinney
& Ong, 2007; Ponterotto & Park-Taylor, 2007; Trimble, 2007; Umaña-Taylor,
2004; Umaña-Taylor & Shin, 2007).
Approaches to measuring ethnic and racial identity range from use of
a single item (Richman, Gaviria, Flaherty, Birz, & Wintrob, 1987) to scales
containing several dimensions (Carter, 1996; Helms, 1990, 1994; Phinney,
1992; Umaña-Taylor, 2004; Weinreich, 1986; Weinreich & Saunderson,
2003). Whatever measurement approach or technique is being developed or
used, one must factor in four domains of inquiry:
77 natality, one’s ancestral genealogy, including parents, siblings,
and grandparents;
77 subjective identification, a declaration of one’s own ethnic or
racial identity. Stephan and Stephan (2000) argued that “the goal
of assessment of race/ethnicity is accuracy from the perspective of
the respondent, and that the accuracy of such a social construct
can only be obtained by individual self-designation” (p. 549);
77 behavioral expressions of identity, as the respondent indicates
preferences for activities germane to his or her ethnic affilia-
tion, such as foods, music, books, and so forth; and
77 situational or contextual influences, with the respondent indi-
cating the situations that call for deliberate expression of the
ethnic affiliation, such as traditional ceremonies, interaction
with family and peers, neighborhood gatherings, and so forth

186       foundations of multicultural psychology

 (Ponterotto & Park-Taylor, 2007; Trimble, 2000; Trimble et al.,
2002; Umaña-Taylor, 2004; Umaña-Taylor & Shin, 2007; Yip
et al., 2014).
At minimum, scales and measures should attempt to capture the essence
of each domain to provide a full and complete profile or silhouette of an
individual’s identity. Helms (1994) added to this suggestion by pointing out
that measures should be tridimensional and include items to tap individual
characteristics, own-group affiliation, and out-group relations.
Simple nominal ethnic and racial procedures for declaring affiliation
and membership have limited usefulness. Although categories are commonly
used, Waters (1990) pointed out that
one cannot tell what this identity means to be an individual, how and
why people choose a particular ethnic identity from a range of possible
choices; how often and in what ways that ethnic identity is used in every-
day life; and how ethnic identity is intergenerationally transferred within
families. (p. 11)
Building a scale on the social identity theory of Tajfel (1982) and the
developmental stages advocated by Erik Erikson, Phinney (1992) created
the Multigroup Ethnic Identity Measure (MEIM), which asks respondents to
indicate their ethnic affiliation twice in the 15-item scale. This widely used
scale has undergone revision and been reduced to six items that evaluate
two components: identity exploration and identity commitment (Phinney
& Ong, 2007).
Until recently, most of the published ethnic and racial identity mea-
sures have asked the respondent to state one ethnic affiliation. However,
because many respondents have more than one ethnic identity, Oetting
and Beauvais (1991) developed a full scale of over 50 items allowing an
individual to “independently express identification or lack of identification”
(p. 663) with several cultural groups. They claimed that “in large-scale sur-
veys of adults, only two basic items may be needed to assess identification
with any one culture reasonably well: (1) Do you live in the ____ way of life?
and (2) Are you a success in the ____ way of life?” (p. 664). The researcher
would fill in the blank space with an ethnic group such as Puerto Rican; the
researcher can further specify the ethnic label by referring to a geographic
locale or some other designation such as a neighborhood, reservation, vil-
lage, or island grouping. Other items can be added to assess such things as
family identification and tradition, cultural event participation, language
preferences, and parental identification to expand the presumed effective-
ness of the scale.
In the 1980s Weinreich put forth his theory and measurement tech-
nique, explicating Identity Structure Analysis (ISA), a complex, highly

ethnic identity and well-being      187

sophisticated approach to assessing individuals’ ethnic identity as well as
their identities with other facets of their lives. If an individual identifies to
some degree with more than one ethnic or racial group, this multiple iden-
tity can be captured with the ISA approach (Weinreich, 1986; Weinreich
& Saunderson, 2003). ISA is grounded in psychodynamic developmental
theory, personal construct psychology, appraisal theory, social construction-
ism, cognitive–affective consistency theories, and symbolic interactionist
theoretical perspectives. According to Weinreich and Saunderson (2003),
“ISA conceptualizes one’s appraisal of social situations as involving one’s
interpretation of their significance to self’s identity from moment to moment.
Appraisal provides and records experiences of situations and events” (p. 20).
ISA can be custom designed to measure identity in an idiographic or nomo-
thetic framework through use of bipolar constructs; thus, the approach can be
tailored for an individual as well as for groups. Indices can be constructed to
measure such constructs as self-image (past, current, and ideal), well-being,
values, role models, reference groups, empathetic identification, and conflict-
ing identification, in addition to evaluation of others and a few other related
domains of identity.
The scales summarized in this section and those described in the lit-
erature are not without criticism. Indeed, researchers and scholars have sub-
jected many of the measures and their corresponding theories to extensive
scrutiny through cross-validation procedures, empirical testing, and theoreti-
cal speculation. For example, Root (2000) maintained that “the current mod-
els do not account for a range of ways in which people construct their core
identities and determine the importance of race in them” (p. 214). Moreover,
Root noted that “researchers have found no reliable method of extrapolating
the core or breadth of one’s identity from one context of identity or from a
response to one question” (p. 212).

Association of Ethnic Identity With Well-Being

A close correspondence has been noted between an individual’s sense

of ethnic identity and subjective well-being (George, 2010). Both constructs
involve personal assessment of the extent to which one identifies with an
ethnic or racial group. Yip and her colleagues (2014) pointed out that “the
contributions of both perspectives result in a common focus on the positive
implications of a coherent sense of racial/ethnic identity for general well-
being” (p. 180).
Subjective well-being can be defined as individuals’ assessment of their
cognitive satisfaction and emotional reactions to their lives. Some scholars
claim that subjective well-being involves global assessment of all aspects of
a person’s life. However, important differences are found in the nature of

188       foundations of multicultural psychology

well-being across multicultural groups. Thus, one might ask whether cognitive–
affective evaluations represent a universal basis for defining well-being across
multicultural groups. To assess well-being subjectively, rather than through a
group normative standard, encompasses assumptions aligned with a cultural
frame valuing personal independence, often labeled as individualism.
The status of well-being research is somewhat culturally situated within
only one system of values and beliefs. For example, Ryff and Keyes (1995)
proposed the following dimensions: self-acceptance, personal growth, life pur-
pose, environmental mastery, autonomy, and positive interpersonal relations.
Diener and Diener (1995) pointed out that subjective well-being can be defined
as “a person’s evaluative reactions to his or her life—either in terms of life
satisfaction (cognitive evaluations) or affect (ongoing emotional reactions)”
(p. 653). They identified three hallmarks of the well-being construct: (a) It
resides within the experience of the individual, (b) it includes positive mea-
sures, and (c) it involves global assessment of all aspects of a person’s life.
Seligman (2011) proposed PERMA as a means to understand happiness
and well-being: positive emotion, engagement, relationship, meaning, and
accomplishments.
All of these approaches are framed within an individuocentric perspec-
tive and thus ignore the richness of “living life well” that exists in countless
ethnocultural populations. On this point, cross-cultural psychologist Michael
Bond (2013) asked, “What factors lead an individual on this planet to assess
himself or herself as satisfied with life?” He contended that “four aspects of
an individual’s ‘life as lived’ will contribute to life satisfaction in any national
culture: sound health, satisfaction with one’s finances, a sense of autonomy,
and happiness” (p. 158).
Limited consideration has been given to the contribution of ethnic and
racial group identification in determining well-being outcomes. Some stud-
ies have found ethnic identity to be associated with a number of positive
outcomes (e.g., Phinney, Horenczyk, Liebkind, & Vedder, 2001). Among a
sample of 161 African American adults, Yap, Settles, and Pratt-Hyatt (2011)
found that relationship centrality and private life satisfaction were medi-
ated by perceptions of belonging. They also found that gender moderated
the strength of each of these mediating effects: Belongingness mediated the
relationships for women but not for men. In two somewhat related stud-
ies Mandara, Gaylord-Harden, Richards, and Ragsdale (2009) and Binning,
Unzueta, Huo, and Molina (2009) found that ethnic identity may be as
important as self-esteem to the mental health of African American adoles-
cents and that those who identified with multiple groups tended to report
either equal or higher psychological well-being and social engagement.
Recent research has now reached a critical mass sufficient to provide
an empirical test of the relationships between the two constructs. Several

ethnic identity and well-being      189

decades ago Phinney (1990) found a reasonable distribution of studies show-
ing a definitive relationship between ethnic identity and psychological well-
being. A meta-analysis used data from 184 studies to systematically explore
the relationship between ethnic identity and well-being among North
American people of color (T. B. Smith & Silva, 2011), with the results indi-
cating a modest but statistically significant average effect size of r = .17 for this
relationship. However, the range of association expressed in effect sizes was
highly variable, and secondary analysis revealed a number of pertinent com-
plexities to these data. The effect sizes were strongest for measures of positive
aspects of well-being, which included happiness, life satisfaction, quality of
life, self-esteem, and self-mastery. Effect sizes were weakest for mental health
symptoms such as depression and anxiety. Ethnic identity appears to function
more efficaciously among people of color by enhancing positive components
of well-being than by buffering against psychopathology.
In a study of over 600 high school students born in the United States,
researchers found among African Americans, White/European Americans,
and Hispanic/Latino(a) Americans a positive association between self-
esteem and ethnic identity (Phinney, Cantu, & Kurtz, 1997). Similar pat-
terns have been observed among Korean American college students: clarity,
pride, and engagement in individuals’ ethnic identity was positively related to
self-esteem and negatively related to depressive symptoms (R. M. Lee, 2005).
However, when these associations were examined over time, no evidence
was found for any longitudinal associations between ethnic identity and self-
esteem (Umaña-Taylor, Vargas-Chanes, Garcia, & Gonzales-Backen, 2008).
In a seminal paper examining cultural perspectives of well-being, Wong
(2011) identified “four pillars” of the good life as meaning, virtue, resilience,
and well-being. Regarding these points, Wong maintained, “Given that there
are cultural differences, subjective well-being still provides a useful index on
how we are doing and how well we live at the individual and national level”
(p. 15). Similarly, Diener and Tov (2009) reported that life satisfaction and
positive affects exist in different forms in many countries, concluding,
The World Poll shows that differences in society can make an enor-
mous difference in levels of subjective well-being. The differences in
well-being between societies are as large as the differences between very
happy and depressed individuals. The role of institutions and societies
should not be underplayed in our attempts to understand and improve
the world. (p. 218)
Allen, Rivkin, and López (2014) agreed, adding,
A significant body of evidence strongly suggests dimensions of well-being
are not the same across cultures, and even what seems to outwardly

190       foundations of multicultural psychology

 appear a common element can instead involve different acts, scripts,
and meanings across multicultural groups. However, despite their central
relevance in addressing the situation, elaboration of indigenous theories
of well-being are generally early in development, largely incomplete, and
often raise many more questions than they answer. In developing this
knowledge base, psychology would do well to remain ever vigilant of the
ecological fallacy, whereby knowledge based on the perspective of an
aggregate is assumed to apply to an individual. (p. 308)
By way of summary, social and behavioral scientists vary in how they
conceptualize and measure the constructs of ethnic and racial identity, which
at their core involve multiple issues of sameness and differentiation. Thus,
the measurement of ethnicity and ethnic identity is no small task, especially
given the debate surrounding its theoretical foundations and its usefulness.
Researchers must consider the “various cultural and structural dimensions of
ethnicity” (Cheung, 1989, p. 72) and “distinguish between general aspects of
ethnic identity that apply across groups and specific aspects that distinguish
groups” (Phinney, 1990, p. 508). To accomplish this we must move away
from viewing ethnic groups as homogeneous entities; in fact, there may be
more heterogeneity within certain ethnic and racial groups than among the
dominant groups in North American society (Cheung, 1993; Trimble, 1990;
Trimble & Dickson, 2005).
This chapter focuses on psychological well-being and its correspon-
dence with ethnic identity. Most of the literature on the subject emphasizes
an individuocentric orientation, with the locus of analysis in an individual’s
assessment of his or her sense of well-being, satisfaction with life, and lev-
els and depths of ethnic identity. More research is necessary to fully under-
stand the contributions of different lifeways and thoughtways to “living
life well.”
This chapter began with an observation by Clifford Geertz (2000),
and the review of literature closes at this point with another of his astute
observations:
As the world becomes more thoroughly interconnected, economically
and politically, as people move about in unforeseen, only partially con-
trollable, and increasingly massive, ways, and new lines are drawn and
old ones erased . . . the catalogue of available identifications expands,
contracts, changes shape, ramifies, involutes, and develops. (p 225)
Accordingly, the only principled way we can meet the challenge posed by the
enlarging catalogue is to engage in a thorough inquiry. The following section
reports an updated meta-analysis of the literature, with a summary of the find-
ings presented in the section after that.

ethnic identity and well-being      191

 QUANTITATIVE SYNTHESIS OF RESEARCH ON THE
ASSOCIATION BETWEEN ETHNIC IDENTITY
AND PERSONAL WELL-BEING

This section consists of a meta-analysis that includes 31 studies beyond

those analyzed by T. B. Smith and Silva (2011). Moreover, additional infor-
mation regarding the racial heterogeneity of samples and the magnitude of
ethnic identity are included in this review. We sought to determine the degree
to which additional data modified the findings of the 2010 publication and
to ascertain whether additional variables affected the findings across studies.
The methods of this meta-analysis are reported in the Appendix of this book.

Description of the Existing Research Literature

We located a total of 215 studies that reported data on 50,717 indi-

viduals’ psychological well-being as a function of their self-reported level of
ethnic identity. As shown in Table 10.1, the number of studies investigating
the topic has increased substantively over the past 2 decades, but few stud-
ies have involved actual mental health clients. All age groups except adults
over age 55 were adequately represented in the literature. Asian Americans
were overrepresented relative to their percentage of the U.S. population.
Native American Indians were underrepresented, with 16 studies including
Native American Indians and only eight being specific to that population.
Only one study evaluated Polynesian Americans, Native Hawaiians.
The typical study involved cross-sectional (correlational) data obtained
from convenience samples of participants who completed a questionnaire
containing measures of ethnic identity (most often the MEIM) and one or
more measures of psychological well-being, typically self-esteem. Thus, the
overall results provide an estimate of the correlation expected when admin-
istering a survey about ethnic identity and well-being to people of color.

Overall Research Findings

Across all types of measures of well-being, the random effects weighted

correlation with ethnic identity was .182 (SE = .009, 95% CI = [.16, .20],
p < .0001). The heterogeneity of the findings was moderately high (I2 = 72,
95% CI = [67, 75], Q(214) = 753, p < .001). Across the 215 research studies, the
observed correlations between individuals’ level of ethnic identity and their
perceptions of well-being ranged from -0.18 to 0.60. However, most of the effect
sizes (80%) fell between 0.0 and 0.35. A positive ethnic identity was associ-
ated with slightly poorer well-being in 10% of the studies, but was strongly

192       foundations of multicultural psychology

 TABLE 10.1
Characteristics of 215 Studies of the Association
Between Ethnic Identity and Well-Being
No. of
Characteristic M studies (k) %
Year of report 2001
Before 1980 0 0
1980–1989 3 1
1990–1999 68 32
2000–2008 121 67
Publication status
Published 93 43
Unpublished 122 57
Sampling procedure
Convenience 200 93
Representative (random selection) 15 7
Research design
Cross-sectional 203 94
Longitudinal 12 6
Population
General community members 62 29
Students 136 63
At-risk group members 14 7
Clinical populations (in treatment) 3 1
Sample size 236
<50 7 3
50–99 55 26
100–199 79 37
200–399 50 23
400–999 18 8
>1,000 6 3
Participant agea 23.0
Children (<13 years) 16 7
Adolescents (13–18 years) 72 34
Young adults (19–29 years) 76 35
Middle-aged adults (30–55 years) 33 15
Senior adults (>55 years) 6 3
Not reported 12 6
Participant gender (% female) 61.8
Participant raceb (%)
African American 36
Asian American 34
Hispanic/Latino(a) American 25
Native American Indian 2
Pacific Islander American 1
Other 2
Note. Not all variables sum to the total number of studies because of missing data.
aAverage age category of participants within studies (not all participants necessarily in the category listed).
bThe racial composition of participants across all studies, calculated by multiplying the number of participants

within studies by the percentage of participants from each racial group and dividing that product by the total
number of participants.

ethnic identity and well-being      193

associated with high levels of well-being in about 10% of the studies. On
average, the strength of individuals’ ethnic identity was found to be modestly
associated with their well-being.

Factors Influencing the Results

Study Characteristics
Meta-analyses can sometimes detect differences in effect size across dif-
ferent types of research studies. We considered the possible influence of the
year in which the study was published, the research design, the type of mea-
sure of ethnic identity, and the type of measure of well-being.
The results obtained across all 215 studies were stable over time. The
correlation between the year of study publication and its effect size was r = -.02
(see Table 10.2). However, the majority of studies were conducted in the
recent past, and substantive social changes typically require more time to
observe any effects. As shown in Table 10.3, the results did not differ across
the type of participant sampling, and although cross-sectional studies did
have higher effect sizes (r = .19) than longitudinal studies (r = .12), this
difference was not statistically significant.
No differences were apparent across the type of measure of ethnic iden-
tity. Although the clear majority of research reviewed here (70%) involved

TABLE 10.2
Random Effects Weighted Correlations of Effect Sizes With Study
and Sample Characteristics
Variable r k
Study characteristics
Year of publication -.02 215
Total number of participants -.06 215
Participants’ characteristics
Age -.17* 186
Average birth yeara .13 186
Gender of clientb -.13 211
Education level -.01 119
Socioeconomic status .05 117
% African Americans -.14 96
% Hispanic/Latino(a) Americans -.14 72
% Asian Americans -.29** 92
% Native American Indians -.53* 16
Average strength of ethnic identityc .25** 136
Note. *p < .05. **p < .01.
aYear of study minus average age of clients at time of study (to estimate cohort effects). bPercentage of

female participants within studies. cAveraged item-level score on the Multiethnic Ethnic Identity Measure
(Phinney, 1992).

194       foundations of multicultural psychology

 TABLE 10.3
Weighted Mean Correlations Across Levels of Several Moderator Variables
in Studies of the Association of Ethnic Identity With Well-Being
Variable Qb p k r+ 95% CI

Data source 2.0 .15

Published 93 .20 [.17, .22]
Unpublished 122 .17 [.15, .19]
Sampling procedure 0.1 .78
Convenience 200 .18 [.16, .20]
Representative (random selection) 15 .17 [.11, .23]
Research design 3.2 .08
Cross-sectional 203 .19 [.17, .21]
Longitudinal 12 .12 [.05, .19]
Population 6.5 .04
General community members 62 .15 [.11, .18]
Students 136 .20 [.18, .22]
At-risk groups or clinical populations 17 .18 [.12, .24]
Socioeconomic Status 0.5 .78
Lower class 29 .16 [.11, .21]
Lower middle class 27 .18 [.13, .24]
Middle class and above 61 .18 [.14, .22]
Gender a 6.1 .02
Female 39 .13 [.08, .18]
Male 12 .25 [.17, .33]
Participant raceb 4.2 .24
African American 58 .19 [.16, .23]
Asian American 61 .16 [.11, .18]
Hispanic/Latino(a) American 34 .16 [.12, .21]
Native American Indian 8 .15 [.07, .26]
Racial composition 8.9 .003
Heterogeneous 49 .23 [.19, .26]
Homogeneous 165 .17 [.15, .19]
Acculturation level 9.1 .03
Low (immigrants) 6 .05 [-.05, .16]
Moderate 28 .15 [.10, .19]
High 40 .19 [.15, .21]
Insufficient information reported 141 .19 [.17, .21]
Ethnic identity measure 1.3 .52
MEIM 150 .19 [.17, .21]
Other, research supported 35 .17 [.12, .21]
Other, homemade 26 .17 [.12, .22]
Well-being types 32.6 >.0001
Mental health symptoms c 23 .09 [.04, .13]
Self-esteem 69 .23 [.21, .26]
General well-being/coping 18 .22 [.16, .27]
Multiple indicators 105 .16 [.14, .19]
Well-being sub-typesd 75.3 >.0001
General mental health symptomsc 16 .11 [.04, .18]
Anxietyc 24 .08 [.02, .13]
Stress/distressc 10 .05 [−.03, .13]
(continues)

ethnic identity and well-being╇╇╇╇╇ 195

 TABLE 10.3
Weighted Mean Correlations Across Levels of Several Moderator Variables
in Studies of the Association of Ethnic Identity With Well-Beingâ•… (Continued)
Variable Qb p k r+ 95% CI

Depression/hopelessnessc 58 .14 [.11, .17]

Other mental health symptomsc 18 .03 [-.03, .10]
Global well-being 37 .20 [.15, .24]
Self-esteem 136 .23 [.21, .25]
Self-mastery/self-control 22 .20 [.14, .25]
Coping skills 18 .16 [.10, .22]
Social support 31 .16 [.11, .20]
Problematic behaviorc 13 .12 [.05, .19]
Multiple indicators (>1 above) 10 .11 [.02, .20]
Note.╇CI = confidence interval; k = number of studies; Qb = Q-value for variance between groups; r+ = random
effects weighted correlation; MEIM = Multigroup Ethnic Identity Measure (Phinney, 1992).
aAnalysis involving only studies with exclusively male or female participants. bAnalysis involving data from

studies with participants of a single racial group. cInverse scaling, such that positive correlations denote less
pathology. dAnalysis involving all effect sizes within studies that contained distinct outcome measures, such
that studies could contribute more than one effect size to the analysis but only one effect size per category.

the MEIM, the results obtained with other measures were of equivalent mag-
nitude. Thus, the averaged results did not appear to depend on how ethnic
identity was operationalized.
However, large differences were observed across the type of well-being
measured (see Table 10.3). On average, the association of ethnic identity
with measures of mental health symptoms was notably lower than with mea-
sures of self-esteem or general well-being. When the data were disaggregated
by the type of measurement used, the same trend was apparent.
Nevertheless, the disaggregated results yielded interesting qualifications.
Measures of symptoms of depression yielded a higher association (r = .13) with
ethnic identity than all other measures of mental health symptoms. Measures of
distress yielded a low correlation (r = .05), perhaps reflecting the temporary nature
of distress relative to the more pervasive nature of other mental health indicators.
Measures of problematic behavior (i.e., the Child Behavioral Checklist, typically
administered in educational settings) yielded an average correlation of similar
size to general mental health symptoms. Measures of self-esteem resulted in the
highest overall correlations with ethnic identity. The direction of all these cor-
relations were standardized for purposes of comparison, such that positive values
indicated that ethnic identity was associated with positive states of well-being
(i.e., fewer problematic behaviors or mental health symptoms).

Participant Characteristics
Characteristics of research participants varied widely across research stud-
ies, prompting investigation of the degree to which those differences might

196 ╇╇╇╇╇ foundations of multicultural psychology

account for differences in the findings across studies. We analyzed the possible
influence of participants’ average age, estimated birth year (to investigate pos-
sible cohort effects), gender composition (% female), racial composition, popu-
lation type (community members, students, at-risk groups, clinical samples),
and ethnic identity level.
The average age of participants moderated the overall results (r = -0.17,
p = .02); studies with younger participants (i.e., adolescents and young adults)
tended to have larger effect sizes than studies of middle-aged and senior adults.
Thus, ethnic identity appeared more relevant to the well-being of younger
populations. This association was found irrespective of age cohort; when the
estimated birth year of participants and the age of participants were regressed
simultaneously, the magnitude of the association for age increased to b = -0.26,
whereas the estimated year of participants’ birth remained low (b = -0.10).
This finding suggested that ethnic identity might be more pertinent to indi-
vidual development than to changing social trends over time.
The differences among samples of community members, students, and at-
risk or clinical populations (Table 10.3) were partially explained by differences
in age. Compared with the other two types of studies, those with students had
higher effect sizes, not unexpected because they consisted of younger partici-
pants. When age and population type were simultaneously regressed on effect
size, age remained the same as it had been for the overall sample (b = -0.17),
and the variables representing population type were lower (b = 0.06 for students
as contrasted with community samples, and b = 0.03 for at-risk and clinical
populations contrasted with community samples). Thus, there were no signifi-
cant differences across population type after accounting for age.
Differences in the overall findings were evident across participant gender
composition. The correlation between studies’ effect sizes and their percent-
age of female participants was r = -0.13 (p =.05), and significant differences
were found between studies using exclusively male versus female participants
(p =.02, see Table 10.3). Taken together, these findings suggest that ethnic
identity may be somewhat more salient to the well-being of males than females.
There were no differences across studies using participants from distinct
racial groups (Table 10.3). This finding seems to indicate that ethnic identity
equivalently predicts well-being among participant cultures. However, the
racial heterogeneity of the research sample did moderate the results. Studies
with populations representing one race reported lower effect sizes (r = .17)
than those with racially mixed samples (r = .23), which could imply that eth-
nic identity is more relevant to well-being in mixed racial settings. This find-
ing was confirmed by the consistent negative direction of the correlations of
effect sizes with the percentages of participants from each racial group. This
trend was particularly notable among samples including Asian American
and Native American Indian participants; studies with relatively higher

ethnic identity and well-being      197

percentages of those populations (more homogeneous samples) tended to
have smaller effect sizes than studies with low percentages of participants
from these populations (Table 10.2).
The strength of participants’ ethnic identity also moderated the results.
Table 10.2 shows a correlation between the average item-level score on the
MEIM (total score divided by the number of items) and the effect size obtained
from that study, r = .25 (p < .01). Studies with participants reporting relatively
higher levels of ethnic identity tended to demonstrate a stronger associa-
tion between ethnic identity and well-being than studies with participants
reporting lower levels of ethnic identity. This finding likely indicates that
ethnic identity is relevant to an individual’s well-being if he or she considers it
somewhat important, but relevance increases when ethnic identify is strongly
affirmed. One’s level of ethnic identity influences perceptions.
Having found that several participant characteristics moderated the
overall results, we next sought to ascertain which of the statistically signifi-
cant variables remained statistically significant in the presence of the others.
We therefore conducted a meta-regression using the variables that had been
found to be statistically significant in the univariate analyses. Random effects
weighted meta-regression is an analogue to multiple regression analysis for
effect sizes. We evaluated two models because data regarding participants’
average level of ethnic identity (average item-level scores measured by the
MEIM) were unavailable in many reports. The first model excluded the level
of ethnic identity in order to maximize the number of studies included in the
analysis. This model better represents the literature (200 studies included),
so it should be the one interpreted unless considering the variable represent-
ing level of ethnic identity, which only appears in the second model. Both
models (see Table 10.4) reached statistical significance (R2 = .19, Q = 48.7,
p < .0001 and R2 = .24, Q = 38.5, p < .0001).
To facilitate interpretation of the data, variables entered into both
models were centered on their means. The B value for the variable labeled
“constant” (Table 10.4) represents the average effect size one would expect
to find if all variables included in the model had the expected mean value.
These values correspond with the overall effect size reported earlier (r = .18).
Interpretation of the B weights (Table 10.4, first column) reveals the extent to
which each study characteristic would be expected to influence the observed
overall effect size when controlling for the presence of the other variables.
Thus, for Model 1, researchers using measures of mental health symptoms
could expect an effect size of r = .108 (.182 - .074), and those using measures
of self-esteem or global well-being would have expected effect sizes of r = .238
(.182 + .056) and r = .236 (.182 + .054), respectively. Overall, the variables
that remained significant in the model were the type of well-being measured,
along with participant gender (with stronger effect sizes among studies with

198       foundations of multicultural psychology

 TABLE 10.4
Random Effects Regression Weights for Study Characteristics
Associated With Effect Sizes
Variable B SE p b
Model 1 (k = 200) R 2 = .19
Constant .182 .009 <.0001
Participants’ average age -.001 .001 .12 -.10
Percentage female participants -.001 .001 .01 -.16
Immigrant statusa -.094 .051 .04 -.13
Homogeneous racial compositionb -.031 .022 .16 -.09
Measures of well-beingc
Mental health symptoms -.074 .028 .009 -.18
Self-esteem .056 .021 .008 .18
Global well-being .054 .033 .10 .11
Model 2 (k = 128) R 2 = .24
Constant .181 .010 <.0001
Participants’ average age -.001 .001 .30 -.09
Percentage female participants -.001 .001 .12 -.13
Immigrant statusa -.066 .072 .36 -.07
Homogeneity of racial compositionb -.035 .025 .17 -.12
Well-being measuresc
Mental health symptoms -.041 .035 .24 -.10
Self-esteem .054 .025 .03 .18
Global well-being .072 .034 .03 .19
Ethnic identity leveld .150 .038 .0001 .33
Note. Separate models were conducted because only 128 reports included information about item-level Multi-
group Ethnic Identity Measure (MEIM; Phinney, 1992) scores. Model 1 better represents the literature and should
be the one used, except when interpreting the effects of ethnic identity level, which was included in Model 2.
aLow acculturation (immigrants) contrasted with studies not reporting information on participant acculturation

level. bSamples of participants of all the same race contrasted with samples of participants from multiple racial
groups. cContrasted with multidimensional measures of well-being. dAverage item-level score on the MEIM.

greater percentages of male participants), and immigrant status (predicting

lower associations between ethnic identity and well-being). In the presence
of the other variables, participant age and sample racial homogeneity no
longer reached statistical significance. In Model 2, the level of ethnic identity
(mean item score on the MEIM) remained a strong moderator of effect size
(b = .33) in the presence of the other variables. The more strongly people of
color endorse ethnic identity, the more relevant it is to their well-being.

Possible Influence of Publication Bias

A majority of studies (122; 57% of the total) in this meta-analysis were
unpublished dissertations. The average effect size obtained across published
versus unpublished studies did not differ (see Table 10.3), so we had no reason
to suspect that unaccounted (unpublished) studies would adversely affect the
findings.

ethnic identity and well-being      199

Figure 10.1. Contour-enhanced funnel plot of effect sizes (Pearson r) by standard
error. The results across studies are highly variable, ranging from somewhat nega-
tive to very strongly positive. Thus, an overall average masks the true heterogeneity
of the data; the association between ethnic identity and well-being is imprecise but
predominantly positive.

As shown in Figure 10.1, effect sizes were relatively evenly distributed

around the average value of r = .18. There were no “missing” corners in the
distribution that would suggest studies excluded unintentionally from analy-
ses. One statistical method to estimate publication bias (Duval & Tweedie,
2000) did not identify any “missing” studies, and Egger’s regression test was
nonsignificant. Therefore, we concluded that publication bias was not a
threat to the results of this meta-analysis.

DISCUSSION AND INTERPRETATION OF THE FINDINGS

The psychosocial literature on ethnic identity and well-being is sub-

stantial. This topic has attracted considerable inquiry, especially in the past
2 decades. Results from the studies, however, have been mixed—they have
been neither persuasive nor robust. On average, ethnic identity explains
about 3.3% of the variance in measures of well-being among people of color.

200       foundations of multicultural psychology

The correlations between the self-reported measures of ethnic identity and
well-being ranged from -.18 to .60, with most between 0.0 and 0.35. The
findings open up an assortment of questions concerning the unexplained
variance.
The wide variability in the research findings confirms the obvious: Many
factors influence both ethnic identity and well-being. Present research prac-
tices have not accounted for that complexity. For instance, few researchers
evaluate variables such as social status that can influence both self-esteem
and ethnic identity. If an ethnic group has been through a history of prejudice
and discrimination, group members could experience a devalued sense of self
(Tajfel, 1982), but if the group has been evaluated positively, commitment is
likely to be strong, contentment high, and involvement in ethnic practices
significant (Phinney, 1991).
In short, decades of scholarship have produced a large body of research
that tells little beyond what could have been guessed intuitively. A positive
relationship exists between self-esteem and ethnic identity, but even that
basic conclusion could be attributed to the wording of measurement items,
which emphasize the certainty and importance of one’s ethnic declaration
and affiliation—for example, “I feel a strong attachment toward my own eth-
nic group.” People who feel strongly affiliated to others would probably also
report higher levels of well-being than individuals who feel unaffiliated or
uncertain about where their allegiance can be placed.

Considerations for Future Research

The sample of studies represented almost all age groups except those
over the age of 55. Asian Americans were the most represented ethnic group,
and Native American Indians, Alaska Natives, and Polynesian Americans
were the least represented in the literature. Some researchers disaggregated
groups into subgroups such as Chinese, Japanese, Korean, and Southeast
Asians. However, most often researchers tended to lump their participants
into one overarching inclusive ethnic label, a practice contributing to ethnic
gloss: the appearance of ethnic and cultural homogeneity that may be inac-
curate (Trimble & Bhadra, 2013).
The tendency for researchers and scholars to gloss over depth of cultural
lifeways and thoughtways raises serious methodological questions. Ethnic gloss
through imprecise categorization of participants provides little or no infor-
mation on the richness and cultural variation within ethnocultural groups
and may ignore the existence of numerous subgroups characterized by unique
cultural traditions. This sorting method minimizes the deep cultural influ-
ences that guide a group member’s thought, feelings, and behavior. Broad
ethnic categorizations can generate biased and flawed scientific research

ethnic identity and well-being      201

as well as promote stereotypes. In addition, use of an ethnic gloss does not
enhance external validity, the potential to generalize findings within and
across subgroups.
An associated finding is that current research practices have inad-
equately accounted for multiethnic individuals, who may affirm multiple
sources of identity, seek ways to pass as members of one group but not the
other, and/or denigrate the value of one group relative to another. Individuals
whose heritage includes multiple ethnic groups may choose to identify with a
particular group regardless of how others may view them, although parental
influences and gender alignment with parents, racism, emotional security,
and several other factors are influential in the complex processes of identifica-
tion (Root, 1994). The data we analyzed did not account for those dynamics,
which require more specific evaluations.
This meta-analysis extended information available on the association
between ethnic identity and psychological well-being by confirming that the
absolute level of ethnic identity does influence that association to a marked
degree. This finding confirms theoretical expectations: Individuals with high
levels of ethnic identity tend to view their ethnic identity as salient to well-
being; individuals with low levels of ethnic identity tend to see it as less
connected to well-being. Hence individuals’ level of ethnic identity should
be more clearly evaluated and understood, including the reasons why ethnic
identity is or is not particularly strong, thus refocusing methods of inquiry to
account for individual variation.
Although research with clinical populations occurs rarely, in the gen-
eral population ethnic identity appears to be more relevant to self-esteem
than to mental health symptoms. According to Tajfel’s (1982) theory of
social identity, one’s social identity strongly influences self-perception, and in
the general population self-perception is more accurately represented by self-
esteem than by psychopathology. Nevertheless, this finding could also sug-
gest that the commonly held assumption that ethnic identity buffers against
distress may be problematic. Analyses of the relationship of ethnic identity
with self-esteem and distress could benefit from more explicit consideration
of detailed theoretical frameworks such as that provided by Tajfel, rather than
the all-too-common practice of correlating sets of variables without a solid
heuristic model and then testing that model. For instance, this meta-analysis
identified differences in the degree to which ethnic identity is associated with
well-being when the participants are examined in ethnically homogeneous
versus heterogeneous settings. This finding is consistent with Tajfel’s social
identity theory, which indicates that identity is more salient in mixed settings
than homogeneous ones; identity must be evaluated relative to the avail-
able social contrasts. Although research in the social psychology literature
has taken care to evaluate situational factors and to explicitly test theories,

202       foundations of multicultural psychology

multicultural psychology research has typically involved broad research ques-
tions without systematic inquiry into the specifics of an association.
We have to be explicit about some questionable practices in the multi-
cultural psychology literature that we reviewed. We found that 93% of the
studies used convenience samples. If the field intends to produce general-
izable research findings, it has to randomly select research participants to
establish external validity. We also found that 94% of the studies involved
cross-sectional surveys. How can the field prove, let alone understand, causal-
ity when relationships are not evaluated across time?
The relationship between ethnic identity and well-being was found to
be minimal among participants with low levels of acculturation to North
American society (i.e., recent immigrants and refugees), but only six studies
evaluated those populations. Thus, we cannot discern with confidence
whether individuals with low acculturation levels, who tend to have ethnically
homogeneous social networks (Phinney et al., 2001), may take their ethnic
identity for granted in terms of their well-being, despite high absolute levels
of in-group identification (Phinney, 2006). Social identity theory (Tajfel,
1982) posits that a group would have to experience contrast (opposition/
oppression) for ethnic identity to be activated as a coping strategy, but recent
immigrants who reside in ethnic enclaves, or who alternatively seek to mini-
mize their differences by trying to fit in with mainstream society, may not
initially activate ethnic identity as a coping strategy. Nevertheless, theories
of acculturation (Berry, 2003) emphasize that multiple processes influence
the affirmation of one’s own ethnicity and its relevance to well-being, such
that an accurate interpretation of our finding would require a more detailed
understanding of the participants’ contexts than is possible in a meta-analysis.
Future research is needed to clarify the relationship between ethnic identity
and well-being as a function of acculturation.
Another challenge facing those who are advancing inquiry into ethnic
and racial identity is that the number of ethnic and racial groups in North
America is increasing not declining, and the populace does not appear to be
assimilating at the rate many demographers and sociologists have predicted.
All over the world geopolitical boundaries are changing because of politi-
cal turmoil, colonialism, and globalization; thus, individuals are changing
their ethnic allegiances and identities as they move to a new environment
or as rearranged boundaries move them into a new environment (Arnett,
2002). Indigenous groups are asserting sovereign rights and demanding rec-
ognition and access to their ancestral lands. Voices that were once suppressed
are demanding to be recognized. Consequently, ethnic groups worldwide are
becoming more independent and visible; this diversity presents new chal-
lenges for the field of ethnic and racial identity, particularly to its relationship
to well-being.

ethnic identity and well-being      203

Suggestions for Practitioners

More now than ever, multicultural awareness is being integrated into

classroom curricula and public discourse. And people are becoming inter-
ested in tracing their genealogy and learning about their ancestral heritage.
For example, Black History Month is one of several popular opportunities for
acknowledging and celebrating diversity and ethnic differences. Many indi-
viduals are coming to believe that they must declare an ethnic background
so that they may join in the celebration. In their work with clients, therapists
can attune to motives such as self-affirmation and sense of belonging that
may be relevant to this exploration of and affiliation with ethnic heritage.
Unfortunately, many clients may be negatively concerned about their
expression of identity; some presenting problems may emanate from feelings
of guilt over assimilation or from a longing to better understand their ethnic
heritage that was wrested away from them when they were adopted or moved
to an area where their ethnic group was misunderstood or victimized. In such
instances, the therapist should explicitly help clients to develop skills to cope
with those circumstances, as well as assist them in valuing their ethnic origins
and expressing them without fear of ostracism and rejection.
The data finding that ethnic identity tends to be more pertinent to
the well-being of younger individuals than older groups aligns with theories
that emphasize the importance of identity development among adolescents
and young adults. Clinicians working with those populations can particularly
attune to client experiences relevant to ethnic identity.
Clinicians must keep in mind the complexity of clients’ identity devel-
opment and expression, including ethnic identity. Our review of the liter-
ature did not resolve uncertainties about the causal relationship between
ethnic identity and well-being. The relationship might be that individu-
als who already have a strong self-esteem are more likely to embrace their
ethnic identity, rather than the other way around. Practitioners should not
presume ethnic identity to be linear or in any way simplistic—even though
that position has been common in the literature. Rather, they should look
more deeply into clients’ troubling conditions for the possible contributions
of ethnic identity to wellness and resilience.

CONCLUSION

Our meta-analysis included 215 published studies involving over

50,000 participants from several ethnic groups. The overall results provide
an estimate of the expected correlation when a survey concerning ethnic
identity and well-being is administered to people of color (r = .18), a value

204       foundations of multicultural psychology

falling not quite midway between mild and moderate according to J. Cohen’s
(1988) guidelines for interpreting effect size. However, that average value
masks variability. A positive ethnic identity was associated with somewhat
lower well-being in 10% of the studies, but was strongly associated with high
levels of well-being in about 10% of the studies.
Several decades of research on the topic of ethnic identity generated
the scatter depicted in Figure 10.1. Investigating global research issues such
as the overall association between ethnic identity and well-being is not par-
ticularly helpful because of multifaceted underlying constructs. We therefore
need to ask more specific research questions. What are the causal pathways?
Under what circumstances does ethnic identity prove most protective against
distress? For the construct of ethnic identity to become deserving of the wide-
spread attention it has received, research efforts must yield information more
useful than the data depicted in Figure 10.1.

ethnic identity and well-being      205

 IV
FOUNDATIONS
FOR THE FUTURE
 11
PHILOSOPHICAL CONSIDERATIONS
FOR THE FOUNDATION OF
MULTICULTURAL PSYCHOLOGY

If we are to achieve a richer culture, rich in contrasting values, we must

recognize the whole gamut of human potentialities, and so weave a less
arbitrary social fabric, one in which each diverse human gift will find a
fitting place.
—Margaret Mead (1963, p. 218)

People’s experiences require interpretation. Research data require inter-

pretation. In a book otherwise about experience and data, this chapter addresses
the topic of interpretation. This shift from evidence to ideas requires explana-
tion, which we offer in the form of an anecdote.
At a recent professional conference, a participant asked a prominent
panel member why he had dropped his successful career in neuropsychology
to teach philosophy—of all things! The question was asked with a tone of
unmistakable incredulity. Although not directly raising the issue of sanity,
the question seemed based on an underlying disbelief that a rational indi-
vidual could possibly make such a decision. Why philosophy?
The question “Why philosophy?” articulated what many individuals in
the audience had thought privately. Is the “big money” not in neuropsychology?
After all, the panelist was quite famous in neuropsychology. Why study ideas
rather than save lives? Is philosophy not the antithesis of neuroscience?
Hundreds of challenges could have been offered, but the point of this anecdote

[Link]
Foundations of Multicultural Psychology: Research to Inform Effective Practice, by T. B. Smith and J. E. Trimble
Copyright © 2016 by the American Psychological Association. All rights reserved.

209
is that each challenge would have required the panelist to justify his position—
in contrast to alternatives. Each alternative would align with a particular
worldview—a philosophy. Philosophy is omnipresent, acknowledged or not.
The premise of this book, that data should inform clinical practice, is
based on multiple assumptions. So are mental health treatments. So are cul-
tural worldviews. Multicultural counseling and psychology seeks to understand
and question these assumptions.

WHY ASK QUESTIONS?

When asked sincerely and thoughtfully, questions seek to bridge gaps

between worldviews. Questions seek interpretation. They seek clarity in terms
that we already understand. Thus, questions expose the assumptions and values
of the questioner. What we already know or assume influences what we believe
we need to find out. And when we believe we know already, we do not ask. Or,
we ask in a way that limits or precludes actually bridging worldviews.
In the introductory anecdote, the incredulous participant did not ask
why the famous panelist had started out in neuropsychology in the first place—
assuming intrinsic value in the study of neuropsychology. The audience had
many assumptions about such things as the optimal means for advancing
knowledge (empiricism vs. rationalism), the usefulness of neuroscience over
philosophy (pragmatism or perhaps utilitarianism), or the individual’s own per-
sonal interests (psychological egoism). Questions reveal underlying assump-
tions and values. The process of identifying assumptions and evaluating ideals
relative to alternatives is the work of philosophy. Seen in this light, psychology
itself is an attempt to apply philosophy to understand and improve human
experience (Robinson, 1995).

HOW DOES PHILOSOPHY RELATE TO

MULTICULTURAL PSYCHOLOGY?

Why include a chapter on philosophical considerations in a book about

multicultural research? We purposefully use the term philosophy rather than
the terms conceptualization or framework to emphasize the readers’ obligation
to engage the material through questioning and critical thinking. We seek to
prompt thoughtful analysis (Machado & Silva, 2007). And we do so by ques-
tioning assumptions in contemporary multicultural psychology. No doubt
some individuals who have read previous chapters in this book bypassed this
chapter after a single glance at a title including the word philosophical. To
them, we can offer no explanation. To the hurried readers who dared to skim
this far into the introductory section, we offer no promises. To the one who

210       foundations of multicultural psychology

continues, we offer more questions. The point of this chapter is that multi­
cultural psychologists should ask questions like the incredulous conference
participant. When questions are asked, assumptions and values can be iden-
tified. When questions are asked, dialogue occurs. And dialogue explicitly
informed by values and assumptions is the essence of multicultural psychology.

WHY DO ASSUMPTIONS MATTER?

Assumptions embedded within ideas have substantive power, especially

if they remain unchallenged. They influence decisions and actions without the
benefit of thoughtful evaluation. Adverse consequences of assumptions can be
obvious: Presidential decisions to support the invasion of Cuba in 1962 or of
Iraq in 2003 gave more weight to presuppositions than to contraindications.
Such examples are easily apparent. But equally apparent are pathologies in
which individuals assume fallacies about their own worth on the basis of evalu-
ations of others (e.g., not “good enough”) and minimize evidence contradicting
their assumptions. Assumptions influence thoughts and behaviors until identi-
fied and contrasted with alternatives.
Psychotherapy can challenge faulty assumptions, but reflecting and iden-
tifying assumptions are not the exclusive responsibility of clients. Clinicians,
students, and instructors assuming certain tenets about multicultural psycho­l­
ogy can also benefit from self-evaluation. The data presented in the preceding
chapters of this book indicate that several of the assumptions in multicultural
psychology research have been proven unreliable. Or perhaps some of the
assumptions are correct but the data were unreliable. Which assumptions or
which data sets are accurate? To what degree? Under which circumstances?
Questioning prevailing assumptions in multicultural psychology may gener-
ate new explanations better aligned with the research data—and may also
yield data characterized by greater reliability.
Multicultural psychology does not benefit from maintaining unreliable
assumptions. Multicultural psychology benefits when we strive to align our
assumptions with the needs and experiences of historically disadvantaged
populations.

IS IT TIME TO ASK HARD (PHILOSOPHICAL)

QUESTIONS IN MULTICULTURAL PSYCHOLOGY?

Examining one’s own limitations can be painful. Would it not be better

for a book promoting multicultural psychology to remain positive, rather than
ask questions that might be uncomfortable? We desire to improve multi­cultural
psychology, but we understand that questioning traditional practices may

philosophical considerations      211

provoke varied reactions. We intend no offense. But even the most cherished
ideas in multicultural psychology fail to explain the vast complexity of reality,
so a re-vision and revision of our profession’s ideas is inevitable, no matter how
presently influential they may be.
In the spirit of working collaboratively, we offer a historical analogy.
Like the scientific and philosophical traditions it broke from in the late 1800s,
early work in psychology relied exclusively on intellectuals from Europe and
North America to the exclusion of alternative worldviews. Assumptions
about race and ethnicity influenced the early practice of psychology in a
myriad of ways, many harmful (D. W. Sue & Sue, 2013). Women and people
of color who received indoctrination as psychologists initially experienced
little freedom to question those assumptions without incurring marginaliza-
tion (Guthrie, 2004), but over several decades harder and harder questions
about race and racism kept coming. For instance, psychologists recognizing
inaccurate assumptions opposed racial segregation (K. B. Clark & Clark,
1939) and other forms of prejudgment based on stereotypes (Allport, 1954).
Multicultural psychology exists today on the basis of scholarship that uncov-
ered and corrected assumptions.
We honor the women and men who questioned prevailing paradigms in
psychology and those of the subsequent generation who built multicultural
psychology on their work (i.e., elders recognized at the National Multicultural
Conference and Summit). After many decades of struggle, multicultural psy-
chology is here to stay (D. W. Sue, Bingham, Porché-Burke, & Vasquez, 1999).
However, much work remains to be done, with the rising generation of gradu-
ate students needing better preparation to effectively negotiate the complex
social realities of our time. In their interest and in the interest of the com-
munities they will serve, will we now collectively improve multicultural psy-
chology by asking harder and harder (philosophical) questions that challenge
our previous assumptions?

SOME QUESTIONS TO CONSIDER (WITH HOPES

FOR MANY MORE TO COME)

In our collective efforts to enhance multicultural psychology, open dia-

logue should not only accelerate its improvement but also model the process
it seeks to promote: learning from differences. The field has matured in recent
decades, but the complexity of multiculturalism will ensure that the disci-
pline will continue to expand over the next century and beyond. Thousands
of questions remain unanswered. Although we provide tentative responses to
the three “example questions” we ask in this section, we recognize that it is
the process of asking questions that has greatest worth. Questions can prompt
additional queries and responses in an iterative cycle.

212       foundations of multicultural psychology

What “Is” Multicultural Psychology?

Numerous definitions and descriptions of multicultural psychology

are available. Their consistent theme is that psychology must embrace the
whole gamut of human potentialities, as the esteemed cultural anthropologist
Margaret Mead (1963) suggested.
Most descriptions of multicultural psychology are aspirational: They
describe what ought to be. In his seminal book Multiculturalism as a Fourth
Force, Paul Pedersen (1999) pointed out, “Multiculturalism refers to a new per-
spective in mainstream psychology characterized as a fourth force complement-
ing the three other theoretical orientations in psychology, i.e., psychodynamic
theory, existentialistic theory, and cognitive–behavioral theory, addressing the
needs of culturally diverse populations” (p. 113). He added, “Multiculturalism
recognizes the complexity of culture” (p. 113). In addition, Pedersen quoted
eminent cross-cultural psychologist John Berry (1991): “Multiculturalism is
meant to create a socio-political context within which individuals can develop
healthy identities and mutually positive intergroup attitudes” (p. 24).
We share these aspirations. We also recognize that we have not yet
arrived at a point where undergraduate students are as familiar with multi­
cultural psychology (the “fourth force”) as they are with behavioral, psycho­
dynamic, and humanistic psychology. And multicultural psychological research
rarely accounts for “the complexity of culture.” We are still largely discuss-
ing antibias strategies rather than creating sociocultural contexts conducive
to “mutually positive intergroup attitudes.” In short, an obvious gap exists
between multicultural psychology as practiced and as frequently defined. It is
time to bridge that gap, starting with an evaluation of reality. Taking inven-
tory of what multicultural psychology “is” in the real world can help deter-
mine where we are relative to what it “ought” to be, with the aim of achieving
the envisioned “ought.”
When we, the authors, have spoken with psychologists who are unaware
of our affiliation with multicultural psychology, we have sensed three gen-
eral approaches to the topic, with a fourth, atypical approach. Polite, surface
acknowledgement is by far the most common response. Yes, multicultural-
ism is important, but when the conversation starts to go further, no substan-
tive methods, theories, or even rationale are mentioned. A second approach
involves strong and apparently genuine affirmation of multiculturalism, with
general principles understood but disconnected from their application; peo-
ple “talk the talk” but do not know how to “walk the walk.” A third approach
is silence, a disengagement rooted in apathy or skepticism. We are thankful
we encounter the latter approach with decreasing frequency. The rarest of all,
thus not yet one of the three “general” approaches, is to successfully practice
multicultural psychology.

philosophical considerations      213

 We occasionally meet people who engage in genuine collaboration with
local communities. They understand complex personal, situational, political,
and historical influences. They have stretched their methodological, theo-
retical, and analytic skills to the point that they have acquired new perspec-
tive and skills. So the envisioned aspirations are possible to attain. We see it.
We would like to see it more. For that to happen, future scholarship has to
address what multicultural psychology means to the people who are attempt-
ing to apply it (“What ‘is’ multicultural psychology?”). Understanding what
people perceive multicultural psychology to be will be essential to lifting their
vision of what it can become.

How Will We More Equitably Serve Historically Oppressed Populations?

Government initiatives to fund mental health care for economically

disadvantaged populations can help improve access to services, but reliance
on such programs will be insufficient to meet the vast need. Presently we see
few graduate students interested in serving impoverished communities; as
students they increasingly incur debt that precludes their entertaining such
a notion. Understandably, graduates often seek the highest paying positions
available. Professional psychology graduate programs can help by seeking
investments and endowments to help offset tuition costs. Even more benefi-
cial, graduate programs can emphasize a service-oriented mentality, provid-
ing practicum and/or externship experiences in high need areas. Internship
sites serving disadvantaged populations could be promoted (Casas, Park, &
Cho, 2010).
Individual practitioners can also take personal responsibility for better
meeting the needs of impoverished individuals needing care. Most psycholo-
gists in private practice have a few spare hours in their caseload. Reduced
rates (sliding scales) and networking in local communities can increase client
service utilization. Rather than merely challenging others to serve impover-
ished groups, we can take action ourselves. To paraphrase a better statement,
we need not ask what our community can do for us, but what we can do for
our community.

What Aims Should Multicultural Psychology Work to Achieve?

Multicultural psychology has broad ambitions: no less than to eradicate

prejudice and discrimination and to promote the well-being of historically
oppressed populations. Nevertheless, it may be useful to specify aims that
can contribute to meeting those long-term objectives. For instance, what
essential aspects of applied mental health services might bring about the
greatest improvements in traditional practices? Within the delimited sphere

214       foundations of multicultural psychology

of mental health services (including prevention and wellness initiatives),
four specific aims seem most desirable: (a) reduction of mental health dispar-
ities, (b) access to services by those in need, (c) retention of those receiving
services, and (d) improved outcomes to the satisfaction of those receiving
services.
Other praiseworthy initiatives, such as promoting multicultural com-
petence among therapists, should not divert the primary focus from these
four aims. In fact, the ultimate purpose of therapist multicultural compe-
tence is to address the four aims, although most contemporary research into
this competence relates only indirectly to them. The same could be said of
applied psychological research concerning perceived racism, acculturation,
ethnic identity, and so forth. Although the relationship of ethnic identity to
well-being is interesting, the application of that knowledge to enhance the
efficacy of therapy or prevention initiatives with at-risk youth is of more use.
Multicultural psychology cannot continue to influence the broader profes-
sion (Pedersen, 1999; D. W. Sue et al., 1999) by drifting into interesting
topics at the expense of the essential ones.
Many authors affirm social justice as an objective for mental health
service providers (Toporek, 2006). Social justice is a far nobler objective
than the four comparatively mundane aims proposed here. However, these
four aims could be considered the focal point of social justice work within
psychology, serving as concrete benchmarks for progress in mental health set-
tings toward the broader aim of social justice, a paradigm worthy of our efforts
and therefore of our questions.

QUESTIONING A PARADIGM:
RECONCEPTUALIZING SOCIAL JUSTICE

Early in this chapter we emphasized that assumptions pervasively influ-

ence both collectives and individuals. We then asked three “example ques-
tions” about multicultural psychology. We now focus on the topic of social
justice to explore possible assumptions at the paradigmatic level and to sug-
gest a possible alternative conceptualization. As with the example questions
discussed earlier, the content of this inquiry matters much less than the pro-
cess. Any paradigm influential in multicultural psychology could and should
be similarly questioned.

How Central Is Social Justice to Multicultural Psychology?

Few concepts can rival the influence of social justice on contemporary

multicultural psychology. Commonly defined as the application of the concept

philosophical considerations      215

of justice to a societal level, social justice is more specifically set out by Rawls
(1999):
Each person possesses an inviolability founded on justice that even the
welfare of society as a whole cannot override. For this reason justice
denies that the loss of freedom for some is made right by a greater good
shared by others. (pp. 3–4)
Social justice, because it aims to promote equity by eradicating discrimination
and poverty, has profoundly shaped multicultural psychology (Arredondo &
Perez, 2003; Leong, Comas-Díaz, Hall, McLoyd, & Trimble, 2014; Toporek,
Gerstein, Fouad, Roysircar, & Israel, 2006). Social justice perspectives have
moved multicultural psychology forward in many essential ways, including
(a) equity in power structures, (b) orientation toward action, and (c) empower-
ment of community.
Why question a paradigm with benefits so obvious and so widely rec-
ognized? Although the concept of social justice is repeated often in multi-
cultural psychology literature, most authors merely mention it in passing,
at times seeming to use it as a type of code word to show familiarity with
contemporary parlance, with little relevance actually shown in the research.
Genuine adherence to the concept of social justice does occur (D’Andrea
& Daniels, 2010), but infrequently (Baluch, Pieterse, & Bolden, 2004). Are
there assumptions in present articulations of social justice that might con-
strain its influence or preclude its widespread application?

What Assumptions Might Be Embedded in the Concept of Social Justice?

The concept of social justice has been invoked by a variety of scholars

advocating for change in psychology (e.g., Arredondo & Perez, 2003; Vera
& Speight, 2003) and higher education generally (Worthington, Hart, &
Khairallah, 2010), so any generic examination of the concept will fail to
represent all perspectives. Although a systematic examination of all possible
assumptions within a social justice paradigm would fill an entire volume, we
restrict our list to three strengths that frequently receive attention in applied
psychology.
77 Primacy of power. Attending to power (i.e., resources, social
influence) is the optimal focus of scholarship, with an aim to
promote equality through social change.
77 Role of advocacy. Mental health professionals should advocate
for oppressed people.
77 Emphasis on empowerment. Power plus advocacy lead to
empowerment—giving voice to the voiceless; bottom-up pro-
cesses are key.

216       foundations of multicultural psychology

 To continue the conceptual analysis, we could examine relevant phil-
osophical positions sharing similar assumptions. However, few authors in
psychology mention underlying philosophy. Moreover, social justice is multi­
faceted (Toporek et al., 2006). Nevertheless, the three assumptions seem
related to the following philosophical concepts: (a) dialectical materialism
and critical pedagogy, (b) praxis and political activism, and (c) liberation
psychology. Although these positions are rarely cited in manuscripts, their
links with social justice have been well articulated (Ivey & Collins, 2003;
Ivey & Zalaquett, 2009; Vera & Speight, 2003).

Which of These Assumptions Might Conflict With Pragmatic Realities

in (North American) Multicultural Psychology?

In our review we found that much of the multicultural psychology lit-

erature did not align with the three assumptions just listed. For example,
far from giving credence to “the primacy of power,” researchers hardly ever
measure or even operationalize issues of power (i.e., resources, social influ-
ence), even those clearly relevant to their investigation. They pay even
less attention to advocacy for social change. Few publications are driven
by “bottom up” community involvement. Undermining the field’s intended
emphasis on social justice, the variables commonly measured in contempo-
rary multicultural psychology research (e.g., assessment validity, ethnic and
racial identity, acculturation) are ancillary to issues of power, advocacy, and
empowerment.
Although this neglect may be excused in research (because detach-
ment from reality is sometimes attributed to inhabitants of ivory towers),
we find the same trends in literature describing clinical practice, with allu-
sions to issues of power, advocacy, and empowerment but few specifics (Vera
& Speight, 2003). Recently, a prominent multicultural psychology leader
lamented that her work with disadvantaged communities was not valued by
her peers and that she knew of few psychologists who shared her passion or
role as advocate, despite prevailing rhetoric. Rarely do we hear of true com-
munity empowerment in the literature. Fields such as development studies,
social work, and social anthropology have developed traditions conducive
to participatory action research (designed to promote community empower-
ment by meeting people’s needs and answering their questions); however,
we find hardly any of that work in multicultural psychology. Despite years of
appeals in the literature, the promotion of social justice seems incongruent
with present practices.
The most obvious reason for such incongruence is reluctance of adher-
ents to practice the principles (Baluch et al., 2004; Speight & Vera, 2004).

philosophical considerations      217

But why is social justice not more commonly practiced when the entire field
seems to be based on it? The following reasons may apply:
77 Issues of power, empowerment, and so forth, may be too abstract
or complex for psychologists to address in therapy or research;
these concepts may need to be grounded in lived experience
(Gergen, 1995) and operationalized (Cooren, 2006).
77 North American psychologists have been raised in a capitalist
society that obfuscates power dynamics, even those that are
obvious to others. People are rarely paid to be social justice
advocates, so they rarely engage in the work.
77 Practitioners may perceive liberation psychology as a theory,
rather than as a worldview for engagement with reality.
77 People who select a career in the mental health professions may
prefer working with individuals and small groups, rather than
dealing with macro-level issues.
77 Methods of social change beyond advocacy and political activ-
ism may be overlooked by individuals who narrowly interpret
liberation psychology. Not all contexts or personalities are com-
patible with advocacy or activism; multiple methods can be
effective in raising awareness or facilitating desirable change.
In addition to these practical reasons why the concept of social justice
has been characterized more by verbal posturing than the intended actions,
we suggest that part of that problem stems from the underlying philosophy.
Materialism, a philosophy informing liberation psychology, praxis, and criti-
cal pedagogy, was a response to oppression, with roots in European intellectual
paradigms. The underlying assumptions are reactionary and thus delimited,
not necessarily aligning with cultures of other origins. That is, philosophical
materialism adds a layer of interpretation that can preclude acceptance of other
worldviews, an objective of multicultural psychology. Philosophical materialism
is a popular worldview among intellectuals, but many indigenous cultures abide
by different conceptualizations. Academics and advocates commonly interpret
experiences of indigenous populations through the lens of materialism, although
they criticize comparable interpretation through the lens of capitalism, individ-
ualism, and so forth. Such interpretation must be distinguished from acceptance
of an indigenous cultural worldview as worthy on its own merits.
Although Marx, Fanon, Freire, Martín-Baró, and others advocating
philosophical materialism both criticized power dynamics and emphasized
relations across all of humanity (humanization), many proponents of social
justice do little more than condemn those who abuse power. Motivated
by indignation, justified anger, they become “like the oppressors, mimick-
ing their patterns of domination and dehumanization” (Gaztambide, 2009,

218       foundations of multicultural psychology

p. 216). Freire (1973) termed this reactionary stance naïve transitivity. His
urging to transcend reactivity aligns with even novice students’ perceptions
about animated social justice advocates pounding the pulpit at professional
conferences: They seem to be selling a version of psychology insufficiently
self-reflective to earn admiration. A genuinely multicultural psychology would
not fix attention on symbols of status (money, influence) at the expense of
omitting other important contexts (social/intimate, holistic/spiritual).
So long as social justice is obtained, a pragmatist would embrace any
means to achieve that end. But if the end is not being achieved (which is cer-
tainly the case), the pragmatist would explore reasons why and replace those
most likely negating efficacy. For instance, if philosophical materialism is an
uncertain fit with many indigenous cultures of Africa, Asia, Australia, North
and South America, and the Pacific Islands, other motivations for promoting
justice may be more desirable for multicultural psychology.

Section Summary: Values and Assumptions

Values and assumptions pervade multicultural psychology—as they do

every discipline. Although we cannot escape values and assumptions, we can
seek to identify them and improve on them if possible.
In this section we have explored the concept of social justice, demonstrat-
ing its relevance to multicultural psychology, listing three assumptions embed-
ded within it, identifying some philosophical underpinnings, and finally listing
possible limitations of those assumptions in the practice of multicultural psychol-
ogy. Any idea or theory prominent in multicultural psychology can be similarly
evaluated. However, the purpose of such deconstruction must be reconstruction:
to retain what is useful and improve on the rest. To that effort we now turn.

How Can Social Justice Be Reconceptualized Through

a Relational Paradigm?

Karl Marx articulated philosophical materialism in response to his era,

the industrial revolution of the latter 1800s. Abuses of power were universal.
Protections for citizens were minimal, and protections for disenfranchised
groups were either nonexistent or ignored. Certainly oppression of the power­
less had always existed—and will always exist—but through materialism
the masses gained the allegiance of scholars, who articulated their plight and
proposed means for their empowerment.
Equal opportunity remains an aim to be sought. However, social and
global dynamics have changed with the times. International cultural exchange
now occurs at unprecedented levels. Electronic networks provide enhanced
connection with global as well as local diversity, with opportunities for

philosophical considerations      219

more equity in access to knowledge. Our collective interests now depend
on multiculturalism and internationalism, requiring that we move beyond
a self-preservation mentality. Just as Marx provided a powerful response to
the needs of his time, contemporary multicultural psychology would benefit
from a philosophy (not merely a set of loosely connected arguments about
power and privilege) that directly responds to an increasingly Internet-based
society with its need for unity amid diversity. Although the laudable work
of Martín-Baró, Aron, and Corne (1994) on liberation psychology provides
philosophical underpinning for social justice, our thesis is that a broader rela-
tional paradigm can provide a more useful philosophical foundation that is
compatible with that work but not dependent on materialism.
Humans are innately social beings. A relational paradigm asserts that
social interactions are central, not tangential, to psychology (Gergen, 1995;
Jordan, 2010; Slife & Wiggins, 2009). Individuals’ interactions with others
form the structure, process, and content of their lives. Primary relationships
(childhood and current) influence other relationships, which vary in impor-
tance across time and across contexts. Each person has a unique pattern of
relationships that is constantly changing, but clear similarities can be found
across individuals and even across cultures because some similar contexts are
shared, most notably physiologic but also environmental, linguistic, historic,
and so forth (D. Cohen, 2001). New relationships and repaired relationships
alter individuals’ perceptions, emotions, cognitions, and behaviors, which all
influence other relationships in interactive processes.
Aspects of a relational paradigm can be found across history. The phi-
losophy of Confucius emphasized social roles. Aristotle conceived humans as
primarily political, by which he meant interactive (Robinson, 1995), neces-
sary parts of the whole—the polis (community or city). Our ultimate interests
are those of the community.
The relational paradigm is a contemporary movement that links with
the tenets of feminism (J. B. Miller, 1986), interpersonal psychotherapy
(Weissman, Markowitz, & Klerman, 2000), object relations (Clarke,
Hahn, & Hoggett, 2008), symbolic interactionism (Charon, 2001),
social constructionism (Gergen, 2009), liberation psychology (Martín-
Baró et al., 1994), and philosophies advanced by Levinas (1969, 1998)
and Bakhtin (1981) among others. The term relational paradigm denotes
a broad worldview, a meta-theory. A variety of synonymous terms have
been used in the literature: relational meta-theory (Lerner & Overton,
2008), relationism (Overton & Ennis, 2006b), relational ontology (Slife,
2004), relationality (Slife & Wiggins, 2009), relational perspective (T. B.
Smith & Draper, 2004; Weissman et al., 2000), and relational methodologi-
cal research approaches (Trimble & Mohatt, 2006). Although specific asser-
tions and assumptions differ (e.g., Oliver, 2001), the core principles align

220       foundations of multicultural psychology

to emphasize reciprocal effects embedded in interpersonal and intergroup
interactions. To understand people, we must understand their relationships.
The following seven general principles help to clarify how individuals are
best understood, not solely as individuals but also as interactive agents in the
context of multiple relationships past, present, and potential.

Connectedness (Mutual Edification)

An innate yearning for attachment with others characterizes human life
(Cassidy & Shaver, 2008). Social engagement provides information exchange
and learning (Bandura, 1977), but at deeper levels it can be emotionally ful-
filling and mutually edifying. Interpersonal intimacy provides meaning and
purpose in life. In fact, an insufficient social network affects longevity as much
as light smoking and much more than alcoholism, obesity, and hypertension
(Holt-Lunstad, Smith, & Layton, 2010). The absence of genuine intimacy
coincides with psychological disturbance and poor health (S. Cohen, 2004;
Holt-Lunstad, Smith, et al., 2015). The principle of mutual edification pro-
vides a philosophical and psychological basis for the aim of multicultural
psychology to eradicate oppression and segregation: People who are socially
distanced experience negative outcomes; people who are socially integrated
experience enhanced well-being.

Holism
A relational paradigm emphasizes contextualization (Overton & Ennis,
2006a). We cannot see the parts without seeing the whole, and we cannot
understand the whole without considering the parts. To understand psycho-
logical processes, we have to learn about both specific events and their con-
texts. Multicultural psychology, with its emphasis on contextualization, has
provided a holistic perspective previously absent from individualistic concep-
tualizations of human experience.

Interactive Volition
Individuals possess an innate will and volition called agency (Adams
& Markus, 2001; Magyar-Moe & López, 2008; R. N. Williams, 1992), but
that volition interacts with the environment (Robichaud, 2006). External
forces, such as sociopolitical oppression, clearly influence and restrict an
individual’s choices. Nevertheless, we retain the power to work to mod-
ify external environments (e.g., combat oppression). We are not free from
external influences, including our own relationship history and culture,
but we are free to change our perspective, repair damaged relationships,
form new relationships, strengthen our own abilities, and work to modify
the environment.

philosophical considerations      221

Becoming
Relational development is ongoing. Personal identity and capacity
evolve as relationships evolve. Whereas most of psychology fixes its focus on
the present, a relational paradigm emphasizes ongoing processes and poten-
tials: the ontology of becoming (Overton & Ennis, 2006b).

Self-in-Relation
In a relational paradigm, the self is seen not as a fixed entity, isolated and
independent, but as a highly complex and fluid pattern, a self-in-relation (Adams
& Markus, 2001; Kaplan, 1986). People understand themselves through their
interactions with others, who serve as points of comparison and contrast across
circumstances and across time (Overton & Ennis, 2006a). For instance, ethnic
identity develops not only through emulation of desired models but also by
contrast with other ethnic groups, particularly oppressive ones (Tajfel, 2010).

Responsibility to Others (Moral Sensibility)

Whenever people interact, they influence one another, even if implicitly.
Thus, people remain responsible to one another for their influence (Gergen,
2009; Levinas, 1969). Given this responsibility, a relational paradigm advocates
an other-engagement (meaningfully interacting in ways mutually beneficial) and
a we-consciousness (explicit attentiveness to the relationship; Levinas, 1998).
Other-engagement and we-consciousness diminish self-interest (Stapel &
Koomen, 2001), which helps keep interpersonal and intergroup interactions
benign rather than oppressive. Thus, therapists maintain not only focus on the
client but also vigilance for effects on the client of their own actions and assump-
tions (Richardson, Fowers, & Guignon, 1999). Sensibility and responsibility to
the client constitute the essence of multicultural counseling competence.

Rights
Interpersonal relationships occur across disparate contexts, including dif-
ferent nations and legal systems. However, crossing a geopolitical boundary
should not change the core human. Thus, the notion of human rights is neces-
sarily grounded in relationships, not in myriad contexts. People bear rights with
them wherever they go, irrespective of organizational policy or national law.
From this perspective a right denotes a deserved protection. Rights “follow
the person” (irrespective of national and organizational boundaries) because
people remain vulnerable to others wherever the location. Human vulnerabili-
ties necessitate protection, so rights link to vulnerabilities (Harré & Robinson,
1995). Social institutions that protect human rights (i.e., government, pro-
fessional organizations such as the American Psychological Association)
hold accountable anyone prepared to compromise others’ well-being or take

222       foundations of multicultural psychology

advantage of their vulnerabilities. Multiculturalism has helped to promote rec-
ognition of human vulnerability and to promote accompanying protections.

How Might a Relational Paradigm Benefit Multicultural Psychology?

An abstract metatheory, like the relational paradigm, is useful to practi-

tioners and researchers to the extent that it facilitates interpretation of lived
experience and research data. A relational paradigm has clear implications
for social justice and community empowerment, as alluded to in the pre-
ceding section. For example, the Miami Youth Development Project applies
a relational approach to promoting social justice by relying on contextual
resources and the relationships of youth to their parents, peers, teachers, and
mentors (Lerner & Overton, 2008).
Most scholars who advise about the conduct of mental health treat-
ments and research with populations other than their own devote attention
to the principles and codes of professional ethical standards and norms; that
is, they are concerned about what is right and wrong, good or bad, harmless
or harmful, intrusive or nonintrusive, and an assortment of other moral and
humanistic considerations. Scholars have expanded on normative profes-
sional standards to include often unstated ethical principles and guidelines
that focus on the importance of establishing firm collaborative relationships
with community leaders, especially in conducting research with ethno­
cultural groups (Fisher et al., 2002; G. V. Mohatt, 1989).
It is time to place the collaboration concept in the center of inquiry
and work out its importance for community research and intervention.
Although some would see it as merely a tool or strategy to getting the
“real” work of behavioral science done, our strong preference is to view
the research relationship in community research and intervention as a
critical part of the “real” work itself. (Trickett & Espino, 2004, p. 62)

A relational paradigm also has clear implications for psychotherapy

(Gelso, 2011; Slife, 2004; Slife & Wiggins, 2009; T. B. Smith & Draper, 2004;
Wachtel, 2008). This chapter cannot include the many ways a relational
paradigm can improve clinical practices, but the basic tenets are obvious:
building interpersonal trust with the client, exploring clients’ relationship
patterns to gain insights into positive and negative coping, strengthening
clients’ social skills and intimacy with others, involving others in the cli-
ents’ efforts to improve, attending to countertransference, modeling desir-
able interpersonal interactions in the here and now, and so forth. “Research
studies demonstrate that it is the relationship between the client and the
psychotherapist, more than any other factor, which determines the effective-
ness of psychotherapy” (Clarkson, 2003, p. 4).

philosophical considerations      223

 The specific implications of a relational paradigm for multicultural psy-
chology are too many to list, but we briefly highlight the relevance of a rela-
tional paradigm to the construct of ethnic identity as one example that may
suggest possibilities for other topics. Research and theory focused on identity
development have received much attention in the literature, but this scholar-
ship typically involves assumptions associated with individualism: Identity is
often assumed to be a trait, something an individual “possesses” (e.g., noting
that Ms. Kim has a strong Korean American identity). In contrast, a rela-
tional paradigm would emphasize the dynamic shared nature of identity (e.g.,
examining Ms. Kim’s relationships with her grandparents, workmates, etc.,
and attending to how those interactions invoke and suppress her perceptions
and actions relevant to cultural values as a Korean American). From this per-
spective, scholarship on identity development should attend to social influ-
ences, primarily family socialization (e.g., L. L. Liu & Lau, 2013; Trimble,
2005). Identity undergoes challenges and redefinitions as social encounters
broaden outside the home, but then stabilizes as social interactions become
predictable and controllable.
For example, a person’s complex identity as bisexual Catholic female
accountant with a learning disability is dependent on her interactions
with other women, bisexuals, Catholics, accountants, and individuals
with learning disabilities, who provide essential modeling and sources of
comparison, and people who share none of those attributes, who serve
as sources of contrast. If the woman has had positive key interactions
with others about her gender, she will likely strongly affirm her identity
as a woman. But if she has had negative interactions with others about
her learning disability and has failed to meet a positive role model with
a learning disability, she may likely minimize or avoid openly acknowl-
edging that aspect of her experience. Identity parallels social inter­
actions. Models of racial identity, gender identity, spiritual identity, etc.
may therefore benefit from incorporating interpersonal-level variables
such as socialization and predictability of interactions. (T. B. Smith &
Draper, 2004, pp. 319–320)

Although identity development theories often mention social dynamics, rel-

evant research has remained steeped in individualism, measuring the indi-
vidual without regard to social context.
Similarly, many other variables in multicultural psychology (e.g., the
effects of racism on well-being) have clear social foundations that have typi-
cally been ignored because of assumptions embedded in individualism. A
relational paradigm attempts to balance prevailing notions with alternative
explanations hopefully representative of lived experience.
We have many reasons to believe that a relational paradigm aligns well with
multicultural psychology (Comstock et al., 2008; Fay, 1996). Conceptualizations

224       foundations of multicultural psychology

and assumptions based on a relational paradigm may offer several advantages
over existing conceptualizations and assumptions based on alternative para-
digms, detailed in the following paragraphs.

Congruence With Psychological Perspectives

A relational paradigm fits well within social and applied psychology, thus
corresponding to psychotherapy better than philosophical paradigms originally
conceived by scholars in economics, comparative literature, sociology, or polit-
ical science. For example, a psychotherapist can work with the relational con-
cept of collective/group well-being (Peterson, Park, & Sweeney, 2008) while
also attending to socioeconomic power (for an alternative conceptualization
see Gergen, 1995).

Congruence With Well-Being

Having sufficient resources to sustain life with reasonable predictability
is essential to well-being (Diener & Oishi, 2000; Howell & Howell, 2008).
Once individuals have sufficient material resources, the basis for human
well-being is interpersonal relationships (e.g., Dwyer, 2000; Magyar-Moe
& López, 2008; Peterson et al., 2008). This research finding, consistent
across world cultures (e.g., Haller & Hadler, 2006), has necessarily focused
inquiry on social factors associated with well-being. Personal and collective
happiness is largely a function of the quality of interpersonal relationships
(Myers, 2008, 2012). A multicultural psychology incorporating principles
of connectedness, holism, becoming, and so forth, could improve current
efforts to promote well-being (N. V. Mohatt, Fok, Burket, Henry, & Allen,
2011). A multicultural psychology informed by these relational principles
can easily integrate with positive psychology, a possibility open for explora-
tion (Pedrotti, Edwards, & López, 2009). A relational paradigm aligns with
the psychology of well-being.

Congruence With a Primary Cause of Trauma and Mental Illness

A relational paradigm is not restricted to a positive psychology focus on
well-being, although that is its strength (Magyar-Moe & López, 2008). When
negative or unpredictable, relationships yield harmful psychological conse-
quences, sometimes terribly destructive ones. Psychological damage results
from violations of intimacy and dignity: incest, rape, verbal abuse, spousal
infidelity, and similar interpersonal desecration unfortunately common among
individuals seeking mental health services. Mental illness that is not directly
explainable by neurochemistry has social underpinnings. The psychology of
abuse, trauma, and pathology fit within a relational paradigm.

philosophical considerations      225

Congruence With Means to Promote Social Change
Multicultural psychologists explicitly promote change in their profes-
sion and in society (Ivey & Zalaquett, 2009; Totikidis & Prilleltensky, 2006).
Change must involve the social world, particularly relationships, if it is to be
sustained. Confrontation and political advocacy can transform institutional
policies, but ultimately, individuals need to adopt a different worldview for
change to persist. For instance, people may continue to tell racist jokes in
private despite an antiracism policy, but they typically stop telling racist jokes
once peers frown rather than smile. Real changes occur when social networks
reinforce stated organizational values. Multicultural psychology seeks not
merely policy change but genuine social inclusion, with efforts that address
many social levels being the most effective.

Congruence With Cultural Values

A relational paradigm seems aligned with the values of many cultures
worldwide. Indigenous African worldviews, Native American Indian world-
views, Central and South American worldviews, Asian worldviews, and Pacific
Islander worldviews tend to emphasize family relationships over individualism.
A relational paradigm challenges and stretches individualistic cultures,
but it does not necessarily conflict with them. Even in the most extreme indi-
vidualistic cultures, genuine interpersonal intimacy has remained a cohesive
force (i.e., families); thus, these cultures already attend to relational issues to
some degree. A relational paradigm provides a bridge between individualistic
and collectivistic cultures. It also provides explicit affirmation of cultural
values not adequately represented in mainstream psychology.

Congruence With the Notion of Intersectionality

Race and culture interact with gender and sexual orientation, which inter-
act with family structures and geographic region, among many other variables.
Multiculturalism increasingly attends to these intersections (McNeill, 2009;
T. B. Smith & Draper, 2004). A relational paradigm offers a framework from
which to conceptualize and operationalize the complex intersections of human
diversity, which ultimately have social meanings, functions, and consequences.
We are not merely groups and not merely individuals; investigation of inter­
sections necessitates holistic reasoning.

Clarification Summary of Benefits of a Relational Paradigm

for Social Justice Work
Earlier we listed three strengths of the concept of social justice for multi-
cultural psychology: its emphasis on power, action, and empowerment. These

226       foundations of multicultural psychology

strengths do not require a paradigm associated with philosophical material-
ism. Holding people accountable for abuses of power does not necessitate
critical pedagogy, but it does require the moral principle of accountability.
Accountability pervades the relational paradigm. Its emphasis on interdepen-
dence checks self-interest, the primary reason for abuses of power. Moreover,
the whole notion of power remains grounded in lived experiences arising from
intergroup and interpersonal exchange (Gergen, 1995). Thus, the conceptual-
ization of power is made explicitly relevant to applied psychology when power
is viewed through a relational lens.
Similarly, the need to take action against oppression does not necessarily
require political activism, but it does require the principle of social responsibil-
ity. Responsibility to others is keenly felt through a we-consciousness. Action
orientation characterizes the principle of other-engagement. Interaction
necessarily entails action.
Likewise, empowerment of oppressed peoples can occur not only through
raising liberation consciousness but also through integrating social networks.
A relational paradigm affirms individual, family, and group rights and their
associated protections by emphasizing that separate but equal is not equal. It is
insufficient for groups to assert their own rights. Protest may receive attention,
but it does not necessarily promote engagement among parties. Equality is not
equality without social network integration.1 A relational paradigm promotes
interactions as equals (e.g., Oliver, 2001). When justice is disallowed, steps are
taken to reengage dialogue, such as affirmations of equality, appeals to moral
sensibilities, appeals to influential third parties, and explanations for refusals
to submit to injustice. A relational paradigm seeks genuine integration and
equity that includes but extends beyond the economic and political.
Within a relational paradigm, a primary motive is mutual engagement
as equals and rejection of the roles of oppressor or oppressed. This approach
seeks to change the contexts that led to the power imbalance in the first place
and to replace the disempowering notions that oppressed groups too often
internalize when reacting from defensive postures. Rather than promote social
justice in terms of “us versus them,” mutual edification provides motivation
for continued engagement across divides. Thus, a relational paradigm sustains
action against oppression because the motivation transcends self-interest.
A relational paradigm and philosophical materialism both attend to issues
of power, access, status, coercion, and so forth (see Table 11.1). One cannot
accurately conceptualize individuals, families, or groups without those concepts,
but to those important concepts a relational paradigm adds sources of affiliation,
ideals, and so forth, such as abilities, gender, geographic region, race, religion,

1Social integration does not necessarily entail assimilation or acculturation. Engagement across difference
is one component of well-being.

philosophical considerations      227

 TABLE 11.1
Comparison of Two Philosophical Positions for Fostering
Social Justice Through Multicultural Psychology
Philosophical materialism Relational paradigm

Primary aspiration Equality Mutual enrichment

Primary emphasis Access to power Holistic well-being
Conceptualizations Critical pedagogy, dialectical Accountability,
of power materialism interdependence,
moral principles
Action orientation Advocacy, praxis, transformation Social responsibility,
of systems protection of human
rights
Empowerment Liberation psychology, grassroots Social network integration,
political activism skill development
Level of primary Macro level systems (and other Intergroup, interpersonal
focus levels as appropriate)
Motivations Emancipation, fighting against Engagement as equals,
oppression to obtain justice dismantling oppressed/
oppressor roles
Note. Common features include the following: emphasis on local community empowerment, action-oriented
promotion of equality and self-determination, explicit opposition to all forms of oppression, and attention to
human conditions and contexts, including issues of power, opportunity, status, coercion, and so forth.

and sexual orientation, that are only indirectly addressed by philosophical

materialism. Material considerations are crucial, even paramount, in desperate
situations, and they overlap with social considerations. Thus, a relational para-
digm maintains the focus on poverty, inequity, oppression, and so forth (the
strength of materialism), while contextualizing those issues in lived experience.
For these reasons, we propose that for applied psychology a relational
paradigm is preferable to alternative social justice conceptualizations. Spe­
cifically, we believe that social justice (an aim of multicultural psychology)
can be better measured, evaluated, and promoted within a relational paradigm
relative to prevailing conceptualizations based on philosophical materialism.
Whether or not future scholarship takes up this issue, we have attempted
to emphasize the point that ideational foundations do matter. Assumptions
influence outcomes.

What Assumptions and Limitations Must Be Expected

for a Relational Paradigm?

Every approach has assumptions and limitations. Just as the strengths

of materialism led us to uncover its possible weaknesses, the strengths of a
relational paradigm also point to its weaknesses. The following limitations
characterize a relational paradigm.

228       foundations of multicultural psychology

Psychological Explanations Can Obfuscate the Value of Other Perspectives
Human behavior can be explained at numerous levels, from the micro
(neurochemical) to the macro (environmental). A relational paradigm clearly
emphasizes interpersonal and intergroup exchanges at the expense of other
levels of explanation. Relational theorists have tended to ignore biologically
oriented research findings, such as those of neuropsychology. Macro issues
such as warfare and access to health care are relevant to but clearly distanced
from purely social causes. And with few exceptions, relational scholarship
shies away from the traditionally influential cornerstones of psychology, such
as comparative psychology and radical behaviorism, as well as technically
oriented dimensions, such as computer simulations of human cognition.
Excessive reliance on the interpersonal level of explanation, even if congru-
ent with the worldviews of mental health professionals, artificially constrains
attention when other mechanisms (e.g., ambient pollution or neurochemistry)
may be more pertinent.

Complexity Restricts the Isolation of Variables

Although a relational paradigm accounts for the contextuality of human
experience, the resulting complexity diminishes the likelihood of isolating
explanatory variables. Even when a specific characteristic is isolated, the con-
ditions affecting that characteristic are potentially infinite. Contextuality
cannot coexist with simplicity; causality becomes difficult to explain.2
Reliance on correlation more than causal models has been one of the
primary weaknesses of psychological research. Nevertheless, we are just now
reaching a point where statistics may enable sufficient complexity in our
data collection and analyses to move beyond correlation. Social network
analyses have increased in their complexity and utility over time (e.g.,
Borgatti, Mehra, Brass, & Labianca, 2009; Kirke, 2007), but additional sim-
plification of the tools for conducting social network statistics is necessary
before graduate curricula in psychology will routinely cover those statistics.
Nevertheless, given the explosion of Internet-based inquiry and statistical
modeling of social networks, we see their widespread use as an eventuality:
Analytics possible through supercomputers will examine trillions of paths
of social influence such that the network shape and directional flow become
apparent and open to inquiry. Until such statistical tools become widely
available, however, research conducted within a relational paradigm must
rely on traditional methods for attempting to provide causal explanations
(Kuhn, 1996).

2Nevertheless, true experimental designs can be used within a relational paradigm. Social psychologists
use a remarkable variety of research methods that could be adopted in multicultural psychology.

philosophical considerations      229

Including Notions of Morality Would Require Discourse
Beyond Current Parameters
If human interactions are fundamentally moral, as a relational paradigm
affirms, morality must be addressed by psychology. Except for the publication
of self-regulating codes of ethics, the profession of psychology has largely
sidestepped morality. In fact, psychology was originally developed in par-
tial reaction against the notion of morality and the institutions, religious
and aristocratic, that overtly enforced their own versions of morality. Thus,
psychologists may have difficulties accepting the emphasis of the relational
paradigm on moral issues. Nevertheless, psychologists constantly confront
questions of meaning, not merely questions of description.
Some scholars have argued for decades that psychology’s discomfort with
moral conceptualizations has not served its interests. Specifically, they have
emphasized that all scientific inquiry remains influenced by human values;
thus, openly acknowledging those values is in the profession’s best interest.
Theory and research should be contextualized (Slife & Williams, 1995). And
psychology will benefit from greater self-awareness and accompanying trans-
parency. Multicultural psychology advocates for greater self-awareness and
transparency as well.
A paramount fear is that entertaining professional discourse about moral-
ity would paralyze the field. Would progress not degenerate into the morass of
debate and counter-accusation? With little prior experience engaging moral
issues, this could happen. Researchers have been systematically taught to hide
personal values in professional writing. To overcome discomfort in debating
the value, meaning, and purpose of their work, psychology researchers would
have to first recognize that questions of value, meaning, and purpose are in fact
the most important questions. Discussions of whether variable X correlates
with or even causes variable Y becomes appropriate in psychotherapy only after
we understand the ramifications of messing with X and Y for a particular client.
Justification for research should be based on arguments about value, meaning,
and purpose, with those justifications subject to challenges and refutation. We
have to invoke the “so what?” question much more often with our own work
and with the work that appears in multicultural psychology journals.

Section Summary: Values and Assumptions

Social justice is a value—a value based on assumptions about human
dignity and fairness. Work that promotes principles of justice and fairness is
a moral endeavor. Multicultural psychology embodies that work; thus, it is a
moral as well as a professional and empirical movement.
Multicultural psychology seeks to promote the well-being of historically
oppressed people. The more clearly multicultural psychology can articulate
its objectives and its proposed means to achieve them, the more support it is

230       foundations of multicultural psychology

likely to draw. Contemporary struggles for equity, such as the denunciation
of racial microaggressions (D. W. Sue et al., 2007), require this articulation
in the face of opposition or, more prevalent, apathy. Ultimately, psycholo-
gists will embrace and infuse multiculturalism in psychotherapy (and in their
personal lives) to the extent that it becomes recognized as the right thing to
do. Moral sensibility is embedded in social relations.
In this section we have attempted to delineate how a relational para-
digm may provide grounding for social justice work in multicultural psycho­l­­
ogy. Assumptions and values of the relational paradigm include the aim of
mutual edification, the necessity of holistic thinking, the existence of human
volition and agency, the developmental perspective of becoming, the under-
standing of identity as a self-in-relation, the inescapability of responsibility
for others, and the existence of human rights—moral obligations to protect
human vulnerabilities.
The objective of multicultural psychology is not to achieve mere toler-
ance, the “recognition” of differences (see Oliver, 2001). Rather, multi­cultural
psychology has sought to promote human well-being through self-affirmation
and, although not articulated as such, other-affirmation. If multicultural psy-
chology seeks these broad aims, its work extends beyond even social justice.
It may rightly advocate for any salutary principle, such as reconciliation,
personal sacrifice for the well-being of others, and deepened interchange
across apparent and genuine ideational differences. Applied psychologists
are already in the business of promoting values (e.g., Magyar-Moe & López,
2008); multicultural psychologists can promote values that benefit disenfran-
chised populations.
A relational conceptualization of multicultural psychology espouses
values that promote mutual enrichment. Whereas the concept of self-
affirmation may presently be popular, a relational conceptualization includes
the paired concept of other-affirmation. Individuals do not exist in a social
vacuum. Affirmation of self yields reciprocity through affirmation of others. In
other words, when people engage in other-affirmation (e.g., schoolteachers
who empower students in an otherwise harsh environment), we call them
praiseworthy (affirming the person who affirms others), but praise for oneself
without genuine engagement with others we call narcissism. So it is with
multicultural psychology: Our work is insufficient if we merely affirm multi-
cultural voices, each one calling out its own music. Expression is far prefera-
ble to voicelessness, but cacophony attracts few listeners. No, self-affirmation
of culture, race, gender, or any other partitioned aspect of human identity is
an aim too delimited for multicultural psychology, even if justice were techni-
cally achieved because no one restricted expression.
Continuing the metaphor of vocal music, a relational conceptualization
offers multicultural psychology the equivalent of a music school. Voices can

philosophical considerations      231

tune to surroundings. The quality of individuals’ and groups’ expressions can
improve. In a music school the ear can be taught to hear others’ tone and
timbre and to recognize the themes and motifs already native to their inflec-
tions. Schools of music enable compositions, orchestrations, and production
of quality performances that generate an audience. Unity amid diversity
can be attained through multicultural psychology, a school for relationships
learned through experience.
A relational conceptualization of multicultural psychology seeks justice
toward the aim of mutual enrichment. Voices must not merely be heard but
understood, appreciated, and joined.

BRINGING IT ALL TOGETHER

Psychology consists of ideas about human experience. Those ideas stem

from underlying philosophies and their associated assumptions. For most of
its history, psychology has presumed the experiences, worldviews, and phi-
losophies of cultures with origins in Western Europe, to the detriment of
others (D. W. Sue, 2015). Multicultural psychology has sought to represent
people previously excluded from mainstream dialogue, and it has brought
attention to significant ideas, such as those covered in previous chapters of
this volume (e.g., multicultural competence). Those ideas can be refined not
only through improvements in empirical methods (e.g., assessment, partici-
pant selection, theory testing) but also through appraisal of their underlying
values and assumptions (Machado & Silva, 2007; Slife & Williams, 1995).
Just as the assumptions of psychology have benefitted from scrutiny with a
multicultural perspective (D. W. Sue, 2015; D. W. Sue & Sue, 2013), the
assumptions of multicultural psychology should benefit from evaluation.
Questioning assumptions, a few readers have likely wondered whether
this chapter about philosophy was necessary in a book otherwise about
data. We ask these readers to consider our intended messages. We hope that
many readers who have previously been skeptical about the relevance of
a broad concept such as justice to psychology have become aware that
real-world psychological implications and applications are generated
by such principles as “injustice anywhere is a threat to justice everywhere”
(quoted from Martin Luther King Jr.’s “Letter from Birmingham Jail”).
Such ideas motivated millions during the civil rights movement. Such
ideas also motivate our work as mental health professionals: If a rela-
tional paradigm enables me to recognize my interconnectedness, I will
be more likely to act when I see injustice, but if I fixate on the inequities,
I may be more likely to react from a stance akin to naive transitivity. Is
this contrast merely a nuance, too subtle to be consequential? Ask about a

232       foundations of multicultural psychology

person’s motives, and you will be in a better position to understand resulting
actions and reactions. That sounds like the work of a psychologist.
Too few mental health professionals have addressed the philosophi-
cal foundations of multicultural psychology, with the notable exceptions
focused on research methods and guiding principles (Cauce, 2011; David,
Okazaki, & Giroux, 2014; Gone, 2011; Ponterotto, 2010). Questions posed
in this chapter openly challenge commonplace apathy about conceptual
analysis. Table 11.2 contains some steps to consider. Of all people, multi-
cultural psychologists should be keenly aware of our own assumptions and
values.

TABLE 11.2
Example Components of Conceptual Analysis
Conceptual analysis
component Rationale
Define constructs precisely. Specification is essential to all subsequent steps of
conceptual analysis. Problems arise from imprecision.
Identify the level(s) of Most constructs are pertinent or valid at only one level
analysis to be under- of explanation, but multiple levels of explanation are
taken (macro to micro, typically necessary in psychology. Constructs should
abstract to concrete). not be generalized beyond their realistic limits.
Identify essential parts of Breaking down constructs can help distinguish truly
the construct. essential features, circumstances, and so forth,
including parts that must be included for the concept
to remain viable.
Identify how the parts Dynamics must be observed and considered,
relate to one another particularly relationships between parts that are not
and to the whole. explained by the proposed theory or construct.
Identify strengths and Construct application requires understanding of
limitations. when and how it works most and least effectively.
Strengths and limitations can be paired: A strength
can be a limitation; a limitation can be a strength.
Identify a concrete case Practice requires understanding of when, where, and
that demonstrates the how concepts fit and do not fit in the real world,
construct, and contrast including exceptions that remain unexplained.
it with contradictory or
hypothetical cases.
Identify alternative Understanding of similar constructs should be used
explanations. to inform analysis. This includes existing relevant
theories and other disciplines that have addressed
similar issues, possibly using other terms.
Identify metaphysics, Every construct has foundational assumptions: for
epistemology, ontology, example, the nature of reality, ways we can know
and so forth. about the world, and so forth.
Consider real-world Constructs have many possible ramifications: people
consequences. who will benefit or be harmed, possible misinterpre-
tations and misuses, consequences of ignoring it,
and so forth.

philosophical considerations      233

 No field can solve astoundingly complex social situations through a
single lens. Many voices contribute to effective solutions. Scholarly synthesis
and contrast, rather than reverential adherence to a few popular ideas, pro-
motes the aims of multicultural psychology. We are reminded of the words of
a Mexican Nobel laureate in literature:
What sets worlds in motion is the interplay of differences, their attrac-
tions and repulsions. Life is plurality, death is uniformity. By suppressing
differences and peculiarities, by eliminating different civilizations and
cultures, progress weakens life and favors death. The ideal of a single
civilization for everyone, implicit in the cult of progress and technique,
impoverishes and mutilates us. Every view of the world that becomes
extinct, every culture that disappears, diminishes a possibility of life.
(Paz, 1985, p. 117)
Across human history, few societies have been multicultural. What can
we learn from those societies that have been? Across human history, no age
has been so globally networked as the present. How will we of the present
age connect while retaining cultural plurality? Across human history, no age
has had greater intellectual and material resources. How will we foster mutual
enrichment? Questions expand the vista of multicultural psychology.

234       foundations of multicultural psychology

 12
FIRMING UP THE FOUNDATION
FOR AN EVIDENCE-BASED
MULTICULTURAL PSYCHOLOGY

Psychotherapy and related mental health services are based on decades of

research (Norcross, 2011). Mental health professionals understand the benefits
of research in improving treatment practices and fostering individuals’ well-
being. Indeed, evidence-based practices have become the standard for the
profession (American Psychological Association [APA] 2005 Presidential
Task Force on Evidence-Based Practice, 2006).
Multicultural psychology should also be based on research evidence
(G. Bernal & Domenech Rodríguez, 2012; Cauce, 2011; Leong, Comas-Díaz,
Nagayama Hall, McLoyd, & Trimble, 2014; G. C. N. Hall & Yee, 2014). A
primary purpose of this book was to evaluate the degree to which multicultural
psychology is being built on a solid research foundation. This book covered
only select topics within multicultural psychology that were specific to race
and ethnicity, but in general, the conclusions were positive. Although there
are still pressing needs for greater empirical evidence stemming from a lack of

[Link]
Foundations of Multicultural Psychology: Research to Inform Effective Practice, by T. B. Smith and J. E. Trimble
Copyright © 2016 by the American Psychological Association. All rights reserved.

235
attention to multiculturalism in some segments of the mental health profes-
sions, remarkable gains have been made in recent decades. And those gains
deserve acknowledgement. Multiculturalism is no longer a peripheral issue
in many professional circles, and increased influence in both research and
practice is the current trajectory. Looking to the future, the continued influ-
ence of multicultural psychology on psychotherapy and other mental health
treatments seems certain (Pedersen, 1999).
Notwithstanding the noble ambitions and rapidly increasing influence
of multicultural psychology, it seems wise to not extend the reach of advocacy
too far beyond the available research evidence (Huey, Tilley, Jones, & Smith,
2014). Many details of multicultural issues relevant to mental health and
well-being remain presumed or unknown. We have much to learn.

RECOMMENDATIONS FOR STRENGTHENING THE

RESEARCH FOUNDATION OF MULTICULTURAL PSYCHOLOGY

To solidify the research foundation of multicultural psychology, four areas

for improvement can be considered: (a) maintaining focus on the essential
dependent variables; (b) attending to meta-analytic data to prioritize inde-
pendent variables; (c) increasing specificity and improving research methods;
and (d) developing, testing, and refining theories. These recommendations
deserve serious attention (see also Awad & Cokley, 2010; B. M. Byrne
et al., 2009; Huey et al., 2014; M. J. Miller & Sheu, 2008; Ponterotto, 2010;
Worthington, Soth-McNett, & Moreno, 2007). They will help solidify knowl-
edge foundational for the construction of a truly multicultural psychology.

Focus on the Essential Dependent Variables

in Mental Health Service Provision

Empirical data can and should inform multicultural psychology (Cauce,

2011), but as pointed out in the previous chapter (Chapter 11), data are gen-
erated and interpreted on the basis of assumptions about what is important
and what is not (see also Slife, 2009). And because values and assumptions
differ, research agendas vary. Even a casual perusal of the multicultural psy-
chology literature quickly confirms wide variability in the topics covered.
Although diversity of thought can improve the profession, scattered research
agendas can be problematic if they inadequately address the most consequen-
tial dependent variables. For mental health professionals, those would be
variables within the scope of their influence that would have the greatest
impact on the clients they serve. In the previous chapter we proposed four:
(a) reduction of mental health disparities, (b) access to services by those in

236       foundations of multicultural psychology

need, (c) retention of those receiving services, and (d) improved outcomes
to the satisfaction of those receiving services.
Does the field adequately address those four aims? To evaluate the
amount of research covering the four aims, we conducted a content analysis
of the applied psychology literature relevant to race, ethnicity, or culture cover-
ing a 6-year period. We sought to evaluate scholarship explicitly addressing
psychotherapy and related mental health services, so we limited our search to
the PsycINFO database using a string of keywords specific to mental health
treatments crossed with variants of the terms race, ethnicity, and culture. Our
searches yielded 3,484 hits, of which 805 turned out to be irrelevant on
further examination; 895 tangentially mentioned race, ethnicity, or culture
in the abstract but the manuscript was primarily relevant to another topic;
and 239 were book reviews. Using established procedures for content analysis
(Krippendorff, 2012), coders assigned the remaining 1,545 manuscripts to
33 content categories, with manuscripts having content that overlapped
categories (i.e., both social justice and ethical issues) being coded in a separate
column from manuscripts having content pertinent to a single category. This
procedure enabled calculation of percentages representative of multiple catego-
ries. In the end, however, the percentages observed for the single-category
studies were virtually the same as those of the multiple-category studies, so we
report only the former for the sake of parsimony. Two coders independently
reviewed every manuscript, with the data averaged between coders reported
in Table 12.1 for 22 of the 33 content-area categories because the remaining
11 categories had only a few manuscripts each.
Table 12.1 provides a snapshot of the content for publications mention-
ing race, ethnicity, or culture that were specific to mental health treatments.

TABLE 12.1
Topics Relevant to Race, Ethnicity, or Culture in the Literature Specific
to Mental Health Treatment
More common topics % Less common topics %
Diagnosis rates by race/ethnicity 13 Perceived racial/ethnic discrimination 3
Therapist multicultural competence 12 Ethnic identity 3
Cultural adaptations to treatment 10 Ethical issues relevant to race/ 2
Client expectations about treatment 9 ethnicity
Therapist multicultural education/ 8 Multicultural supervision 2
training Racial identity 2
Client treatment outcomes by race/ 8 Client–therapist racial/ethnic match 1
ethnicity Individualism/collectivism 1
Client utilization of treatment 7 Social justice 1
Assessment/measurement issues 4 Client participation in treatment 1
Attitudes toward mental illness 4 Client spirituality 1
Therapist prejudice/bias 4 Color-blind racial attitudes 1
Acculturation 4

firming up the foundation╇╇╇╇╇ 237

Topics central to mental health treatment were represented more often than
topics peripheral to mental health treatment. The topic most often addressed
was diagnostic rates by race or ethnicity, which is directly relevant to the first
of the four proposed aims of the field. Therapist multicultural competence and
cultural adaptations to treatment were the second and third most commonly
addressed topics, which are both relevant to the other three aims. When consid-
ering the whole, about three fourths of the literature could be construed to align
with one or more of the four proposed aims presented earlier in this chapter.
Some scatter characterizes the multicultural psychology literature, but at pres-
ent, two recommendations for improvement of research focus may suffice.
First, greater attention can be given to client outcomes. Only 8%
of manuscripts addressed client outcomes in treatment, a percentage that
seems low given (a) the importance of documenting client improvement
in treatment and (b) the fact that many of the manuscripts retrieved did
not contain data, meaning that fewer than 8% of manuscripts actually
evaluated client outcomes. We urge greater attention in the multicultural
psychology literature to clients’ outcomes in treatment (see also Lau, Chang,
& Okazaki, 2010).
Second, researchers may have to more frequently evaluate the degree
of client participation in treatment (i.e., completion and drop-out rates).
The fact that only 1% of manuscripts addressed client participation in treat-
ment seems problematic. The third of the four proposed aims appears to be
neglected. Nevertheless, there may justifiable reasons why client participa-
tion in mental health treatments has not been examined more frequently.
Specifically, the data in Chapter 5 of this book suggested that racial discrepancies
in client participation rates may be less severe than previously imagined.
Overall, aligning multicultural psychology research with the experiences
of individuals in need of mental health services will greatly improve the utility
of present scholarship. Whereas 1,545 manuscripts that addressed race, ethnic-
ity, or culture and that were specific to mental health treatments were identi-
fied in the PsycINFO database over a 6-year period, over 40,000 manuscripts
that addressed race, ethnicity, or culture during that same time frame did not
mention mental health treatment (no mention of the root terms psychotherapy,
treatment, therapy, intervention, program, service, clinic, counseling, patient, or client).
It would appear that although issues of race, ethnicity, and culture are being
widely attended to, proportionately little research in psychology addresses
dependent variables essential to mental health services.

Synthesize Research Findings and Prioritize Efforts Using Those Data

Prioritizing research foci involves both conceptual and empirical con-

siderations. Systematic literature reviews and meta-analyses prove useful in

238       foundations of multicultural psychology

making those decisions. Whereas thousands of systematic literature reviews
and meta-analyses have been conducted in the general psychology literature
(Cooper & Koenka, 2012), relatively few have been conducted in multicultural
psychology.
As one step toward informing the field of priorities for investigation,
the meta-analyses reported in this book can be considered together, with the
data portrayed in a single display. Figure 12.1 depicts the effect sizes found
within the several meta-analyses, all converted to the metric of Cohen’s d.
When depicted in this manner, several points can be made after sounding a
caution.1 At one level of interpretation, Figure 12.1 provides the numbers
of manuscripts (k) with data on each topic. A comparison of the number of
manuscripts with data makes it clear that studies of therapist multicultural
competence were underrepresented in the literature relative to other topics;
a field cannot establish its credibility on solely 16 studies. Conversely, other
topics may have received relatively excessive research attention. Specifically,
a substantial investment has been made to investigate strength of ethnic
identity, yet ethnic identity accounts for an average of 3.3% of the variance in
the well-being of people of color—and the direction of causality has not been
established (see Chapter 10). The overall mental health literature on ethnic
identity consists of several thousands of research studies, which represent
millions of hours of effort expended (hours of work by authors × number of
authors × number of studies, plus hours spent by participants). In the future
it may be more advantageous to invest research efforts in variables with dem-
onstrated causality that account for substantial variance in well-being, such
as socioeconomic conditions (e.g., McKee-Ryan, Song, Wanberg, & Kinicki,
2005), coping strategies (e.g., Ano & Vasconcelles, 2005), relationship and
family qualities (e.g., Proulx, Helms, & Buehler, 2007), and variables asso-
ciated with positive psychology (Sin & Lyubomirsky, 2009). Professionals

1
The data in the several meta-analyses cannot be compared against one another without an accurate
understanding of the context of each meta-analysis. The topics are entirely distinct in content, measure-
ment, and methodology, just as apples are distinct from oranges, even though both can be digested.
Given the severity of concerns about data misinterpretation, we hesitated to generate Figure 12.1.
Nevertheless, we trust the reader to become informed by reviewing each meta-analysis and thus avoiding
uninformed comparisons. As one of several errors that can be made when considering Figure 12.1, a
reader observing the very large effect size found when client experiences in therapy are associated with
client ratings of their therapist’s multicultural competence (the third row in Figure 12.1) could falsely
conclude that that topic is the most important or (even worse) the most effective. Those conclusions
would be erroneous because the information presented in Chapter 3 made it clear that that large effect
size is highly problematic for several reasons, particularly when contrasted with the small effect size
obtained when therapists’ rate their own multicultural competence. In this instance, the benefit of the
graph is the clear contrast between the two types of measurement, not the pretentiously large effect
size observed when using one problematic measure. Another source of error in interpreting Figure 12.1
would be a failure to account for the size of the corresponding literature. The numbers of studies (k) for
the data from Chapter 3 (the second and third rows) are much fewer than the numbers of studies in the
other meta-analyses; only seven studies evaluated therapists’ perceptions and only 11 studies evaluated
clients’ perceptions.

firming up the foundation      239

240       foundations of multicultural psychology 0 0.2 0.4 0.6 0.8 1 1.2 1.4
Chapter 2
Therapist mulcultural educaon & mulcultural competence (k = 23) A
Chapter 3
Therapist mulcultural competence rated by therapist (k = 7)
Therapist mulcultural competence rated by client (k = 11)
Chapter 4
Treatment ulizaon discrepancy by race: Overall (k = 130) A
Treatment ulizaon discrepancy for Asian Americans (k = 31)
Chapter 5
A
Treatment a‰endance discrepancy by race: Overall (k = 48)
Treatment a‰endance discrepancy by African Americans (k = 33)
A
Treatment compleon discrepancy by race: Overall (k = 31)
Treatment compleon discrepancy by African Americans (k = 17)
Chapter 6
Treatment parcipaon by racial match with therapist: Overall (k = 53)
Treatment parcipaon by racial match: Asian Americans (k = 16)
Chapter 7
A
Outcomes in culturally adapted treatments: Overall (k = 79)
B
Outcomes in culturally adapted treatments: Highly adapted (k = 13)
Chapter 8
Acculturaon and mental health services: Overall (k = 107)
Acculturaon and mental health services: Immigrants (k = 16)
Chapter 9 A
Perceived racism associated with well being (k = 81)
Chapter 10
Ethnic identy associated with well-being: Overall (k = 215)
Ethnic identy associated with self esteem (k = 136)
Ethnic identy associated with mental health (k = 126)

Figure 12.1. Effect sizes (Cohen’s d) and 95% confidence intervals for meta-analyses reported in chapters of this book. A = adjusted for
publication bias; B = studies with at least six of eight recommended cultural adaptations; k = number of studies.
intent on enhancing the well-being of people of color will benefit from
attending to variables such as these to a greater degree than variables that
have proven less consequential.
Acculturation to North American society has also been widely examined,
with over 1,500 relevant studies appearing in PsycINFO. Yet the review in
Chapter 8 found fewer than two dozen studies investigating immigrants’
experiences in mental health treatments. The data indicate the obvious:
The construct of acculturation is most salient for immigrant populations.
Professionals can therefore align with the data and either analyze differences
within samples (including curvilinear associations) or discontinue the com-
mon practice of administering generic measures of acculturation when level
of acculturation is unlikely to account for variance in the dependent measure
administered to individuals with moderate or high levels of acculturation.
Culturally adapted mental health treatments are increasingly recognized
as evidence-based alternatives to traditional treatments (Cardemil, 2010a;
Huey et al., 2014). According to the meta-analysis in Chapter 7, clients
appear to benefit most when the adaptations are systematic (following estab-
lished guidelines) and multifaceted. That is, the more a treatment aligns with
the cultural worldviews and experiences of a client, the more effective the
treatment is likely to be.
In general, clients engage in and attend or complete mental health treat-
ments at similar rates across race (Chapters 4 and 5). The clear exceptions are
that Asian Americans are much less likely than other groups to enter profes-
sional treatment and are much more likely to remain in treatment when the
therapist is of the same race or ethnicity, and African Americans are much
less likely than other groups to remain in treatment. Attention to the expla-
nations for those discrepancies can facilitate the implementation of needed
steps to remove them.
In short, the field of multicultural psychology will benefit from attending
closely to systematic literature reviews and meta-analyses. At present, the
lack of clinical studies (research with actual mental health clients) appears to
be the greatest gap in the overall corpus of scholarship (see also Lau et al., 2010;
Worthington et al., 2007). Future meta-analytic reviews will continue to
redirect efforts in the ongoing process of improvement based on data.

Specify Constructs and Improve Research Methods

Multicultural psychology seeks to inform mental health professionals of

the essential variability in human conditions and experiences. Nevertheless,
that intention is undercut by the high degree of variability that characterizes
research findings in multicultural psychology, as evidenced by the scatterplots
(and large I2 values) accompanying each meta-analysis in this book. Effect

firming up the foundation      241

sizes vary in any field. In our analyses, they varied so much that the random
effects weighted averages reported in some chapters could be criticized as
informative fiction. Yes, the averaged effect sizes represent the best overall
estimate, but a single research study might yield anything, from moderately
positive to moderately negative results. Under these circumstances, trust
in the findings of any single quantitative study, particularly those with few
participants, becomes restricted to a degree approaching disregard.
Wide variations in circumstances and participant characteristics already
introduce multiple explanations for the variability of research findings in multi­
cultural psychology, but when both known sources of variation and threats
to internal validity are unaccounted for in studies, when research procedures
are inadequately described, and when there are no standard approaches to
evaluating a given phenomenon, ascertaining the “true effect size” is highly
improbable (J. Miller & Schwarz, 2011). For our estimates to become more
reliable, our measurements and our methods must improve.
Greater specificity is one solution. Multicultural psychology is complex,
but its literature is characterized by frequent use of categorical assumptions.
Categories of race, ethnicity, gender, sexual orientation, religion, and so forth,
are assumed to suffice as independent variables, yet we have known for decades
about extreme variability within groups (e.g., Phinney, 1996) and overlap across
groups (e.g., biracial identification; Young, Sanchez, & Wilton, 2013). As one
negative consequence among many, graduate classes in multicultural psychol-
ogy often provide categorical group-level descriptions implicitly portrayed as
sufficient. Categories and categorical thinking can be useful heuristics for the
completely uninformed, but at what cost do we perpetuate group stereotypes
among individuals being trained as experts in human behavior?
Scholars affirm that professionals should attend to proximal variables, such
as the meanings and consequences of ethnicity, not merely categorical ethnicity
(e.g., S. Sue, 1988). Categorical operationalization of race, ethnicity, and
culture is too imprecise to yield more than general trends. As important as
the constructs of race, ethnicity, and culture are, an entire field cannot help
but crawl forward while relying so heavily on generic, distal variables. It is time
to break apart categorical approaches and precisely describe aspects of experi-
ence, such as the multiple components of racial socialization or cultural values of
holism, useful to improving mental health and quality of life. Researchers should
specify their questions, measurements, and analyses, rather than continue to rely
on the easily generated but largely ineffective check box.
Another solution is to use methods that explicitly account for variability
and control for potential confounds. Presently, much of the multicultural psy-
chology literature consists of surveys and correlational studies, with a relative
absence of research designs that account for sources of error and differences
across contexts. Multicultural psychology is important. It is so important that

242       foundations of multicultural psychology

the research deserves to be done well (e.g., Awad & Cokley, 2010; B. M.
Byrne et al., 2009; Lau et al., 2010; Ponterotto, 2010). With that consideration
in mind, university research advisors, journal editorial board members, and
anyone now informed by the scatterplots in this book can commit to elevate
expectations for research quality in multicultural psychology.
Improving the quality of quantitative research need not entail inordinate
amounts of effort. Six steps should have a consequential impact: (a) precisely
define constructs, optimally broken down into subcomponents; (b) evalu-
ate proximal variables rather than distal or proxy variables; (c) investigate
sources of within-group and between-group variability; (d) control for likely
confounds, randomly select participants, and randomly assign participants
to conditions when ethical or feasible; (e) use very conservative estimates
of effect size when conducting a power analysis to estimate the number of
participants needed in a study; and (f) have theory guide research questions
and test theory in the research. Other recommendations specific to qualita-
tive research can also be followed (e.g., Ponterotto, 2010), but steps such as
these six will enhance the utility of quantitative research efforts. Believing
the sixth recommendation, to attend to theory, to be particularly important,
we elaborate on that recommendation.

Develop, Test, and Refine Theory

In conducting our meta-analytic reviews of multicultural psychology

research (Chapters 2 to 10), we observed that few manuscripts involved research
questions explicitly based on relevant theories, and even fewer attempted to
evaluate theories. Those that did were predominantly the notable leaders in
the field and their graduate students. Although the work of those leaders is
truly admirable and has attracted followers, the work of the followers does not
elaborate much on what has already been proposed. Collectively, we seem
reluctant to critically engage theory and push one another to reconsider and
refine. Metaphorically, it seems that we are more interested in riding comfort-
ably along on the bandwagon than in terrain exploration.
As pointed out in the previous chapter (Chapter 11), we are often
reluctant to engage in the terrain exploration of theory development chiefly
because we are unaccustomed to doing so, not because we are unsuited to the
work. And the benefits of theory development are substantial. At a time when
laboratory experimentation was the norm, a revolutionary young physicist,
Albert Einstein, engaged in thought experiments (Ono, 1982). Revolutions
in science begin with novel conceptualizations (Kuhn, 2012). At a time
when most multicultural psychology research involves self-report surveys, we
need both more laboratory experimentation and more thought experiments.
Trailblazing enables progress.

firming up the foundation      243

 At this point it may be helpful to consider a concrete example of the
need for explanatory theories in multicultural psychology. A conceptual
challenge to both categorical notions of race, ethnicity, and culture and to
the attempts at specificity intended to correct for the problems of that cate­
gorization, as discussed in the previous section, comes from intersectionality
(e.g., McNeill, 2009). Intersectionality involves a holistic perspective inclu-
sive of the multiplicity of identities and experiences that cannot easily be
categorized or specified without committing misrepresentation. An individual
is simultaneously racial, ethnic, and cultural along with having multiple other
attributes (age, gender, etc.). Although we applaud emerging scholarship that
accounts for intersectionality (e.g., Miville & Ferguson, 2014), much work
remains to be done. How will such complexity be operationalized in research
and in professional training? How will intersectionality be put into practice?
We anticipate the development of relevant theories.
Given the overall need for theory development, where should scholars
in multicultural psychology begin? Which of the many topics and challenges
in the field should be targeted in the future? In considering how to narrow
recommendations, it seemed appropriate to solicit input from renowned
scholars in the field. They have been engaged in the work for decades, and
their broad vision can inform speculation about pressing future needs. We
therefore requested the opinions of several leaders in multicultural psychology
about what the profession should consider for the future, given demographic
projections and likely social conditions. We received the following responses,
which we have edited for brevity.
77 “I would gravitate to the challenges related to the three P’s:
power, privilege, and person. We need to attend to social and
economic disparities.”
77 “We need to address the role of religion in people’s lives. Under-
funding of traditionally ethnic religious institutions is limiting
access to important support systems. This leads to loss of tra-
ditions, creates schisms between generations, and presents in
counseling as loss of direction, diminished hope, and struggle
to find meaning in life.”
77 “A challenge to all of the helping professions is the recruitment
and training of clinicians who are members of diverse groups
(broadly and inclusively defined). The overall field runs the risk
of failing to keep pace with demographics trends. Collectively,
we are training a different population of therapists (e.g., upper-
middle class) than the clients we serve.”
77 “Problems often arise in supervision and consultation. I’m also
reminded that issues of oppression extend to political issues,
such as voter suppression.”

244       foundations of multicultural psychology

 77 “All counseling is likely to be multicultural in some sense and
to some degree, given the multiple identities and affiliations
that each of us has. So counselors must think about and address
culture and its implications from a dynamic perspective. To date,
our approach to culture has been much too static.”
77 “A topic that comes to mind is internalization of hatred, mani-
fested as complacency.”
77 “My prediction is that we’re going to need to do more macro
systemic advocacy around political and economic issues, as there
simply won’t be enough social services for most people in need.”
77 “Your question reminded of a client I saw last week. She was
hungry. She could not obtain food stamps for two to three more
days. How do I manage that in counseling? She said, ‘You look
like you do OK. Can you tell me, how do I rise out of poverty?’
That is very humbling. What is our role in that scenario?”
77 “Sexual and gender minorities (individuals who identify as
lesbian, gay, bisexual, or transgender) will be more open with
their families of origin about LGBT status than they might have
been in prior generations. This creates new opportunities for
family closeness, but it also creates more of a need for extended
family members to work out their acceptance of each other.”
77 “Endangered languages are becoming even more endangered.
People who speak an endangered language will feel torn between
their duties or desires to cultivate their cultural inheritance and
their desires to operate in a dominant culture.”
77 “White people continue to occupy policy, judicial, and power
broker positions. As psychologists and counselors, what role do
we play in addressing those inequities?”
77 “We need to address gender issues related to gender identity
and to violence among intimate partners.”
77 “Younger people did not experience the racism of the past, so they
may feel it is irrelevant to their lives. Older Whites may fear
becoming a minority group with diminished power.”
77 “Three recommendations: (1) indigenous peoples and their
growing awareness of settler decolonization issues, (2) lingering
assimilation efforts and racial microaggressions, and (3) sensitiv-
ity to unique historical issues of diverse groups.”
Distilling the experts’ informal comments, we make two observations.
First, several of the experts recommended that the field focus on issues of
power, particularly socioeconomic status and poverty, although other recom-
mended topics could also be conceptualized in terms of power inequities. In

firming up the foundation      245

Chapter 11, we described a philosophical framework from which this kind of
work may gain more traction than attempts based on philosophical materi-
alism have thus far generated in the mental health professions. Irrespective
of the paradigms or methods used, the primary point that several experts in
the field are making is that we must no longer act as if power inequities do
not affect mental health. They do, and we must attend to them.
Second, beyond the topics of power and poverty, there was no consensus on
priorities for the future. Many challenges plead for solutions. Encompassing all
of human diversity, multicultural psychology is so broad that organization and
prioritization proves daunting. Sorting and sifting through myriad important
topics will require our best efforts, so we repeat our recommendations to focus
squarely on the four dependent variables most essential to the work of mental
health professionals (listed above) and to consider research data when selecting
among the many independent variables that call for our attention. And we
emphasize the central role that strong theories can play in explaining the
relationships among variables, providing essential guidance for targeting
underlying causes, rather than surface-level descriptions.
Admittedly, the complexity of multicultural psychology resists the
artificial imposition of theoretical concepts. No theory can account for all
exceptions, so it seems unlikely for a single approach to unify all aspects of
the field’s complexity (Cauce, 2011). Nevertheless, it is abundantly clear that
the field needs theories that go beyond description to explanation. We need
strong theories that move beyond what questions to answer how and eventually
why questions. This line of scholarship can focus on (a) micro-theories, those
delimited by circumstances or attributes; and (b) middle-range theories, those
that explain events common to multiple settings but do not claim universal-
ity (Merton, 1968). Thus far, theory development has been limited to a few
topics in multicultural psychology, yet it will be obvious to those familiar with
the multicultural psychology literature how very much those theories have
contributed to the field. Existing theories have to be tested and subsequently
refined, but even further upstream, efforts at theory development have not
even come close to keeping pace with the torrent of need. If the complexity
of human diversity is to be described in terms that promote well-being and
decrease psychopathology, that work has a place in multicultural psychology.

CONCLUSION

Multicultural psychology has great potential. Metaphorically, multi­

cultural psychology can be compared to a construction project, not a single
edifice but a vast complex of buildings. In a word, it is a community, a com-
munity in which all are welcomed, none excluded. It has the potential to

246       foundations of multicultural psychology

 EXHIBIT 12.1
Data-Based Implications for Mental Health Practitioners
Therapists benefit from obtaining experience and education on multicultural competence
(see Chapter 2, this volume).
Therapists working with clients of color should remain attuned to client perceptions,
which predict client experiences in therapy much better than the therapist’s self-
perceptions of multicultural competence (Chapter 3).
People of color underutilize mental health services, but the racial discrepancies are
lower than commonly assumed, except for Asian Americans. Clinicians can take
steps to increase the relevance, availability, and appeal of their services for people
of color, particularly Asian Americans (Chapter 4).
People of color have a small risk of prematurely terminating mental health services.
African Americans are at greatest risk, but even that risk is lower than commonly
assumed (Chapter 5). Clients with a therapist of their own race or ethnicity are
somewhat less likely to prematurely terminate mental health services, particularly
Asian Americans (Chapter 6). Clinicians can address factors that could decrease
the risk of premature termination.
The more a treatment aligns with the cultural experiences of a client, the better the
clients’ outcome (Chapter 7).
Clinicians should attend to the perceptions of and experiences with mental health
services for populations with low levels of acculturation (e.g., first generation
immigrants; Chapter 8).
Experiences of racism are negatively associated with the well-being of people of color,
such that clinicians should empower clients’ reporting racist encounters (Chapter 9).
Strength of ethnic identity is positively associated with well-being, such that clinicians
should attend to the interactions of that identity with psychological functioning
(Chapter 10).
Research results vary widely (all chapters). Averaged data (Figure 12.1) provides
general direction, but clinicians should understand the worldviews and experiences
of individual clients and not impose their own preconceptions, even when those
come from data. In particular, practitioners have to learn directly from clients from
groups inadequately represented in the literature: Native American Indians, Alaska
Natives, Polynesian Americans, and Arab Americans.

be a force for unity amidst diversity. It has the potential to improve mental
health practices (see Exhibit 12.1) and to unite North American practices
with those found elsewhere. It is a large community indeed.
The community of mental health professionals relies on evidence-based
practices (APA 2005 Presidential Task Force on Evidence-Based Practice,
2006). To what degree is multicultural psychology supported by research
(G. C. N. Hall & Yee, 2014)? On the basis of research data, cultural adapta-
tions to mental health services result in greater client improvement than
traditional mental health services (“treatment as usual”); in fact, the more
aligned with clients’ culture, the more effective the treatment. On the basis of
research data, psychologists should account for level of acculturation among
immigrant populations in mental health services and should address experi-
ences of racism and level of ethnic identity among people of color generally.

firming up the foundation      247

On the basis of research data, mental health professionals should understand
that Asian Americans do not enter mental health services to the same extent
as other groups nor remain in treatment as long as other groups if the thera-
pist is not of their same racial or ethnic background; findings indicate that
steps beyond current practices must be taken to meet the mental health needs
of Asian Americans (e.g., S. Sue, Cheng, Saad, & Chu, 2012). On the basis
of similar research data, mental health professionals should understand that
because African American clients discontinue treatment at rates higher than
those of other groups, steps beyond current practices must be taken to effec-
tively engage and retain individual clients (e.g., Snowden, 2012). On the
basis of research data, therapists benefit when receiving education designed
to enhance multicultural competence. And on the basis of research data,
mental health professionals should understand that their own perceptions of
multicultural competence may not align with the experiences of their clients of
color but that clients of color do perceive therapist multicultural competence
to be very strongly associated with their own experiences in treatment.
In summary, with thousands of manuscripts now providing both quan-
titative and qualitative data, the research basis for multicultural psychology
appears to be in place. With the foundation cleared and soon to be solidified,
construction is underway. Construction workers are needed.

248       foundations of multicultural psychology

 APPENDIX: GENERAL METHODS
OF THE META-ANALYSES
(CHAPTERS 2–10)

Chapters 2 to 10 in this book report the results of meta-analyses. Those

meta-analyses involved similar methods, so a detailed description of the
meta-analytic methods in each chapter proved redundant.
This Appendix provides information specified in professional guide-
lines for publishing meta-analyses (APA Publications and Communications
Board Working Group on Journal Article Reporting Standards, 2008; Moher,
Liberati, Tetzlaff, Altman, & the PRISMA Group, 2009). Procedures used
in the meta-analyses reported in this book were adapted to those professional
guidelines; several meta-analyses had already commenced prior to the
publication of the professional guidelines. A supplemental file available
online ([Link] provides a detailed compari-
son of the methods used in the meta-analyses reported in this book com-
pared with the Meta-Analysis Reporting Standards (APA Publications
and Communications Board Working Group on Journal Article Reporting
Standards, 2008) and Preferred Reporting Items for Systematic Reviews and
Meta-Analyses (Moher et al., 2009).

GENERAL ELIGIBILITY CRITERIA

We restricted our meta-analytic reviews to data collected in the United

States or Canada because cultural values and mental health practices may
systematically differ in other nations. When the research question was spe-
cific to people of color, data from White/European Americans and Canadians
were excluded from the analyses (Chapters 5–8), but biracial and multiracial
individuals were included. We analyzed manuscripts written in English because
our research team lacked sufficient coders fluent in other languages. All quan-
titative research designs except case studies and single-N studies were eligible
for inclusion.
Given the relative dearth of previous systematic reviews in the field
of multicultural psychology, our intention1 was to review broad constructs,
rather than restrict analyses to a particular conceptualization or measure-
ment of a given construct. This approach appeared justified because we found

1We affirm that our decisions and analyses were not influenced by external parties and that we are solely
responsible for the contents of the book. Brigham Young University funded the research assistants.

249
inconsistent terminology in the literature: Different authors used slightly dif-
ferent terms when intending similar meaning. For instance, authors referred
to the degree of client participation in mental health services (Chapter 5)
using a variety of terms (e.g., usage, attendance, completion), such that had
we restricted our review to certain synonyms, we would have inaccurately
excluded manuscripts with pertinent data. We preferred to err on the side of
conceptual similarity rather than on the side of precision, given that the field
has yet to conclude which operationalization of complex constructs would be
preferable among the many available.
In our reviews, we defined mental health services as any treatment, out­
patient or inpatient, explicitly intended to improve emotional or psychological
well-being, whether provided by a psychologist, psychiatrist, psychiatric nurse,
clinical social worker, or mental health counselor. When a manuscript con-
tained information about mental health treatments provided by both general
practitioner physicians and mental health specialists, we extracted only the
information pertaining to mental health specialists. We excluded studies that
were limited to substance abuse treatments because of the complicating issues
of mandated versus voluntary treatment, multiple types of substances with
different impacts, dynamics of substance abuse dependence, and relapse rates
that differ from treatments for mental health concerns. When a manuscript
contained information about both mental health services and substance abuse
treatment, we coded only the data pertaining to services provided by men-
tal health professionals. Otherwise, we did not restrict manuscripts on any
particular mental health problem or treatment modality. Prevention-oriented
interventions with at-risk populations or community members were included
but coded separately from treatments with clinical populations.
Our several meta-analyses took substantial time to complete, with stag-
gered cutoff dates for data inclusion, as follows: Chapter 2 (multicultural edu-
cation for therapists) by October 2008; Chapter 3 (therapist multicultural
competence) by June 2012; Chapter 4 (mental health service utilization) by
April 2008; Chapter 5 (participation in mental health services) by April 2008;
Chapter 6 (ethnic matching) by March 2013; Chapter 7 (culturally adapted
treatments) by July 2012; Chapter 8 (acculturation and mental health services)
by July, 2008; Chapter 9 (racism and well-being) by June 2010; and Chapter 10
(ethnic identity and well-being) by August 2009. Because findings may have
changed over time, we coded and analyzed the year of publication.

INFORMATION SOURCES AND SEARCH STRATEGIES

To locate relevant published and unpublished studies, at least three and

up to eight members of the research team conducted extensive searches of the
following electronic databases: PsycINFO, PsycArticles, Science Citation

250       appendix

Index, Social Sciences Abstracts, Social Sciences Citation Index, and Digital
Dissertations. In these electronic searches, lists of synonymous words and
phrases were cross-referenced with one another. For instance, in the meta-
analysis of acculturation and mental health services (Chapter 8), the word
root acculturat was crossed with a list of word roots and phrases relevant to
mental health services: client, counsel, treatment, clinic, therapy, psychotherapy,
session, intake, screening, “mental health treatment,” “psychological service,”
and “mental health service” using the Boolean OR to include all terms. Full
search strings for all chapters are available from the first author. After iden-
tifying studies that met coding criteria, a team member examined the refer-
ences cited within those manuscripts to locate additional articles that fit the
inclusion criteria but were not initially located in the electronic databases.
Attempts were also made to contact authors who had published three or
more articles on the topic of the meta-analysis to request information regard-
ing other (unpublished) studies to include in the review. In three instances,
solicitations for unpublished manuscripts were posted on professional list-
servs. Reference lists of all studies included within each chapter are published
online ([Link]

DATA COLLECTION PROCESSES

A research team consisting of a university faculty member, at least one

graduate student, and at least three undergraduate students coded manuscripts
after receiving extensive training on relevant procedures. Team members
worked in pairs to enhance the accuracy of coding decisions and data entry.
Each article was coded by two separate pairs of coders. In addition to effect
size data, coders extracted information about possible moderating variables,
including participant characteristics (e.g., age, gender, race) and study char-
acteristics (e.g., study setting, study design, sample size). The interrater agree-
ment of initial coding decisions was evaluated for categorical variables using
Cohen’s kappa and for continuous variables using intraclass correlation coef-
ficients (ICC) generated in one-way random effects models for single mea-
sures. Across all meta-analyses, the average coefficients obtained for coding
pairs were acceptably high, as follows: Chapter 2 (multicultural education
for therapists) kappa = 0.77, ICC = 0.89; Chapter 3 (therapist multicultural
competence) kappa = 0.72, ICC = 0.94; Chapter 4 (mental health service uti-
lization) kappa = 0.74, ICC = 0.95; Chapter 5 (participation in mental health
services) kappa = 0.83, ICC = 0.82; Chapter 6 (ethnic matching) kappa =
0.78, ICC = 0.96; Chapter 7 (cultural adaptations to treatment) kappa = 0.75,
ICC = 0.97; Chapter 8 (acculturation and mental health services) kappa =
0.86, ICC = 0.98; Chapter 9 (racism and well-being) kappa = 0.83, ICC = .95;

appendix      251
and Chapter 10 (ethnic identity and well-being) kappa = 0.85, ICC = 0.89.
Discrepancies between pairs were resolved by a third round of review by at least
one member from both original pairs. The first author adjudicated unresolved
disagreements.

DATA ANALYSES

We used meta-analytic software to transform statistical estimates derived

from a variety of metrics (e.g., t, F, and p values) to one of three types of effect
sizes: (a) Cohen’s d for meta-analyses involving differences between groups
or conditions (Chapters 2, 5–7), (b) Pearson’s r for meta-analyses evaluating
associations between variables, with the data analyzed after Fisher’s z transfor-
mation (Chapters 3, 8–10), and (c) odds ratios for the meta-analysis involv-
ing a binary comparison of mental health service utilization, with the data
analyzed after natural log transformation (Chapter 4).
When a study contained multiple effect sizes, the values were aver-
aged (weighted by the standard error or number of participants included in
each analysis) to compute an aggregate effect size for that particular study,
such that each study contributed only one data point to the calculation of
the omnibus effect size. Because the variables measured in studies (a) had
been imprecisely defined across research studies, (b) were influenced by many
factors beyond those measured within studies, and thus (c) were expected
to yield effect size estimates that differed across individual participants and
across individual studies, inverse variance weighted random effects models
were used in analyzing the data using the 2006 version of macros for SPSS
developed by Lipsey and Wilson (2001). Random effects models allowed for
generalization beyond the studies included in the analyses.
For the analyses conducted in several of the chapters, we expected the
results to differ by participant race, so in those instances we conducted multi-
variate meta-analyses (Becker, 2000) using all effect sizes in studies by racial
or ethnic group, with those analyses accounting for within-study correlations
among effect sizes. The multivariate analyses were specific to data from African
Americans, Asian Americans, and Hispanic/Latino(a) Americans, but data
specific to Native Americans or Alaska Natives were insufficient across stud-
ies so were combined with the data from unspecified and “other” racial cat-
egories. Because participant race remained unspecified in manuscripts when
authors used the catchall term “ethnic minorities” when describing clients
with ancestry primarily outside Europe, the “unspecified/other” racial category
analyzed in the multivariate analyses represented a generic contrast of limited
interpretability.

252       appendix

 REFERENCES

Abreu, J. M., Gim Chung, R. H., & Atkinson, D. R. (2000). Multicultural counsel-
ing training: Past, present, and future directions. The Counseling Psychologist, 28,
641–656. [Link]
Acevedo, M. C., Reyes, C. J., Annett, R. D., & López, E. M. (2003). Assessing language
competence: Guidelines for assisting persons with limited English proficiency in
research and clinical settings. Journal of Multicultural Counseling and Development,
31, 192–204. [Link]
Adams, G., & Markus, H. R. (2001). Culture as patterns: An alternative approach
to the problem of reification. Culture & Psychology, 7, 283–296. [Link]
org/10.1177/1354067X0173002
Ægisdóttir, S., & Gerstein, L. H. (2010). International counseling competencies: A new
frontier in multicultural training. In J. G. Ponterotto, J. M. Casas, L. A. Suzuki, &
C. Alexander (Eds.), Handbook of multicultural counseling (3rd ed., pp. 175–188).
Thousand Oaks, CA: Sage.
Ajzen, I., & Fishbein, M. (1977). Attitude–behavior relations: A theoretical analysis
and review of empirical research. Psychological Bulletin, 84, 888–918. [Link]
org/10.1037/0033-2909.84.5.888
Alegría, M., Canino, G., Ríos, R., Vera, M., Calderón, J., Rusch, D., & Ortega, A. N.
(2002). Inequalities in use of specialty mental health services among Latinos,
African Americans, and non-Latino Whites. Psychiatric Services, 53, 1547–1555.
[Link]
Alegría, M., Chatterji, P., Wells, K., Cao, Z., Chen, C. N., Takeuchi, D., . . .
Meng, X. L. (2008). Disparity in depression treatment among racial and ethnic
minority populations in the United States. Psychiatric Services, 59, 1264–1272.
[Link]
Alegría, M., Polo, A., Gao, S., Santana, L., Rothstein, D., Jimenez, A., . . . Normand,
S. L. (2008). Evaluation of a patient activation and empowerment intervention
in mental health care. Medical Care, 46, 247–256. [Link]
MLR.0b013e318158af52
Alladin, W. J. (2002). Ethnic matching in counselling. In S. Palmer (Ed.), Multi-
cultural counselling: A reader (pp. 175–180). London, England: Sage.
Allen, J., Lewis, J., & Johnson-Jennings, M. (2015). Well-being and health. In
P. Pedersen, J. Draguns, W. Lonner, J. E. Trimble, & M. Sharron-del Rio (Eds.),
Counseling across cultures (7th ed., pp. 435–455). Thousand Oaks, CA: Sage.
Allen, J., Rivkin, I. D., & López, E. D. S. (2014). Health and well-being. In F. T. L.
Leong, L. Comas-Díaz, G. C. Nagayama Hall, V. C. McLoyd, & J. E. Trimble
(Eds.), APA handbook of multicultural psychology: Vol. 1. Theory and research
(pp. 299–311). Washington, DC: American Psychological Association.

253
Allport, G. W. (1954). The nature of prejudice. Oxford, England: Addison-Wesley.
Alvidrez, J. (1999). Ethnic variations in mental health attitudes and service use
among low-income African American, Latina, and European American
young women. Community Mental Health Journal, 35, 515–530. [Link]
org/10.1023/A:1018759201290
Alvidrez, J., Snowden, L. R., & Patel, S. G. (2010). The relationship between stigma
and other treatment concerns and subsequent treatment engagement among
Black mental health clients. Issues in Mental Health Nursing, 31, 257–264. http://
[Link]/10.3109/01612840903342266
American Psychological Association. (1986). Accreditation handbook. Washington, DC:
Author.
American Psychological Association. (1993). Guidelines for providers of psycho-
logical services to ethnic, linguistic, and culturally diverse populations. American
Psychologist, 48, 45–48. [Link]
American Psychological Association. (1994). Guidelines and principles for accreditation
of programs in professional psychology. Washington, DC: Author.
American Psychological Association. (2003). Guidelines on multicultural education,
training, research, practice, and organizational change for psychologists. American
Psychologist, 58, 377–402.
American Psychological Association. (2005). APA presidential task force on enhancing
diversity: Final report. Washington, DC: Author.
American Psychological Association. (2009). Report of the APA task force on the imple-
mentation of the multicultural guidelines. Washington, DC: Author.
American Psychological Association. (2014). Proposed standards of accreditation in
health service psychology. Retrieved from [Link]
Amiot, C. E., de la Sablonnière, R., Terry, D. J., & Smith, J. R. (2007). Integration of social
identities in the self: Toward a cognitive–developmental model. Personality and Social
Psychology Review, 11, 364–388. [Link]
Ancis, J. R. (2003). Teaching multicultural competencies using the Internet and other
technologies. In D. B. Pope-Davis, H. L. K. Coleman, & R. L. Toporek (Eds.),
Handbook of multicultural competencies in counseling & psychology (pp. 575–588).
Thousand Oaks, CA: Sage. [Link]
Ancis, J. R., & Ali, S. R. (2005). Multicultural counseling training approaches:
Implications for pedagogy. In C. Z. Enns & A. L. Sinacore (Eds.), Teaching
and social justice: Integrating multicultural and feminist theories in the classroom
(pp. 85–97). Washington, DC: American Psychological Association. http://
[Link]/10.1037/10929-005
Andersen, R., & Newman, J. F. (1973). Societal and individual determinants of med-
ical care utilization in the United States. The Milbank Memorial Fund Quarterly.
Health and Society, 51, 95–124. [Link]
Angold, A., Erkanli, A., Farmer, E. M. Z., Fairbank, J. A., Burns, B. J., Keeler, G.,
& Costello, E. J. (2002). Psychiatric disorder, impairment, and service use in

254       references

 rural African American and White youth. Archives of General Psychiatry, 59,
893–901. [Link]
Ano, G. G., & Vasconcelles, E. B. (2005). Religious coping and psychological adjust-
ment to stress: A meta-analysis. Journal of Clinical Psychology, 61, 461–480.
[Link]
APA 2005 Presidential Task Force on Evidence-Based Practice. (2006). Evidence-
based practice in psychology. American Psychologist, 61, 271–285. [Link]
org/10.1037/0003-066X.61.4.271
APA Publications and Communications Board Working Group on Journal Article
Reporting Standards. (2008). Reporting standards for research in psychology:
Why do we need them? What might they be? American Psychologist, 63, 839–851.
[Link]
Arnett, J. J. (2002). The psychology of globalization. American Psychologist, 57,
774–783. [Link]
Arredondo, P., & Perez, P. (2003). Expanding multicultural competence through social
justice leadership. The Counseling Psychologist, 31, 282–289. [Link]
10.1177/0011000003031003003
Arredondo, P., Rosen, D. C., Rice, T., Perez, P., & Tovar-Gamero, Z. G. (2005). Multi­
cultural counseling: A 10-year content analysis of the Journal of Counseling &
Development. Journal of Counseling & Development, 83, 155–161. [Link]
org/10.1002/j.1556-6678.2005.tb00592.x
Arredondo, P., & Toporek, R. (2004). Multicultural counseling competencies = ethical
practice. Journal of Mental Health Counseling, 26, 44.
Arredondo, P., Toporek, R., Brown, S. P., Jones, J., Locke, D. C., Sanchez, J., &
Stadler, H. (1996). Operationalization of the multicultural counseling compe-
tencies. Journal of Multicultural Counseling and Development, 24, 42–78. http://
[Link]/10.1002/j.2161-1912.1996.tb00288.x
Arredondo, P., & Tovar-Blank, Z. G. (2014). Multicultural competencies: A dynamic
paradigm for the 21st century. In F. T. L. Leong, L. Comas-Díaz, G. C. Nagayama
Hall, V. C. McLoyd, & J. E. Trimble (Eds.), APA handbook of multicultural psychol-
ogy: Vol. 2. Applications and training (pp. 19–34). Washington, DC: American
Psychological Association.
Asnaani, A., & Hofmann, S. G. (2012). Collaboration in multicultural therapy:
Establishing a strong therapeutic alliance across cultural lines. Journal of Clinical
Psychology, 68, 187–197. [Link]
Atkinson, D. R., & Wampold, B. E. (1982). A comparison of the Counselor Rating
Form and the Counselor Effectiveness Rating Scale. Counselor Education and
Supervision, 22, 25–36. [Link]
Awad, G. H., & Cokley, K. O. (2010). Designing and interpreting quantitative
research in multicultural counseling. In J. G. Ponterotto, J. M. Casas, L. Suzuki,
& C. Alexander (Eds.), Handbook of multicultural counseling (pp. 385–396).
Thousand Oaks, CA: Sage.

references      255
Baca, L., Ridley, C. R., Kelly, S. M., Sue, S., Takooshian, H., Grimes, T. R., & Taylor, L.
(2007). Future directions for practitioners, trainers, and researchers: Interdisciplin-
ary perspectives. In G. B. Esquivel, E. C. López, & S. G. Nahari (Eds.), Handbook
of multicultural school psychology: An interdisciplinary perspective (pp. 659–694).
Mahwah, NJ: Erlbaum.
Bakhtin, M. M. (1981). Dialogic imagination: Four essays. Austin: University of Texas
Press.
Baluch, S. P., Pieterse, A. L., & Bolden, M. A. (2004). Counseling psychology and
social justice: Houston . . . we have a problem. The Counseling Psychologist,
32, 89–98. [Link]
Bandura, A. (1977). Social learning theory. Oxford, England: Prentice-Hall.
Barrera, M., Jr., & Castro, F. G. (2006). A heuristic framework for the cultural adap-
tation of interventions. Clinical Psychology: Science and Practice, 13, 311–316.
[Link]
Barrera, M., Jr., Castro, F. G., Strycker, L. A., & Toobert, D. J. (2013). Cultural
adaptations of behavioral health interventions: A progress report. Journal of
Consulting and Clinical Psychology, 81, 196–205. [Link]
a0027085
Barrett, M. S., Chua, W.-J., Crits-Christoph, P., Gibbons, M. B., & Thompson, D.
(2008). Early withdrawal from mental health treatment: Implications for psycho-
therapy practice. Psychotherapy: Theory, Research, Practice, Training, 45, 247–267.
[Link]
Barth, F. (Ed.). (1969). Ethnic groups and boundaries. Boston, MA: Little, Brown.
Becker, B. J. (2000). Multivariate meta-analysis. In H. E. A. Tinsley & S. D. Brown
(Eds.), Handbook of applied multivariate statistics and mathematical modeling
(pp. 499–525). San Diego, CA: Academic Press. [Link]
B978-012691360-6/50018-5
Benish, S. G., Quintana, S., & Wampold, B. E. (2011). Culturally adapted psycho-
therapy and the legitimacy of myth: A direct-comparison meta-analysis. Journal
of Counseling Psychology, 58, 279–289. [Link]
Berkman, L. F., Glass, T., Brissette, I., & Seeman, T. E. (2000). From social integra-
tion to health: Durkheim in the new millennium. Social Science & Medicine, 51,
843–857. [Link]
Bernal, G., Bonilla, J., & Bellido, C. (1995). Ecological validity and cultural sensitiv-
ity for outcome research: Issues for the cultural adaptation and development of
psychosocial treatments with Hispanics. Journal of Abnormal Child Psychology, 23,
67–82. [Link]
Bernal, G., & Domenech Rodríguez, M. M. (2012). Cultural adaptations: Tools for
evidence-based practice with diverse populations. Washington, DC: American
Psychological Association. [Link]
Bernal, G., & Flores-Ortiz, Y. (1982). Latino families in therapy: Engagement and
evaluation. Journal of Marital and Family Therapy, 8, 357–365. [Link]
org/10.1111/j.1752-0606.1982.tb01458.x

256       references

Bernal, G., Jiménez-Chafey, M. I., & Domenech Rodríguez, M. M. (2009). Cultural
adaptation of treatments: A resource for considering culture in evidence-based
practice. Professional Psychology: Research and Practice, 40, 361–368. http://
[Link]/10.1037/a0016401
Bernal, M. E., & Castro, F. G. (1994). Are clinical psychologists prepared for service
and research with ethnic minorities? Report of a decade of progress. American
Psychologist, 49, 797–805. [Link]
Bernal, M. E., & Knight, G. P. (Eds.). (1993). Ethnic identity: Formation and transmission
among Hispanics and other minorities. Albany: State University of New York Press.
Bernal, M. E., Saenz, D. S., & Knight, G. P. (1991). Ethnic identity and adaptation
of Mexican American youths in school settings. Hispanic Journal of Behavioral
Sciences, 13, 135–154. [Link]
Berry, J. W. (1980). Social and cultural change. In H. C. Triandis & R. Brislin (Eds.),
Handbook of cross-cultural psychology: Vol. 5. Social (pp. 211–279). Boston, MA:
Allyn & Bacon.
Berry, J. W. (1991). Understanding and managing multiculturalism: Some possible
implications of research in Canada. Psychology and Developing Societies, 3, 17–49.
[Link]
Berry, J. W. (1994). Acculturation and psychological adaptation: An overview. In
A. Bouvy, F. J. R. Van de Vijver, P. Boski, & P. Schmitz (Eds.), Journeys into
cross-cultural psychology (pp. 129–141). Berwyn, PA: Swets & Zeitlinger.
Berry, J. W. (1995). Psychology of acculturation. In N. R. Goldberger & J. B. Veroff
(Eds.), The culture and psychology reader (pp. 457–488). New York: New York
University Press.
Berry, J. W. (1997). Immigration, acculturation, and adaptation. Applied Psychology,
46, 5–34.
Berry, J. W. (2003). Conceptual approaches to acculturation. In K. M. Chun, P. Balls
Organista, & G. Marín (Eds.), Acculturation: Advances in theory, measurement,
and applied research (pp. 17–37). Washington, DC: American Psychological
Association.
Berry, J. W., & Kim, U. (1986). Acculturation and mental health. In P. Dasen, J. W.
Berry, & N. Sartorius (Eds.), Applications of cross-cultural psychology to healthy
human development (pp. 207–236). Beverly Hills, CA: Sage.
Berry, J. W., Poortinga, Y. H., Segall, M. H., & Dasen, P. R. (1992). Cross-cultural
psychology: Research and applications. New York, NY: Cambridge University
Press.
Berry, J. W., & Sam, D. L. (1996). Acculturation and adaptation. In J. W. Berry, M.
H. Segall, & C. Kagitcibasi (Eds.), Handbook of cross-cultural psychology: Vol. 3.
Social behavior and applications (pp. 291–326). Boston, MA: Allyn & Bacon.
Berry, J. W., Trimble, J. E., & Olmedo, E. (1986). Assessment of acculturation. In W. J.
Lonner & J. W. Berry (Eds.), Field methods in cross-cultural research (pp. 291–324).
Newbury Park, CA: Sage.

references      257
Binning, K. R., Unzueta, M. M., Huo, Y. J., & Molina, L. E. (2009). The interpreta-
tion of multiracial status and its relation to social engagement and psychological
well-being. Journal of Social Issues, 65, 35–49. [Link]
4560.2008.01586.x
Birman, D. (2011). Migration and well-being: Beyond the macrosystem. Psychosocial
Intervention, 20, 339–341. [Link]
Birman, D., & Simon, C. D. (2014). Acculturation research: Challenges, complexi-
ties, and possibilities. In F. T. L. Leong, L. Comas-Díaz, G. C. Nagayama Hall,
V. C. McLoyd, & J. E. Trimble (Eds.), APA handbook of multicultural psychology:
Vol. 1. Theory and research (pp. 207–230). Washington, DC: American Psycho-
logical Association.
Bojuwoye, O., & Sodi, T. (2010). Challenges and opportunities to integrating tra-
ditional healing into counselling and psychotherapy. Counselling Psychology
Quarterly, 23, 283–296. [Link]
Bond, M. H. (2013). The pan-culturality of well-being: But does culture fit into the
equation? Asian Journal of Social Psychology, 16, 158–162. [Link]
ajsp.12024
Bonham, V. L., Warshauer-Baker, E., & Collins, F. S. (2005). Race and ethnicity in
the genome era: The complexity of the constructs. American Psychologist, 60,
9–15. [Link]
Borgatti, S. P., Mehra, A., Brass, D., & Labianca, G. (2009). Network analysis
in the social sciences. Science, 323, 892–895. [Link]
1165821
Bosworth, H. B., Parsey, K. S., Butterfield, M. I., McIntyre, L. M., Oddone, E. Z.,
Stechuchak, K. M., & Bastian, L. A. (2000). Racial variation in wanting and
obtaining mental health services among women veterans in a primary care clinic.
Journal of the National Medical Association, 92, 231–236.
Broman, C. L. (1987). Race differences in professional help seeking. American Journal
of Community Psychology, 15, 473–489.
Brown, T. N. (2003). Critical race theory speaks to the sociology of mental health:
Mental health problems produced by racial stratification. Journal of Health and
Social Behavior, 44, 292–301. [Link]
Bynum, W. F., & Porter, R. (Eds.). (2013). Companion encyclopedia of the history of
medicine. New York, NY: Routledge.
Byrne, B. M., Oakland, T., Leong, F. T., van de Vijver, F. J., Hambleton, R. K., Cheung,
F. M., & Bartram, D. (2009). A critical analysis of cross-cultural research and
testing practices: Implications for improved education and training in psychol-
ogy. Training and Education in Professional Psychology, 3, 94–105. [Link]
org/10.1037/a0014516
Byrne, D. E. (1971). The attraction paradigm. New York, NY: Academic Press.
Cabassa, L. J. (2003). Measuring acculturation: Where we are and where we need to
go. Hispanic Journal of Behavioral Sciences, 25, 127–146. [Link]
0739986303025002001

258       references

Cabral, R. R., & Smith, T. B. (2011). Racial/ethnic matching of clients and therapists
in mental health services: A meta-analytic review of preferences, perceptions, and
outcomes. Journal of Counseling Psychology, 58, 537–554. [Link]
a0025266
Cachelin, F. M., & Striegel-Moore, R. H. (2006). Help seeking and barriers to
treatment in a community sample of Mexican American and European Amer-
ican women with eating disorders. International Journal of Eating Disorders, 39,
154–161.
Cachelin, F. M., Veisel, C., Barzegarnazari, E., & Striegel-Moore, R. H. (2000). Dis-
ordered eating, acculturation, and treatment-seeking in a community sample of
Hispanic, Asian, Black, and White women. Psychology of Women Quarterly, 24,
244–253. [Link]
Calabrese, J. D. (2008). Clinical paradigm clashes: Ethnocentric and political barri-
ers to Native American efforts at self-healing. Ethos, 36, 334–353. [Link]
org/10.1111/j.1548-1352.2008.00018.x
Campbell, L. F., Norcross, J. C., Vasquez, M. J., & Kaslow, N. J. (2013). Recognition
of psychotherapy effectiveness: The APA resolution. Psychotherapy, 50, 98–101.
[Link]
Cardemil, E. V. (2010a). The complexity of culture: Do we embrace the challenge or
avoid it? The Scientific Review of Mental Health Practice, 7(2), 41–47.
Cardemil, E. V. (2010b). Cultural adaptations to empirically supported treatments:
A research agenda. The Scientific Review of Mental Health Practice, 7(2), 8–21.
Carter, R. T. (1996). Exploring the complexity of racial identity attitude measures.
In G. R. Sodowsky & J. C. Impara (Eds.), Multicultural assessment in counseling
and clinical psychology (pp. 193–223). Lincoln, NE: Buros Institute of Mental
Measurements.
Carter, R. T. (2003). Becoming racially and culturally competent: The racial–cultural
counseling laboratory. Journal of Multicultural Counseling and Development, 31,
20–30. [Link]
Carter, R. T. (2007). Racism and psychological and emotional injury recognizing and
assessing race-based traumatic stress. The Counseling Psychologist, 35, 13–105.
[Link]
Carter, R. T., & Pieterse, A. (2013, October). A review and critique of the current status
of race-related scholarship in counseling psychology. Paper presented at the Diversity
Challenge Conference, Chestnut Hill, MA.
Carter, R. T., & Qureshi, A. (1995). A typology of philosophical assumptions in
multicultural counseling and training. In J. G. Ponterotto, J. M. Casas, & C. M.
Alexander (Eds.), Handbook of multicultural counseling (pp. 239–262). Thousand
Oaks, CA: Sage.
Cartwright, B. Y., Daniels, J., & Zhang, S. (2008). Assessing multicultural competence:
Perceived versus demonstrated performance. Journal of Counseling & Develop-
ment, 86, 318–322. [Link]

references      259
Casas, J. M., Park, Y. S., & Cho, B. (2010). The multicultural and international-
ization counseling psychology movements: When all is said and done, it’s all
multicultural, isn’t it? In J. G. Ponterotto, J. M. Casas, L. A. Suzuki, & C. M.
Alexander (Eds.), Handbook of multicultural counseling (3rd ed., pp. 189–200).
Thousand Oaks, CA: Sage.
Case, L., & Smith, T. B. (2000). Ethnic representation in a sample of the literature of
applied psychology. Journal of Consulting and Clinical Psychology, 68, 1107–1110.
[Link]
Cassidy, J., & Shaver, P. R. (Eds.). (2008). Handbook of attachment: Theory, research,
and clinical applications (2nd ed.). New York, NY: Guilford Press.
Castillo, L. G., Brossart, D. F., Reyes, C. J., Conoley, C. W., & Phoummarath, M. J.
(2007). The influence of multicultural training on perceived multicultural coun-
seling competencies and implicit racial prejudice. Journal of Multicultural Counsel-
ing and Development, 35, 243–255. [Link]
tb00064.x
Castro, F. G., Barrera, M., Jr., & Holleran Steiker, L. K. (2010). Issues and chal-
lenges in the design of culturally adapted evidence-based interventions.
Annual Review of Clinical Psychology, 6, 213–239. [Link]
annurev-clinpsy-033109-132032
Cates, J. T., Schaefle, S. E., Smaby, M. H., Maddux, C. D., & Le Beauf, I. (2007). Com-
paring multicultural with general counseling knowledge and skill competency
for students who completed counselor training. Journal of Multicultural Coun-
seling and Development, 35, 26–39. [Link]
tb00047.x
Cauce, A. M. (2011). Is multicultural psychology a-scientific? Diverse methods for
diversity research. Cultural Diversity and Ethnic Minority Psychology, 17, 228–233.
[Link]
Chao, R. C., & Nath, S. R. (2011). The role of ethnic identity, gender roles, and
multicultural training in college counselors’ multicultural counseling compe-
tence: A mediation model. Journal of College Counseling, 14, 50–64. [Link]
org/10.1002/j.2161-1882.2011.tb00063.x
Chao, R. C., Wei, M., Good, G. E., & Flores, L. Y. (2011). Race/ethnicity, color-
blind racial attitudes, and multicultural counseling competence: The moderating
effects of multicultural counseling training. Journal of Counseling Psychology, 58,
72–82. [Link]
Charon, J. M. (2001). Symbolic interactionism: An introduction, an interpretation, an
integration. Upper Saddle River, NJ: Prentice Hall.
Chen, S., Sullivan, N. Y., Lu, Y. E., & Shibusawa, T. (2003). Asian Americans and
mental health services: A study of utilization patterns in the 1990s. Journal of
Ethnic & Cultural Diversity in Social Work, 12, 19–42. [Link]
J051v12n02_02

260       references

Chen, S. W., & Davenport, D. S. (2005). Cognitive–behavioral therapy with Chinese
American clients: Cautions and modifications. Psychotherapy: Theory, Research,
Practice, Training, 42, 101–110. [Link]
Cheung, Y. W. (1989). Making sense of ethnicity and drug use: A review and sugges-
tions for future research. Social Pharmacology, 3, 55–82.
Cheung, Y. W. (1993). Approaches to ethnicity: Clearing roadblocks in the study
of ethnicity and substance use. International Journal of the Addictions, 28,
1209–1226.
Chow, J. C., Jaffee, K., & Snowden, L. (2003). Racial/ethnic disparities in the use of
mental health services in poverty areas. American Journal of Public Health, 93,
792–797. [Link]
Chowdhary, N., Jotheeswaran, A. T., Nadkarni, A., Hollon, S. D., King, M., Jordans,
M. J., . . . Patel, V. (2014). The methods and outcomes of cultural adaptations of
psychological treatments for depressive disorders: A systematic review. Psycho-
logical Medicine, 44, 1131–1146. [Link]
Cirese, S. (1985). Quest: A search for self. New York, NY: HOH, Rurehard, & Winston.
Clark, K. B., & Clark, M. K. (1939). Segregation as a factor in the racial identifica-
tion of Negro pre-school children. Journal of Experimental Education, 8, 161–163.
[Link]
Clark, R., Anderson, N. B., Clark, V. R., & Williams, D. R. (1999). Racism as a
stressor for African Americans. A biopsychosocial model. American Psychologist,
54, 805–816. [Link]
Clarke, S., Hahn, H., & Hoggett, P. (Eds.). (2008). Object relations and social rela-
tions: The implications of the relational turn in psychoanalysis. London, England:
Karnac Books.
Clarkson, P. (2003). The therapeutic relationship. London, England: Whurr.
Cohen, D. (2001). Cultural variation: Considerations and implications. Psychological
Bulletin, 127, 451–471. [Link]
Cohen, J. (1988). Statistical power analysis for the behavioral sciences. New York, NY:
Routledge.
Cohen, S. (2004). Social relationships and health. American Psychologist, 59, 676–684.
[Link]
Coleman, H. L. K. (2004). Multicultural counseling competencies in a pluralistic
society. Journal of Mental Health Counseling, 26, 56–66.
Coleman, H. L. K., Wampold, B. E., & Casali, S. L. (1995). Ethnic minorities’ ratings
of ethnically similar and European American counselors: A meta-analysis.
Journal of Counseling Psychology, 42, 55–64. [Link]
0167.42.1.55
Coleman, M. N. (2006). Critical incidents in multicultural training: An examina-
tion of student experiences. Journal of Multicultural Counseling and Development,
34, 168–182. [Link]

references      261
Comas-Díaz, L. (2006). Cultural variation in the therapeutic relationship. In
C. Goodheart, A. Kazdin, & R. Sternberg (Eds.), Evidence-based psychotherapy:
Where practice and research meet (pp. 81–105). Washington, DC: American
Psychological Association. [Link]
Comas-Díaz, L., & Jacobsen, F. M. (2001). Ethnocultural allodynia. The Journal of
Psychotherapy Practice and Research, 10, 246–252.
Comstock, D. L., Hammer, T. R., Strentzsch, J., Cannon, K., Parsons, J., & Salazar,
G., II. (2008). Relational-cultural theory: A framework for bridging relational,
multicultural, and social justice competencies. Journal of Counseling & Develop-
ment, 86, 279–287. [Link]
Conner, K. O., Koeske, G., & Brown, C. (2009). Racial differences in attitudes toward
professional mental health treatment: The mediating effect of stigma. Journal of
Gerontological Social Work, 52, 695–712. [Link]
01634370902914372
Connolly Gibbons, M. B., Rothbard, A., Farris, K. D., Wiltsey Stirman, S., Thompson,
S. M., Scott, K., . . . Crits-Christoph, P. (2011). Changes in psychotherapy utiliza-
tion among consumers of services for major depressive disorder in the community
mental health system. Administration and Policy in Mental Health and Mental Health
Services Research, 38, 495–503. [Link]
Constantine, M. G. (2001). Multiculturally-focused counseling supervision: Its rela-
tionship to trainees’ multicultural counseling self-efficacy. The Clinical Supervisor,
20, 87–98. [Link]
Constantine, M. G. (2002). Predictors of satisfaction with counseling: Racial and
ethnic minority clients’ attitudes toward counseling and ratings of their coun-
selors’ general and multicultural counseling competence. Journal of Counseling
Psychology, 49, 255–263. [Link]
Constantine, M. G. (2007). Racial microaggressions against African American cli-
ents in cross-racial counseling relationships. Journal of Counseling Psychology,
54, 1–16. [Link]
Constantine, M. G., & Ladany, N. (2000). Self-report multicultural counseling com-
petence scales: Their relation to social desirability attitudes and multicultural
case conceptualization ability. Journal of Counseling Psychology, 47, 155–164.
[Link]
Constantine, M. G., & Ladany, N. (2001). New visions for defining and assessing
multicultural counseling competence. In J. G. Ponterotto, J. M. Casas, L. A.
Suzuki, & C. Alexander (Eds.), Handbook of multicultural counseling (2nd ed.,
pp. 482–498). Thousand Oaks, CA: Sage.
Constantine, M. G., & Sue, D. W. (2006). Addressing racism: Facilitating cultural com-
petence in mental health and educational settings. Hoboken, NJ: Wiley.
Cook, B. L., Doksum, T., Chen, C. N., Carle, A., & Alegría, M. (2013). The role of
provider supply and organization in reducing racial/ethnic disparities in mental
health care in the U.S. Social Science & Medicine, 84, 102–109. [Link]
org/10.1016/[Link].2013.02.006

262       references

Cookson, J. (2012, June 7). How much do you know about U.S. immigrants? CNN
World. Retrieved from [Link]
immigration-nation-what-you-need-to-know/
Cooper, H., Hedges, L. V., & Valentine, J. C. (2009). The handbook of research synthe-
sis and meta-analysis. New York, NY: Russell Sage Foundation.
Cooper, H., & Koenka, A. C. (2012). The overview of reviews: Unique challenges
and opportunities when research syntheses are the principal elements of new
integrative scholarship. American Psychologist, 67, 446–462. [Link]
org/10.1037/a0027119
Cooren, F. (2006). The organizational world as a plenum of agencies. In E. J. Van
Every (Ed.), Communication as organizing: Empirical and theoretical explorations in
the dynamic of text and conversation (pp. 81–100). Mahwah, NJ: Erlbaum.
Corning, A. F., & Malofeeva, E. V. (2004). The application of survival analysis to
the study of psychotherapy termination. Journal of Counseling Psychology, 51,
354–367. [Link]
Corrigan, P. (2004). How stigma interferes with mental health care. American Psycholo-
gist, 59, 614–625. [Link]
Crook-Lyon, R., O’Grady, K. A., Smith, T. B., Jensen, D. R., Golightly, T., &
Potkar, K. A. (2012). Addressing religious and spiritual diversity in gradu-
ate training and multicultural education for professional psychologists. Psy-
chology of Religion and Spirituality, 4, 169–181. [Link]
a0026403
Cross, W. (1991). Shades of black: Diversity in African American identity. Philadelphia,
PA: Temple University Press.
Cuéllar, I., Arnold, B., & Maldonado, R. (1995). Acculturation Rating Scale for
Mexican Americans–II: A revision of the original ARSMA scale. Hispanic
Journal of Behavioral Sciences, 17, 275–304.
Czopp, A. M., Mark, A. Y., & Walzer, A. S. (2014). Prejudice and racism. In F. T. L.
Leong, L. Comas-Díaz, G. C. Nagayama Hall, V. C. McLoyd, & J. E. Trimble
(Eds.), APA handbook of multicultural psychology: Vol. 1. Theory and research
(pp. 361–377). Washington, DC: American Psychological Association.
D’Andrea, M., & Daniels, J. (2010). Promoting multiculturalism, democracy, and
social justice in organizational settings: A case study. In J. G. Ponterotto, J. M.
Casas, L. A. Suzuki, & C. M. Alexander (Eds.), Handbook of multicultural coun-
seling (3rd ed., pp. 591–602). Thousand Oaks, CA: Sage.
D’Andrea, M., Daniels, J., & Heck, R. (1991). Evaluating the impact of multicultural
counseling training. Journal of Counseling & Development, 70, 143–150. http://
[Link]/10.1002/j.1556-6676.1991.tb01576.x
D’Andrea, M., & Heckman, E. (2008). A 40-year review of multicultural counsel-
ing outcome research: Outlining a future research agenda for the multicultural
counseling movement. Journal of Counseling & Development, 86, 356–363.
[Link]

references      263
David, E. J. R., Okazaki, S., & Giroux, D. (2014). A set of guiding principles to
advance multicultural psychology and its major concepts. In F. T. L. Leong,
L. Comas-Díaz, G. C. Nagayama Hall, V. C. McLoyd, & J. E. Trimble (Eds.),
APA handbook of multicultural psychology: Vol. 1. Theory and research (pp. 85–104).
Washington, DC: American Psychological Association.
Dawson, J. L. M. (1971). Theory and research in cross-cultural psychology. Bulletin
of the British Psychological Society, 24, 291–306.
Delgado, R., & Stefancic, J. (2012). Critical race theory. New York: New York Uni-
versity Press.
DeNeve, K. M., & Cooper, H. (1998). The happy personality: A meta-analysis of 137
personality traits and subjective well-being. Psychological Bulletin, 124, 197–229.
[Link]
de Smith, M. J., Goodchild, M. F., & Longley, P. A. (2007). Geospatial analysis: A
comprehensive guide to principles, techniques and software tools (2nd ed.). Leicester,
England: Troubador.
Diala, C., Muntaner, C., Walrath, C., Nickerson, K. J., LaVeist, T. A., & Leaf, P. J.
(2000). Racial differences in attitudes toward professional mental health care
and in the use of services. American Journal of Orthopsychiatry, 70, 455–464.
[Link]
Diaz-Martinez, A., Interian, A., & Waters, D. (2010). The integration of CBT, multi-
cultural and feminist psychotherapies with Latinas. Journal of Psychotherapy
Integration, 20, 312–326. [Link]
Dickson, G. L., Argus-Calvo, B., & Tafoya, N. G. (2010). Multicultural counselor
training experiences: Training effects and perceptions of training among a sam-
ple of predominately Hispanic students. Counselor Education and Supervision, 49,
247–265. [Link]
Dickson, G. L., & Jepsen, D. A. (2007). Multicultural training experiences as predic-
tors of multicultural competencies: Students’ perspectives. Counselor Education
and Supervision, 47, 76–95. [Link]
Diener, E., & Diener, M. (1995). Cross-cultural correlates of life satisfaction and self-
esteem. Journal of Personality and Social Psychology, 68, 653–663. [Link]
org/10.1037/0022-3514.68.4.653
Diener, E., & Oishi, S. (2000). Money and happiness: Income and subjective well-being
across nations. In E. M. Suh (Ed.), Culture and subjective well-being (pp. 185–218).
Cambridge, MA: The MIT Press.
Diener, E., & Tov, W. (2009). Well-being on planet Earth. Psychological Topics, 18,
213–219.
Domenech Rodríguez, M. M., & Bernal, G. (2012). Frameworks, models, and guide-
lines for cultural adaptation. In G. Bernal & M. M. Domenech Rodríguez (Eds.),
Cultural adaptations: Tools for evidence-based practice with diverse populations
(pp. 23–44). Washington, DC: American Psychological Association. http://
[Link]/10.1037/13752-002

264       references

Dressler, W. W., Oths, K. S., & Gravlee, C. C. (2005). Race and ethnicity in public
health research: Models to explain health disparities. Annual Review of Anthropol-
ogy, 34, 231–252. [Link]
Drinane, J. M., Owen, J., Adelson, J. L., & Rodolfa, E. (2014). Multicultural compe-
tencies: What are we measuring? Psychotherapy Research, 1–10. [Link]
10.1080/10503307.2014.983581
Dumas, J. E., Moreland, A. D., Gitter, A. H., Pearl, A. M., & Nordstrom, A. H. (2008).
Engaging parents in preventive parenting groups: Do ethnic, socioeconomic,
and belief match between parents and group leaders matter? Health Education &
Behavior, 35, 619–633. [Link]
Dunn, T. W., Smith, T. B., & Montoya, J. A. (2006). Multicultural competency instru-
mentation: A review and analysis of reliability generalization. Journal of Coun-
seling & Development, 84, 471–482. [Link]
tb00431.x
Duval, S., & Tweedie, R. (2000). Trim and fill: A simple funnel-plot-based method of
testing and adjusting for publication bias in meta-analysis. Biometrics, 56, 455–463.
[Link]
Dwyer, D. (2000). Interpersonal relationships. Florence, KY: Taylor & Frances/Routledge.
Escobar, J., & Vega, W. (2000). Mental health and immigration’s AAAs: Where are
we and where do we go from here? Journal of Nervous and Mental Disorders, 188,
736–740.
Evans, B., & Whitfield, J. (Eds.). (1988). Black males in the United States: An anno-
tated bibliography from 1967–1987. Washington, DC: American Psychological
Association.
Falicov, C. J. (2009). Commentary: On the wisdom and challenges of culturally attuned
treatments for Latinos. Family Process, 48, 292–309. [Link]
j.1545-5300.2009.01282.x
Fay, B. (1996). Contemporary philosophy of social science: A multicultural approach.
Malden, MA: Blackwell.
Festinger, L. (1954). A theory of social comparison processes. Human Relations, 7,
117–140. [Link]
Fischer, A. R., & Shaw, C. M. (1999). African Americans’ mental health and percep-
tions of racist discrimination: The moderating effects of racial socialization expe-
riences and self-esteem. Journal of Counseling Psychology, 46, 395–407. http://
[Link]/10.1037/0022-0167.46.3.395
Fischer, E. H., & Farina, A. (1995). Attitudes toward seeking professional psycho-
logical help: A shortened form and considerations for research. Journal of College
Student Development, 36, 368–373.
Fisher, C. B., Hoagwood, K., Boyce, C., Duster, T., Frank, D. A., Grisso, T., . . . Zayas,
L. H. (2002). Research ethics for mental health science involving ethnic minor-
ity children and youths. American Psychologist, 57, 1024–1040. [Link]
org/10.1037/0003-066X.57.12.1024

references      265
Flaskerud, J. H. (1986). The effects of culture-compatible intervention on the utili-
zation of mental health services by minority clients. Community Mental Health
Journal, 22, 127–141. [Link]
Flaskerud, J. H., & Liu, P. Y. (1990). Influence of therapist ethnicity and language
on therapy outcomes of Southeast Asian clients. International Journal of Social
Psychiatry, 36, 18–29. [Link]
Flicker, S. M., Waldron, H. B., Turner, C. W., Brody, J. L., & Hops, H. (2008). Ethnic
matching and treatment outcome with Hispanic and Anglo substance-abusing
adolescents in family therapy. Journal of Family Psychology, 22, 439–447. http://
[Link]/10.1037/0893-3200.22.3.439
Fouad, N. A. (2006). Multicultural guidelines: Implementation in an urban counsel-
ing psychology program. Professional Psychology: Research and Practice, 37, 6–13.
[Link]
Fouad, N. A., & Arredondo, P. (2007). Becoming culturally oriented: Practical advice for
psychologists and educators. Washington, DC: American Psychological Association.
[Link]
Freire, P. (1973). Education for critical consciousness. New York, NY: Seabury.
Fuertes, J. N., Stracuzzi, T. I., Bennett, J., Scheinholtz, J., Mislowack, A., Hersh, M.,
& Cheng, D. (2006). Therapist multicultural competency: A study of therapy
dyads. Psychotherapy: Theory, Research, Practice, Training, 43, 480–490. http://
[Link]/10.1037/0033-3204.43.4.480
Gans, H. (2003, March 7). Identity. The Chronicle of Higher Education, p. B4.
Garcia, S. M., Tor, A., & Gonzalez, R. (2006). Ranks and rivals: A theory of com-
petition. Personality and Social Psychology Bulletin, 32, 970–982. [Link]
org/10.1177/0146167206287640
Garland, A. F., Hough, R. L., Landsverk, J. A., McCabe, K., Yeh, M., Ganger, W., &
Reynolds, B. J. (2000). Racial and ethnic variations in mental health care utili-
zation among children in foster care. Children’s Services: Social Policy, Research,
& Practice, 3, 133–146. [Link]
Garroutte, E. M., Kunovich, R. M., Jacobsen, C., & Goldberg, J. (2004). Patient sat-
isfaction and ethnic identity among American Indian older adults. Social Science
& Medicine, 59, 2233–2244. [Link]
Gaztambide, D. J. (2009). Religion as a wellspring of healing and liberation: Toward a
liberation psychology of religion (Unpublished doctoral dissertation). Union Theo-
logical Seminary, New York, NY.
Gaztambide, D. J. (2012). Addressing cultural impasses with rupture resolution strat-
egies: A proposal and recommendations. Professional Psychology: Research and
Practice, 43, 183–189. [Link]
Geertz, C. (2000). Available light: Anthropological reflections on philosophical topics.
Princeton, NJ: Princeton University Press.

266       references

Gelso, C. J. (2011). The real relationship in psychotherapy: The hidden foundation of
change. Washington, DC: American Psychological Association. [Link]
org/10.1037/12349-000
George, L. K. (2010). Still happy after all these years: Research frontiers on subjec-
tive well-being in later life. Journal of Gerontology: Social Sciences, 10, 1–9. http://
[Link]/10.1093/geronb/gbq006
Gergen, K. J. (1995). Relational theory and the discourses of power. In K. J. Gergen
(Ed.), Management and organization: Relational alternatives to individualism
(pp. 29–50). Brookfield, VT: Avebury/Ashgate.
Gergen, K. J. (2009). Relational being: Beyond self and community. New York, NY:
Oxford University Press.
Gladwell, M. (2006). The tipping point: How little things can make a big difference. New
York, NY: Hachette Digital.
Glass, G. (1976). Primary, secondary, and meta-analysis of research. Educational
Researcher, 5, 3–8. [Link]
Golding, J. M., & Wells, K. B. (1990). Social support and use of mental health ser-
vices by Mexican Americans and non-Hispanic Whites. Basic and Applied Social
Psychology, 11, 443–458. [Link]
Gómez, L. E., & López, N. (2013). Mapping race: Critical approaches to health disparities
research. New Brunswick, NJ: Rutgers University Press.
Gone, J. P. (2011). Is psychological science a-cultural? Cultural Diversity and Ethnic
Minority Psychology, 17, 234–242. [Link]
Gone, J. P., & Trimble, J. E. (2012). American Indian and Alaska Native mental health:
Diverse perspectives on enduring disparities. Annual Review of Clinical Psychol-
ogy, 8, 131–160. [Link]
Gong, L. (2007). Ethnic identity and identification with the majority group: Rela-
tions with national identity and self-esteem. International Journal of Intercultural
Relations, 31, 503–523. [Link]
Gordon, M. M. (1978). Human nature, class, and ethnicity. New York, NY: Oxford
University Press.
Graves, T. D. (1967). Acculturation, access, and alcohol in a tri-ethnic community.
American Anthropologist, 69, 306–321. [Link]
3-4.02a00030
Greeley, A. M. (1974). Ethnicity in the United States. New York, NY: Wiley.
Griner, D., & Smith, T. B. (2006). Culturally adapted mental health intervention:
A meta-analytic review. Psychotherapy: Theory, Research, Practice, Training, 43,
531–548. [Link]
Grusec, J. E., & Hastings, P. D. (2007). Handbook of socialization: Theory and research.
New York, NY: Guilford Press.

references      267
Guthrie, R. V. (1976). Even the rat was White: A historical view of psychology. New York,
NY: Harper & Row.
Guthrie, R. V. (2004). Even the rat was White: A historical view of psychology (2nd ed.).
Upper Saddle River, NJ: Pearson.
Haasen, C., Demiralay, C., & Reimer, J. (2008). Acculturation and mental distress
among Russian and Iranian migrants in Germany. European Psychiatry, 23, 10–13.
[Link]
Hall, C. C. I. (1997). Cultural malpractice. The growing obsolescence of psychology
with the changing U.S. population. American Psychologist, 52, 642–651. http://
[Link]/10.1037/0003-066X.52.6.642
Hall, G. C. N. (2001). Psychotherapy research with ethnic minorities: Empirical,
ethical, and conceptual issues. Journal of Consulting and Clinical Psychology, 69,
502–510. [Link]
Hall, G. C. N., & Yee, A. H. (2014). Evidence-based practice. In F. T. L. Leong,
L. Comas-Díaz, G. C. Nagayama Hall, V. C. McLoyd, & J. E. Trimble (Eds.), APA
handbook of multicultural psychology: Vol. 2. Applications and training (pp. 59–79).
Washington, DC: American Psychological Association.
Haller, M., & Hadler, M. (2006). How social relations and structures can produce
happiness and unhappiness: An international comparative analysis. Social
Indicators Research, 75, 169–216. [Link]
6297-y
Hannigan, T. (1990). Traits, attitudes, and skills that are related to intercultural effec-
tiveness and their implications for cross-cultural training: A review of the lit-
erature. International Journal of Intercultural Relations, 14, 89–111. [Link]
org/10.1016/0147-1767(90)90049-3
Harré, R., & Robinson, D. N. (1995). On the primacy of duties. Philosophy, 70, 513.
[Link]
Harrell, S. P. (2000). A multidimensional conceptualization of racism-related
stress: Implications for the well-being of people of color. American Journal of
Orthopsychiatry, 70, 42–57. [Link]
Harris, H. W., Blue, H. C., & Griffith, E. H. (Eds.). (1995). Racial and ethnic identity.
New York, NY: Routledge.
Harris, P. M. (1998). Attrition revisited. American Journal of Evaluation, 19, 293–305.
[Link]
Hayes, J. A., Owen, J., & Bieschke, K. J. (2015). Therapist differences in symptom
change with racial/ethnic minority clients. Psychotherapy. Advance online pub-
lication. [Link]
Heath, A. E., Neimeyer, G. J., & Pedersen, P. B. (1988). The future of cross-cultural
counseling: A Delphi poll. Journal of Counseling & Development, 67, 27–30.
[Link]
Helms, J. E. (1990). (Ed.) Black and White racial identity: Theory, research, and practice.
Westport, CT: Greenwood Press.

268       references

Helms, J. E. (1994). The conceptualization of racial identity and other “racial” con-
structs. In E. J. Trickett, R. J. Watts, & D. Birman (Eds.), Human diversity:
Perspectives on people in context (pp. 285–311). San Francisco, CA: Jossey-Bass.
Helms, J. E. (2007). Some better practices for measuring racial and ethnic iden-
tity constructs. Journal of Counseling Psychology, 54, 235–246. [Link]
org/10.1037/0022-0167.54.3.235
Helms, J. E., Jernigan, M., & Mascher, J. (2005). The meaning of race in psychology
and how to change it: A methodological perspective. American Psychologist, 60,
27–36. [Link]
Henderson, R. C., Williams, P., Gabbidon, J., Farrelly, S., Schauman, O., Hatch,
S., . . . the MIRIAD study group. (2014). Mistrust of mental health services:
Ethnicity, hospital admission and unfair treatment. Epidemiology and Psychiatric
Sciences, 1–8. [Link]
Henrich, J., Heine, S. J., & Norenzayan, A. (2010). The weirdest people in the
world? Behavioral and Brain Sciences, 33, 61–83. [Link]
S0140525X0999152X
Hinton, D. E., Rivera, E. I., Hofmann, S. G., Barlow, D. H., & Otto, M. W. (2012).
Adapting CBT for traumatized refugees and ethnic minority patients: Examples
from culturally adapted CBT (CA-CBT). Transcultural Psychiatry, 49, 340–365.
[Link]
Holt-Lunstad, J., Smith, T. B., Baker, M., Harris, T., & Stephenson, D. (2015).
Loneliness and social isolation as risk factors for mortality: A meta-analytic
review. Perspectives on Psychological Science, 10, 227–237. [Link]
10.1177/1745691614568352
Holt-Lunstad, J., Smith, T. B., & Layton, J. B. (2010). Social relationships and mor-
tality risk: A meta-analytic review. PLoS Medicine, 7, e1000316–e1000316.
[Link]
Howe, C., Heim, D., & O’Connor, R. (2014). Racism, identity and psychological
well-being: A longitudinal perspective on politically embattled relations. Ethnic
and Racial Studies, 37, 2457–2474. [Link]
35057
Howell, R. T., & Howell, C. J. (2008). The relation of economic status to subjective
well-being in developing countries: A meta-analysis. Psychological Bulletin, 134,
536–560. [Link]
Hoyt, W. T., Warbasse, R. E., & Chu, E. Y. (2006). Construct validation in counsel-
ing psychology research. The Counseling Psychologist, 34, 769–805. [Link]
org/10.1177/0011000006287389
Huey, S. J., Jr., & Polo, A. J. (2008). Evidence-based psychosocial treatments for
ethnic minority youth. Journal of Clinical Child and Adolescent Psychology, 37,
262–301. [Link]
Huey, S. J., Jr., Tilley, J. L., Jones, E. O., & Smith, C. A. (2014). The contribution of
cultural competence to evidence-based care for ethnically diverse populations.
Annual Review of Clinical Psychology, 10, 305–338. [Link]
annurev-clinpsy-032813-153729

references      269
Hunt, L. M., Schneider, S., & Comer, B. (2004). Should “acculturation” be a variable
in health research? A critical review of research on U.S. Hispanics. Social Science
& Medicine, 59, 973–986. [Link]
Hurwich-Reiss, E., Rindlaub, L. A., Wadsworth, M. E., & Markman, H. J. (2014).
Cultural adaptation of a family strengthening intervention for low-income
Spanish-speaking families. Journal of Latina/o Psychology, 2, 21.
Hwang, W. C. (2006). The psychotherapy adaptation and modification framework:
Application to Asian Americans. American Psychologist, 61, 702–715. http://
[Link]/10.1037/0003-066X.61.7.702
Hwang, W. C. (2009). The formative method for adapting psychotherapy (FMAP): A
community-based developmental approach to culturally adapting therapy. Profes-
sional Psychology: Research and Practice, 40, 369–377. [Link]
a0016240
Hwang, W. C., & Myers, H. F. (2007). Major depression in Chinese Americans: The
roles of stress, vulnerability, and acculturation. Social Psychiatry and Psychiatric
Epidemiology, 42, 189–197. [Link]
Ibaraki, A. Y., & Hall, G. C. N. (2014). The components of cultural match in
psychotherapy. Journal of Social and Clinical Psychology, 33, 936–953. http://
[Link]/10.1521/jscp.2014.33.10.936
Imel, Z. E., Baldwin, S., Atkins, D. C., Owen, J., Baardseth, T., & Wampold, B. E.
(2011). Racial/ethnic disparities in therapist effectiveness: A conceptualization
and initial study of cultural competence. Journal of Counseling Psychology, 58,
290–298. [Link]
Ivey, A. E., & Collins, N. M. (2003). Social justice: A long-term challenge for coun-
seling psychology. The Counseling Psychologist, 31, 290–298. [Link]
10.1177/0011000003031003004
Ivey, A. E., & Zalaquett, C. P. (2009). Psychotherapy as liberation: Multicultural coun-
seling and psychotherapy (MCT) contributions to the promotion of psychological
emancipation. In J. L. Chin (Ed.), Diversity in mind and in action: Vol. 3. Social
justice matters. (pp. 181–199). Santa Barbara, CA: Praeger/ABC-CLIO.
Jackson, L. C., Schmutzer, P. A., Wenzel, A., & Tyler, J. D. (2006). Applicability of
cognitive–behavior therapy with American Indian individuals. Psychotherapy:
Theory, Research, Practice, Training, 43, 506–517. [Link]
3204.43.4.506
Jahoda, G. (1979). A cross-cultural perspective on experimental social psychology.
Personality and Social Psychology Bulletin, 5, 142–148. [Link]
014616727900500203
Jahoda, G. (1988). J’Accuse. In M. Bond (Ed.), The cross-cultural challenge to social
psychology (pp. 86–95). Newbury Park, CA: Sage.
Jones, J. M. (2003). Constructing race and deconstructing racism: A cultural psychol-
ogy approach. In G. Bernal, J. E. Trimble, A. K. Burlew, & F. T. Leong (Eds.),
Handbook of racial and ethnic minority psychology (pp. 276–290). London, England:
Sage. [Link]

270       references

Jones, L. K. (1974). The Counselor Evaluation Rating Scale: A valid criterion of
counselor effectiveness? Counselor Education and Supervision, 14, 112–116. http://
[Link]/10.1002/j.1556-6978.1974.tb00852.x
Jordan, J. V. (2010). Relational–cultural therapy. Washington, DC: American Psycho-
logical Association.
Kaplan, A. (1986). The “self-in-relation”: Implications for depression in women. Psy-
chotherapy: Theory, Research, Practice, Training, 23, 234–242. [Link]
10.1037/h0085603
Kaslow, N. J., Rubin, N. J., Bebeau, M. J., Leigh, I. W., Lichtenberg, J. W., Nelson,
P. D., . . . Smith, I. L. (2007). Guiding principles and recommendations for
the assessment of competence. Professional Psychology: Research and Practice, 38,
441–451. [Link]
Katz, A. D., & Hoyt, W. T. (2014). The influence of multicultural counseling compe-
tence and anti-Black prejudice on therapists’ outcome expectancies. Journal of
Counseling Psychology, 61, 299–305. [Link]
Katz, M. H., & Hauck, W. W. (1993). Proportional hazards (Cox) regression. Journal
of General Internal Medicine, 8, 702–711. [Link]
Katz, P. A., & Taylor, D. A. (Eds.). (1988). Eliminating racism: Profiles in controversy.
New York, NY: Plenum Press. [Link]
Kazdin, A. E. (1996). Dropping out of child psychotherapy: Issues for research and
implications for practice. Clinical Child Psychology and Psychiatry, 1, 133–156.
[Link]
Kazdin, A. E. (2008). Evidence-based treatment and practice: New opportunities to
bridge clinical research and practice, enhance the knowledge base, and improve
patient care. American Psychologist, 63, 146–159. [Link]
066X.63.3.146
Keita, G., & Petersen, A. (Eds.). (1996). Blacks in the United States: Abstracts of the
psychological and behavioral Literature, 1987–1995. Washington, DC: American
Psychological Association.
Kelly, T. A., & Strupp, H. H. (1992). Patient and therapist values in psycho-
therapy: Perceived changes, assimilation, similarity, and outcome. Journal of
Consulting and Clinical Psychology, 60, 34–40. [Link]
006X.60.1.34
Kiev, A. (1972). Transcultural psychiatry. New York, NY: Free Press.
Kim, B. S. K., & Abreu, J. M. (2001). Acculturation measurement: Theory, current
instruments, and future directions. In J. Ponterotto, M. Casas, L. Suzuki, &
C. Alexander (Eds.), Handbook of multicultural counseling (2nd ed.; pp. 394–
424). Thousand Oaks, CA: Sage.
Kim, B. S. K., Atkinson, D. R., & Umemoto, D. (2001). Asian cultural values and the
counseling process: Current knowledge and directions for future research. The
Counseling Psychologist, 29, 570–603. [Link]
Kim, U., Yang, K., & Hwang, K. (Eds.). (2006). Indigenous and cultural psychology.
New York, NY: Springer. [Link]

references      271
Kirke, D. M. (2007). Social network analysis and psychological research. The Irish
Journal of Psychology, 28(1–2), 53–61. [Link]
10446248
Kiselica, M. S. (1998). Preparing Anglos for the challenges and joys of multi­
culturalism. The Counseling Psychologist, 26, 5–21. [Link]
0011000098261001
Kivisto, P., & Nefzger, B. (1993). Symbolic ethnicity and American Jews: The rela-
tionship of ethnic identity to behavior and group affiliation. The Social Science
Journal, 30, 1–12. [Link]
Kleinman, A., Eisenberg, L., & Good, B. (1978). Culture, illness, and care: Clinical
lessons from anthropologic and cross-cultural research. Annals of Internal Medi-
cine, 88, 251–258. [Link]
Knox, S., Adrians, N., Everson, E., Hess, S., Hill, C., & Crook-Lyon, R. (2011). Clients’
perspectives on therapy termination. Psychotherapy Research, 21, 154–167. http://
[Link]/10.1080/10503307.2010.534509
Knox, S., Burkard, A. W., Johnson, A. J., Suzuki, L. A., & Ponterotto, J. G. (2003).
African American and European American therapists’ experiences of address-
ing race in cross-racial psychotherapy dyads. Journal of Counseling Psychology, 50,
466–481. [Link]
Krippendorff, K. (2012). Content analysis: An introduction to its methodology (3rd ed.).
Thousand Oaks, CA: Sage.
Kuhn, T. S. (1996). The structure of scientific revolutions (3rd ed.). Chicago, IL: University
of Chicago Press. [Link]
Kuhn, T. S. (2012). The structure of scientific revolutions (50th anniversary ed.). Chicago,
IL: University of Chicago Press.
LaFromboise, T. D., Coleman, H. L. K., & Gerton, J. (1993). Psychological impact of
biculturalism: Evidence and theory. Psychological Bulletin, 114, 395–412. http://
[Link]/10.1037/0033-2909.114.3.395
LaFromboise, T. D., Coleman, H. L. K., & Hernandez, A. (1991). Development and
factor structure of the Cross-Cultural Counseling Inventory—Revised. Professional
Psychology: Research and Practice, 22, 380–388. [Link]
7028.22.5.380
Lambert, M. J. (2007). What we have learned from a decade of research aimed at
improving psychotherapy outcome in routine care. Psychotherapy Research, 17,
1–14. [Link]
Lambert, M. J. (2010). Prevention of treatment failure: The use of measuring, monitoring,
and feedback in clinical practice. Washington, DC: American Psychological Associa-
tion. [Link]
Landis, D., & Brislin, R. W. (Eds.). (2013). Handbook of intercultural training: Vol. 1.
Issues in theory and design. New York, NY: Elmsfeld.
Langer, N. (1999). Culturally competent professionals in therapeutic alliances enhance
patient compliance. Journal of Health Care for the Poor and Underserved, 10, 19–26.
[Link]

272       references

La Roche, M. J. (2012). Cultural psychotherapy: Theory, methods, and practice. Atlanta,
GA: Sage.
La Roche, M. J., & Christopher, M. S. (2008). Culture and empirically supported
treatments: On the road to a collision? Culture & Psychology, 14, 333–356. http://
[Link]/10.1177/1354067X08092637
La Roche, M. J., D’Angelo, E., Gualdron, L., & Leavell, J. (2006). Culturally sen-
sitive guided imagery for allocentric Latinos: A pilot study. Psychotherapy:
Theory, Research, Practice, Training, 43, 555–560. [Link]
3204.43.4.555
La Roche, M. J., & Lustig, K. (2010). Cultural adaptations: Unpacking the meaning
of culture. The Scientific Review of Mental Health Practice, 7(2), 26–30.
La Roche, M. J., & Lustig, K. (2013). Being mindful about the assessment of culture:
A cultural analysis of culturally adapted acceptance-based behavior therapy
approaches. Cognitive and Behavioral Practice, 20, 60–63. [Link]
[Link].2012.04.002
Lassiter, P. S., Napolitano, L., Culbreth, J., & Ng, K. (2008). Developing multicultural
competence using the structured peer group supervision model. Counselor Educa-
tion and Supervision, 47, 164–178. [Link]
tb00047.x
Lau, A. S. (2006). Making the case for selective and directed cultural adaptations of
evidence-based treatments: Examples from parent training. Clinical Psychology:
Science and Practice, 13, 295–310. [Link]
00042.x
Lau, A. S., Chang, D. F., & Okazaki, S. (2010). Methodological challenges in treat-
ment outcome research with ethnic minorities. Cultural Diversity and Ethnic
Minority Psychology, 16, 573–580. [Link]
Lazarus, R. S., & Folkman, S. (1984). Stress, appraisal, and coping. New York, NY:
Springer-Verlag.
Lee, D. L., & Ahn, S. (2011). Racial discrimination and Asian mental health: A
meta-analysis. The Counseling Psychologist, 39, 463–489. [Link]
0011000010381791
Lee, R. M. (2005). Resilience against discrimination: Ethnic identity and other-
group orientation as protective factors for Korean Americans. Journal of Coun-
seling Psychology, 52, 36–44. [Link]
Lee, R. M., Chalk, L., Conner, S. E., Kawasaki, N., Jannetti, A., LaRue, T., & Rodolfa,
E. (1999). The status of multicultural counseling training at counseling intern-
ship sites. Journal of Multicultural Counseling and Development, 27, 58–74. http://
[Link]/10.1002/j.2161-1912.1999.tb00215.x
Lee, R. M., Yoon, E., & Liu-Tom, H. T. (2006). Structure and measurement of
acculturation/enculturation for Asian Americans using the ARSMA-II. Mea-
surement and Evaluation in Counseling and Development, 39, 42–55.
Leong, F. T. L. (2011). Cultural accommodation model of counseling. Journal of
Employment Counseling, 48, 150–152. [Link]
2011.tb01099.x

references      273
Leong, F. T. L., Comas-Díaz, L., Nagayama Hall, G. C., McLoyd, V. C., & Trimble, J. E.
(Eds.). (2014). APA handbook of multicultural psychology (Vols. 1–2). Washington,
DC: American Psychological Association. [Link]
Leong, F. T. L., Holliday, B. G., Trimble, J. E., Padilla, A. M., & McCubbin, L. D. (2012).
Ethnic minority psychology. In I. B. Weiner (Ed.), Handbook of psychology: Vol. 1.
Ethnic minority psychology (2nd ed., pp. 530–561). Hoboken, NJ: Wiley.
Leong, F. T. L., Kim, H. H., & Gupta, A. (2011). Attitudes toward professional coun-
seling among Asian-American college students: Acculturation, conceptions of
mental illness, and loss of face. Asian American Journal of Psychology, 2, 140–153.
[Link]
Leong, F. T. L., & Whitfield, J. R. (Eds.). (1992). Asians in the United States: Abstracts of
the psychological and behavioral literature, 1967–1991. Washington, DC: American
Psychological Association.
Lerner, R. M., & Overton, W. F. (2008). Exemplifying the integrations of the rela-
tional developmental system: Synthesizing theory, research, and application to
promote positive development and social justice. Journal of Adolescent Research,
23, 245–255. [Link]
Levinas, E. (1969). Totality and infinity: An essay on exteriority. Pittsburgh, PA: Duquesne
University Press.
Levinas, E. (1998). Entre nous: On thinking-of-the-other. New York, NY: Columbia
University Press.
Li, L. C., & Kim, B. S. (2004). Effects of counseling style and client adherence to Asian
cultural values on counseling process with Asian American college students.
Journal of Counseling Psychology, 51, 158–167. [Link]
0167.51.2.158
Lipsey, M. W., & Wilson, D. B. (2001). Practical meta-analysis. Thousand Oaks, CA:
Sage.
Liu, L. L., & Lau, A. S. (2013). Teaching about race/ethnicity and racism matters: An
examination of how perceived ethnic racial socialization processes are associated
with depression symptoms. Cultural Diversity and Ethnic Minority Psychology, 19,
383–394. [Link]
Lo, C. C., Cheng, T. C., & Howell, R. J. (2014). Access to and utilization of health ser-
vices as pathway to racial disparities in serious mental illness. Community Mental
Health Journal, 50, 251–257. [Link]
Lonner, W. J. (2000). On the growth and continuing importance of cross-cultural
psychology. Eye on Psi Chi, 4(3), 22–26.
López, S. R., Barrio, C., Kopelowicz, A., & Vega, W. A. (2012). From documenting to
eliminating disparities in mental health care for Latinos. American Psychologist,
67, 511–523. [Link]
López, S. R., Grover, K. P., Holland, D., Johnson, M. J., Kain, C. D., Kanel, K., . . . Rhyne,
M. C. (1989). Development of culturally sensitive psychotherapists. Professional
Psychology: Research and Practice, 20, 369–376. [Link]
7028.20.6.369

274       references

Machado, A., & Silva, F. J. (2007). Toward a richer view of the scientific method.
The role of conceptual analysis. American Psychologist, 62, 671–681. http://
[Link]/10.1037/0003-066X.62.7.671
Magyar-Moe, J., & López, S. J. (2008). Human agency, strengths-based development,
and well-being. In W. B. Walsh (Ed.), Biennial review of counseling psychology
(Vol. 1, pp. 157–175). New York, NY: Routledge/Taylor & Francis.
Mandara, J., Gaylord-Harden, N. K., Richards, M. H., & Ragsdale, B. L. (2009).
The effects of changes in racial identity and self-esteem on changes in African
American adolescents’ mental health. Child Development, 80, 1660–1675. http://
[Link]/10.1111/j.1467-8624.2009.01360.x
Maramba, G. G., & Nagayama Hall, G. C. (2002). Meta-analyses of ethnic match as a
predictor of dropout, utilization, and level of functioning. Cultural Diversity and Eth-
nic Minority Psychology, 8, 290–297. [Link]
Martín-Baró, I. M., Aron, A., & Corne, S. (1994). Writings for a liberation psychology.
Cambridge, MA: Harvard University Press.
McCabe, K., Yeh, M., Lau, A., & Argote, C. B. (2012). Parent–child interaction ther-
apy for Mexican Americans: Results of a pilot randomized clinical trial at follow-
up. Behavior Therapy, 43, 606–618. [Link]
McCubbin, L. D., & Marsella, A. (2009). Native Hawaiians and psychology: The
cultural and historical context of indigenous ways of knowing. Cultural Diversity
and Ethnic Minority Psychology, 15, 374–387. [Link]
McCutcheon, S. R., & Imel, Z. E. (2009). Valuing diversity implementing our best
intentions. The Counseling Psychologist, 37, 760–771. [Link]
0011000009334094
McGuire, T. G., Alegría, M., Cook, B. L., Wells, K. B., & Zaslavsky, A. M. (2006).
Implementing the Institute of Medicine definition of disparities: An application
to mental health care. Health Services Research, 41, 1979–2005. [Link]
org/10.1111/j.1475-6773.2006.00583.x
McGuire, T. G., & Miranda, J. (2008). New evidence regarding racial and ethnic
disparities in mental health: Policy implications. Health Affairs, 27, 393–403.
[Link]
McKee-Ryan, F., Song, Z., Wanberg, C. R., & Kinicki, A. J. (2005). Psychological
and physical well-being during unemployment: A meta-analytic study. Journal
of Applied Psychology, 90, 53–76.
McNeill, B. W. (Ed.). (2009). Intersections of multiple identities: A casebook of evidence-
based practices with diverse populations. New York, NY: Routledge/Taylor & Francis.
McPherson, M., Smith-Lovin, L., & Cook, J. M. (2001). Birds of a feather: Homoph-
ily in social networks. Annual Review of Sociology, 27, 415–444. [Link]
org/10.1146/[Link].27.1.415
Mead, M. (1963). Sex and temperament in three primitive societies. New York, NY:
Morrow.
Mendenhall, M., & Oddou, G. (1985). The dimensions of expatriate acculturation:
A review. The Academy of Management Review, 10, 39–47.

references      275
Mendoza, R. (1984). Acculturation and sociocultural variability. In J. L. Martinez
& R. Mendoza (Eds.), Chicano Psychology (2nd ed., pp. 61–75). Orlando, FL:
Academic Press. [Link]
Merton, R. K. (1968). Social theory and social structure (Rev ed.). New York, NY: Free
Press.
Meyer, O. L., & Takeuchi, D. T. (2014). Help seeking and service utilization. In
F. T. L. Leong, L. Comas-Díaz, G. C. Nagayama Hall, V. C. McLoyd, &
J. E. Trimble (Eds.), APA handbook of multicultural psychology: Vol. 2. Applica-
tions and training (pp. 529–541). Washington, DC: American Psychological
Association.
Miller, J., & Schwarz, W. (2011). Aggregate and individual replication probabil-
ity within an explicit model of the research process. Psychological Methods, 16,
337–360. [Link]
Miller, J. B. (1986). Toward a new psychology of women (2nd ed.). Boston, MA: Beacon
Press.
Miller, M. J. (2007). A bilinear multidimensional measurement model of Asian Amer-
ican acculturation and enculturation: Implications for counseling interventions.
Journal of Counseling Psychology, 54, 118–131. [Link]
0167.54.2.118
Miller, M. J., & Sheu, H. (2008). Conceptual and measurement issues in multicultural
psychology research. In S. D. Brown & R. W. Lent (Eds.), Handbook of counseling
psychology (4th ed., pp. 103–120). New York, NY: Wiley.
Miller, N. B. (1982). Social work services to urban Indians. In J. W. Green (Ed.),
Cultural awareness in the human services (pp. 157–183). Englewood Cliffs, NJ:
Prentice Hall.
Miranda, J., Bernal, G., Lau, A., Kohn, L., Hwang, W. C., & LaFromboise, T. (2005).
State of the science on psychosocial interventions for ethnic minorities. Annual
Review of Clinical Psychology, 1, 113–142. [Link]
clinpsy.1.102803.143822
Miville, M. L., & Ferguson, A. D. (Eds.). (2014). Handbook of race–ethnicity and gender
in psychology. New York, NY: Springer. [Link]
8860-6
Mohatt, G. V. (1989). The community as informant or collaborator? American Indian
and Alaska Native Mental Health Research, 2(3), 64–70. [Link]
aian.0203.1989.64
Mohatt, N. V., Fok, C. C. T., Burket, R., Henry, D., & Allen, J. (2011). Assessment
of awareness of connectedness as a culturally-based protective factor for Alaska
native youth. Cultural Diversity and Ethnic Minority Psychology, 17, 444–455.
[Link]
Moher, D., Liberati, A., Tetzlaff, J., Altman, D. G., & the PRISMA Group. (2009).
Preferred reporting items for systematic reviews and meta-analyses: The
PRISMA statement. Annals of Internal Medicine, 151, 264–269, W64. http://
[Link]/10.7326/0003-4819-151-4-200908180-00135

276       references

Mollen, D., Kelly, S. M., & Ridley, C. R. (2011). Therapeutic change: The raison
d’être for counseling competence. The Counseling Psychologist, 39, 918–927. http://
[Link]/10.1177/0011000011405221
Morgan, H. W. (1981). Drugs in America. Syracuse, NY: Syracuse University Press.
Moynihan, D. P. (1993). Pandaemonium: Ethnicity in international politics. New
York, NY: Oxford University Press.
Mui, A. C., & Kang, S. Y. (2006). Acculturation stress and depression among
Asian immigrant elders. Social Work, 51, 243–255. [Link]
sw/51.3.243
Murdock, N. L., Edwards, C., & Murdock, T. B. (2010). Therapists’ attributions
for client premature termination: Are they self-serving? Psychotherapy: Theory,
Research, Practice, Training, 47, 221–234. [Link]
Myers, D. G. (2008). Will money buy happiness? In S. J. López (Ed.), Positive psychology:
Exploring the best in people: Vol. 4. Pursuing human flourishing (pp. 37–56). Westport,
CT: Praeger/Greenwood.
Myers, D. G. (2012). Human connections and the good life: Balancing individu-
ality and community in public policy. In S. Joseph (Ed.), Positive psychology
in practice (pp. 641–657). Hoboken, NJ: Wiley. [Link]
9780470939338.ch38
National Research Council. (2004). Measuring racial discrimination. Washington, DC:
National Academies Press.
Negy, C., Shreve, T. L., Jensen, B. J., & Uddin, N. (2003). Ethnic identity, self-
esteem, and ethnocentrism: A study of social identity versus multicultural theory
of development. Cultural Diversity and Ethnic Minority Psychology, 9, 333–344.
[Link]
Neighbors, H. W., Caldwell, C., Williams, D. R., Nesse, R., Taylor, R. J., Bullard, K.
M., . . . Jackson, J. S. (2007). Race, ethnicity, and the use of services for mental
disorders: Results from the National Survey of American Life. Archives of
General Psychiatry, 64, 485–494. [Link]
Neville, H. A., Spanierman, L. B., & Lewis, J. A. (2012). The expanded psycho-
social model of racism: A new model for understanding and disrupting racism
and White privilege. In N. A. Fouad, J. A. Carter, & L. M. Subich (Eds.), APA
handbook of counseling psychology: Vol. 2. Practice, interventions, and applications
(pp. 333–360). Washington, DC: American Psychological Association. http://
[Link]/10.1037/13755-014
Newell, M. L., Nastasi, B. K., Hatzichristou, C., Jones, J. M., Schanding, G. T., &
Yetter, G. (2010). Evidence on multicultural training in school psychology: Rec-
ommendations for future directions. School Psychology Quarterly, 25, 249–278.
[Link]
Norcross, J. C. (2011). Psychotherapy relationships that work (2nd ed.). New York, NY: Oxford
University Press. [Link]

references      277
Norcross, J. C., & Lambert, M. J. (2011). Evidence-based therapy relationships. In
J. C. Norcross (Ed.), Psychotherapy relationships that work (2nd ed., pp. 3–21).
New York, NY: Oxford University Press.
Norcross, J. C., & Wampold, B. E. (2011). What works for whom: Tailoring psycho-
therapy to the person. Journal of Clinical Psychology, 67, 127–132. [Link]
org/10.1002/jclp.20764
Ober, A. M., Granello, D. H., & Henfield, M. S. (2009). A synergistic model to enhance
multicultural competence in supervision. Counselor Education and Supervision, 48,
204–221. [Link]
O’Donohue, W., & Benuto, L., (2010). The many problems with cultural sensitivity.
Scientific Review of Mental Health Practice, 7, 34–37.
Oetting, E. R., & Beauvais, F. (1991). Orthogonal cultural identification theory: The
cultural identification of minority adolescents. International Journal of Addictions,
25, 655–685.
Oh, Y., Koeske, G. F., & Sales, E. (2002). Acculturation, stress, and depressive
symptoms among Korean immigrants in the United States. The Journal of Social
Psychology, 142, 511–526. [Link]
Okun, M. A., Stock, W. A., Haring, M. J., & Witter, R. A. (1984). Health and
subjective well-being: A meta-analysis. International Journal of Aging & Human
Development, 19, 111–132. [Link]
Oliver, K. (2001). Witnessing: Beyond recognition. Minneapolis: University of Minnesota
Press.
Olmedo, E. (1979). Acculturation: A psychometric perspective. American Psychologist,
34, 1061–1070. [Link]
Olmedo, E., & Walker, V. (Eds.). (1990). Hispanics in the United States: Abstracts of the
psychological and behavioral literature, 1980–1989. Washington, DC: American
Psychological Association.
Ono, Y. A. (1982). How I created the theory of relativity. Physics Today, 35, 45–47.
[Link]
Overton, W. F., & Ennis, M. D. (2006a). Cognitive–developmental and behavior–
analytic theories: Evolving into complementarity. Human Development, 49,
143–172. [Link]
Overton, W. F., & Ennis, M. D. (2006b). Relationism, ontology, and other concerns.
Human Development, 49, 180–183. [Link]
Owen, J. J., Imel, Z., Adelson, J., & Rodolfa, E. (2012). ‘No-show’: Therapist racial/
ethnic disparities in client unilateral termination. Journal of Counseling Psychology,
59, 314–320. [Link]
Owen, J. J., Leach, M. M., Wampold, B., & Rodolfa, E. (2011a). Client and therapist
variability in clients’ perceptions of their therapists’ multicultural competencies.
Journal of Counseling Psychology, 58, 1–9. [Link]

278       references

Owen, J. J., Leach, M. M., Wampold, B., & Rodolfa, E. (2011b). Multicultural
approaches in psychotherapy: A rejoinder. Journal of Counseling Psychology, 58,
22–26. [Link]
Owen, J. J., Tao, K., Leach, M. M., & Rodolfa, E. (2011). Clients’ perceptions of
their psychotherapists’ multicultural orientation. Psychotherapy, 48, 274–282.
[Link]
Padgett, D. K., Patrick, C., Burns, B. J., & Schlesinger, H. J. (1994). Women and
outpatient mental health services: Use by Black, Hispanic, and White women
in a national insured population. Journal of Mental Health Administration, 21,
347–360. [Link]
Padilla, A. (1980). The role of cultural awareness and ethnic loyalty in acculturation. In
A. Padilla (Ed.), Acculturation: Theory, models and some new findings (pp. 47–84).
Boulder, CO: Westview.
Paniagua, F. A., & Yamada, A. M. (Eds.). (2013). Handbook of multicultural mental
health: Assessment and treatment of diverse populations. San Diego, CA: Academic
Press.
Paradies, Y. (2006). A systematic review of empirical research on self-reported
racism and health. International Journal of Epidemiology, 35, 888–901. http://
[Link]/10.1093/ije/dyl056
Parra Cardona, J. R., Domenech Rodríguez, M. M., Forgatch, M., Sullivan, C., Bybee,
D., Holtrop, K., . . . Bernal, G. (2012). Culturally adapting an evidence-based
parenting intervention for Latino immigrants: The need to integrate fidelity and
cultural relevance. Family Process, 51, 56–72. [Link]
5300.2012.01386.x
Pascoe, E. A., & Smart Richman, L. (2009). Perceived discrimination and health:
A meta-analytic review. Psychological Bulletin, 135, 531–554. [Link]
org/10.1037/a0016059
Paz, O. (1985). The labyrinth of solitude and other writings. New York, NY: Grove Press.
Pedersen, P. B. (1999). Multiculturalism as a fourth force. Philadelphia, PA: Brunner/
Mazel.
Pedersen, P. B. (2000). A handbook for developing multicultural awareness (3rd ed.).
Alexandria, VA: American Counseling Association.
Pedersen, P. B. (2004). 110 experiences for multicultural learning. Washington, DC:
American Psychological Association.
Pedersen, P. B. (2008). Ethics, competence, and professional issues in cross-cultural
counseling. In P. B. Pedersen, J. G. Draguns, W. J. Lonner, & J. E. Trimble (Eds.),
Counseling across cultures (6th ed., pp. 5–20). Thousand Oaks, CA: Sage. http://
[Link]/10.4135/9781483329314.n1
Pedersen, P. B., Draguns, J. G., & Lonner, W. J. (Eds.). (1976). Counseling across
cultures. Honolulu: University of Hawaii Press.

references      279
Pedersen, P. B., Draguns, J. G., Lonner, W. J., & Trimble, J. E. (Eds.). (2008). Coun-
seling across cultures (6th ed.). Thousand Oaks, CA: Sage. [Link]
10.4135/9781483329314
Pedrotti, J. T., Edwards, L. M., & López, S. J. (2009). Positive psychology within a
cultural context. In S. J. López & C. R. Snyder (Eds.), Oxford handbook of posi-
tive psychology (2nd ed., pp. 49–57). New York, NY: Oxford University Press.
Peters, J. L., Sutton, A. J., Jones, D. R., Abrams, K. R., & Rushton, L. (2006). Com-
parison of two methods to detect publication bias in meta-analysis. JAMA, 295,
676–680. [Link]
Peterson, C., Park, N., & Sweeney, P. J. (2008). Group well-being: Morale from a
positive psychology perspective. Applied Psychology, 57, 19–36. [Link]
org/10.1111/j.1464-0597.2008.00352.x
Phinney, J. S. (1990). Ethnic identity in adolescents and adults: Review of research. Psy-
chological Bulletin, 108, 499–514. [Link]
Phinney, J. S. (1991). Ethnic identity and self-esteem: A review and integration.
Hispanic Journal of Behavioral Sciences, 13, 193–208. [Link]
07399863910132005
Phinney, J. S. (1992). The multigroup ethnic identity measure: A new scale for use
with diverse groups. Journal of Adolescent Research, 7, 156–176. [Link]
org/10.1177/074355489272003
Phinney, J. S. (1996). When we talk about American ethnic groups, what do we
mean? American Psychologist, 51, 918–927. [Link]
066X.51.9.918
Phinney, J. S. (2003). Ethnic identity and acculturation. In K. M. Chun, P. B. Organista,
& G. Marín (Eds.), Acculturation: Advances in theory, measurement, and applied
research (pp. 63–81). Washington, DC: American Psychological Association.
[Link]
Phinney, J. S. (2006). Ethnic identity exploration in emerging adulthood. In J. J. Arnett
& J. L. Tanner (Eds.), Emerging adults in America: Coming of age in the 21st century
(pp. 117–134). Washington, DC: American Psychological Association.
Phinney, J. S., Cantu, C. L., & Kurtz, D. A. (1997). Ethnic and American iden-
tity as predictors of self-esteem among African American, Latino, and White
adolescents. Journal of Youth and Adolescence, 26, 165–185. [Link]
10.1023/A:1024500514834
Phinney, J. S., Horenczyk, G., Liebkind, K., & Vedder, P. (2001). Ethnic identity, immi-
gration, and well-being: An interactional perspective. Journal of Social Issues, 57,
493–510. [Link]
Phinney, J. S., & Ong, A. (2007). Conceptualization and measurement of ethnic
identity: Current status and future directions. Journal of Counseling Psychology,
54, 271–281. [Link]
Pieterse, A. L. (2009). Teaching antiracism in counselor training: Reflections on a
course. Journal of Multicultural Counseling and Development, 37, 141–152. http://
[Link]/10.1002/j.2161-1912.2009.tb00098.x

280       references

Pieterse, A. L., Evans, S. A., Risner-Butner, A., Collins, N. M., & Mason, L. B.
(2009). Multicultural competence and social justice training in counseling psy-
chology and counselor education: A review and analysis of a sample of multi-
cultural course syllabi. The Counseling Psychologist, 37, 93–115. [Link]
org/10.1177/0011000008319986
Pieterse, A. L., Todd, N. R., Neville, H. A., & Carter, R. T. (2012). Perceived racism and
mental health among Black American adults: A meta-analytic review. Journal
of Counseling Psychology, 59, 1–9. [Link]
Pinquart, M., & Sörensen, S. (2000). Influences of socioeconomic status, social net-
work, and competence on subjective well-being in later life: A meta-analysis.
Psychology and Aging, 15, 187–224. [Link]
Piselli, A., Halgin, R. P., & MacEwan, G. H. (2011). What went wrong? Therapists’
reflections on their role in premature termination. Psychotherapy Research, 21,
400–415. [Link]
Ponterotto, J. G. (1997). Multicultural counseling training: A competency
model and national survey. In D. B. Pope-Davis, & H. L. K. Coleman (Eds.),
Multicultural counseling competencies: Assessment, education and training,
and supervision (pp. 111–130). Thousand Oaks, CA: Sage. [Link]
10.4135/9781452232072.n5
Ponterotto, J. G. (1998). Charting a course for research in multicultural counsel-
ing training. The Counseling Psychologist, 26, 43–68. [Link]
0011000098261004
Ponterotto, J. G. (2010). Qualitative research in multicultural psychology: Philo-
sophical underpinnings, popular approaches, and ethical considerations. Cultural
Diversity and Ethnic Minority Psychology, 16, 581–589. [Link]
a0012051
Ponterotto, J. G., & Austin, R. (2005). Emerging approaches to training psycholo-
gists to be culturally competent. In R. T. Carter (Ed.), Handbook of racial–cultural
psychology and counseling: Vol. 2. Training and practice (pp. 19–35). Hoboken, NJ:
Wiley.
Ponterotto, J. G., Casas, J. M., Suzuki, L. A., & Alexander, C. M. (Eds.). (2010).
Handbook of multicultural counseling (3rd ed.). Thousand Oaks, CA: Sage.
Ponterotto, J. G., Fuertes, J. N., & Chen, E. C. (2000). Models of multicultural counsel-
ing. In R. W. Lent (Ed.), Handbook of counseling psychology (3rd ed., pp. 639–669).
Hoboken, NJ: Wiley.
Ponterotto, J. G., & Park-Taylor, J. (2007). Racial and ethnic identity theory, measure-
ment, and research in counseling psychology: Present status and future directions.
Journal of Counseling Psychology, 54, 282–294. [Link]
0167.54.3.282
Pope-Davis, D. B., Coleman, H. L. K., Liu, W. M., & Toporek, R. L. (2003). Handbook
of multicultural competencies in counseling and psychology. Thousand Oaks, CA:
Sage.

references      281
Pope-Davis, D. B., Liu, W. M., Toporek, R. L., & Brittan-Powell, C. S. (2001). What’s
missing from multicultural competency research: Review, introspection, and
recommendations. Cultural Diversity and Ethnic Minority Psychology, 7, 121–138.
[Link]
Presnell, A., Harris, G., & Scogin, F. (2012). Therapist and client race/ethnicity
match: An examination of treatment outcome and process with rural older
adults in the deep south. Psychotherapy Research, 22, 458–463. [Link]
/10.1080/10503307.2012.673022
Priester, P. E., Jones, J. E., Jackson-Bailey, C. M., Jana-Masri, A., Jordan, E., & Metz,
A. J. (2008). An analysis of content and instructional strategies in multicultural
counseling courses. Journal of Multicultural Counseling and Development, 36, 29–39.
[Link]
Proulx, C. M., Helms, H. M., & Buehler, C. (2007). Marital quality and personal
well-being: A meta-analysis. Journal of Marriage and Family, 69, 576–593. http://
[Link]/10.1111/j.1741-3737.2007.00393.x
Ramos, Z., & Alegría, M. (2014). Cultural adaptation and health literacy refinement
of a brief depression intervention for Latinos in a low-resource setting. Cultural
Diversity and Ethnic Minority Psychology, 20, 293–301. [Link]
a0035021
Rawls, J. (1999). A theory of justice. Cambridge, MA: Harvard University Press.
Reis, B. F., & Brown, L. G. (1999). Reducing psychotherapy dropouts: Maximizing
perspective convergence in the psychotherapy dyad. Psychotherapy: Theory,
Research, Practice, Training, 36, 123–136. [Link]
Reynolds, A. L. (1997). Using the multicultural change intervention matrix (MCIM)
as a multicultural counseling training model. In D. B. Pope-Davis & H. L. K.
Coleman (Eds.), Multicultural counseling competencies: Assessment, education
and training, and supervision (pp. 209–226). Thousand Oaks, CA: Sage. http://
[Link]/10.4135/9781452232072.n9
Reynolds, A. L. (2011). Understanding the perceptions and experiences of faculty
who teach multicultural counseling courses: An exploratory study. Training
and Education in Professional Psychology, 5, 167–174. [Link]
a0024613
Richards, G. (2012). “Race,” racism and psychology: Towards a reflexive history (2nd ed.).
New York, NY: Routledge/Taylor & Francis.
Richardson, F. C., Fowers, B. J., & Guignon, C. B. (1999). Re-envisioning psychology:
Moral dimensions of theory and practice. San Francisco, CA: Jossey-Bass.
Richman, J. A., Gaviria, M., Flaherty, J. A., Birz, S., & Wintrob, R. M. (1987). The
process of acculturation: Theoretical perspectives and an empirical investigation
in Peru. Social Science & Medicine, 25, 839–847. [Link]
9536(87)90042-6
Ridley, C. R., Espelage, D. L., & Rubinstein, K. J. (1997). Course development
in multicultural counseling. In D. B. Pope-Davis & H. L. K. Coleman (Eds.),
Multicultural counseling competencies: Assessment, education and training, and

282       references

 supervision (pp. 131–158). Thousand Oaks, CA: Sage. [Link]
9781452232072.n6
Ridley, C. R., Mendoza, D. W., & Kanitz, B. E. (1994). Multicultural training:
Reexamination, operationalization, and integration. The Counseling Psychologist,
22, 227–289. [Link]
Ridley, C. R., Mollen, D., & Kelly, S. M. (2011a). Beyond microskills: Toward a model
of counseling competence. The Counseling Psychologist, 39, 825–864. [Link]
org/10.1177/0011000010378440
Ridley, C. R., Mollen, D., & Kelly, S. M. (2011b). Counseling competence: Appli-
cation and implications of a model. The Counseling Psychologist, 39, 865–886.
[Link]
Ridley, C. R., & Shaw-Ridley, M. (2011). Multicultural counseling competencies:
An analysis of research on clients’ perceptions: Comment on Owen, Leach,
Wampold, and Rodolfa (2011). Journal of Counseling Psychology, 58, 16–21. http://
[Link]/10.1037/a0022221
Rivera, F. I. (2007). Contextualizing the experience of young Latino adults: Accul-
turation, social support, and depression. Journal of Immigrant and Minority Health,
9, 237–244. [Link]
Robichaud, D. (2006). Steps toward a relational view of agency. In E. J. Van Every
(Ed.), Communication as organizing: Empirical and theoretical explorations in the
dynamic of text and conversation (pp. 101–114). Mahwah, NJ: Erlbaum.
Robinson, D. N. (1995). An intellectual history of psychology (3rd ed.). Madison:
University of Wisconsin Press.
Rodríguez, N., Mira, C. B., Paez, N. D., & Myers, H. F. (2007). Exploring the complexi-
ties of familism and acculturation: Central constructs for people of Mexican origin.
American Journal of Community Psychology, 39, 61–77. [Link]
s10464-007-9090-7
Rogers, M. R., & O’Bryon, E. C. (2014). Multicultural training models and curric-
ulum. In F. T. L. Leong, L. Comas-Díaz, G. C. Nagayama Hall, V. C. McLoyd,
& J. E. Trimble (Eds.), APA handbook of multicultural psychology: Vol. 2. Appli-
cations and training (pp. 659–679). Washington, DC: American Psychological
Association.
Rogers-Sirin, L. (2008). Approaches to multicultural training for professionals: A
guide for choosing an appropriate program. Professional Psychology: Research and
Practice, 39, 313–319. [Link]
Rooney, S. C., Flores, L. Y., & Mercier, C. A. (1998). Making multicultural educa-
tion effective for everyone. The Counseling Psychologist, 26, 22–32. [Link]
org/10.1177/0011000098261002
Root, M. P. P. (1994, Summer). Reasons racially mixed persons identify as people of
color. Focus: Notes from the Psychological Study of Ethnic Minority Issues, 8, 14–16.
Root, M. P. P. (2000). Rethinking racial identity development. In P. Spickard & W. J.
Burroughs (Eds.), We are a people: Narrative and multiplicity in constructing ethnic
identity (pp. 205–220). Philadelphia, PA: Temple University Press.

references      283
Rosenstock, I. M. (1966). Why people use health services. The Milbank Memorial
Fund Quarterly, 44, 94–127. [Link]
Rosenthal, R. (1991). Meta-analysis: A review. Psychosomatic Medicine, 53, 247–271.
[Link]
Rudmin, F. W. (2003). Critical history of the acculturation psychology of assimila-
tion, separation, integration, and marginalization. Review of General Psychology,
7, 3–37. [Link]
Rudmin, F. W. (2009). Constructs, measurements and models of acculturation and
acculturative stress. International Journal of Intercultural Relations, 33, 106–123.
[Link]
Rudmin, F. W., & Ahmadzadeh, V. (2001). Psychometric critique of acculturation
psychology: The case of Iranian migrants in Norway. Scandinavian Journal of
Psychology, 42, 41–56. [Link]
Ruglass, L. M., Hien, D. A., Hu, M. C., Campbell, A. N., Caldeira, N. A., Miele,
G. M., & Chang, D. F. (2014). Racial/ethnic match and treatment outcomes
for women with PTSD and substance use disorders receiving community-
based treatment. Community Mental Health Journal, 50, 811–822. [Link]
org/10.1007/s10597-014-9732-9
Ryff, C. D., & Keyes, C. L. M. (1995). The structure of psychological well-being
revisited. Journal of Personality and Social Psychology, 69, 719–727. [Link]
org/10.1037/0022-3514.69.4.719
Salzman, M. (2001). Cultural trauma and recovery: Perspectives from terror man-
agement theory. Trauma, Violence, & Abuse, 2, 172–191. [Link]
10.1177/1524838001002002005
Sarason, S. B. (1974). The psychological sense of community: Prospects for a community
psychology. Oxford, England: Jossey-Bass.
Schmitt, M. T., Branscombe, N. R., Postmes, T., & Garcia, A. (2014). The con-
sequences of perceived discrimination for psychological well-being: A meta-
analytic review. Psychological Bulletin, 140, 921–948. [Link]
a0035754
Schnittker, J., & McLeod, J. D. (2005). The social psychology of health dispari-
ties. Annual Review of Sociology, 31, 75–103. [Link]
soc.30.012703.110622
Schwartz, S. J., Pantin, H., Sullivan, S., Prado, G., & Szapocznik, J. (2006). Nativity
and years in the receiving culture as markers of acculturation in ethnic enclaves.
Journal of Cross-Cultural Psychology, 37, 345–353. [Link]
0022022106286928
Schwartz, S. J., Unger, J. B., Zamboanga, B. L., & Szapocznik, J. (2010). Rethinking
the concept of acculturation: Implications for theory and research. American
Psychologist, 65, 237–251. [Link]
Seligman, M. E. P. (2011). Flourish: A visionary new understanding of happiness and
well-being. New York, NY: Free Press.

284       references

Sellers, R. M., & Shelton, J. N. (2003). The role of racial identity in perceived
racial discrimination. Journal of Personality and Social Psychology, 84, 1079–1092.
[Link]
Sevig, T., & Etzkorn, J. (2001). Transformative training: A year-long multicultural
counseling seminar for graduate students. Journal of Multicultural Counseling and
Development, 29, 57–72. [Link]
Sharf, J., Primavera, L. H., & Diener, M. J. (2010). Dropout and therapeutic alli-
ance: A meta-analysis of adult individual psychotherapy. Psychotherapy: Theory,
Research, Practice, Training, 47, 637–645. [Link]
Sheu, H. B., & Lent, R. W. (2007). Development and initial validation of the
Multicultural Counseling Self-Efficacy Scale–Racial Diversity Form. Psycho-
therapy: Theory, Research, & Practice, 44, 30–45. [Link]
3204.44.1.30
Shin, S., Chow, C., Camacho-Gonsalves, T., Levy, R. J., Allen, I. E., & Leff, H. S.
(2005). A meta-analytic review of racial–ethnic matching for African American
and Caucasian American clients and clinicians. Journal of Counseling Psychology,
52, 45–56. [Link]
Simons, H. W., Berkowitz, N. N., & Moyer, R. J. (1970). Similarity, credibility, and
attitude change: A review and a theory. Psychological Bulletin, 73, 1–16. http://
[Link]/10.1037/h0028429
Simpson, J. A., & Weiner, E. S. (1989). The Oxford English dictionary (2nd ed., Vol. 7).
Oxford, England: Clarendon Press.
Sin, N. L., & Lyubomirsky, S. (2009). Enhancing well-being and alleviating depres-
sive symptoms with positive psychology interventions: A practice-friendly meta-
analysis. Journal of Clinical Psychology, 65, 467–487. [Link]
jclp.20593
Slife, B. D. (2004). Taking practice seriously: Toward a relational ontology. Journal of
Theoretical and Philosophical Psychology, 24, 157–178. [Link]
h0091239
Slife, B. D. (2009). A primer of the values implicit in counseling research methods.
Counseling and Values, 53, 8–21. [Link]
tb00110.x
Slife, B. D., & Wiggins, B. J. (2009). Taking relationship seriously in psychotherapy:
Radical relationality. Journal of Contemporary Psychotherapy, 39, 17–24. http://
[Link]/10.1007/s10879-008-9100-6
Slife, B. D., & Williams, R. (1995). What’s behind the research? Discovering hidden
assumptions in the behavioral sciences. Thousand Oaks, CA: Sage. [Link]
org/10.4135/9781483327372
Smith, T. B. (2010). Culturally congruent practices in counseling and psychother-
apy: A review of research. In J. G. Ponterotto, J. M. Casas, L. A. Suzuki, & C. M.
Alexander (Eds.), Handbook of multicultural counseling (3rd ed., pp. 439–450).
Thousand Oaks, CA: Sage.

references      285
Smith, T. B., Constantine, M. G., Dunn, T. W., Dinehart, J. M., & Montoya,
J. A. (2006). Multicultural education in the mental health professions: A meta-
analytic review. Journal of Counseling Psychology, 53, 132–145. [Link]
org/10.1037/0022-0167.53.1.132
Smith, T. B., & Draper, M. (2004). Understanding individuals in their context:
A relational perspective of multicultural counseling and psychotherapy. In
T. B. Smith (Ed.), Practicing multiculturalism: Affirming diversity in counseling
and psychology (pp. 313–323). Boston, MA: Allyn & Bacon.
Smith, T. B., Richards, P. S., Granley, M., & Obiakor, F. (2004). Practicing multicultur-
alism: An introduction. In T. B. Smith (Ed.), Practicing multiculturalism: Affirming
diversity in counseling and psychology (pp. 3–16). Boston, MA: Allyn & Bacon.
Smith, T. B., Rodríguez, M. D., & Bernal, G. (2011). Culture. Journal of Clinical
Psychology, 67, 166–175. [Link]
Smith, T. B., & Silva, L. (2011). Ethnic identity and personal well-being of people
of color: A meta-analysis. Journal of Counseling Psychology, 58, 42–60. http://
[Link]/10.1037/a0021528
Smith, W. A., Allen, W. R., & Danley, L. L. (2007). “Assume the position . . . you
fit the description”: Psychosocial experiences and racial battle fatigue among
African American male college students. American Behavioral Scientist, 51,
551–578. [Link]
Snowden, L. R. (2012). Health and mental health policies’ role in better understanding
and closing African American–White American disparities in treatment access
and quality of care. American Psychologist, 67, 524–531. [Link]
a0030054
Snowden, L. R., Storey, C., & Clancy, T. (1989). Ethnicity and continuation in treatment
at a Black community mental health center. Journal of Community Psychology,
17, 111–118. [Link]
JCOP2290170202>[Link];2-2
Snowden, L. R., & Yamada, A. M. (2005). Cultural differences in access to care. Annual
Review of Clinical Psychology, 1, 143–166. [Link]
clinpsy.1.102803.143846
Social Science Research Council. (1954). Acculturation: An exploratory formulation.
American Anthropologist, 56, 973–1002.
Soheilian, S. S., Inman, A. G., Klinger, R. S., Isenberg, D. S., & Kulp, L. E. (2014).
Multicultural supervision: Supervisees’ reflections on culturally competent super-
vision. Counselling Psychology Quarterly, 27, 379–392. [Link]
09515070.2014.961408
Spanierman, L. B., Oh, E., Heppner, P. P., Neville, H. A., Mobley, M., Wright,
C. V., . . . Navarro, R. (2011). The multicultural teaching competency scale:
Development and initial validation. Urban Education, 46, 440–464. [Link]
[Link]/10.1177/0042085910377442
Spanierman, L. B., Poteat, V. P., Wang, Y., & Oh, E. (2008). Psychosocial costs
of racism to White counselors: Predicting various dimensions of multicultural

286       references

 counseling competence. Journal of Counseling Psychology, 55, 75–88. http://
[Link]/10.1037/0022-0167.55.1.75
Speight, S. L., & Vera, E. M. (2004). A social justice agenda: Ready, or not? The Coun-
seling Psychologist, 32, 109–118. [Link]
Spindler, L., & Spindler, G. (1967). Male and female adaptations in culture change:
Menominee. In R. Hunt (Ed.), Personalities & cultures (pp. 56–78). New York,
NY: Natural History Press.
Stapel, D. A., & Koomen, W. (2001). I, we, and the effects of others on me: How self-
construal level moderates social comparison effects. Journal of Personality and
Social Psychology, 80, 766–781. [Link]
Statistics Canada. (2011). 2011 census profile. Retrieved from [Link]
[Link]
Steele, C. (2010). Whistling Vivaldi: And other clues to how stereotypes affect us. New York,
NY: Norton.
Steffen, P. R., Smith, T. B., Larson, M., & Butler, L. (2006). Acculturation to Western
society as a risk factor for high blood pressure: A meta-analytic review. Psychosomatic
Medicine, 68, 386–397. [Link]
Steinberg, S. (1981). The ethnic myth: Race, ethnicity, and class in America. New York,
NY: Atheneum.
Stephan, C. W., & Stephan, W. G. (2000). The measurement of racial and ethnic
identity. International Journal of Intercultural Relations, 24, 541–552. [Link]
org/10.1016/S0147-1767(00)00016-X
Sue, D. W. (2005). Racism and the conspiracy of silence. The Counseling Psychologist,
33, 100–114. [Link]
Sue, D. W. (2013). Race talk: The psychology of racial dialogues. American Psychologist,
68, 663–672. [Link]
Sue, D. W. (2015). Race talk and the conspiracy of silence: Understanding and facilitating
difficult dialogues on race. New York, NY: Wiley.
Sue, D. W., Arredondo, P., & McDavis, R. J. (1992a). Multicultural competencies/
standards: A pressing need. Journal of Counseling and Development, 70, 477–486.
Sue, D. W., Arredondo, P., & McDavis, R. J. (1992b). Multicultural counseling com-
petencies and standards: A call to the profession. Journal of Multicultural Coun-
seling and Development, 20, 64–88. [Link]
tb00563.x
Sue, D. W., Bernier, J. E., Durran, A., Feinberg, L., Pedersen, P., Smith, E. J., & Vasquez-
Nuttall, E. (1982). Position paper: Cross-cultural counseling competencies. The
Counseling Psychologist, 10, 45–52. [Link]
Sue, D. W., Bingham, R. P., Porché-Burke, L., & Vasquez, M. (1999). The diversification
of psychology: A multicultural revolution. American Psychologist, 54, 1061–1069.
[Link]
Sue, D. W., Capodilupo, C. M., Torino, G. C., Bucceri, J. M., Holder, A. M. B.,
Nadal, K. L., & Esquilin, M. (2007). Racial microaggressions in everyday life:

references      287
 Implications for clinical practice. American Psychologist, 62, 271–286. http://
[Link]/10.1037/0003-066X.62.4.271
Sue, D. W., & Sue, D. (2013). Counseling the culturally diverse: Theory and practice
(6th ed.). Hoboken, NJ: Wiley.
Sue, S. (1977). Community mental health services to minority groups. Some
optimism, some pessimism. American Psychologist, 32, 616–624. [Link]
org/10.1037/0003-066X.32.8.616
Sue, S. (1988). Psychotherapeutic services for ethnic minorities. Two decades of research
findings. American Psychologist, 43, 301–308. [Link]
066X.43.4.301
Sue, S. (1998). In search of cultural competence in psychotherapy and counseling.
American Psychologist, 53, 440–448. [Link]
Sue, S. (1999). Science, ethnicity, and bias: Where have we gone wrong? American
Psychologist, 54, 1070–1077. [Link]
Sue, S. (2003). In defense of cultural competency in psychotherapy and treatment. Amer-
ican Psychologist, 58, 964–970. [Link]
Sue, S. (2010). Cultural adaptations in treatment. The Scientific Review of Mental
Health Practice, 7(2), 31–32.
Sue, S., Cheng, J. K. Y., Saad, C. S., & Chu, J. P. (2012). Asian American mental
health: A call to action. American Psychologist, 67, 532–544. [Link]
org/10.1037/a0028900
Sue, S., Fujino, D. C., Hu, L., Takeuchi, D. T., & Zane, N. W. S. (1991). Community
mental health services for ethnic minority groups: A test of the cultural respon-
siveness hypothesis. Journal of Consulting and Clinical Psychology, 59, 533–540.
[Link]
Sue, S., & Okazaki, S. (2009). Asian-American educational achievements: A phenom-
enon in search of an explanation. Asian American Journal of Psychology, 1, 45–55.
[Link]
Sue, S., & Zane, N. (1987). The role of culture and cultural techniques in psycho-
therapy. A critique and reformulation. American Psychologist, 42, 37–45. http://
[Link]/10.1037/0003-066X.42.1.37
Sue, S., Zane, N., Levant, R. F., Silverstein, L. B., Brown, L. S., Olkin, R., &
Taliaferro, G. (2006). How well do both evidence-based practices and treat-
ment as usual satisfactorily address the various dimensions of diversity? In
J. C. Norcross, L. E. Beutler, & R. F. Levant (Eds.), Therapy in the real world:
Effective treatments for challenging problems (pp. 329–374). Washington, DC:
American Psychological Association.
Sue, S., Zane, N., Nagayama Hall, G. C., & Berger, L. K. (2009). The case for cultural
competency in psychotherapeutic interventions. Annual Review of Psychology,
60, 525–548. [Link]
Suinn, R. M., Khoo, G., & Ahuna, C. (1995). The Suinn-Lew Asian Self-Identity
Acculturation Scale: Cross-cultural information. Journal of Multicultural Counsel-

288       references

 ing and Development, 23, 139–148. [Link]
tb00269.x
Suls, J., Martin, R., & Wheeler, L. (2002). Social comparison: Why, with whom, and
with what effect? Current Directions in Psychological Science, 11, 159–163. http://
[Link]/10.1111/1467-8721.00191
Sussman, L. K., Robins, L. N., & Earls, F. (1987). Treatment-seeking for depression
by Black and White Americans. Social Science & Medicine, 24, 187–196. http://
[Link]/10.1016/0277-9536(87)90046-3
Swartz, M. S., Wagner, H. R., Swanson, J. W., Burns, B. J., George, L. K., & Padgett,
D. K. (1998). Administrative update: Utilization of services comparing use of
public and private mental health services: The enduring barriers of race and age.
Community Mental Health Journal, 34, 133–144.
Swift, J. K., & Callahan, J. L. (2011). Decreasing treatment dropout by addressing
expectations for treatment length. Psychotherapy Research, 21, 193–200. http://
[Link]/10.1080/10503307.2010.541294
Swift, J. K., Callahan, J. L., & Levine, J. C. (2009). Using clinically significant change
to identify premature termination. Psychotherapy: Theory, Research, Practice,
Training, 46, 328–335. [Link]
Swift, J. K., & Greenberg, R. P. (2012). Premature discontinuation in adult psychother-
apy: A meta-analysis. Journal of Consulting and Clinical Psychology, 80, 547–559.
[Link]
Tajfel, H. (1982). Social identity and intergroup relations. Cambridge, England: Cambridge
University Press.
Tajfel, H. (2010). Social identity and intergroup relations (Vol. 7). Cambridge, England:
Cambridge University Press.
Talebi, H. (2006). An examination of differences in time to treatment, pretreatment symp-
tomatology, and attrition rates in a sample of maltreated European American and Latino
youth (Unpublished doctoral dissertation). University of California, Santa Barbara.
Tesser, A., & Campbell, J. (1982). Self-evaluation maintenance and the percep-
tion of friends and strangers. Journal of Personality, 50, 261–279. [Link]
org/10.1111/j.1467-6494.1982.tb00750.x
Thomas, K. C., & Snowden, L. R. (2001). Minority response to health insurance cov-
erage for mental health services. Journal of Mental Health Policy and Economics, 4,
35–41.
Thompson, C. E., & Neville, H. A. (1999). Racism, mental health, and mental health
practice. The Counseling Psychologist, 27, 155–223. [Link]
0011000099272001
Thompson, R. H. (1989). Theories of ethnicity: A critical appraisal. New York, NY:
Greenwood Press.
Tomlinson-Clarke, S., & Clarke, D. (2010). Culturally focused community-centered
service learning: An international cultural immersion experience. Journal of Multi­
cultural Counseling and Development, 38, 166–175. [Link]
j.2161-1912.2010.tb00124.x

references      289
Toporek, R. (Ed.). (2006). Handbook for social justice in counseling psychology: Leadership,
vision, and action. Thousand Oaks, CA: Sage.
Toporek, R. L., Gerstein, L. H., Fouad, N. A., Roysircar, G., & Israel, T. (Eds.).
(2006). Handbook for social justice in counseling psychology: Leadership, vision, and
action. Thousand Oaks, CA: Sage.
Toporek, R. L., & Vaughn, S. R. (2010). Social justice in the training of professional
psychologists: Moving forward. Training and Education in Professional Psychology,
4, 177–182. [Link]
Torres, L. (2010). Predicting levels of Latino depression: Acculturation, acculturative
stress, and coping. Cultural Diversity and Ethnic Minority Psychology, 16, 256–263.
[Link]
Totikidis, V., & Prilleltensky, I. (2006). Engaging community in a cycle of praxis: Multi­
cultural perspectives on personal, relational, and collective wellness. Commu-
nity, Work & Family, 9, 47–67. [Link]
Tran, N., & Birman, D. (2010). Questioning the model minority: Studies of Asian
American academic performance. Asian American Journal of Psychology, 1,
106–118. [Link]
Triandis, H. C. (1976). Variations in Black and White perceptions of the social envi-
ronment. Oxford, England: University of Illinois Press.
Triandis, H. C., Kashima, Y., Shimada, E., & Villareal, M. (1986). Acculturation indices
as a means of confirming cultural differences. International Journal of Psychology, 21,
43–70. [Link]
Trickett, E. J., & Espino, S. L. (2004). Collaboration and social inquiry: Multiple
meanings of a construct and its role in creating useful and valid knowledge.
American Journal of Community Psychology, 34, 1–69. [Link]
B:AJCP.0000040146.32749.7d
Trimble, J. E. (1988). Multilinearity of acculturation: Person–situation interactions. In
D. Keats, D. Munro, & L. Mann (Eds.), Heterogeneity in cross-cultural psychology
(pp. 173–186). Berwyn, PA: Swets & Zeitlinger.
Trimble, J. E. (1990). Ethnic specification, validation prospects, and the future of
drug use research. International Journal of the Addictions, 25, 149–168.
Trimble, J. E. (1995). Ethnic minorities. In R. Coombs & D. Ziedonis (Eds.),
Handbook on drug abuse prevention: A comprehensive strategy to prevent the
abuse of alcohol and other drugs (pp. 379–410). Needham Heights, MA: Allyn
& Bacon.
Trimble, J. E. (2000). Social psychological perspectives on changing self-identification
among American Indians and Alaska Natives. In R. H. Dana (Ed.), Handbook
of cross-cultural and multicultural personality assessment (pp. 197–222). Mahwah,
NJ: Erlbaum.
Trimble, J. E. (2005). An inquiry into the measurement of racial and ethnic identity.
In R. Carter (Ed.), Handbook of racial–cultural psychology and counseling: Theory
and research (Vol. 1, pp. 320–359). New York, NY: Wiley.

290       references

Trimble, J. E. (2007). Prolegomena for the connotation of construct use in the measure-
ment of ethnic and racial identity. Journal of Counseling Psychology, 54, 247–258.
[Link]
Trimble, J. E., & Bagwell, W. (Eds.). (1995). North American Indians and Alaska Natives:
Abstracts of psychological and behavioral literature, 1967–1995. Washington, DC:
American Psychological Association.
Trimble, J. E., & Bhadra, M. (2013). Ethnic gloss. In K. Keith (Ed.), Encyclopedia
of cross-cultural psychology (pp. 500–504). New York, NY: Wiley. [Link]
org/10.1002/9781118339893.wbeccp204
Trimble, J. E., & Dickson, R. (2005). Ethnic identity. In C. B. Fisher & R. M. Lerner
(Eds.), Applied developmental science: An encyclopedia of research, policies, and pro-
grams. Thousand Oaks: Sage. [Link]
Trimble, J. E., Helms, J. E., & Root, M. P. (2002). Social and psychological perspec-
tives on ethnic and racial identity. In G. Bernal, J. Trimble, K. Burlew, & F. Leong
(Eds.), Handbook of racial and ethnic minority psychology (pp. 239–275). Thousand
Oaks, CA: Sage.
Trimble, J. E., King, J., Morse, G. S., & Thomas, L. R. (2014). North American Indian
and Alaska Native spirituality and psychotherapy. In P. S. Richards (Ed.), Hand-
book of psychotherapy and religious diversity (2nd ed.). Washington, DC: American
Psychological Association.
Trimble, J. E., & Mohatt, G. V. (2006). Coda: The virtuous and responsible researcher
in another culture. In J. E. Trimble & C. B. Fisher (Eds.), Handbook of ethical and
responsible research with ethnocultural populations and communities (pp. 325–334).
Thousand Oaks, CA: Sage.
Trimble, J. E., Scharrón-del-Río, M. R., & Hill, J. S. (2012). Ethical considerations
in the application of cultural adaptation models with ethnocultural popula-
tions. In G. Bernal & M. M. D. Rodríguez (Eds.), Cultural adaptations: Tools for
evidence-based practice with diverse populations (pp. 45–67). Washington, DC:
American Psychological Association. [Link]
Trimble, J. E., & Vaughn, L. (2013). Cultural measurement equivalence. In K. Keith
(Ed.), Encyclopedia of cross-cultural psychology (pp. 313–319). New York, NY: Wiley.
Tsai, J., Ying, Y. W., & Lee, P. A. (2000). The meaning of “being Chinese” and “being
American”. Journal of Cross-Cultural Psychology, 31, 302–332. [Link]
10.1177/0022022100031003002
Tummala-Narra, P. (2009). Teaching on diversity: The mutual influence of students
and instructors. Psychoanalytic Psychology, 26, 322–334. [Link]
a0016444
Umaña-Taylor, A. J. (2004). Ethnic identity and self-esteem: Examining the role of
social context. Journal of Adolescence, 27, 139–146. [Link]
adolescence.2003.11.006
Umaña-Taylor, A. J., & Shin, N. (2007). An examination of ethnic identity and self-
esteem with diverse populations: Exploring validation by ethnicity and geography.
Cultural Diversity & Ethnic Minority Psychology, 13, 178–186.

references      291
Umaña-Taylor, A. J., Vargas-Chanes, D., Garcia, C. D., & Gonzales-Backen, M.
(2008). A longitudinal examination of Latino adolescents’ ethnic identity, coping
with discrimination, and self-esteem. The Journal of Early Adolescence, 28, 16–50.
[Link]
U.S. Census Bureau. (2010). Overview of race and Hispanic origin, 2010. Washington,
DC: Government Printing Office.
U.S. Census Bureau. (2011). Income, poverty, and health insurance coverage in the United
States: 2010. Retrieved from [Link]
U.S. Department of Health and Human Services, Office of the Surgeon General.
(2001). Mental health: Culture, race, and ethnicity—A supplement to mental health:
A report of the Surgeon General. Rockville, MD: Author.
Utsey, S. O., Grange, C., & Allyne, R. (2006). Guidelines for evaluating the racial and
cultural environment of graduate training programs in professional psychology. In
M. G. Constantine & D. W. Sue (Eds.), Addressing racism: Facilitating cultural com-
petence in mental health and educational settings (pp. 213–232). Hoboken, NJ: Wiley.
Valentine, A., DeAngelo, D., Alegría, M., & Cook, B. L. (2014). Translating dispari-
ties research to policy: A qualitative study of state mental health policymakers’
perceptions of mental health care disparities report cards. Psychological Services,
11, 377–387. [Link]
van den Berghe, P. (1981). The ethnic phenomenon. New York, NY: Elsevier.
van Loon, A., van Schaik, A., Dekker, J., & Beekman, A. (2013). Bridging the gap
for ethnic minority adult outpatients with depression and anxiety disorders by
culturally adapted treatments. Journal of Affective Disorders, 147, 9–16. http://
[Link]/10.1016/[Link].2012.12.014
Vasquez, M. J. T. (2012). Psychology and social justice: Why we do what we do.
American Psychologist, 67, 337–346. [Link]
Vera, E. M., & Speight, S. L. (2003). Multicultural competence, social justice, and
counseling psychology: Expanding our roles. The Counseling Psychologist, 31,
253–272. [Link]
Vereen, L. G., Hill, N. R., & McNeal, D. T. (2008). Perceptions of multicultural
counseling competency: Integration of the curricular and the practical. Journal
of Mental Health Counseling, 30, 226–236.
Vinokurov, A., Trickett, E. J., & Birman, D. (2002). Acculturative hassles and immigrant
adolescents: A life-domain assessment for Soviet Jewish refugees. The Journal of
Social Psychology, 142, 425–445. [Link]
Wachtel, P. L. (2008). Relational theory and the practice of psychotherapy. New York,
NY: Guilford Press.
Walker, I., & Smith, H. J. (Eds.). (2001). Relative deprivation: Specification, development,
and integration. West Nyack, NJ: Cambridge University Press. [Link]
10.1017/CBO9780511527753
Waters, M. C. (1990). Ethnic options: Choosing identities in America. Berkeley: University
of California Press.

292       references

Watkins, M. P., & Portney, L. G. (2009). Foundations of clinical research: Applications
to practice (3rd ed.). New York, NY: Pearson/Prentice Hall.
Webb, C. A., DeRubeis, R. J., & Barber, J. P. (2010). Therapist adherence/competence
and treatment outcome: A meta-analytic review. Journal of Consulting and Clinical
Psychology, 78, 200–211. [Link]
Weinrach, S. G., & Thomas, K. R. (2002). A critical analysis of the multicultural
counseling competencies: Implications for the practice of mental health coun-
seling. Journal of Mental Health Counseling, 24, 20–35.
Weinrach, S. G., & Thomas, K. R. (2004). The AMCD multicultural counseling
competencies: A critically flawed initiative. Journal of Mental Health Counseling,
26, 81–93.
Weinreich, P. (1986). The operationalization of identity theory in racial and eth-
nic relations. In J. Rex & D. Mason (Eds.), Theories of race and ethnic relations
(pp. 299–320). Cambridge, England: Cambridge University Press. [Link]
org/10.1017/CBO9780511557828.016
Weinreich, P., & Saunderson, W. (Eds.). (2003). Analyzing identity: Cross-cultural,
societal, and clinical contexts. New York, NY: Routledge.
Weissman, M. M., Markowitz, J. C., & Klerman, G. L. (2000). Comprehensive guide
to interpersonal psychotherapy. New York, NY: Basic Books.
Wendt, D. C., Gone, J., & Nagata, D. (2014). Potentially harmful therapy and multi-
cultural counseling: Bridging two disciplinary discourses. The Counseling Psycholo-
gist. [Link]
Whaley, A. L. (2001). Cultural mistrust and mental health services for African
Americans: A review and meta-analysis. The Counseling Psychologist, 29,
513–531. [Link]
Whitbeck, L. B. (2006). Some guiding assumptions and a theoretical model for
developing culturally specific preventions with Native American people.
Journal of Community Psychology, 34, 183–192. [Link]
20094
Whiting, B. B. (1963). Six cultures: Studies of child rearing. Oxford, England: Wiley.
Wierzbicki, M., & Pekarik, G. (1993). A meta-analysis of psychotherapy dropout.
Professional Psychology: Research and Practice, 24, 190–195. [Link]
10.1037/0735-7028.24.2.190
Williams, D. R., & Mohammed, S. A. (2009). Discrimination and racial disparities
in health: Evidence and needed research. Journal of Behavioral Medicine, 32,
20–47. [Link]
Williams, D. R., Neighbors, H. W., & Jackson, J. S. (2003). Racial/ethnic discrimina-
tion and health: Findings from community studies. American Journal of Public
Health, 93, 200–208. [Link]
Williams, R. N. (1992). The human context of agency. American Psychologist, 47,
752–760. [Link]

references      293
Wong, P. T. P. (2011). Positive psychology 2.0: Towards a balanced interactive model
of the good life. Canadian Psychology/Psychologie Canadienne, 52, 69–81. http://
[Link]/10.1037/a0022511
Worthington, R. L., & Dillon, F. R. (2011). Deconstructing multicultural counsel-
ing competencies research: Comment on Owen, Leach, Wampold, and Rodolfa
(2011). Journal of Counseling Psychology, 58, 10–15. [Link]
a0022177
Worthington, R. L., Hart, J., & Khairallah, T. S. (2010). Counselors as diversity
change agents in higher education. In J. G. Ponterotto, J. M. Casas, L. A.
Suzuki, & C. M. Alexander (Eds.), Handbook of multicultural counseling (3rd ed.,
pp. 563–576). Thousand Oaks, CA: Sage.
Worthington, R. L., Soth-McNett, A. M., & Moreno, M. V. (2007). Multicultural
counseling competencies research: A 20-year content analysis. Journal of Coun-
seling Psychology, 54, 351–361. [Link]
Yap, S. C. Y., Settles, I. H., & Pratt-Hyatt, J. S. (2011). Mediators of the relationship
between racial identity and life satisfaction in a community sample of African
American women and men. Cultural Diversity and Ethnic Minority Psychology, 17,
89–97. [Link]
Yeh, C. J. (2003). Age, acculturation, cultural adjustment, and mental health
symptoms of Chinese, Korean, and Japanese immigrant youths. Cultural Diver-
sity & Ethnic Minority Psychology, 9, 34–48. [Link]
9809.9.1.34
Yeh, C. J., Parham, T. A., Gallardo, M. E., & Trimble, J. E. (2011). Culturally adap-
tive counseling skills: Demonstrations of evidence-based practices. Thousand Oaks,
CA: Sage.
Yeh, M., McCabe, K., Hough, R. L., Lau, A., Fakhry, F., & Garland, A. (2005). Why
bother with beliefs? Examining relationships between race/ethnicity, parental
beliefs about causes of child problems, and mental health service use. Journal of
Consulting and Clinical Psychology, 73, 800–807. [Link]
006X.73.5.800
Yinger, J. M. (1986). Intersecting strands in the theorization of race and ethnic rela-
tions. In J. Rex & D. Mason (Eds.), Theories of race and ethnic relations (pp. 20–41).
Cambridge, England: Cambridge University Press. [Link]
CBO9780511557828.003
Yip, T., Douglass, S., & Sellers, R. M. (2014). Ethnic and racial identity. In
F. T. L. Leong, L. Comas-Díaz, G. C. Nagayama Hall, V. C. McLoyd, & J. E. Trim-
ble (Eds.), APA handbook of multicultural psychology: Vol. 1. Theory and research
(pp. 179–205). Washington, DC: American Psychological Association.
Yoo, H. C., & Pituc, S. T. (2013). Assessments of perceived racial stereotypes,
discrimination, and racism. In K. F. Geisinger, B. A. Bracken, J. F. Carlson,
J. C. Hansen, N. R. Kuncel, S. P. Reise, & M. C. Rodriguez (Eds.), APA handbook
of testing and assessment in psychology: Vol. 2. Testing and assessment in clinical
and counseling psychology (pp. 427–451). Washington, DC: American Psycho-
logical Association. [Link]

294       references

Yoon, E., Chang, C.-T., Kim, S., Clawson, A., Cleary, S. E., Hansen, M., . . . Gomes,
A. M. (2013). A meta-analysis of acculturation/enculturation and mental
health. Journal of Counseling Psychology, 60, 15–30. [Link]
a0030652
Young, D. M., Sanchez, D. T., & Wilton, L. S. (2013). At the crossroads of race:
Racial ambiguity and biracial identification influence psychological essentialist
thinking. Cultural Diversity and Ethnic Minority Psychology, 19, 461–467. http://
[Link]/10.1037/a0032565
Yutrzenka, B. A. (1995). Making a case for training in ethnic and cultural diversity
in increasing treatment efficacy. Journal of Consulting and Clinical Psychology, 63,
197–206. [Link]
Zane, N., & Ku, H. (2014). Effects of ethnic match, gender match, acculturation,
cultural identity, and face concern on self-disclosure in counseling for Asian
Americans. Asian American Journal of Psychology, 5, 66–74. [Link]
org/10.1037/a0036078
Zárate, M. A., Quezada, S. A., Shenberger, J. M., & Lupo, A. K. (2014). Reduc-
ing racism and prejudice. In F. T. L. Leong, L. Comas-Díaz, G. C. Nagayama
Hall, V. C. McLoyd, & J. E. Trimble, (Eds.), APA handbook of multicultural
psychology: Vol. 2. Applications and training (pp. 593–606). Washington, DC:
American Psychological Association. [Link]
Zea, M. C., Asner-Self, K. K., Birman, D., & Buki, L. P. (2003). The abbreviated multi­
dimensional acculturation scale: Empirical validation with two Latino/Latina
samples. Cultural Diversity & Ethnic Minority Psychology, 9, 107–126.

references      295
 INDEX

Accountability, in relational Alaska Natives

paradigm, 227 cultural adaptations to treatment
Acculturation, 9, 145–164 for, 130
current research on, 241 increase in research on, 12
enculturation vs., 147 Alegría, M., 110, 112
existing research literature on, Ali, S. R., 23, 44
153–155 Allen, J., 190–191
factors influencing research findings Alvidrez, J., 72
on, 156–160 American Indian and Alaska Native
future research on, 161–163 Mental Health Research, 6
interpretation of research findings American Psychological Association
on, 160–161 (APA), 6, 21, 25, 52, 127, 143
measurement of, 150–151 American Psychologist, 73
and mental health service Ancis, J. R., 23, 44
utilization, 85–86 Anxiety, 135
need for therapists to consider, 247 APA. See American Psychological
previous research findings on, Association
150–153 Aristotle, 220
recommendations for therapists Aron, A., 220
on, 163 Arredondo, P., 51
relevant theory on, 147–150 Asian American Journal of Psychology, 6
research findings on, 155–156 Asian Americans
Acculturative stress, 145–146 acculturation of, 151
Advocacy conceptualization of mental health
in multicultural psychology problems among, 68
literature, 217 culturally adapted treatments with,
role of, in social justice, 216 140–141
African Americans importance of ethnic identity to,
and ethnic identity, 189, 190 197–198
increase in research on, 11–12 increase in research on, 12
mental health service utilization by, increasing treatment completion
76–81, 83–87, 90–91 by, 127
mistrust of mental health services mental health service utilization by,
by, 70 76–81, 84–87, 94, 248
and racial/ethnic matching of clients and racial/ethnic matching of clients
and therapists, 117, 118, and therapists, 121, 124–126
121, 126 racism studies with, 171, 179
racism studies with, 171, 179 treatment attendance by, 104, 108
treatment attendance by, 104, 108 treatment completion by, 106,
treatment completion by, 106, 109, 109, 241
241, 248 Assumptions
Age categorical, 242
and acculturation, 158 challenging of, 211–212
and culturally adapted treatments, 140 embedded in social justice, 216–219
and ethnic identity, 197 in psychological research, 236
Agency, personal, 221 Austin, R., 21–22

297
Awareness, as multicultural counseling Coleman, M. N., 26
competency, 51 Collectivism
addressed in treatment, 130
Barrett, M. S., 98 in relational paradigm, 226
Barth, F., 184–185 Community, in multicultural
Beauvais, F., 187 psychology, 220
Becoming, process of, 222 Community awareness initiatives, 93
Bernal, M. E., 186 Confucius, 220
Bernier, J. E., 53 Connectedness, in relational
Berry, John W., 148, 149, 213 paradigm, 221
Best practices, 41 Constantine, M. G., 50
Biases, 51, 63 Contextualization, in relational
Binning, K. R., 189 paradigm, 221
Biopsychosocial model, 169 Cookson, John, 146–147
Birz, S., 151 Corne, S., 220
Bond, Michael, 189 The Counseling Psychologist, 51
Brislin, R. W., 22 Counseling skills, 62
Burns, B. J., 70 Critical race theory, 169
Cross-Cultural Counseling Inventory—
Cabassa, L. J., 148–149 Revised (CCCI–R), 58
Cachelin, F. M., 71, 85
Cultural diversity, 7
Carter, Robert T., 45, 172, 184, 186
Cultural Diversity and Ethnic Minority
Categorical assumptions, 242
Psychology, 6
CCCI–R (Cross-Cultural Counseling
Cultural Diversity and Mental Health, 6
Inventory—Revised), 58
Culturally adapted mental health
Census estimates, 89
services, 129–144
Chao, R. C., 24–25, 26
as alternative to traditional
Chen, S., 71
treatments, 241
Cheung, Y. W., 191
conceptual issues relevant to,
Chow, J. C., 71, 72, 93
Clancy, T., 111 131–134
Clarkson, P., 223 existing research literature on,
Client(s) 136–138
access of, to mental health services, future research on, 142–143
8–9 interpretations of findings on,
expectations of, 96–98, 99, 113, 163 141–142
as focus of therapy, 14–15 previous research on, 134–135
language spoken by therapists and, recommendations for therapists
123–124, 126, 127, 132 on, 143
racial heterogeneity of, 89–90 research findings on, 138–141
therapist’s multicultural competencies traditional vs., 72, 247
rated by, 40, 60–62 and treatment completion/
Client outcomes attrition, 112
influenced by therapist’s multicultural Cultural values, 226
counseling competencies, Czopp, A. M., 167
59–60
need for future research on, 238 Depression, 135
Clinical practice, multicultural Diala, C., 93
education in, 43–45 Diener, E., 189, 190, 225
Cohen, J., 76, 79, 88, 108 Diener, M., 189

298       index

Discrimination Even the Rat Was White (Robert
lack of access to mental health Guthrie), 17
services as, 67–68 Evidence-based practices (EBPs),
as subjective, 170 235–248
Draper, M., 224 cultural adaptations to, 130–131
Durran, A., 53 focus on essential dependent
Dynamic sizing, 132 variables in, 236–238
improving research methods for,
EBPs. See Evidence-based practices 241–243
Ecological validity model, 133–134 and meta-analyses, 13
Einstein, Albert, 243 and recommendations for therapists,
Emergency mental health services, 72 246–248
Empirically supported treatments theory development for, 243–246
(ESTs), 13 use of research data in, 238–241
Empowerment Expectations, in treatment, 96–98,
in multicultural psychology 113, 163
literature, 217 Expert opinion, 15
in social justice, 216 External validity, 13–14
through social network
integration, 227 Familismo, 52
Enculturation, 147 Family
including in culturally adapted
Escobar, J., 150
treatment, 130
Espino, S. L., 223
mental health problems addressed
ESTs (empirically supported
in, 69–70
treatments), 13
Feinberg, L., 53
Ethical principles, 223
Festinger, L., 168
Ethnic gloss, 9n1, 201–202
First Nations peoples, 12
Ethnic identity, 181–205
Flaherty, J. A., 151, 186
and acculturation, 151
Flores, L. Y., 24–25
complexity in development of, 14
Freire, P., 219
current research on, 239
existing literature on, 192, 193 Gans, Herbert, 184
factors influencing research findings Garland, A. F., 71
on, 194–200 Gaviria, M., 151
future research on, 201–203 Gaylord-Harden, N. K., 189
interpretation of research findings Gaztambide, D. J., 218–219
on, 200–201 Geertz, Clifford, 181, 191
measurement of, 185–188 George, L. K., 70
need for therapists to consider, 247 Glass, G., 12
origins of constructs of, 182–184 Good, G. E., 24–25
recommendations for therapists Gordon, M. M., 184
on, 204 Graduate classes, on multiculturalism, 23
relevant theory on, 184–185 Grusec, J. E., 147
research findings on, 192, 194 Guthrie, Robert, 17
and well-being, 10, 188–191
Ethnicity, race vs., 183 Handbook of Multicultural Psychology, 167
Ethnic matching of clients and therapists. Harrell, S. P., 169
See Racial/ethnic matching of Harris, P. M., 100, 112
clients and therapists Hastings, P. D., 147

index      299
Health insurance coverage, 71 Journal of Black Psychology, 6
Helms, Janet E., 183, 187 Journal of Cross-Cultural Psychology, 6
Hispanic Journal of Behavioral Sciences, 6
Hispanics and Latinos(as) Kashima, Y., 151
acculturation of, 151 Kazdin, A. E., 111, 130
cultural adaptations to treatment Kelly, S. M., 40
for, 130 Keyes, C. L. M., 189
and ethnic identity, 190 Kiev, Ari, 145
heterogeneity among, 116 King, Martin Luther, Jr., 232
increase in research on, 12 Knight, G. P., 186
mental health service utilization by, Knowledge, as multicultural counseling
76–79, 82, 84–87 competency, 51–52
and racial/ethnic matching of clients Korean Americans, 190
and therapists, 117, 121
treatment attendance by, 108 Ladany, N., 37, 41, 50
treatment completion by, 106, 109 Lambert, M. J., 113, 134
Holism, in relational paradigm, 221 Language
Homophily, 117 spoken by clients and therapists, 123,
Human behavior, theories 126, 127, 132
explaining, 229 use of culturally appropriate, 133,
Human rights, 222–223 143
Huo, Y. J., 189 Latinos(as). See Hispanics and
Latinos(as)
Identity “Letter from Birmingham Jail” (Martin
continual development of, 222 Luther King Jr.), 232
defined, 182 Local referral networks, 93
and development of ethnic López, E. D. S., 190–191, 225
identity, 204
and social interactions, 224 Mandara, J., 189
Identity Structure Analysis (ISA), Mark, A. Y, 167
187–188 Martín-Baró, I. M., 220
Imagined ethnicity, 184 Marx, Karl, 219
Immigration, 146–147 Materialism. See Philosophical
Immigration status, 71 materialism
Individualism MCC(s). See Multicultural counseling
in identity research, 224 competency(-ies)
in mental health services, 11 McLeod, J. D., 169, 170
and social interactions, 226 Mead, Margaret, 209, 213
and subjective well-being, 189 Measurement
valued in Western psychology, 130 of acculturation, 150–151
Instructors, recommendations for, 41–46 of ethnic identity, 185–188
Interactive volition, 221 of racism/prejudice, in studies, 178
Internal validity, 13–14 MEIM (Multigroup Ethnic Identity
International Association for Measure), 187
Cross-Cultural Psychology, 6 Mendenhall, M., 149
Intersectionality, 226, 244 Mendoza, R., 148
ISA (Identity Structure Analysis), Mental health concerns
187–188 multicultural conceptualizations of,
68–69
Jaffee, K., 71 poverty as risk factor for, 71
Jahoda, G., 5 and relational paradigm, 225

300       index

Mental Health: Culture, Race, and Moral sensibility, 222, 231
Ethnicity in 2001, 73 Moreno, M. V., 53
Mental health insurance coverage, 71 Moynihan, D. P., 182
Mental health services Multicultural awareness, 25
client access to, 8–9 Multicultural counseling competency
defined for meta-analyses, 250 (-ies) (MCC[s]), 49–64
in different locations, 72 defined, 21
emergency, 72 evaluation of, 50, 60–62
improving, for minority existing research literature on, 55, 56
populations, 214 factors influencing research findings
mistrust of, among people of color, 70 on, 57–60
multiculturalism in, 5–6 as focus of multicultural education,
private, 72 42–43
public, 72, 86 as foundation of multicultural
public payment of, 90 education, 24
studies on race/ethnicity and, future research on, 60–63
237–238 high expectations for, 43–44
theoretical aims of, 214–215 importance of, 10–11
voluntary vs. involuntary use of, 72 previous research on, 53–54
Mental health service utilization, 8–9, recommendations for therapists on,
67–94 63–64
existing research literature on, 75–78 relevant theory on, 51–53
future research on, 91–92 research findings on, 55, 57, 59–60
interpretation of research findings Multicultural education and training,
on, 88 21–47
issues of interpretation of findings existing research literature on, 28–30
on, 88 factors influencing research findings
participant characteristics on, 31–35
influencing research on, future research on, 39–41
84–87 interpretation of research findings
previous research on, 72–74 on, 35–36
publication bias influencing research previous research on, 23–25
on, 79–80 recommendations for instructors/
recommendations for therapists on, program directors on, 41–46
93–94 recommendations for therapists
relevant theory on, 68–72 on, 38
research findings on, 76, 78–79 relevant theory on, 22–23
secondary analyses of, 89–91 research findings on, 30–31
study characteristics influencing secondary analyses of, 36–38
research on, 80–84 trends in recent research on, 25–26
Meta-analyses methods, 249–252 Multiculturalism as a Fourth Force
data analyses, 252 (Paul Pedersen), 213
data collection processes, 251–252 Multicultural psychology, 3–18,
general eligibility criteria, 249–250 209–234
information sources and search challenges in, 8–11
strategies, 250–251 conceptual analysis of, 232–234
Miami Youth Development Project, 223 defining, 213–214
Molina, L. E., 189 importance of multicultural
Mollen, D., 40, 42 counseling competencies to, 50
Morality, problems in studying, 230 meta-analyses of, 11–15

index      301
Multicultural psychology, continued racism studies, 176
objectives of, 231 studies on racial/ethnic matching
overview of, 4–8 of clients and therapists,
philosophy of, 209–215 122–124
and relational paradigm, 219–232 treatment participation studies,
and social justice, 215–219 106–107
theoretical aims of, 214–215 Participatory action research, 217
Multicultural training. See Multicultural Pascoe, A. E., 177
education and training Paz, O., 234
Multidimensional framework, for Pedersen, Paul B., 3, 6, 22, 53, 213
acculturation, 151 Performance evaluations, 45
Multiethnic individuals, 202 Philosophical materialism
Multigroup Ethnic Identity Measure influence of, on psychology, 218–219
(MEIM), 187 relational paradigm vs., 228
Mutual edification, 221, 227 Philosophy
development of, for multicultural
Naïve transitivity, 219 psychology, 220
National Association for Multicultural and foundations of multicultural
Education, 41 psychology, 210–215
Native American Indians Phinney, Jean S., 183–184, 185, 187,
acculturation issues faced by, 162 190, 191
and ethnic gloss, 9n1 Pieterse, A. L., 177
importance of ethnic identity to, Ponterotto, J. G., 21–22, 39, 53
197–198 Poverty
increase in research on, 12 and access to mental health
Neville, H. A., 169 services, 71
North America, cultural diversity in, 7 increasing access to mental health
services for those in, 214
Observer ratings and mental health, 245–246
of multicultural counseling Power
competencies, 62–63 abuses of, in industrial
of multicultural education revolution, 219
effectiveness, 36–37 and accountability, 227
Oddou, G., 149 inequities in, and mental health,
Oetting, E. R., 187 245–246
Olmedo, E., 150 in multicultural psychology
Other-affirmation, 231 literature, 217
Other-engagement, 222 obfuscation of, 218
Outreach programs, 93 sharing of, and multiculturalism, 43
and social justice, 216
Padgett, D. K., 70, 92, 93 Pratt-Hyatt, J. S., 189
Padilla, A., 151 Prejudice. See Racism and prejudice
Participant characteristics, effects of Private mental health services, 72
acculturation studies, 158–159 Program directors, recommendations
culturally adapted treatment studies, for, 41–46
140–141 Pseudo-ethnicity, 184
ethnic identity studies, 196–199 PsychINFO, 11–12, 237
mental health service utilization Psychological distress
studies, 84–87 ethnic identity as buffer against, 202
multicultural education studies, 33 and racism, 178–179

302       index

Psycho-social conceptualism, Racial identity, 183. See also Ethnic
of racism, 169 identity
Publication bias Racism and prejudice, 167–180
correction of, 41 existing literature on, 172–174
Publication bias, effects of factors influencing research findings
acculturation studies, 160 on, 174–176
culturally adapted treatment future research on, 178–179
studies, 140 interpretations of research findings
ethnic identity studies, 199–200 on, 176–177
mental health service utilization measurement of, in studies, 178
studies, 79–80 need for therapists to consider, 247
multicultural counseling competency previous research on, 169–172
studies, 58–59 recommendations for therapists on
multicultural education studies, handling, 179–180
34–35 relevant theory on, 168–169
racism studies, 174 research findings on, 174
studies on racial/ethnic matching of secondary analyses of, 177
clients and therapists, 124 and well-being, 10
treatment participation studies, 106, Ragsdale, B. L., 189
107–108 Rawls, J., 215–216
Public mental health services, 72, 86 Relational paradigm, 219–232
Public payment, of mental health assumptions and limitations of,
services, 90 228–232
benefits to multicultural psychology
Questions, value of asking, 210, 212 from, 223–228
defining, 220–221
Race, ethnicity vs., 183 philosophical materialism vs., 228
Race-based traumatic stress injury, 170 social justice reconceptualized
Racial battle fatigue, 170 through, 219–223
Racial diversity, 33, 37–38 Relationships
Racial/ethnic matching of clients and importance of, in relational
therapists, 9, 115–128 paradigm, 220
as culturally adapted mental health influence of, on identity, 224
service, 132 and social change, 226
existing research literature on, Relative deprivation theory, 168
119–121 Research data, 15
future research on, 125–126 Research design, 242–243
interpretation of research findings Responsibility to others, 222, 227
on, 124–125 Retrospective surveys, 40–41
previous research on, 117–119 Richards, M. H., 189
recommendations for therapists on, Richman, J. A., 151, 171, 176
126–127 Ridley, C. R., 22–23, 40, 42
relevant theory on, 117 Rights, human, 222–223
research findings across racial groups, Rivkin, I. D., 190–191
121–124 Root, M. P. P., 188
Racial groups Rural areas, mental health services in, 72
differing mental illness rates Ryff, C. D., 189
among, 70
heterogeneity in, 14 Saenz, D. S., 186
socioeconomic differences among, 71 Salzman, M., 163

index      303
Saunderson, W., 188 Social networks
Schmitt, M. T., 176–177 advances in research on, 229
Schnittker, J., 169, 170 empowerment through integration
Self-affirmation, 231 of, 227
Self-concept negative outcomes due to lack of, 221
as a fluid pattern, 222 Social psychology, 117
and identity, 182 Social science, 182–183
Self-esteem, 196, 202, 204 Social Science Research Council, 147
Self-in-relation, 222 Socioeconomic factors
Self-report surveys in access to mental health services, 71
for evaluation of multicultural in treatment completion/
counseling competencies, 50 attrition, 111
in multicultural psychology Sociorace, 183
research, 243 Soth-McNett, A. M., 53
used in multicultural education Standardized approaches,
studies, 40–41 in treatment, 11
validity of, 62 Steele, C., 169
Seligman, M. E. P., 189 Stephan, C. W., 186
Sellers, R. M., 170 Stephan, W. G., 186
Settles, I. H., 189 Stigma, associated with mental health
Shelton, J. N., 170 service utilization, 70, 90–91
Shimada, E., 151 Storey, C., 111
Silva, L., 192, 210, 232 Striegel-Moore, R. H., 71
Simpson, J. A., 182 Study characteristics, effects of
Skill development, 24 acculturation studies, 156–158
Skills, as multicultural counseling culturally adapted treatment studies,
competency, 52–53 140–141
Smart Richman, L., 177 ethnic identity studies, 194–196
Smith, E. J., 53 mental health service utilization
Smith, T. B., 26, 192, 224 studies, 80–84
Snowden, L. R., 71, 111 multicultural education studies,
Social change, 226 31–33
Social comparison theory, 168 racism studies, 174–176
Social identity theory, 186–187, studies on racial/ethnic matching
202, 203 of clients and therapists,
Social interactions 122–124
and identity, 224 treatment participation studies,
and individualism, 226 106–107
Social justice, 215–223 Subjective well-being, 188
assumptions embedded in, 216–219 Suburban areas, mental health services
as central to multicultural in, 72
psychology, 215–216 Sue, D. W., 51, 53
defining, 215–216 Sue, S., 94, 132
and multicultural psychology, Swanson, J. W., 70
215–219 Swartz, M. S., 70
as objective of therapists, 215
real-world applications of, 232 Tajfel, H., 186–187, 202
reconceptualization of, through Teacher education, 41
relational paradigm, 219–223 Theoretical basis, of multicultural
as value, 230 education, 37

304       index

Theory development, research on, Treatment goals, client-generated, 143
243–246 Treatment participation, 8–9, 95–114
Therapeutic alliance existing research literature on,
and effectiveness of treatment, 223 102, 103
as factor in treatment completion/ future research on, 110–112
attrition, 98 interpretation of research findings
Therapists on, 108–110
acculturation recommendations for, need to future research on, 238
163 previous research on, 97–101
cultural awareness of, 51 recommendations for therapists on,
cultural knowledge of, 51–52 112–113
culturally adapted mental health relevant theory on, 96–97
services recommendations research findings across measures of,
for, 143 102, 104–108
cultural resources available to, 134 and therapist’s multicultural
ethnic identity recommendations competence, 100
for, 204 Triandis, H. C., 151
evidence-based practice recommen- Trickett, E. J., 223
dations for, 246–248 Trimble, J. E., 148
implicit biases of, 51, 63 Tripartite model, of multicultural
language spoken by clients and, 123, counseling competencies, 51
126, 127, 132
mental health service utilization Universalism
recommendations for, 93–94 and culturally adapted mental health
multicultural concerns of, 17–18 services, 130–131
multicultural counseling competen- in mental health services, 5
cies of, 52–53, 63–64 in multicultural education, 23
multicultural education of, 38 Unzueta, M. M, 189
racial/ethnic matching of clients and Urban areas, mental health services
therapists, 126–127 in, 72
and racism/prejudice in clients,
179–180 Value(s)
social justice as objective of, 215 promotion of, 231
treatment participation recommen- social justice as, 230
dations for, 112–113 Van den Berghe, Pierre, 185
Tov, W., 190 Vasquez-Nuttall, E., 53
Traditional mental health services Vega, W., 150
addressing cultural factors in, 134 Villareal, M., 151
contextual variables minimized Volition, interactive, 221
in, 130
culturally adapted vs., 72, 247 Wagner, H. R., 70
Trauma, 225 Walzer, A. S., 167
Treatment attendance, 97, 110. See also Waters, M. C., 187
Treatment participation We-consciousness, 222
Treatment completion/attrition, 97–99. Wei, M., 24–25
See also Treatment participation Weiner, E. S., 182
measures of, 111 Weinreich, P., 187–188
and racial/ethnic matching of clients WEIRD (Western, educated,
and therapists, 118 industrialized, rich, and
treatment attendance vs., 110 democratic) populations, 17

index      305
Well-being Wong, P. T. P., 190
and ethnic identity, 10, 188–191, 198 Workplace programs, for multicultural
and racism, 10, 170, 176–177 education and training, 40
and relational paradigm, 225 Worldviews
subjective, 188 and acculturation, 146
White Cloud Journal of American Indian/ culturally adapted treatments
Alaska Native Mental Health, 6 considering, 132–133, 143, 241
White/European Americans and and racial/ethnic matching of clients
Canadians and therapists, 116, 125–126
and ethnic identity, 190 in relational paradigm, 226
mental service utilization by, 76–79 Worthington, R. L., 53
as normative reference group, 74, 101
and racial/ethnic matching of clients Yap, S. C. Y., 189
and therapists, 127 Yip, T., 186–187, 188
Wintrob, R. M., 151 Yoon, Eunju, 152

306       index

 ABOUT THE AUTHORS

Timothy B. Smith, PhD, is a professor and department chair of the

Department of Counseling Psychology at Brigham Young University in
Provo, Utah. His research on multicultural psychology, spirituality, and
quality relationships has received several national awards, including from
Division 45 (Society for the Psychological Study of Culture, Ethnicity
and Race) of the American Psychological Association (APA). He is a
Fulbright Scholar and a Fellow of APA (Division 17, Society of Counseling
Psychology).

Joseph E. Trimble, PhD, is a Distinguished University Professor and professor

of psychology at Western Washington University and a President’s Professor
at the Center for Alaska Native Health Research at the University of Alaska
Fairbanks. He has written over 140 publications on multicultural topics in
psychology, including 20 books. His excellence in teaching and research
awards for his work in the field of multicultural psychology include the Janet E.
Helms Award for Mentoring and Scholarship in Professional Psychology;
the Distinguished Elder Award from the National Multicultural Conference
and Summit; the Henry Tomes Award for Distinguished Contributions to the

307
Advancement of Ethnic Minority Psychology; the International Lifetime
Achievement Award for Multicultural and Diversity Counseling awarded by
the University of Toronto’s Ontario Institute for Studies in Education; the
2013 Francis J. Bonner, MD Award from the Department of Psychiatry at
Massachusetts General Hospital; and the 2013 Elizabeth Hurlock Beckman
Award.

308       about the authors