Scandinavian Journal of Psychology, 2002, 43, 335–351

Outcomes for children with lesbian or gay parents.

Blackwell Science Ltd

A review of studies from 1978 to 2000

NORMAN ANDERSSEN, CHRISTINE AMLIE and ERLING ANDRÉ YTTERØY
Department of Psychosocial Science, University of Bergen, Norway

Anderssen, N., Amlie, C. & Ytterøy, E. A. (2002). Outcomes for children with lesbian or gay parents. A review of studies from 1978 to 2000.
Scandinavian Journal of Psychology, 43, 335 – 351.
Twenty-three empirical studies published between 1978 and 2000 on nonclinical children raised by lesbian mothers or gay fathers were reviewed
(one Belgian/Dutch, one Danish, three British, and 18 North American). Twenty reported on offspring of lesbian mothers, and three on offspring
of gay fathers. The studies encompassed a total of 615 offspring (age range 1.5 – 44 years) of lesbian mothers or gay fathers and 387 controls,
who were assessed by psychological tests, questionnaires or interviews. Seven types of outcomes were found to be typical: emotional functioning,
sexual preference, stigmatization, gender role behavior, behavioral adjustment, gender identity, and cognitive functioning. Children raised by
lesbian mothers or gay fathers did not systematically differ from other children on any of the outcomes. The studies indicate that children
raised by lesbian women do not experience adverse outcomes compared with other children. The same holds for children raised by gay men,
but more studies should be done.
Key words: Children, lesbians, gays, psychological outcomes, review.
Norman Anderssen, University of Bergen, Department of Psychosocial Science, Christiesgt.12, N-5015 Bergen, Norway. E-mail:
[email protected]

INTRODUCTION counseling, and providing therapy, and it is important that

this knowledge be valid. Further, a sound knowledge base
In the Scandinavian (Denmark, Norway, and Sweden) and in this area is warranted given the increasing demand for
other Western industrialized countries more and more chil- counseling regarding custody questions, adoption, and foster
dren live in families where the mother or the father or both care issues. Theoretically, research on outcomes for children
openly identify themselves as lesbian or gay. A body of raised by their lesbian or gay parents may shed light on
research has now assessed the outcomes for these children, prominent theories of individual differences in personal
and the present paper reviews empirical works on children and social development, for example concerning various
raised by lesbian mothers or gay fathers published between aspects of gender development and the development of sex-
1978 and 2000. ual preferences (see Golombok, 1999, for a review). The
The prevalence of children living with a lesbian or gay importance of having both mother and father as role fig-
parent, whether openly lesbian /gay or not, runs into several ures (social learning theory) or identification figures (psycho-
thousands in each Scandinavian country, however it is dynamic perspectives) may be explored by looking at children
estimated (see Strommen, 1989, and Bozett, 1987, for ex- who do not have both a mother and a father (Patterson,
amples of estimates in the USA). In a survey of 2983 lesbian 1992).
women and gay men in Norway, 10% reported having Systematic research on children of lesbian and gay parents
children, and 5% reported living with their children now began to appear in major professional journals in 1978 (Pat-
(Hegna, Kristiansen & Moseng, 1999). terson, 1992). But the total body of empirical evidence is
The Scandinavian courts do not have any history of ruling difficult to review because of the variety of outcome themes
against lesbian mothers and gay fathers in custody cases. and research methods, and because there is often more than
However, the legal recognition of lesbian or gay couples in one published report from the same study. Further, there
these countries does not include the right to be assessed as have been various interpretations of specific studies and of
suitable for adopting children, which is available to hetero- the studies in general. Some reviews on the topic concluded
sexual couples. Therefore, in the current debates of family that there is no harm done to these children. Patterson
policy, including adoption of children by lesbian women or concluded from her review from the early 1990s that “There
gay men, evidence of the psychological outcome for children is no evidence to suggest that psychosocial development
raised by lesbian women and gay men is needed. among children of gay men or lesbians is compromised in
The outcomes for children raised by lesbian mothers any respect relative to that among offspring of heterosexual
or gay fathers are of specific interest to psychologists for parents” (1992, p. 1036). Goodman, Emery and Haugaard
practical and theoretical reasons. Practically, psychologists (1998) stated in the authoritative Handbook of child psycho-
administer and promote specific knowledge through institu- logy that “the results of social science research raise no
tionalized practices such as teaching, writing textbooks, concerns about the development of children raised in

© 2002 The Scandinavian Psychological Associations. Published by Blackwell Publishers, 108 Cowley Road, Oxford OX4 1JF, UK and
350 Main Street, Malden, MA 02148, USA. ISSN 0036-5564.
 14679450, 2002, 4, Downloaded from https://onlinelibrary.wiley.com/doi/10.1111/1467-9450.00302 by Szkola Wyzsza Psychologii Spolecznej, Wiley Online Library on [04/10/2023]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
336 N. Anderssen et al. Scand J Psychol 43 (2002)

lesbian-headed households” (p. 849). Allen and Burrell (1996) Selection criteria
found in their metaanalysis no statistically significant differ- To be included in the review, the material had to be published in an
ences between children of lesbian and gay parents on the available journal or book and based on empirical data collected from
topics they analyzed: emotional well-being and sexual ori- nonclinical samples of children raised by one or two lesbian or gay
entation of the child. Brewaeys and Hall (1997) found no parents, with or without proxy information from parents and teachers,
with or without control groups, with or without children born in a
investigations that could identify adverse effects of lesbian setting of heterosexual marriage or cohabitation (with later change in
motherhood on child development, and reviews published in parental lifestyles), and recruited through self-identified lesbian or gay
Scandinavia conclude similarly (Ernulf & Innala, 1991; parents. Excluded from the review were reports with limited circulation,
Halvorsen & Joner, 1999). However, other reviewers con- such as master and doctoral theses and conference proceedings.
cluded that there is evidence to believe that there is harm, We searched the most common databases within psychology and
related disciplines as of 1 February 2000 with no time limitations
or not enough evidence to reach any conclusion. Belcastro, backwards. The searches were based on the following terms in Eng-
Gramlich, Nicholson, Price and Wilson (1993) concluded in lish: “lesbian”, “gay”, “homosexual(-ity)”, “parents”, “mother(s)”,
their review that the research database (as of 1993) did not “father(s)”, “children”, “families”, and “couples” in the English lan-
contain data of sufficient quality to be able to conclude guage databases Psychlit, Medline, Eric, Article-first, Isi, PapersFirst,
whether children raised by lesbian mothers differ significantly and Proceedings. The literature databases in the other languages
were Danske tidsskrift- og avisartikler and Dansk nationalbiblio-
from children raised by heterosexual mothers. In their grafi (Danish), Bibliographie National Francais (French), Norart:
review Cameron and Cameron (1997) held that the available Norske tidsskriftartikler and Bibsys (Norwegian), and Libris ( Swedish).
empirical evidence regarding sexual orientation and inter- In the non-English searches we utilized fewer and wider terms (e.g.
personal relations in offspring of lesbian mothers or gay “lesbiske” and “bøsser” in Danish).
fathers is far from conclusive. Baumrind (1995) stated that
“Studies to date show few differences among children of
Assessments of the literature
lesbian, gay, and heterosexual couples. Research findings to
date are not definitive” (p. 134). The selected studies were assessed and categorized according to
sample and design, measures and assessments, and findings (see
In sum, there is a need to provide an overview of the Table 1). Categorizations will inevitably simplify and leave out infor-
outcome themes and research methods in this literature. The mation from each study. However, the most important studies are
reviews referred to above cited English-language publica- described in the text. Only offsprings’ and controls’ age, and the
tions, and the studies typically originate in the USA and to type of study (cross-sectional or not) were easy to categorize. To
a lesser extent in Britain. Our review therefore intended to avoid too many “Not reported” in Table 1 we have exercised our
best judgment where the information was not clear but was deduc-
include research from the French and Scandinavian literat- ible (e.g. blind participation).
ure on the subject as well.
The purpose of the present review, thus, was: Sample and design categories. The parents of the children in the
reviewed studies belong to several categories according to various
(1) to describe, categorize, and interpret empirical studies histories of marriage and divorce, cohabitation and living arrangements
on children raised by lesbian mothers or gay fathers; (single or not). This is indicated in Table 1 and partly in the text to
the degree that the information is provided in the reports. Several
(2) to provide an overview of the existing studies (excluding
parts of a research process may be blinded to avoid having research-
multiple reports from the same database). ers, research assistants, or participants consciously or unconsciously
bias the data in any direction. In Table 1 we chose to categorize the
studies according to blinding during three phases of a research project:
METHOD
(1) data collection blind to researcher – whether the researcher
Deciding on the outcomes to review who collected data knew the status (heterosexual, lesbian, gay)
The most common research questions in the field relate to the types of the person from whom he / she collected data;
of outcomes that have been the focus of concern in child custody (2) data scoring blind to researcher – whether the researcher who
cases in the USA and England (Patterson & Redding, 1996; Tasker coded the information knew the status of the person who had
& Golombok, 1997): those labeled as sexual identity issues (gender provided the data;
identity, gender role behavior, and sexual orientation), personal (3) aim of study blind to participants – whether the participant
characteristics other than sexual identity (e.g. psychiatric status, knew the objectives of the study.
self-concept, and intelligence), and social relationships (e.g. peer
relations) (see the review by Patterson, 1992). Here, we organize the In categorizing the measures and assessments in each study we
review according to a modified scheme of these outcomes, regard- tried to stay close to the concepts provided in each study.
less of the motivational base, paradigmatic approach, or origin of
the research questions. We hold that these research questions center Findings categories. The seven findings categories represent to a large
around the following seven outcomes for children, listed according degree what is reported in the studies. “Norms” may mean popu-
to number of studies which assess the outcomes: “emotional func- lation norms (as with the Child Behavior Checklist, used for
tioning”, “sexual preference”, “stigmatization”, “gender role beha- example by Patterson, 1994) or more limited sample norms (as with
vior”, “behavioral adjustment”, “gender identity”, and “cognitive the Harter scale, used by Gershon, Tschann & Jemerin, 1999). In
functioning”. Definitions of the concepts are presented within each other works norms may be based on a trained observer’s experience
of the themes below. (e.g. Green, 1978).

© 2002 The Scandinavian Psychological Associations.

 14679450, 2002, 4, Downloaded from https://onlinelibrary.wiley.com/doi/10.1111/1467-9450.00302 by Szkola Wyzsza Psychologii Spolecznej, Wiley Online Library on [04/10/2023]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
© 2002 The Scandinavian Psychological Associations.

Scand J Psychol 43 (2002)

Table 1. Overview of empirical works on nonclinical offspring raised by lesbian or gay parents, organized alphabetically by authors 1

Sample and design

Number of Aim of Data Data

offspring of study collection scoring
lesbian mothers Number of Ages Sample Study blind to blind to blind to
Work or gay fathers2,3 controls2,3 (years) type type4 participants5 researcher researcher6 Measures and assessments Findings7

Bailey et al., 43 s of gay fa 0 17–43 Convenience Cross- Yes Yes Yes Mailed questionnaires Sexual preference:
1995 (all fa earlier sectional 37 reported to have
mar, 91% sep het preferences
or div today)

Bozett, 1988 19 s and d of 0 14–35 Convenience Cross- No No No Unstructured in-depth Sexual preference:
gay fa (various sectional interviews (grounded 16 reported to have
family histories) theory) het preferences

Brewaeys et al., 30 s and d of 52 s and d 4–8 Register samples Cross- Not Yes Yes Questionnaires to parents: Behavioral adjustment:
1997 lesbian couples of het couples (donor insemination) sectional reported Child Behavior Checklist, No group differences
(from birth) (from birth) and convenience Preschool Activities for sons. Fewer
(donor8) (26 donor and Inventory problems among
26 traditionally daughters of lesbian
conceived) and het (nondonor)
couples
Gender role behavior:
No group differences

Outcomes for children with lesbian or gay parents 337

Chan et al., 55 s and d of 25 s and d of 5– Register sample Cross- No Yes Yes Mailed standardized Emotional functioning:
1998 lesbian couples het couples (mean: (clients of sectional questionnaires to parents No group differences
(from birth) (from birth) 7) sperm bank) and teachers, including: Behavioral adjustment:
and lesbian and het single Child Behavior Checklist, No group differences
single mo (some mo (some earlier Teacher’s Report Form
earlier mar) mar) (all donor)
(all donor)

Flaks et al., 15 s and d 15 s and d 3–8 Convenience Cross- No Partly Yes Standardized questionnaires Emotional functioning:
1995 of lesbian of het couples sectional to parents and teachers, No group differences
couples (from (from birth) including: Child Behavior Behavioral adjustment:
birth) (donor) (traditionally Checklist, Teacher’s Report No group differences
conceived) Form, testing of offspring Cognitive functioning:
with WPPSI-R or WISC-R No group differences

Gershon et al., 76 s and d of 0 11–18 Convenience Cross- No No Yes Standardized questionnaires Emotional functioning:
1999 lesbian mo (67% sectional (interview), including: Harter No differences from
of mo in het Self Perception Profile for norms by Harter
marriage at time Adolescents
of birth)
 14679450, 2002, 4, Downloaded from https://onlinelibrary.wiley.com/doi/10.1111/1467-9450.00302 by Szkola Wyzsza Psychologii Spolecznej, Wiley Online Library on [04/10/2023]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
© 2002 The Scandinavian Psychological Associations.

338
Table 1. Continued

N. Anderssen et al.
Sample and design

Number of Aim of Data Data

offspring of study collection scoring
lesbian mothers Number of Ages Sample Study blind to blind to blind to
Work or gay fathers2,3 controls2,3 (years) type type4 participants5 researcher researcher6 Measures and assessments Findings7

Golombok 37 s and d of 38 s and d 5–17 Convenience Cross- No Partly Yes Structured interviews with Emotional functioning:
et al., 19839 lesbian single of het single sectional mo and with offspring More children with
and nonsingle mo (23/27 (separately); sexual het mo had psychiatric
mo (23/27 mo mo earlier mar) preference assessment symptoms
earlier mar or only for the older part Sexual preference:
cohabiting of sample; standardized No group differences
with a man) questionnaires to mo and Stigmatization:
teachers about offspring No group differences
Gender role behavior:
No group differences
Behavioral adjustment :
No group differences
Gender identity:
No group differences

Golombok 30 s and d of 42 s and d of 3–9 Convenience Cross- Partly Partly Partly Structured interviews and Emotional functioning:
et al., 1997 lesbian mo het single mo sectional questionnaires for mo; No group differences
(from birth) (single since ratings from school teachers; Stigmatization:
(15 single at child’s first year testing of offspring, No group differences
time of data of life) including adaptation of Behavioral adjustment:
collection) Separation Anxiety Test No group differences

Gottman, 1990 35 d of 70 d of 18–44 Not Cross- Not Yes Yes Standardized questionnaires Emotional functioning:
lesbian div mo het div mo reported sectional reported (returned by mail), No group differences
(cohabiting with (35 single, including: Personal on 17 of 18 scales.
another women 35 remarried) Attribute Questionnaire, On well-being scale d
at least some Sexual Orientation Method, of div single mo
point in time) California Psychological indicated more problems
Inventory (18 scales) Sexual preference:
No group differences
Gender role behavior:
No group differences
Gender identity:
No group differences

Scand J Psychol 43 (2002)

Green, 1978 21 s and d 0 5–14 Not Cross- Not No No Structured interviews and Sexual preference:
lesbian reported sectional reported standardized tests, including: 4 of 4 adolescents
nonsingle mo Draw-a-Person test reported het preferences
(all or most div) (sexual preference assessment Stigmatization: Minor
only for the older part incidents of teasing for
of sample) 3 children
Gender role behavior:
No atypical variation
Gender identity: No
atypical variation
 Sample and design

14679450, 2002, 4, Downloaded from https://onlinelibrary.wiley.com/doi/10.1111/1467-9450.00302 by Szkola Wyzsza Psychologii Spolecznej, Wiley Online Library on [04/10/2023]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
Number of Aim of Data Data
offspring of study collection scoring
lesbian mothers Number of Ages Sample Study blind to blind to blind to
Work or gay fathers2,3 controls2,3 (years) type type4 participants5 researcher researcher6 Measures and assessments Findings7

© 2002 The Scandinavian Psychological Associations.

Scand J Psychol 43 (2002)

Green et al., 56 s and d 48 s and d 3–11 Convenience Cross- Not reported Partly Partly Standardized questionnaires Emotional functioning:
198610 of lesbian of nonlesbian, sectional to mo (returned by mail). No group differences
single and single mo Interviews with offspring Stigmatization: No
nonsingle mo (10% never mar) and with mo (separately). group differences
(10% never mar) Testing of offspring, Gender role behavior:
including: WPPSI or No differences for
WISC-R, sexual identity boys, more girls of
tests (Draw-a-Person, lesbian mo preferring
It-Scale for Children), some boy-typical
family relations tests, activities, clothes
self-reported and and future adult roles
mother-reported Gender identity:
peer popularity No group differences
Cognitive functioning:
No group differences

Haack-Møller 13 s and d of 0 14– 31 Convenience Cross- No No No Interviews Sexual preference:

& Møhl, 1984 lesbian mo sectional11 1 of 13 reported
(all in couples homosexual preference
before, some now) Stigmatization: The
children worried
about reactions
from peers, specific
incidents reported

Hoeffer, 1981 20 s and d 20 s and d of 6–9 Not Cross- Not reported No Not Structured interviews Gender role behavior:
of lesbian het, single reported sectional reported with offspring. Testing No group differences
single mo mo (a majority of offspring, including
(a majority sep /div) modified version
sep/div) of Block’s
Toy Preference Test

Outcomes for children with lesbian or gay parents 339

Huggins, 1989 18 s and d of 18 s and d 13–19 Convenience Cross- Not reported Yes Yes Structured interviews. Emotional functioning:
lesbian div mo of het div mo sectional Standardized questionnaire: No group differences
Coopersmith Self-Esteem
Inventory
Javaid, 199312 26 s and d 28 s and d 6–25 Convenience Cross- Not reported No Not Nonstructured clinical Gender role behavior:
of lesbian of het, single sectional reported psychiatric interviews No group differences
nonsingle div mo with offspring and mo
div mo (probably separately)
 14679450, 2002, 4, Downloaded from https://onlinelibrary.wiley.com/doi/10.1111/1467-9450.00302 by Szkola Wyzsza Psychologii Spolecznej, Wiley Online Library on [04/10/2023]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
© 2002 The Scandinavian Psychological Associations.

340
N. Anderssen et al.
Table 1. Continued

Sample and design

Number of Aim of Data Data

offspring of study collection scoring
lesbian mothers Number of Ages Sample Study blind to blind to blind to
Work or gay fathers2,3 controls2,3 (years) type type4 participants5 researcher researcher6 Measures and assessments Findings7

Kirkpatrick 20 s and d 20 s and d 5–12 Convenience Cross- Not reported Yes Yes Semistructured interview Emotional functioning:
et al., 1981 of lesbian of het div, sectional with offspring and with No group differences
div mo single mo mo (separately). Observation Gender identity:
and testing of offspring, No group differences
including: Play-room Cognitive functioning:
observation, WISC, No group differences
Human Figure Drawing

Lewis, 1980 21 s and d 0 9–26 Convenience Cross- No No No In-depth interviews with Stigmatization:
of lesbian sectional children (psychodynamically Children at all ages
nonsingle mo oriented) worried about potential
reactions from peers,
no report of specific
incidents

McCandlish, 7 s and d of 0 1.5–7 Convenience Cross- No No No Structured interviews with Behavioral adjustment:
1987 lesbian couples sectional offspring and mo together. No specific problems
(from birth) (donor) Observations of children Gender identity:
during the interviews No specific problems
(object-relations perspective)

Miller, 1979 14 s and d 0 14–33 Convenience Cross- No No No In-depth interviews Sexual preference:
of gay fa sectional 2 of 14 reported
(various family to be lesbian/gay
histories) Stigmatization:
No specific
incidents reported

Scand J Psychol 43 (2002)

O’Connell, 1993 11 s and d 0 16–23 Convenience Cross- No No No Semistructured interviews Sexual preference:
of lesbian sectional 1 identified as
div mo homosexual
Stigmatization:
Children at
all ages worried
about potential
reactions from
peers, no report
of specific incidents
 Sample and design

14679450, 2002, 4, Downloaded from https://onlinelibrary.wiley.com/doi/10.1111/1467-9450.00302 by Szkola Wyzsza Psychologii Spolecznej, Wiley Online Library on [04/10/2023]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
Number of Aim of Data Data
offspring of study collection scoring
lesbian mothers Number of Ages Sample Study blind to blind to blind to
Work or gay fathers2,3 controls2,3 (years) type type4 participants5 researcher researcher6 Measures and assessments Findings7

© 2002 The Scandinavian Psychological Associations.

Scand J Psychol 43 (2002)

Patterson, 199413 37 s and d of 0 4–9 Convenience Cross- No Partly Partly Standardized questionnaires Emotional functioning:
lesbian mo sectional for mo, including Child No differences on
(26 couples, Behavior Checklist. Child Behavior
7 singles, Standardized questionnaires Checklist. No
4 in joint for children, including differences from
custody Children’s Self-View norms on self-concepts
between Questionnaire. Standard related to aggression,
two mo) open-ended interview social closeness, social
(from birth) of children potency, but more
stress reactions and
higher well-being
Gender role behavior :
No remarkable pattern
Behavioral adjustment:
No differences from
norms

Steckel, 1987 11 s and d of 11 s and d 3–4 Not reported Cross- Not reported No No Structured interview with Emotional functioning:
lesbian couples of het couples sectional parents; Q-sort administered No group differences
to teachers and parents; in degree of
Projective Structured psychopathology
Doll Technique interview or difficulties in
with offspring separation–individuation
processes

Tasker & 25 s and d of 21 s and d of 17–35 Convenience Longitudinal Not reported Partly Partly Baseline details: see Emotional functioning:
Golombok, lesbian mo het mo (19/21 (14 years) Golombok et al. (1983) No group differences
19979,14 follow-up (22/25 by by het couples, above Sexual preference:
of Golombok lesbian couples) these mo Follow-up: Semistructured No group differences,
et al., 1983 no longer single) interviews. Standardized but more variation in
(baseline details questionnaires, including: offspring of lesbian mo

Outcomes for children with lesbian or gay parents 341

above) Trait Anxiety Inventory, Stigmatization: No group
Beck Depression Inventory differences, but a tendency
for children with lesbian
mo to have been teased
more about own sexuality
 14679450, 2002, 4, Downloaded from https://onlinelibrary.wiley.com/doi/10.1111/1467-9450.00302 by Szkola Wyzsza Psychologii Spolecznej, Wiley Online Library on [04/10/2023]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
© 2002 The Scandinavian Psychological Associations.

342
N. Anderssen et al.
Table 1. Continued

Sample and design

Number of Aim of Data Data

offspring of study collection scoring
lesbian mothers Number of Ages Sample Study blind to blind to blind to
Work or gay fathers2,3 controls2,3 (years) type type4 participants5 researcher researcher6 Measures and assessments Findings7

Totals15 61516 38716 1.5–44 Convenience Cross- Yes 1; no, 11; Yes, 6; Yes, 9; Interviews: 16 Emotional functioning:
or not reported, sectional, partly, 1; not no, 11; no, 8; Questionnaires: 11 12 studies
21; register 22; reported, 10 partly, 6; partly, 4; Tests: 7 Sexual preference:
sample, 2 longitudinal, 1 not not Observation: 2 9 studies
reported, 0 reported, 2 Stigmatization: 9 studies
Gender role behavior:
8 studies
Behavioral adjustment:
7 studies
Gender identity: 6 studies
Cognitive functioning:
3 studies

Notes: Abbrievations: s = sons; d = daughters; fa = fathers; mo = mothers; het=heterosexual; mar = married; sep = separated; div = divorced.
1
Only sample sizes, design, measures, and assessments that are relevant and meeting the inclusion criteria for the present review are presented, not necessarily the total array of participants or methods in the study
referred to.
2
The family histories vary considerably. If provided in the study, the table includes information about divorce or not, single status or not, whether lived in lesbian or gay households since birth or not.
3
The number of children is reported (not of parents).
4
By cross-sectional design we mean studies where information is collected at one time point even though the information collected covers a time span (e.g. in studies where persons are asked to give information of
childhood experiences).
5
By participants we mean those who provided information, whether it is offspring, parents, or teachers.
6
Classification in this column is done under the assumption that scoring of questionnaires followed prewritten rules, and accordingly questionnaires are classified as blind to researcher.
7
For purposes of clarity, phrases for the assessed outcomes correspond to the categorization scheme outlined in the present text. Phrases in the original works are not necessarily used.
8
Donor means that the children are conceived through donor insemination.
9
British Longitudinal Study of Lesbian Mother Families.
10
Earlier reports from this study (with fewer subjects and no specific measures reported) seem to be Mandel and Hotvedt (1980) and Hotvedt and Mandel (1982). These two reports are thus not included in the review.
11
This study is a follow-up of a sample of 15 offspring of lesbian mothers who were interviewed and assessed in 1973, published as a psychology thesis at University of Copenhagen (Leick and Nielsen, 1974). The thesis
was not included in this review because of the limited circulation (theses were not included in the review).
12
This study explored stigmatization and sexual preferences as well, but these themes were not included in the review because of imprecise procedures and results descriptions.

Scand J Psychol 43 (2002)

13
Bay Area Families Study.
14
These results are also published in journals, see Golombok and Tasker (1996) and Tasker and Golombok (1995). For purposes of simplicity, we chose to put the full report published in their book in the table. In the
text only the book is referred to.
15
The sums may be higher than total number of studies because several studies utilized combined types of samples, measures, and outcomes.
16
Not included Tasker and Golombok (1997) since this is a follow-up of Golombok et al. (1983).
 14679450, 2002, 4, Downloaded from https://onlinelibrary.wiley.com/doi/10.1111/1467-9450.00302 by Szkola Wyzsza Psychologii Spolecznej, Wiley Online Library on [04/10/2023]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
Scand J Psychol 43 (2002) Outcomes for children with lesbian or gay parents 343

FINDINGS papers and personal networks. Thirty-seven children aged

5–17 years with lesbian mothers were thus assessed and
We found 23 different works described in a large number of compared with 38 children in the same age range of single
journal articles and books. In these works, 615 offspring heterosexual women. The assessments consisted of adapta-
of lesbian mothers or gay fathers were assessed with 387 tions of standardized and validated interviews of the mothers
controls (only baseline numbers are counted in the longitu- and the children (separately), assessing various aspects of
dinal study of Tasker & Golombok, 1997) (see Table 1). personal and family functioning, the child’s history of gender-
The offsprings’ ages varied from 1.5 to 44 years of age. related activities and interests, and friendships. Based on
We found only one empirical study on children raised by transcripts from the interviews, the child’s psychiatric state
lesbian mothers or gay fathers in non-English databases as was rated by an experienced child psychiatrist (blinded to
of February 2000 (Danish). Seventeen of the empirical works the mother’s sexual preference). Further, validated question-
reviewed originated in USA, one partly in USA and partly naires about the child (devised for the Isle of Wight epidemio-
Canada (Miller, 1979), three in England, one in Denmark logical studies) were completed separately by mothers and
(Haack-Møller & Møhl, 1984), and one in Belgium and The teachers assessing the child’s emotions, behavior, and friend-
Netherlands (Brewaeys, Ponjaert, Hall & Golombok, 1997). ships. Several outcomes were compared for the two groups
Of the 23 studies, 20 assessed the offspring of lesbian of children based on the collected information: emotional
mothers, while only three assessed the offspring of gay functioning, behavioral adjustment, sexual preference (with
fathers; 13 had control groups. Two studies drew particip- the pubertal and post-pubertal adolescents), gender role beha-
ants from registers, while all others were either convenience vior, gender identity, and stigmatization. Emotional function-
samples or did not report recruitment procedures. Fourteen ing as assessed in the questionnaires showed no differences
studies reported on adoption status. In these studies 2% of between the offspring groups, whereas the assessments
children of lesbian mothers or gay fathers were adopted based on the mother’s interviews indicated more psychiatric
(8/438) while none were in the control groups (0/259). symptoms in the children of the heterosexual mothers (statist-
With our use of blinding criteria (see above), we see from ically significant).
Table 1 that the study aim was blind for probably only one As can be seen in Table 1, seven other studies assessed
study (may be two), six studies were blinded for the children in the age range 3–12 years regarding emotional
researcher (and six partly blinded), and nine were blinded in functioning (Chan, Raboy & Patterson, 1998; Flaks, Ficher,
coding procedures (and four partly). Masterpasqua & Joseph, 1995; Golombok, Tasker & Mur-
Interviews (more or less structured) were used in 16 stud- ray, 1997; Green, Mandel, Hotvedt, Gray & Smith, 1986;
ies, self-administered questionnaires were used in 11 studies, Kirkpatrick, Smith & Roy, 1981; Patterson, 1994; Steckel,
and testing of offspring in at least seven studies (that is, 1987). All were based in North America except Golombok
exposing offspring to some standardized material other than et al. (1997), which was British. Six of the studies had con-
questionnaires to record responses). Observation was used trol groups, six studies utilized convenience samples, and six
in two studies. had at least some blinding procedures. The measures in these
Emotional functioning was the most often studied out- studies varied considerably, encompassing procedures such
come (12 studies), followed by sexual preference (nine as play-room observation, projective and nonprojective
studies), stigmatization (nine studies), gender role behavior psychological tests, interviews with children and with
(eight studies), behavioral adjustment (seven studies), gender mothers, and questionnaires for teachers and mothers (see
identity (six studies), and cognitive functioning (three studies). Table 1 for details). No differences between the children of
The findings and details from the studies are reviewed below. lesbian mothers and control children were found in any of
A majority of the studies (15) assessed more than one outcome. these studies. Two Californian studies focused specifically
on adolescents and self-esteem. Offspring of lesbian
mothers were not found to differ from offspring of hetero-
(1) Emotional functioning sexual mothers (Huggins, 1989) or from established norms
“Emotional functioning” covers in the present context a (Gershon et al., 1999).
range of phenomena which most generally may be referred Long-term emotional functioning outcomes were assessed
to as the inner life of the participants other than aspects in two studies where participants were the adult offspring
related to sexuality, gender, and cognitive capacities. No of lesbian mothers (and their controls). The study by
indications were reported in the 12 studies that the children Golombok et al. (1983) gives baseline measures in the only
with lesbian mothers had more emotional difficulties than prospective, longitudinal study in the entire field. In the
other children. None of the studies were of offspring of gay follow-up study (in our review listed as a separate study),
fathers. Tasker and Golombok (1997) were able to recruit 25 sons
Golombok, Spencer and Rutter (1983) conducted a study and daughters of lesbian mothers, and 21 sons and daugh-
in Britain where lesbian women and single heterosexual ters of heterosexual mothers, who had all participated in the
women were contacted through gay and single-parent news- baseline 14 years earlier (a follow-up rate of 62%). Outcomes

© 2002 The Scandinavian Psychological Associations.

 14679450, 2002, 4, Downloaded from https://onlinelibrary.wiley.com/doi/10.1111/1467-9450.00302 by Szkola Wyzsza Psychologii Spolecznej, Wiley Online Library on [04/10/2023]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
344 N. Anderssen et al. Scand J Psychol 43 (2002)

assessed at the follow-up were emotional functioning, as peers with heterosexual mothers to identify as lesbian or gay,
measured by the Trait Anxiety Inventory and Beck Depres- or to be attracted to someone of their own gender, “however,
sion Inventory, sexual preferences, and stigmatization (see if they do experience same-gender attraction, they are more
below). No differences between the two groups of offspring likely to pursue a sexual relationship” (p. 132).
were found on the two inventories. Gottman (1990) con- In two studies covering adolescent and adult offspring,
ducted a mail survey among 35 adult daughters of lesbian the groups did not differ on survey measures of sexual ori-
divorced mothers, 35 adult daughters of heterosexual entation (Gottman, 1990) or on interview questions about
divorced mothers who remained single, and 35 adult daugh- romantic crushes and erotic fantasies (Green, 1978; Golom-
ters of heterosexual divorced mothers who remarried/lived bok et al., 1983). In the remaining four studies (three in the
with a man. She reported no differences between the three USA and Canada, one in Denmark) 57 sons and daughters
groups on the measures used (scales from the California of lesbian mothers and gay fathers were interviewed with no
Psychological Inventory). comparison groups (Bozett, 1988; Haack-Møller & Møhl,
1984; Miller, 1979; O’Connell, 1993). Among these, six off-
spring (11%) said they had homosexual preferences (in
(2) Sexual preference Bozett’s study, recorded as “nonheterosexual preferences”).
By “sexual preference” we mean “the individual’s physical
sexual activity with, interpersonal affection for, and erotic
fantasies about members of the same or opposite biological (3) Stigmatization
sex” (after De Cecco, 1981, p. 61, in his definition of sexual Stigmatization, that is, being teased, harassed, or bullied, is
orientation). Sexual preference is one of the outcomes of one of the potential outcomes for children of lesbian mothers
most concern in debates about children growing up with a or gay fathers that worries court judges in the USA most
lesbian mother or gay father. We prefer the term “sexual (Rivera, 1987). Goffman (1963) conceptualizes stigma in
preference” to “sexual orientation”, as does Baumrind (1995), relational terms and defines the one being stigmatized as
to emphasize the nonfixed nature of sexual relations. possessing an attribute that make him or her “reduced in
Of the nine studies examining this outcome, three studied our minds from a whole and usual person to a tainted, dis-
offspring living or having lived with their gay fathers (i.e. all counted one” (p. 12). According to Goffman, persons who
the studies with gay fathers in the present review). None relate to the stigmatized person run the risk of being stigmat-
of the studies reported that sexual preferences in offspring ized as well. Children with lesbian mothers and gay fathers
varied with parental sexual preferences. might run the risk of being stigmatized, since lesbian
Bailey, Bobrow, Wolfe and Mikach (1995) recruited gay women and gay men still represent a stigmatized group of
fathers through advertisements in gay publications in several persons, as indicated by attitude surveys reporting that a
states in the USA. Self-reported sexual preferences from substantial proportion of the adult populations in Norway
their 43 adult sons aged 17– 43 were assessed through and the USA hold negative views of homosexuality and of
questions about sexual orientation (mailed questionnaires lesbians and gays (Statistisk Sentralbyrå, 1997; Herek &
blinded for informants as to the researchers’ specific interest Capitanio, 1996). King and Black (1999) found that in two
in sexual orientation). Thirty-seven of the sons (86%) rated samples of college students in the US Mid-West (a total of
themselves as heterosexual and six as nonheterosexual. 615 students) 15 –19% indicated that they were not willing to
In the longitudinal study of Tasker and Golombok (1997, have a spouse whose mother was a lesbian.
see details above) “sexual orientation” was recorded through We included in this stigmatization category studies that
semistructured interviews. There were no differences between reported on offsprings’ social relations with friends and
the two groups (25 offspring with lesbian mothers and 21 schoolmates. The nine studies that covered the issue of
with heterosexual mothers) on “same-gender sexual attrac- stigmatization of children of lesbian mothers (eight studies)
tions” or “sexual identity”. However, more offspring of les- or gay fathers (one study) found generally that the children
bian mothers reported that they had “considered a lesbian/ were not stigmatized, but they tended to be teased more
gay relationship” as a possibility or still did (14 offspring than their peers.
of lesbian mothers as compared with three offspring of Only one study reported that the children had experienced
heterosexual mothers), and more offspring of lesbian mothers direct negative actions from others due to their parents’
(six) had experienced “same-gender sexual relationships” as alternative choice of sexual partner. In their report from
compared with none among the others. Tasker and Golom- interviews with 13 Danish offspring of lesbian mothers,
bok interpreted these variations as indicating a different Haack-Møller and Møhl (1984) stated that “The relation-
degree of openness in their sample due to the specific family ship to friends have in many instances been problematic,
experiences, offspring of heterosexual mothers being less there have often been direct negative reactions towards the
likely to think of same-gender relationships in terms of children because of their lesbian mother” (p. 317, our trans-
possible sexual relationships. They stated that the offspring lation) (the number of informants with these experiences
of lesbian mothers probably are no more likely than their and the number of instances were not reported). The

© 2002 The Scandinavian Psychological Associations.

 14679450, 2002, 4, Downloaded from https://onlinelibrary.wiley.com/doi/10.1111/1467-9450.00302 by Szkola Wyzsza Psychologii Spolecznej, Wiley Online Library on [04/10/2023]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
Scand J Psychol 43 (2002) Outcomes for children with lesbian or gay parents 345

authors pointed especially to the age period 10 –11 years as In sum, the studies reported few or no incidents of serious
difficult. This was the period when the informants tended to teasing, harassment, and bullying due to having a lesbian
realize that their mother was a lesbian, and they experienced mother or gay father. However, the studies clearly indicate
tension between being loyal to her while being afraid of that the children were concerned about the chance of being
sanctions from friends. One boy reported: “I said that they stigmatized (O’Connell, 1993; Tasker & Golombok, 1997),
had shared [a house] for reasons of economy, because the and the foremost worry was the chance of getting teased
house rent was so high” (p. 316, our translation). about being lesbian or gay oneself. O’Connell reported that
In the 1983 study by Golombok et al. in Britain (see several of her 11 informants experienced shame due to the
details above) “quality of children’s peer relationships” was conflict between the loyalty they felt for the mother (lesbian)
assessed through structured interviews with mothers, and and the need for self-protection, that is, concealing that the
the interview transcripts were later rated blind to the sexual mother was a lesbian.
preference of the mothers. No group differences were In a collection of interviews and stories told by sons and
reported, and the majority in each group were found to be daughters of lesbian women (mainly North American), it
able to make and maintain relationships with people of their was evident that that the interviewed children had invested
own age. In the follow-up of this study 14 years later, the energy to deal with the issue of their friends’ knowing or not
offspring were thoroughly interviewed about “stigmatization knowing that they had a lesbian mother (Rafkin, 1990, not
in school”, the extent of teasing and being bullied, and a research report). Here are two examples. Carey, aged 21,
“integration of family and friends”, about telling friends, said: “Friends would come to my house, and I would run
bringing friends home, and school friends’ response to ahead to check if my mother was home or if she was with
knowledge of their mother being a lesbian (Tasker & her lover” (p. 157). Carl, aged 12, said: “But it is hard some
Golombok, 1997). It was found that there was no higher times. I don’t know what the kids would do if they knew”
prevalence of peer group hostility reported by the offspring (p. 50). Javaid (1993) described “a general attitude of secrecy”
of lesbian mothers. However, more of the male offspring of (p. 243) in her interviews with children of lesbian mothers
lesbian mothers did report teasing about being gay them- (see below).
selves compared with the other males. Further, there was a Bozett (1988), who specifically studied how children
trend for the offspring of lesbian mothers to have been experienced being a child with a gay father, found that chil-
teased more about their mother’s lifestyle than the offspring dren employed various strategies so that they were perceived
of heterosexual mothers. The authors speculated that reports by others as they wanted to be perceived. The strategies
of more frequent teasing of the children of lesbian mothers included setting and controlling the limits for the father’s
may reflect actual occurrences, or may indicate that these expression of homosexuality (“boundary control”), keeping
children recognize and remember such teasing more than it a secret that the father was gay (“nondisclosure”), and
other children. In the study among children aged 3–9 years disclosure to a larger number of people that the father was
(Golombok et al., 1997), “peer acceptance”, a subscale of gay to prepare them to meet the father (“disclosure”).
Pictorial Scale of Perceived Competence and Social Accept- The studies reviewed and the stories referred to may be
ance for Young Children, did not reveal differences between summed up under three points. First, children of lesbian
the children of lesbian and heterosexual mothers in how the mothers or gay fathers generally developed peer relation-
children perceived their own relationships with friends. ships as stable and as good as those of other children.
Interviews of children of lesbian and heterosexual mothers Second, very few were harassed more than other children,
about “peer group relationships” (popularity with same-sex although they were teased somewhat more. Third, they
and other-sex children in school and in neighborhood) and invested energy in other people knowing or not knowing.
the mothers’ ratings of relations between offspring and peers
indicated that children of lesbian mothers did not differ in
peer group relationships (Green et al., 1986). (4) Gender role behavior
In an in-depth psychodynamically oriented study of Gender role behavior means behaviors that are culturally
21 children (aged 9–26) of divorced lesbian mothers in associated with men or with women (after Shively & De
Massachusetts, children were interviewed about their “feelings Cecco, 1977). This definition captures what is meant by the
about their mother’s changed life-style”. The children authors of the seven studies assessing to what degree the
expressed concern about how others might react if they offspring of lesbian mothers (no studies with gay fathers)
knew their mother was a lesbian (Lewis, 1980). No specific deviate from gender role norms. Generally, the studies
incidents of rejection or harassment were, however, reported. found that children with lesbian mothers tended to choose
The same pattern is indicated in the interview studies of the gender-typical activities, toys, and games, much as other
young and adult offspring of gay fathers (Miller, 1979) and children.
lesbian mothers (O’Connell, 1993). In the interview and test- Again, we start with the study by Golombok et al. (1983)
ing study by Green (1978), three of the 21 children reported (see above). To assess “sex role behavior”, they devised a
minor incidents of teasing. 14-item sex role scale used in their structured interviews with

© 2002 The Scandinavian Psychological Associations.

 14679450, 2002, 4, Downloaded from https://onlinelibrary.wiley.com/doi/10.1111/1467-9450.00302 by Szkola Wyzsza Psychologii Spolecznej, Wiley Online Library on [04/10/2023]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
346 N. Anderssen et al. Scand J Psychol 43 (2002)

mothers. The scale was constructed of items about the (playing with trucks, rough-and-tumble play), clothing, and
frequency with which the child had participated in typical future adult roles (as doctor, lawyer, and astronaut).
gender-typed activities (according to earlier studies), such as Patterson (1994), in her standard open-ended interviews
male sports, dressing up, and pretend games like a tea party. of 37 children of lesbian mothers, found no remarkable
A five-item scale was likewise constructed for offspring pattern in “preferences for sex role behavior” (peer friend-
interviews, consisting of questions about favorite activities/ ships, favorite toys, favorite games, and favorite characters).
hobbies, toys and collections, books and comics, sport, and Gottman (1990), surveying adult daughters of lesbian and
television programs. No group differences were found heterosexual mothers, reported no difference in “gender
between the offspring of lesbian and heterosexual mothers, role” as assessed by the Personal Attribute Questionnaire,
for neither boys nor girls, as reported by mothers and as a bipolar masculinity and femininity scale.
reported by offspring.
Hoeffer (1981) compared 20 6–9-year old children of
lesbian, single mothers with 20 children of heterosexual (5) Behavioral adjustment
single mothers in the San Francisco area. “Sex role behavior” The term “adjustment” signals a foundation in prevailing
was assessed by a modified version of Block’s Toy Preference values, which may be defensible or not. For the present
Test and through a structured interview in which the child purpose we use the pragmatic approach and refer to studies
showed and explained eight favorite toys and activities to the that assess overt behaviors. Thus, with the phrase “behavioral
investigator (Toy Selection Interview; the interview data adjustment” we mean the degree to which children behave
were later rated as masculine, feminine, or neutral). No according to expectations about social behavior. Of six stud-
group differences were found. In a study from Belgium and ies that compared the children of lesbian mothers with other
The Netherlands, Brewaeys et al. (1997) compared the gen- children (no studies with gay fathers), none gave indications
der role behavior of 30 children aged 4 – 8 years conceived by of higher prevalence of behavioral problems among children
donor to lesbian couples with 52 children of heterosexual with lesbian mothers.
matched couples (donor insemination, 26 children; tradi- In the studies Golombok et al. (1983) and Golombok
tional conception, 26 children). The donor samples were et al. (1997) the assessments of the children included scores
drawn from registers. Gender role behavior was assessed on “unsociability” (1983), “conduct difficulty” (1983), and
by a validated questionnaire completed by parents (the “behavioral problems” (1997), as reported by both mothers
Preschool Activity Inventory), and no differences between and teachers on the questionnaires. In addition, the child’s
the three groups of children were reported. “behavioral problems” were assessed through standardized
Green (1978) interviewed 21 children of lesbian mothers interviews with the mother (1997). None of the measures
about their toy and game preferences, peer group composi- was found to differentiate between children of lesbian
tion, clothing preference, roles played in fantasy games, and mothers and children of heterosexual mothers.
vocational aspiration, and he also conducted the Draw-a- In the studies by Chan et al. (1998), Flaks et al. (1995),
Person test. No specific atypical patterns were identified. and Patterson (1994), behavioral assessments were based on
(Green assessed gender role behavior and gender identity, reports from mothers on the Child Behavior Checklist
but does not specify which of these measures assessed what (“externalizing scale”), and from the corresponding ques-
concept.) Javaid (1993) interviewed sons and daughters of tionnaire completed by teachers, the Teacher’s Report
lesbian divorced mothers (26 offspring) and heterosexual Form (though not in the Patterson study). Chan et al. and
divorced mothers (28 offspring) about gender role prefer- Flaks et al. did not find evidence of differences between
ences and expectations, for example attitudes toward being children with lesbian mothers and heterosexual mothers,
married and having children. A χ2 analysis of her data and Patterson did not find evidence that children with
(p. 242) conducted by the present reviewers showed no stat- lesbian mothers deviated from norms provided by the
istically significant differences. authors of the scale. Brewaeys et al. (1997) also assessed
Green et al. (1986) assessed 56 children of lesbian mothers the children in their study with Child Behavior Checklist,
and 48 children of heterosexual mothers with “tests of sexual but without distinguishing between externalizing and inter-
identity” (p. 170) and with interviews with children and nalizing scales in the report. (In the present context the
mothers about the children’s favorite games, toys, and activ- reported score was categorized within “behavioral adjust-
ities at home and at school (e.g. children selected between ment” and not “emotional functioning”). There were found
typical sex-typed and neutral activities). It is not clear which no group differences for boys, but fewer problems were
of these measures specifically measured gender role behavior, reported among daughters of lesbian couples and among
as opposed to gender identity. However, the study reported no traditionally conceived daughters of heterosexual couples
differences in the various preferences for boys. While there (compared with daughters of heterosexually couples with
were no differences for many types of preferences among donor insemination).
daughters, more daughters of lesbian mothers than those of A seventh study did not make a comparison between
heterosexual mothers preferred some boy-typical activities groups of children. Nonetheless, from her structured and open-

© 2002 The Scandinavian Psychological Associations.

 14679450, 2002, 4, Downloaded from https://onlinelibrary.wiley.com/doi/10.1111/1467-9450.00302 by Szkola Wyzsza Psychologii Spolecznej, Wiley Online Library on [04/10/2023]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
Scand J Psychol 43 (2002) Outcomes for children with lesbian or gay parents 347

ended interviews with five lesbian couples who had children DISCUSSION
1.5 –7 years of age (donor insemination), McCandlish (1987)
found that the mothers did not report behavioral problems The purpose of this review was to describe, categorize, and
with the children. Neither were behavioral problems of the interpret empirical studies on children raised by lesbian
children noted during the interviews. mothers or gay fathers and to provide an overview of the
existing studies (excluding double reports). The studies
reported surprisingly similar findings. Children raised by
(6) Gender identity lesbian mothers or gay fathers did not systematically differ
By “gender identity” we mean “an individual’s basic con- from other children on any outcomes. The typical outcomes
viction of being male or female” (adapted from Green’s in the 23 studies that met the inclusion criteria for the review
conceptualization, 1974, p. xv). Six studies assessed gender were emotional adjustment (12 studies), sexual preference
identity, all in children of lesbian and heterosexual mothers. (nine studies), stigmatization (nine studies), gender role
None of the studies reported that children of lesbian behavior (eight studies), behavioral adjustment (seven stud-
mothers had specific problems with gender identity. ies), gender identity (six studies), and cognitive functioning
The six studies utilized a variety of methods (see Table 1 (three studies). Of these, only three studies assessed children
for details): structured interviews with offspring about feel- of gay fathers, and they assessed sexual preference (three)
ings about being male or female (Golombok et al., 1983); and stigmatization (one).
Personal Attribute Questionnaire masculinity and feminin- Among the seven outcomes, emotional functioning,
ity scales (Gottman, 1990); interview information about toy behavioral adjustment, and stigmatization might be seen as
and game preference, peer group composition, clothing pre- indicating to what degree the offspring suffer. The opera-
ference, roles played in fantasy games, vocational aspiration, tionalization and measures of these concepts varied con-
and the Draw-a-Person test (Green, 1978; some of these siderably across the studies. The fact that none of them
methods measured gender role behavior); It-Scale for Chil- indicated that the offspring of lesbian mothers had worse
dren, testing aspects of gender identity where the children emotional functioning or more behavioral problems than
used a gender-neutral figure to select from a series of sex- other children supports the notion that the offspring of
typed toys, games, and activity preferences, interview ques- lesbian mothers do not suffer more than other children. Not
tions about wish to be a person of the opposite sex if born surprisingly, the stigmatization measures suggested that chil-
again, and Koppitz system scoring of the first-drawn person dren of lesbian mothers or gay fathers may experience the
on Draw-a-Person (Green et al., 1986, see also some of the stigma attached to persons choosing a lesbian or gay life-
measures described above under gender role behavior); style, as some of the offspring of lesbian mothers reported
historical data, probably from interviews with mothers, being teased more than others. Also, the offspring of lesbian
including characters chosen in fantasy play, sex of favored mothers or gay fathers seemed to invest energy into whether
playmates, sex play, and reports of cross-dressing, Koppitz and to whom to reveal that their mother or father was a
system scoring of the Human Figure Drawing test, semi- homosexual. However, this was probably not of a magnitude
structured playroom interviews with children concerning that hurt the offspring on a long-term basis, as reported by
sex, current interests, and future roles in life (Kirkpatrick the offspring in the follow-up study by Tasker and Golombok
et al., 1981); and interviews with mothers and offspring (1997), and as indicated by the fact that the offspring of
together and observations during interviews (McCandlish, lesbian mothers did not report different emotional function-
1987). ing or more behavioral problems than other children. Open-
ness toward offspring, neighborhood, and school from an
early age might minimalize the issue so that the fear of being
(7) Cognitive functioning stigmatized is reduced. This, however, presupposes a certain
Cognitive functioning means in the present review scores on degree of acceptance in the specific culture/neighborhood.
intelligence as measured by the Wechsler Intelligence Scale Gender identity, gender role behavior, and sexual orienta-
for Children (WISC-R and WISC) or Wechsler Preschool tion were in the studies typically seen as distinct phenomena,
and Primary Scale of Intelligence (WPPSI-R and WPPSI). with frequent reference to Money and Erhardt (1972). The
Three studies included such testing of children (pooled age empirical studies reported no differences between children of
range 3 –12 years), all assessing offspring of lesbian mothers lesbian women or gay men and other children in any of the
(in sum 91 children) as compared with offspring of hetero- three realms, despite the great variety of measures employed.
sexual mothers (in sum 83 children) (Flaks et al., 1995; Commentators on this research do, however, speculate that
Green et al., 1986; Kirkpatrick et al., 1981) (see details in the studies indicate a higher proportion of lesbian/gay off-
Table 1). In Kirkpatrick et al.’s study the testing was con- spring of lesbian mothers or gay fathers than what is
ducted without knowledge of the mother’s sexual preference. believed to be the case in the population at large (Baumrind,
In all three studies no group differences between the children 1995; Wardle, 1997). But again, to the degree that the differ-
of lesbian mothers and the others were reported. ences in proportions actually exist, it may reflect that it is

© 2002 The Scandinavian Psychological Associations.

 14679450, 2002, 4, Downloaded from https://onlinelibrary.wiley.com/doi/10.1111/1467-9450.00302 by Szkola Wyzsza Psychologii Spolecznej, Wiley Online Library on [04/10/2023]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
348 N. Anderssen et al. Scand J Psychol 43 (2002)

easier to be openly lesbian or gay with lesbian mothers or control for experiences believed to be important in children’s
gay fathers rather than more of the offspring developing development that may confound the comparisons of various
homosexual preferences as such. It should also be noted that child groups. Such experiences might be to what degree a
important discussions about the concepts of gender identity male/female adult figure has been present in the children’s
and sexual orientation take place within social construction- family life, to what degree the children have experienced a
ist and other post modern-oriented psychology (e.g. Bem, divorce, whether the children have one or two care-givers,
1995; Bohan, 1993; Richardson, 1984), as in related discip- and when parents “came out” to their children as lesbian or
lines. Here, the very categories of women and men, hetero- gay (see Fitzgerald, 1999; Golombok, 1999).
sexual and homosexual, are questioned. Gender and sexual
orientation outcomes are within these perspectives possible Measures.
to study and understand only in relation to cultural contexts Some of the measures had uncertain validity. For example,
through which the categories are produced and reproduced. the relationship between toy, activity or peer preferences and
The perspectives taken in the reviewed literature seem, how- gender identity is not clear (Goodman et al., 1998). Self-
ever, to be unproblematized and essentialistic. identification as lesbian or gay most often does not happen
Intelligence scores were, as expected, not differentiated by until late adolescence or later in life, and thus studies in
the mother’s sexual preferences. We know of no theories pre- which sexual preference was measured in younger years was an
dicting that intelligence scores are related to parental sexual outcome had methodological difficulties (Patterson, 1992).
preferences, although one might speculate that children with
emotional problems would score lower on intelligence tests Blinding procedures and response bias.
due to investments in maintaining emotional equilibrium. Blinding of data collection, coding, and interpretation was
In sum, none of the reviewed studies reported substantial not conducted in all studies. Participants or researchers may
differences in outcome among groups of children. However, consciously or unconsciously bias data in one or the other
there is a need to discuss the view of Belcastro et al. (1993) direction, and this bias may become stronger when using
and Baumrind (1995) that the data in the empirical studies self-reported recall data.
are insufficient to conclude that no differences exist. Two
related lines of reasoning must be laid out in discussing this, Strengths.
concerning both the research methods and the reasons why Some of the above limitations are inherent in doing studies
outcomes for children of lesbian mothers or gay fathers are on hidden and stigmatized groups, and we may in fact see
an issue. the studies as representing unique and valuable evidence
from groups that until the last decade have been largely
invisible, both in the public and in the psychological liter-
Research methods in the studies ature. Strictly speaking, representative samples of lesbian
The research methods in the studies reviewed are not with- mothers or gay fathers are unattainable, because many of
out weaknesses, as the authors have commented themselves them will not have “come out”; similarly, adequate sample
(see also Fitzgerald, 1999; Goodman et al., 1998), as well as sizes are difficult to establish. Two of the recently published
certain strengths. We will point out three typical weaknesses. studies utilized, however, sample sizes that gave more statist-
ical power (Bailey et al., 1995; Chan et al., 1998), and issues
Samples. of statistical power and effect sizes were discussed (Chan
The samples were typically selected through snowball et al., 1998). The meta-analysis by Allen and Burrell (1996),
techniques or self-selection, which bias the samples toward reporting no differences in emotional well-being and sexual
the more advantaged subjects among children of lesbian preference between children of heterosexual and homo-
mothers or gay fathers. The samples were most often small, sexual parents, was based on analyses of adequate effect
increasing the chance to conclude that no differences exist size and power considerations. The studies that included
between groups when in fact the differences do exist. This matching procedures represent time-consuming and costly
casts doubt on the external validity of the studies, as Belcas- efforts at solving the problems with matching according to
tro et al. (1993) pointed out. Further, parents are often exposure to various family structures. These problems are
poorly matched. Case-control designs, used in several of the of a conceptual and practical nature. For example, single
studies, are well suited for comparing groups of small preval- mothers may live with extended families and thus not fall
ence or groups who are difficult to reach, and it represents within either “single” or “couple” categories, and single
one of the standard designs within epidemiology. It requires, mothers may provide their children with male figures to
however, careful matching between cases (lesbian mother various degrees.
or gay father headed families) and controls (heterosexual Although measures of uncertain validity were utilized in
headed families). For example, many (but not all) lesbian some studies, several of the studies utilized measures with
mothers and gay fathers have had their children within a known reliability and validity, like the Child Behavior
heterosexual relationship, and it is therefore necessary to Checklist (Brewaeys et al., 1997; Chan et al., 1998; Flaks

© 2002 The Scandinavian Psychological Associations.

 14679450, 2002, 4, Downloaded from https://onlinelibrary.wiley.com/doi/10.1111/1467-9450.00302 by Szkola Wyzsza Psychologii Spolecznej, Wiley Online Library on [04/10/2023]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
Scand J Psychol 43 (2002) Outcomes for children with lesbian or gay parents 349

et al., 1995; Patterson, 1994) and the WISC-R (Flaks et al., Future studies
1995; Green et al., 1986) and others have gone far to estab- What is needed is a large research program exploring spe-
lish valid and reliable measures (as Golombok et al., 1983). cific experiences and needs of children and their lesbian
Blinding in studies where persons are assessed face to face mothers and gay fathers, with less emphasis on psychological
complicates recruitment, data collection, and coding proced- outcomes as such. A panel sample of families from all
ures. More than half of the studies did, nevertheless, utilize Scandinavian countries aimed at cultural, cross-sectional and
blinding procedures in one or more phases of the study. longitudinal analyses might be established and followed for
The studies converged on finding no substantial differ- several years. Due to the unambiguous results in the studies
ences between the children groups on the seven outcomes reviewed, we believe that large epidemiological studies with
reviewed, despite the variety of samples, ways of defining more fine-tuned instruments and tests are less needed than
concepts, measures, and procedures. This strengthens the in-depth and process-oriented methods. An approach wel-
validity of the data (Goodman et al., 1998). It is unreasonable coming lesbian women and gay fathers as parents would
to believe that we, in the near future, in any satisfactory way, start with the fact that their children must have experiences
will totally overcome the conceptual and practical problems with their parents’ identity and behavior that are different
in comparing children from the various types of families. from other children. Questions that really speak up relate to
Therefore we have to make use of the available evidence. issues like how one may help these families to overcome
In conclusion, then, although there were methodological prejudice, exploring the experiences of coming out to one’s
weaknesses in the reviewed studies, the data, in our view, children (e.g. Lynch & Murray, 2000), and exploring what
were of good enough quality to make use of them. nonlesbian or nongay families may learn from these families
(e.g. Benkov, 1995). A growing body of literature is studying
these phenomena, but much remains to be done.
Comments on the research questions
The reviewed literature originates primarily in the USA and
What are the ideologies underlying the research themes? Britain. Other studies published in French or Scandinavian
An answer to this question would require historical and languages may have been published in nonindexed literature,
discursive analyses, as the research questions originate in and thus may have been overlooked by us. Our guess is that
prevailing values and ideologies, and we will only briefly the results of the present review may be generalized to Scan-
raise the issue here. We suspect that the research questions dinavian settings, since family structures, economic systems,
in the reviewed works originate in negative views about gender relations, and conceptualizations of sexual categories
homosexual expressions and lifestyles in Western cultures. are basically the same, in contrast to, for example, Arabic
Whether one shares these negative views or resists them, the cultures. However, important differences do also exist
research questions center around comparisons between child between the regions, such as the comparative lack of ethnic
groups. While appreciating the value of many such empirical diversity in Scandinavian countries, and there is a need to
studies, Benkov (1995) also stated: “I was uncomfortable explore specific Scandinavian experiences.
with the existing lesbian mother studies for many reasons, I
don’t see the traditional nuclear family as a normative model
which should set standards for all others to meet. I don’t Conclusion
think it matters whether children grow up to be lesbian or The present review did not reveal evidence that children of
gay, or construct their sense of gender in non-traditional lesbian mothers differed from other children on emotional
ways. In both these respects, the studies took up rather than adjustment, sexual preference, stigmatization, gender role
questioned homophobic and heterosexist assumptions” behavior, behavioral adjustment, gender identity, or cognit-
(p. 53; see also Clarke, 2000; Fitzgerald, 1999; Pollack, ive functioning. The studies reported surprisingly similar
1990). The normative status of the nuclear family within findings despite the variety of conceptual and methodolo-
developmental psychology probably has a strong influence gical approaches. In sum, the findings support the idea that
on which research questions are phrased and supported fin- lesbian women should be allowed to be considered suitable for
ancially. As Burman (1994) has pointed out, normalization adoption. For men, there are too few studies to provide sub-
ideologies and adjustment rhetorics are typical within stantive evidence, although the same probably holds for them.
the field. Thus, the field might reinforce fear of homosexual
relations by the very focus on the outcomes studied. On the
other hand, in a short historical perspective, it is interesting REFERENCES
to note that during the 1950s, 1960s, and 1970s research on
adult lesbian women and gay men focused on their adjustment Allen, M. & Burrell, N. (1996). Comparing the impact of homo-
and comparison with heterosexuals (e.g. Hooker, 1957), and sexual and heterosexual parents on children: Meta-analysis of
existing research. Journal of Homosexuality, 32, 19 –35.
many would argue that this research was important in
Bailey, J. M., Bobrow, D., Wolfe, M. & Mikach, S. (1995). Sexual
removing disease notions of lesbian women and gay men. orientation of adult sons of gay fathers. Developmental Psycho-
The time now seems to have come for their children. logy, 31, 124 –129.

© 2002 The Scandinavian Psychological Associations.

 14679450, 2002, 4, Downloaded from https://onlinelibrary.wiley.com/doi/10.1111/1467-9450.00302 by Szkola Wyzsza Psychologii Spolecznej, Wiley Online Library on [04/10/2023]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
350 N. Anderssen et al. Scand J Psychol 43 (2002)

Baumrind, D. (1995). Commentary on sexual orientation: Research socioemotional development of children of lesbian and single
and social policy implications. Developmental Psychology, 31, heterosexual mothers. Journal of Child Psychology and Psychiatry,
130–136. 38, 783–791.
Belcastro, P. A., Gramlich, T., Nicholson, T., Price, J. & Wilson, R. Goodman, G. S., Emery, R. E. & Haugaard, J. J. (1998). Develop-
(1993). A review of data based studies addressing the affects of mental psychology and law: Divorce, child maltreatment, foster
homosexual parenting on children’s sexual and social function- care, and adoption. In W. Damon, I. E. Sigel & K. A. Renninger
ing. Journal of Divorce and Marriage, 20, 105 –122. (Eds.), Handbook of child psychology (vol. 4, pp. 775–874). New
Bem, S. L. (1995). Dismantling gender polarization and compulsory York: Wiley.
heterosexuality: Should we turn the volume down or up? Journal Gottman, J. S. (1990). Children of gay and lesbian parents. In F. W.
of Sex Research, 32, 329–334. Bozett & M. B. Sussman (Eds.), Homosexuality and family rela-
Benkov, L. (1995). Lesbian and gay parents: From margin to center. tions (pp. 177–196). New York: Haworth Press.
Journal of Feminist Family Therapy, 7, 49– 64. Green, R. (1974). Sexual identity conflicts in children and adults.
Bohan, J. S. (1993). Regarding gender: Essentialism, construction- New York: Basic Books.
ism, and feminist psychology. Psychology of Women Quarterly, Green, R. (1978). Sexual identity of 37 children raised by homosexual or
17, 5–21. transsexual parents. American Journal of Psychiatry, 135, 692–697.
Bozett, F. W. (1987). Gay fathers. In F. W. Bozett (Ed.), Gay and Green, R., Mandel, J. B., Hotvedt, M. E., Gray, J. & Smith, L.
lesbian parents (pp. 3 –22). London: Praeger. (1986). Lesbian mothers and their children: A comparison with
Bozett, F. W. (1988). Social control of identity by children of gay solo parent heterosexual mothers and their children. Archives of
fathers. Western Journal of Nursing Research, 10, 550 –565. Sexual Behavior, 15, 167–184.
Brewaeys, A. & Hall, E. V. v. (1997). Review. Lesbian motherhood: Haack-Møller, A. & Møhl, H. (1984). Børn af lesbiske mødre
The impact on child development and family functioning. Jour- [Children of lesbian mothers]. Dansk Psykolog Nyt, 38, 316–318.
nal of Psychosomatic Obstetrics and Gynaecology, 18, 1–16. Halvorsen, M. S. & Joner, A. (1999). Barns oppvekst i lesbiske
Brewaeys, A., Ponjaert, I., Hall, E. V. v. & Golombok, S. (1997). parforhold. Tidsskrift for Norsk Psykologforeninng, 36, 514–522.
Donor insemination: Child development and family functioning Hegna, K., Kristiansen, H. W. & Moseng, B. U. (1999). Levekår og
in lesbian mother families. Human Reproduction, 12, 1349–1359. livskvalitet blant lesbiske kvinner og homofile menn [Living
Burman, E. (1994). Deconstructing developmental psychology. conditions and life quality among lesbian women and men]
London: Routledge. (vol. 1). Oslo: Norsk institutt for forskning om oppvekst, velferd
Cameron, P. & Cameron, K. (1997). Did the APA misrepresent the og aldring (NOVA – Norwegian social research).
scientific literature to courts in support of homosexual custody? Herek, G. M. & Capitanio, J. P. (1996). “Some of my best friends”:
Journal of Psychology, 131, 313 –332. Intergroup contact, concealable stigma, and heterosexuals’
Chan, R. W., Raboy, B. & Patterson, C. J. (1998). Psychosocial attitudes toward gay men and lesbians. Personality and Social
adjustment among children conceived via donor insemination Psychology Bulletin, 22, 412 – 424.
by lesbian and heterosexual mothers. Child Development, 69, Hoeffer, B. (1981). Children’s acquisition of sex-role behavior in
443–457. lesbian-mother families. American Journal of Orthopsychiatry,
Clarke, V. (2000). “Stereotype, attack and stigmatize those who dis- 51, 536 – 544.
agree”: Employing scientific rhetoric in debates about lesbian Hooker, E. (1957). The adjustment of the male overt homosexual.
and gay parenting. Feminism and Psychology, 10, 152 –159. Journal of Projective Techniques, 21, 18–31.
De Cecco, J. P. (1981). Definition and meaning of sexual orienta- Hotvedt, M. E. & Mandel, J. B. (1982). Children of lesbian mothers.
tion. Journal of Homosexuality, 6(4), 51– 67. In P. William, J. D. Weinreich, J. C. Gonsiorek & M. E. Hotvedt
Ernulf, K. E. & Innala, S. M. (1991). Homosexualla män och les- (Eds.), Homosexuality. Social, psychological, and biological
biska kvinnor som föräldrar: En sammanfatning av aktuell issues (pp. 275 –285). London: Sage Publications.
forskning. Nordisk Sexologi, 9, 65 –74. Huggins, S. L. (1989). A comparative study of self-esteem of adoles-
Fitzgerald, B. (1999). Children of lesbian and gay parents: A review cent children of divorced lesbian mothers and divorced hetero-
of the literature. Marriage and Family Review, 29, 57–75. sexual mothers. Journal of Homosexuality, 18, 123 –135.
Flaks, D. K., Ficher, I., Masterpasqua, F. & Joseph, G. (1995). Javaid, G. A. (1993). The children of homosexual and heterosexual
Lesbians choosing motherhood: A comparative study of lesbian single mothers. Child Psychiatry and Human Development, 23,
and heterosexual parents and their children. Developmental 235 –248.
Psychology, 31, 105 –114. King, B. R. & Black, K. N. (1999). Extent of relational stigmatiza-
Gershon, T. D., Tschann, J. M. & Jemerin, J. M. (1999). Stigmatiza- tion of lesbians and their children by heterosexual college
tion, self-esteem, and coping among the adolescent children of students. Journal of Homosexuality, 37, 65 – 81.
lesbian mothers. Journal of Adolescent Health, 24, 437– 445. Kirkpatrick, M., Smith, C. & Roy, R. (1981). Lesbian mothers and
Goffman, E. (1963/1981). Stigma. Notes on the management of their children: A comparative study. American Journal of
spoiled identity. Harmondsworth: Pelican. Orthopsychiatry, 51, 545 –551.
Golombok, S. (1999). New family forms: Children raised in solo Leick, N. & Nielsen, J. (1974). Om lesbiske familier [On lesbian
mother families, lesbian mother families, and in families created families]. København Universitet, København.
by assisted reproduction. In L. Balter & C. S. Tamis-LeMonda Lewis, K. G. (1980). Children of lesbians: their point of view. Social
(Eds.), Child psychology. A handbook of contemporary issues Work (May), 198 –203.
(pp. 429–446). Philadelphia: Psychology Press. Lynch, J. M. & Murray, K. (2000). For the love of the children: The
Golombok, S., Spencer, A. & Rutter, M. (1983). Children in lesbian and coming out process for lesbian and gay parents and stepparents.
single-parent households: Psychosexual and psychiatric Journal of Homosexuality, 39, 1–24.
appraisal. Journal of Child Psychology and Psychiatry, 24, 551–572. Mandel, J. B. & Hotvedt, M. E. (1980). Lesbians as parents.
Golombok, S. & Tasker, F. (1996). Do parents influence the sexual A preliminary comparison of heterosexual and homosexual
orientation of their children? Findings from a longitudinal study mothers and their children. Huisarts and Praktijk, 31–34.
of lesbian families. Developmental Psychology, 32, 3 –11. McCandlish, B. M. (1987). Against all odds: Lesbian mother family
Golombok, S., Tasker, F. & Murray, C. (1997). Children raised in dynamics. In F. W. Bozett (Ed.), Gay and lesbian parents
fatherless families from infancy: Family relationships and the (pp. 23 –36). London: Praeger.

© 2002 The Scandinavian Psychological Associations.

 14679450, 2002, 4, Downloaded from https://onlinelibrary.wiley.com/doi/10.1111/1467-9450.00302 by Szkola Wyzsza Psychologii Spolecznej, Wiley Online Library on [04/10/2023]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
Scand J Psychol 43 (2002) Outcomes for children with lesbian or gay parents 351

Miller, B. (1979). Gay fathers and their children. Family Coordinator Rivera, R. R. (1987). Legal issues in gay and lesbian parenting.
(October), 544–552. In F. W. Bozett (Ed.), Gay and lesbian parents (pp. 199–227).
Money, J. & Ehrhardt, A. A. (1972). Man and woman boy and girl. London: Praeger.
The differentiation and dimorphism of gender identity from Statistisk Sentralbyrå (1997). Verdiundersøkelsen, 1996: Vi er blitt
conception to maturity. London: Johns Hopkins University Press. mer liberale. Ukens statistikk (33).
O’Connell, A. (1993). Voices from the heart: The developmental Shively, M. G. & De Cecco, J. P. (1977). Components of sexual
impact of a mother’s lesbianism on her adolescent children. identity. Journal of Homosexuality, 3, 41– 48.
Smith College Studies in Social Work, 63, 281–299. Steckel, A. (1987). Psychosocial development of children of lesbian
Patterson, C. J. (1992). Children of lesbian and gay parents. Child mothers. In F. W. Bozett (Ed.), Gay and lesbian parents (pp. 75–
Development, 63, 1025 –1042. 85). London: Praeger.
Patterson, C. J. (1994). Children of the lesbian baby boom. In B. Greene Strommen, E. F. (1989). “You’re a what?”: Family member reactions
& G. M. Herek (Eds.), Lesbian and gay psychology. Theory, research, to the disclosure of homosexuality. Journal of Homosexuality,
and clinical applications (pp. 156 –175). London: Sage Publications. 18, 37–58.
Patterson, C. J. & Redding, R. E. (1996). Lesbian and gay families Tasker, F. & Golombok, S. (1995). Adults raised as children in
with children: Implications of social science research for policy. lesbian families. American Journal of Orthopsychiatry, 203–
Journal of Social Issues, 52, 29 –50. 215.
Pollack, S. (1990). Lesbian parents: Claiming our visibility. In J. P. Tasker, F. L. & Golombok, S. (1997). Growing up in a lesbian family.
Knowles & E. Cole (Eds.), Motherhood. A feminist perspective Effects on child development. London: Guilford Press.
(pp. 181–194). London: Haworth Press. Wardle, L. D. (1997). The potential impact of homosexual parent-
Rafkin, L. (1990). Different mothers. Sons and daughters of lesbians ing on children. University of Illinois Law Review, 833 –920.
talk about their lives. Pittsburgh: Cleis Press.
Richardson, D. (1984). The dilemma of essentiality in homosexual Received 31 August 2000, accepted 4 May 2001
theory. Journal of Homosexuality, 9, 79 –90.

© 2002 The Scandinavian Psychological Associations.