Socioeconomic Disadvantage and

Child Development
Vonnie C. McLoyd
Duke University

Recent research consistently reports that persistent pov- Whites). Poor individuals living in high-poverty commu-
erty has more detrimental effects on IQ, school achieve- nities, as compared with their counterparts residing in
ment, and socioemotional functioning than transitory communities with lower rates of poverty, are disadvan-
poverty, with children experiencing both types of poverty taged by reduced accessibility to jobs, high-quality public
This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.

generally doing less well than never-poor children. and private services (e.g., child care, schools, parks, com-
This document is copyrighted by the American Psychological Association or one of its allied publishers.

Higher rates of perinatal complications, reduced access munity centers), and informal social supports. They also
to resources that buffer the negative effects of perinatal are subject to increased exposure to life-threatening, per-
complications, increased exposure to lead, and less nicious environmental stressors such as street violence,
home-based cognitive stimulation partly account for di- homelessness, illegal drugs, and negative role models
minished cognitive functioning in poor children. These (Duncan, 1991; Jargowsky, 1994; Shinn & Gillespie,
factors, along with lower teacher expectancies and 1994; Zigler, 1994). The level of economic deprivation
poorer academic-readiness skills, also appear to contrib- associated with childhood poverty also has increased.
ute to lower levels of school achievement among poor Whereas in 1975, 32% of all poor children ages zero to
children. The link between socioeconomic disadvantage five years lived in " d e e p " poverty (i.e., families with
and children's socioemotional functioning appears to be income 50% below the poverty threshold), by 1993, the
mediated partly by harsh, inconsistent parenting and ele- percentage had risen to 47% (Bronfenbrenner et al.,
vated exposure to acute and chronic stressors. The impli- 1996). In addition, there is some suggestion that since
cations of research findings for practice and policy are the 1970s, poverty has become more difficult to escape
considered. (i.e., more chronic; Rodgers & Rodgers, 1993), although
analyses yield different findings on this point, depending
on how poverty and its chronicity are measured (Dun-

C hildhood poverty is far from a rarity in America, can & Rodgers, 1991).
and in certain ethnic and racial minority groups, Several factors have contributed to the deterioration
it is epidemic. As of 1994, 22% of American chil- over the past two decades of American children's eco-
dren lived in families with cash incomes below the pov- nomic well-being. These factors include (a) sluggish eco-
erty threshold (U.S. Bureau of the Census, 1996). In nomic growth that has stagnated and eroded income, es-
addition to being more economically disadvantaged than pecially among young families; (b) significant loss of
their counterparts in other Western industrialized coun- low-skill, high-wage jobs due to a decline in manufactur-
tries (see Figure 1; Smeeding, 1992), American children ing industries (see Figure 1); (c) erosion of government
today are faring less well than their American counter- transfers benefiting children; and (d) increases in the
parts three decades ago. Following a period of decline number of children living with single mothers, especially
in the rate of official childhood poverty in America from never-married and teenage mothers. The risk of poverty
27% in 1959 to 14% in 1969, the rate rose slightly in single-mother families is high for many reasons, in-
throughout the 1970s and increased sharply between cluding low wages for women, unfavorable economic
1979 and 1984 from 17% to 22%. A temporary decline conditions, the low educational attainment of many single
between 1984 and 1989 was followed by a gradual in- mothers, and low rates and levels of child support from
crease (Danziger & Danziger, 1993). Children under 6 fathers (Bartfeld & Meyer, 1994; Danziger & Danziger,
years of age are at higher risk of being poor than are 1993; Garfinkel & McLanahan, 1986; Huston, 1991).
children ages 6 - 1 7 years, largely because their parents
are younger and command lower wages (see Figure 2;
Bronfenbrenner, McClelland, Wethington, Moen, & Ceci, Vonnie C. McLoyd,Departmentof Psychology:Social and Health Sci-
1996). ences, Duke University.
Increases in the prevalence of poverty have been I thank Sheba Shakir,Jamila Ponton,and Dina Greenbergfor their
accompanied by changes in the nature of poverty. During bibliographic and editorial assistance.
the past two decades or so, poverty has become more Correspondence concerning this article should be addressed to
Vonnie C. McLoyd,who is now at the Center for Human Growth and
geographically concentrated in inner-city neighborhoods Development, University of Michigan, 300 North Ingalls, Ann Arbor,
(among poor African Americans but not among poor MI 48109. Electronicmail may be sent to [email protected].

February 1998 • American Psychologist 185

Copyright 1998 by the American Psychological Association, Inc. 0003-066X/98/$2.00
Vol. 53, No. 2, 185-204
 erty and, if they are poor, to live in areas of concentrated
poverty (Duncan & Rodgers, 1988; Jargowsky, 1994).
Institutional barriers and structural factors have contrib-
uted to these racial and ethnic disparities. For example,
because of long-standing discrimination in education,
employment, and housing, African Americans, as com-
pared with Whites, have lower levels of education; hold
jobs with higher rates of displacement and lower rates
of reemployment following layoffs; and are less likely to
reside in suburbs and nonmetropolitan areas, where most
growth in entry-level jobs is occurring (Fusfeld & Bates,
1984; James, 1985; W. J. Wilson, 1996). Higher rates of
mother-headed families also have given rise to higher
poverty rates among African American children (Egge-
been & Lichter, 1991), a disparity that has been linked
to the shrinking pool of marriageable (i.e., employed or
This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.

earning wages above the poverty threshold) African

This document is copyrighted by the American Psychological Association or one of its allied publishers.

American men brought on by their deteriorating eco-

Vonnie C. nomic status (Center for the Study of Social Policy, 1986;
McLoyd Garfinkel & McLanahan, 1986; W. J. Wilson & Necker-
Photo by J.im Wallace,
Duke University man, 1986).
Photography
A confluence of factors related to America's chang-
ing ethnic and racial composition portends a rise in the
proportion of poor children who are from racial and eth-
Although the majority of poor children in the United nic minority backgrounds. Driven by higher immigration
States are of European ancestry, rates of childhood pov- rates among Latinos and Asians than among non-Latino
erty among African American, Native American, and cer- Whites of European or Middle Eastern descent and by
tain groups of Latino children typically are two to three slightly higher fertility rates among immigrants as com-
times higher than that of non-Latino White children pared with native-born Americans, the past decade ush-
(Reddy, 1993; U.S. Bureau of the Census, 1996). African ered in dramatic increases in the proportional representa-
American and Puerto Rican children are more likely than tion of racial and ethnic minorities in the American popu-
non-Latino White children to experience persistent pov- lation. Even sharper increases occurred in the proportion

Figure 1
Child Poverty Rates in Selected Western Industrialized Countries in the Mid-1980s

~_~ 20.4%
,-1

a.o~
o,3

9.3% 9.0%
7.4%
4.5% 3.8%
2.8%
1.6%

Note. Children include those under 18 years of age. This figure is based on data presented in Smeeding {1992).

186 February 1998 • American Psychologist

 Figure 2
Child Poverty Rates and the State of the U.S. Economy
This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.
This document is copyrighted by the American Psychological Association or one of its allied publishers.

Note. The gray areas refer to periods of economic recession. From The State of Americans: This Generation and the Next (p. 151), by U. Bronfenbrenner, P.
McClelland, E. Wethington, P. Moen, and S. Ceci, 1996, New York: Free Press. Copyright 1996 by Urie Bronfenbrenner, Peter McClelland, Elaine Wethington,
Phyllis Moen, and Stephen J. Ceci. Reprinted with permission of The Free Press, a Division of Simon & Schuster.

of racial and ethnic minorities in the child population, highlight some of the important questions that beckon
such that by 1992, one in three American children under empirical study. I begin with a discussion of conceptual
five years of age was an ethnic minority (66% non-Latino and methodological issues that puts in perspective the
Whites, 15% African Americans, 14% Latinos, 4% review of research findings that follows. The second and
Asians or Pacific Islanders, and 1% American Indians; major section of this article focuses on investigations of
Martin & Midgley, 1994; U.S. Bureau of the Census, the effects of poverty and low socioeconomic status
1994). Growth among racial and ethnic minorities is pro- (SES) on children's cognitive functioning, academic
jected to exceed that of non-Latino Whites for several achievement, and socioemotional development; the pro-
decades (U.S. Bureau of the Census, 1994). Post-1979 cesses by which these outcomes emerge; and factors that
immigrants are economically and educationally disad- buffer negative effects. Primary attention is given to re-
vantaged relative to native-born Americans and to immi- search studies of the postinfancy period. In the final sec-
grants who entered the United States before 1980. They tion, I discuss the implications of research findings for
may be less able than their predecessors to improve their policy and practice.
economic status once they gain experience in the U.S.
job market, first, because they are entering an American BACKGROUND AND FRAMING ISSUES
economy that is less robust and opportunity-laden than Advances in the Study of Socioeconomic
in earlier times and, second, because a significant propor-
tion are subject to racial and ethnic discrimination evoked
Disadvantage
by their darker skin color (e.g., Dominicans, Haitians, Taken as a whole, recent research on poor children is
Mexicans; Portes & Zhou, 1993; Rumbaut, 1994). distinguished by several conceptual and methodological
Research on poor children burgeoned during the advances, although, as noted below, its intellectual moor-
mid-1980s and the 1990s, prompted by sharp increases in ings include some of the concepts and assumptions that
childhood poverty and the rising amplitude of the welfare held sway during previous eras. The organization and the
reform debate. In this article, I selectively review what content of this article reflect these advances. First,
has been learned from this recent surge of scholarship and whereas research conducted during the 1960s and the

February 1998 • American Psychologist 187

 1970s conceptualized socioeconomic disadvantage as a not a percentage of the median income nor a percentile,
static phenomenon, recent research emphasizes the life- it is theoretically possible for everyone to be above the
course dynamics of poverty and low income, wherein poverty threshold. The threshold does not reflect how far
the timing and the duration of poverty are of primary below (or above) the threshold people fall (poverty gap).
importance. Spurred by national survey research pub- Increasingly popular in contemporary research is use of
lished in the 1980s documenting the volatility of income the official poverty threshold to derive an income-needs
in American families (Duncan, 1984), this shift in focus ratio, a unit of measurement that indicates the amount
owes much to the concepts and the tenets of life-course of poverty or affluence (an income-needs ratio of 1.0
theory and event-history models in which causal pro- indicates a household income that is equal to the official
cesses are seen as time-variant (Elder, 1974, 1997). A poverty threshold; e.g., Brooks-Gunn, Klebanov, & Liaw,
life-course perspective prompts questions such as (a) Is 1995; Duncan et al., 1994).
a spell of poverty that occurs in early life more damaging Unlike poverty status, SES signifies an individual's,
than one that occurs later? (b)What are the factors that a family' s, or a group's ranking on a hierarchy according
mediate the effects of poverty on development? and (c) to its access to or control over some combination of
Do causal processes differ by age? Undergirding the re- valued commodities such as wealth, power, and social
cent focus on the timing of poverty is the venerable and status (Mueller & Parcel, 1981). Dispute exists among
This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.

widespread belief in the primacy of early experience (J. social scientists about how SES should be defined and
This document is copyrighted by the American Psychological Association or one of its allied publishers.

Smith, Brooks-Gunn, & Klebanov, 1997; Zigler & Ber- measured, but there is agreement that parental occupa-
man, 1983). This n o t i o n - - a major cornerstone of Head tion, parental education, family income, prestige, power,
Start (Zigler & Berman, 1983)--was popularized during and a certain lifestyle are important components of SES
the 1960s by Bloom's (1964) influential treatise in which (House, 1981). In addition to being multidimensional
he argued that the effects of the environment on intelli- rather than unidimensional and to denoting relative posi-
gence and other human characteristics are greatest during tion rather than status defined by an absolute standard,
the early and most rapid periods of development of the SES is considerably less volatile than poverty status. Dur-
characteristics and, furthermore, that to ameliorate the ing adulthood, SES indicators such as educational attain-
effects of environmental deprivation, intervention should ment and occupational status are less likely to change
occur as early in life as possible. In queries about the markedly from one year to the next than is income relative
effects of the duration of poverty, researchers generally to need (Duncan, 1984).
espouse an additive or "cumulative deficit" model as These distinctions between SES and income poverty
articulated during the 1960s. This model assumes that are conceptually important and are viewed as crucial
longer exposure to poverty and its environmental corre- for public-policy discussions (Duncan, Yeung, Brooks-
lates will result in more adverse consequences (Deutsch, Gunn, & Smith, in press). Some research has indicated,
1973). for example, that poverty and income status have effects
In keeping with the emergence of more complex on children's development independent of parental educa-
and dynamic conceptions of poverty, contemporary re- tion (Duncan et al., 1994), although it is not yet known
search has directed attention to the impact of the context how changes in poverty and income status act synergisti-
of poverty (e.g., neighborhood-level poverty). This fo- cally with more stable indicators of SES to influence
cus reflects a trend within developmental psychology development (Huston, McLoyd, & Garcia Coil, 1994).
toward an ecological approach that takes account of Policy analysts regard the SES-income distinction as
influences at multiple levels of proximity to children. critical, partly on the presumption that it is generally
Keen interest in the effects of the economic character easier to design and implement programs that alter family
of neighborhoods on children was evoked by Bronfen- income (e.g., increasing welfare benefits, tax credits, and
brenner's (1986) conceptual work but especially by minimum wage) than programs that modify family char-
W. J. Wilson's (1987) analysis of historical changes in acteristics that mark social class (Duncan et al., in press).
the spatial concentration of poverty in inner-city African In sum, use of the federal poverty threshold as a unit of
American neighborhoods wrought by structural changes measurement is advantageous because it enables re-
in the economy and his speculations about the impact searchers to link child outcomes to the poverty gap, to
of this trend on children's expectancies, attitudes, distinguish the unique contributions of income versus
norms, and development (Brooks-Gunn, Duncan, Kleb- other components of SES, to more readily generalize the
anov, & Sealand, 1993; Chase-Lansdale & Gordon, findings to "officially" poor individuals, and to contrib-
1996; Duncan, Brooks-Gunn, & Klebanov, 1994; ute to policy discussions. Although an income-needs
Jencks & Mayer, 1990). ratio is an improvement over a poor-nonpoor dichotomy,
A second and related advance in the study of socio- vigorous debate continues regarding the validity of the
economic disadvantage and child development is more official poverty index itself (Citro & Michael, 1995;
precise measurement of economic well-being and rigor- Hauser & Carr, 1995; Haveman, 1987; Ruggles, 1990;
ous differentiation of income poverty from low SES. Re- Vaughan, 1993).
cent studies of poverty most commonly define poverty A third advance, also related to the first, is the shift
by cash income, using the federal poverty threshold as a toward greater explanatory emphasis in contemporary
marker. Because the index is an absolute dollar amount, research. Descriptions of the effects of poverty and low

188 February 1998 • American Psychologist

 income have given way to analyses of processes by which siblings), and most studies lack careful and direct mea-
such effects come about. Researchers are increasingly surement of potentially important dimensions in the envi-
sensitive to the fact that mediating pathways can vary ronment (Hoffman, 1985). Another limitation is that esti-
with age, race, gender, and ethnicity, or at least that one mates of heritability (and, therefore, of environmental
cannot assume that the processes demonstrated in one contributions) are highly specific to the population of
group apply to others. However, despite high rates of individuals and the population of environments sampled
poverty among certain groups of Latinos, Native Ameri- (Scarr, 1981). Children living in poverty are exposed to
cans, and Asian Americans, it remains unclear whether more extreme environmental conditions than are those
many of the findings of recent studies are generalizable living in more comfortable circumstances. Such extreme
to these groups because most recent research on poor conditions may exert a stronger influence, relative to ge-
children has focused on African Americans or ethnically netic makeup, than is the case for children in relatively
and racially diverse samples. Moreover, as was true dur- affluent families. In an investigation of African American
ing the 1960s and the 1970s, research focusing specifi- and White twins from a range of economic backgrounds,
cally on poor non-Latino Whites is remarkably sparse for example, estimates of heritability of school achieve-
(McLoyd, 1997). ment test performance were considerably lower for chil-
A fourth advance is increasing study of longitudinal dren living in poor neighborhoods than for children living
This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.

samples, many of which are nationally representative in more affluent areas (Scarr, 1981). If more behavioral
This document is copyrighted by the American Psychological Association or one of its allied publishers.

(e.g., the National Longitudinal Study of Youth--Child genetics studies included the full range of environments
Supplement [NLSY-CS] and the Panel Study of Income experienced by affluent and poor children in America
Dynamics [PSID]). In general, studies based on these (most focus on nonpoor children), they would probably
samples provide more externally valid estimates of ef- yield larger estimates of environmental influences but
fects and complement both national statistics (Sherman, would still provide a limited understanding of what as-
1994) and cross-sectional studies based on smaller, non- pects of the environment are important or what processes
representative samples of convenience. Furthermore, lon- are operating.
gitudinal data are essential to address new questions that For these and other reasons, behavioral genetics
have emerged from the shifts in foci delineated above. studies cannot stand alone. No one method is suPerior to
For example, although, in general, studies of preschoolers all others for all purposes; each method has strengths and
tend to report larger income effects on ability and weaknesses. Converging evidence from multiple methods
achievement outcomes than do studies of adolescents, is more scientifically sound than that from any one
longitudinal data across the entire period of childhood method. Notwithstanding the potential for genetic-envi-
are needed to draw definitive conclusions about the devel- ronment confounding in studies conducted within biolog-
opmental significance of the timing of economic depriva- ically related families, collectively, investigations using
tion (Duncan & Brooks-Gunn, 1997; Duncan et al., in a variety of methods provide strong evidence that income
press). poverty and related experiences influence children's cog-
nitive, academic, and socioemotional functioning through
Income and Poverty: Estimating Their environmental processes that go well beyond genetically
Causal Influences on Development transmitted attributes (Huston, McLoyd, & Garcia Coil,
Behavioral geneticists point out that researchers studying 1997). Included in this corpus of studies are (a) longitudi-
the relations among income (including SES), home envi- nal studies with statistical controls for genetically based
ronment, and child outcomes within biologically related shared variance (e.g., analyses of effects of poverty on
families typically ignore the possibility that income itself, children's IQ that control for maternal IQ, maternal edu-
the environments provided by parents, and children's de- cation, or both; e.g., Duncan et al., 1994; Korenman,
velopmental outcomes are influenced by genetic factors. Miller, & Sjaastad, 1995), (b) studies that control for
Therefore, they assert that the relations among these sets enduring family characteristics (and, hence, any stable,
of variables cannot be attributed solely to environmental genetically based family attributes) by using children as
influences, nor can income effects on development be their own controls or by comparing siblings (e.g., Cur-
assumed to be causal (Rowe & Rodgers, 1997; Scarr, rie & Thomas, 1995; Duncan et al., in press), (c) studies
1993). These are important and legitimate cautionary of the effects of naturally occurring income changes (e.g.,
notes. Still, it does not follow, as some behavioral geneti- Garrett, Ng'andu, & Ferron, 1994), (d) experimental ma-
cists have asserted (Rowe & Rodgers, 1997), that only nipulations of income (Salkind & Haskins, 1982), (e)
behavioral genetics studies can provide strong tests of experimental manipulations of home environments
environmental influences on children's development. (Olds & Kitzman, 1993), and (f) experimental studies of
Such studies can make a contribution to understanding preschool interventions (Royce, Darlington, & Murray,
the effects of income on development, but they are not 1983). Several of these studies are reviewed in this arti-
without limitations. cle. (For a more detailed discussion of these issues and
For example, some of the pivotal assumptions of research studies, see Huston et al., 1997.)
behavioral genetics methods are dubious (e.g., that family Scholars who are not behavioral geneticists also
environments are equally similar for monozygotic and have expressed concern that conventional analytic strate-
dizygotic twins or for twins, full siblings, and step- gies used to estimate the effects of parental income on

February 1998 • American Psychologist 189

 children's development (i.e., regression models with con- income and poverty status were more powerful predictors
trois for some parental and demographic characteristics) of IQ scores than was maternal education. Moreover,
overstate the true effect of income because of the mutual poverty status at three years of age predicted children's
association that parental income and child outcomes IQ at age five, even with age-three IQ controlled. Chil-
share with unmeasured parental characteristics. Mayer dren in Duncan et al.'s IHDP sample were low-birth
(1997) used several approaches to address this "omitted weight, premature infants (55% African Americans, 11%
variables" problem. Following attempts to assess the Hispanics, and 34% non-Hispanic Whites); therefore,
causal effect of parental welfare receipt on child out- caution should be taken in generalizing the results to
comes (Duncan & Yeung, 1995; Gottschalk, 1992), one normal-birth weight samples. Nonetheless, the IHDP is
of her approaches involved comparing the effect of paren- critically important for understanding the effects of pov-
tal income before the occurrence of a child outcome with erty because it is the only developmental data set that
the effect of parental income after the occurrence of a has combined high-quality measurement of develop-
child outcome. The rationale was that if income after the mental outcomes with longitudinal data on family eco-
outcome predicted the outcome, "this must mean that nomic status and neighborhood conditions.
income after the outcome is a proxy for parental charac- Adding another important detail to this picture is
teristics that existed before the outcome" (Mayer, 1997, growing evidence that income effects on IQ are nonlinear
This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.

p. 9). Findings from these analyses, all based on data (Duncan & Brooks-Gunn, 1997; J. Smith et al., 1997),
This document is copyrighted by the American Psychological Association or one of its allied publishers.

from the NLSY and the PSID, led Mayer to conclude although this effect is not always found (Duncan et al.,
that conventional estimates of the effects of parental in- 1994). In a particularly well-designed study using data
come on children's early cognitive functioning, years of from the IHDP and the NLSY, J. Smith et al. found
education, and young men's earnings are not biased to that the positive impact of family income on children's
any significant degree, although they appear to overstate cognitive development (as indicated by IQ scores and
the effects of parental income on children's behavior PPVT scores) was much larger among children in fami-
problems, teenage motherhood, school dropout, and sin- lies with incomes below or near the poverty line than
gle parenthood. Controversy about how best to deal with among children in middle-class or affluent families. Dun-
the problem of unmeasured parental characteristics when can et al. (in press) reported a similar nonlinear relation
randomized experiments are precluded can be expected between income and completed years of education.
to continue until some common understanding is achieved
about the relative merits of different approaches.
Duration, Timing, and Neighborhood Context of
Poverty
EFFECTS OF POVERTY AND LOW Persistent poverty is consistently found to have more ad-
SOCIOECONOMIC STATUS ON verse effects than transitory poverty on the cognitive de-
CHILDREN'S DEVELOPMENT velopment of preschool children, with children experi-
Cognitive Functioning During Early encing both types of poverty scoring lower than never-
poor children (Duncan et al., 1994; Korenman et al.,
Childhood 1995; J. Smith et al., 1997; Zill et al,, 1995). Effect sizes
Researchers have assessed the impact of poverty and SES are substantial. For instance, in Duncan et al.'s IHDP
on numerous indicators of cognitive functioning during sample, five-year-olds in chronic poverty had adjusted
early childhood, but the most prominent of these indica- mean IQs about three quarters of a standard deviation
tors are IQ test scores and, more recently, performance (nine IQ points) lower than nonpoor children. Studies
on the Peabody Picture Vocabulary Test (PPVT; Duncan have found no relation between timing of poverty and
et al., 1994; J. Smith et al., 1997; Zill, Moore, Smith, young children's IQ scores, verbal skills, or receptive
Stief, & Coiro, 1995). The PPVT yields an estimate of vocabulary (Duncan et al., 1994; J. Smith et al., 1997),
children's receptive (hearing) vocabulary and verbal abil- but this may be due to the truncated nature of the timing
ity and is a good predictor of school performance and variable in these investigations (timing categories limited
literacy (J. Smith et al., 1997). to the preschool and early school years).
A number of studies that controlled for maternal IQ, Several studies have estimated the effects of
maternal education, and a host of other maternal charac- schools' socioeconomic composition on children's cog-
teristics and behaviors (e.g., age and behaviors during nitive skills, but examinations of the link between cogni-
pregnancy) have reported significant effects of poverty tive skills and the socioeconomic composition of chil-
on children's cognitive and verbal skills (Korenman et dren's neighborhoods, as distinct from the composition
al., 1995; Liaw & Brooks-Gunn, 1994; J. Smith et al., of the schools children attend, are remarkably rare
1997). For example, Duncan et al.'s (1994) investigation (Jencks & Mayer, 1990). Rarer still are studies that docu-
based on longitudinal data from the Infant Health and ment the pathways by which neighborhood characteris-
Development Program (IHDP) found that family income tics affect children' s development (Tienda, 1991). In the
and poverty status were significant predictors of IQ IHDP sample, the proportion of poor neighbors was unre-
scores in five-year-olds, even after accounting for mater- lated to children's IQ scores at five years of age, but
nal education, family structure, ethnicity, and other differ- having a larger proportion of affluent neighbors was asso-
ences between low- and high-income families. Family ciated with higher IQ scores. This effect, considerably

190 February 1998 • American Psychologist

 smaller than the effects of family income and poverty perinatal complications (e.g., Escalona, 1984; Sameroff,
status on IQ, may reflect a salutary influence of neighbor- 1986; Werner & Smith, 1982). Premature children who
hood resources (e.g., higher quality kindergartens and are born into poverty but are resilient at three years of age
increased presence of parks, youth organizations, and (i.e., functioning well in terms of cognitive competence,
positive role models), but Duncan et al. were unable to behavioral competence, health status, and growth status)
test this or competing hypotheses. have significantly more protective agents in their home
environments (e.g., lower density and safer play environ-
Mediators of Family-Level Income and ments, higher levels of acceptance by their caregivers,
Socioeconomic Status Effects increased availability of learning materials, and more va-
Physical health status. Impairment of children' s riety) than do their nonresilient counterparts (Bradley et
physical health status at birth and diminution of chil- al., 1994).
dren's access to resources that buffer the negative effects Lead poisoning is another significant pathway
of perinatal complications are major pathways by which whereby poor children's cognitive functioning is under-
socioeconomic disadvantage hinders cognitive develop- mined. Elevated levels of lead in the blood are associated
ment during the early years. Prematurity, especially ex- with cognitive deficits, lower school achievement, and
tremely low birth weight (less than 1,250 grams), in- long-term impairment of neurological function (Nee-
This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.

creases infants' risk for a series of respiratory, neurologi- dleman, Schell, Bellinger, Leviton, & Allred, 1990). Poor
This document is copyrighted by the American Psychological Association or one of its allied publishers.

cal, and cognitive problems, including birth asphyxia, children, especially those living in inner-city areas, have
apnea, cerebral palsy, seizure disorders, visual and motor- higher levels of lead in their blood than do nonpoor chil-
coordination problems, mental retardation, and learning dren, which is due to higher rates of residence in older
disabilities (Bradley et al., 1994; Crooks, 1995). In- housing that contains lead paint and lead-soldered pipes
creased severity of illness during the perinatal period and to increased exposure to industrial and automobile air
predicts lower performance on IQ tests and tests of verbal pollution. Lead poisoning also is more prevalent among
skills (Sameroff, 1986; Siegel, 1982). Early first-trimester African American children than White children, a dispar-
initiation of prenatal care is associated with a reduction in ity traceable to long-standing housing discrimination that
low birth weight (Frank, Strobino, Salkever, & Jackson, has relegated African Americans in disproportionate
1992). Because of inadequate nutrition and substandard, numbers to poor, urban neighborhoods where inadequate
delayed, or total lack of prenatal care, poor infants are housing units and local-level industrialization are concen-
overrepresented in premature samples (gestational age of trated (Crooks, 1995; Massey, 1994).
36 weeks or less and birth weight of less than 2,500 Home-based cognitive stimulation. The ef-
grams). A disproportionate number of premature infants fects of socioeconomic disadvantage on early cognitive
are born to adolescent mothers, although questions re- functioning also are mediated through differences in the
main about whether this is solely a consequence of inade- levels of learning and academic and language stimulation
quate prenatal care (more than 50% of adolescent mothers that children receive in their home environments. Poverty,
receive no prenatal care) or whether maternal age itself low levels of maternal education, and demographic corre-
is a risk factor (Crooks, 1995; Randolph & Adams-Taylor, lates of these variables are associated with less cognitive
in press). stimulation (e.g., academic and language) in the home
Poor infants' health status also is hampered by environment, as assessed by the widely used HOME in-
higher rates of prenatal exposure to drugs, both illegal strument (Bee et al., 1982; Clarke-Stewart & Apfel, 1978;
(e.g., cocaine) and legal (e.g., nicotine, alcohol). This Deutsch, 1973; Siegel, 1982; J. Smith et al., 1997). Provi-
exposure increases perinatal complications such as re- sion of cognitively stimulating experiences in the home
ductions in birth weight, head circumference, and length has repeatedly been linked to children's IQ (Bradley et
of gestation, all of which are risk factors for delayed al., 1989, 1994; Deutsch, 1973), although this association
cognitive development, especially if the infants are poor often is weaker among African Americans and Latinos
(Hawley & Disney, 1992; Korenman et al., 1995; than Whites, probably because of cultural bias in assess-
Meisels & Plunkett, 1988; Siegel, 1982). Drug-exposed ment instruments (Clarke-Stewart & Apfel, 1978; Sug-
children, for example, show impaired organizational land et al., 1995). Moreover, preschoolers' IQ scores de-
skills (the ability to complete test items that require the cline (Sameroff, Seifer, Barocas, Zax, & Greenspan,
children to attend to several objects simultaneously or to 1987) and their home environments become less stimulat-
structure the task) and language skills during their early ing (Brooks-Gunn et al., 1995) as the number of stressful
years of life, although they are still generally within the life events and conditions increases (e.g., high family
normal range (Hawley & Disney, 1992). Poverty and low density, parental unemployment).
SES increase the probability that perinatal complications In more direct assessments of mediating influences,
will result in longer term developmental problems. An the quality of children's home environments--especially
interaction between birth status and SES-poverty is a differences in the provision of learning stimulation (as
common finding, particularly in studies of cognitive de- indicated by, e.g., presence of toys that teach color, size,
velopment, and reflects the fact that economically disad- and shape) as opposed to differences in parental
vantaged children are afforded fewer social, educational, warmth--has been found to account for a substantial
and material resources that buffer the negative effects of portion of the effects of family income and maternal

February 1998 • American Psychologist 191

 education on children's performance on IQ tests, the findings. Studies with income data for only a few years
PPVT, and other tests of cognitive functioning (Duncan of childhood (e.g., 2 - 3 years) suggest that income from
et al., 1994; J. Smith et al., 1997). Especially compelling nonwelfare sources (e.g., employment, assets) and in-
are longitudinal data indicating that within-family de- come from welfare sources have similarly positive effects
creases in income-needs ratios and increases in the dura- on daughters' educational attainment (M. S. Hill & Dun-
tion of poverty predict both deterioration in the quality can, 1987). However, those studies based on income data
of children's home environments and declines in chil- covering a large portion of childhood (e.g., 6 - 1 0 years)
dren's IQ scores (Dubow & Ippolito, 1994; Garrett et report negative effects of welfare income on children's
al., 1994). Improvements over time in family income completed years of schooling and economic status (Cor-
have the strongest effects on the quality of children's coran, Gordon, Laren, & Solon, 1992; Duncan & Yeung,
home environments if children are born poor or spend 1995) or suggest that welfare income improves children's
more time in poverty (Garrett et al., 1994). These effects educational attainment less than child support payments
remain after controlling for maternal (education, age, and (Knox & Bane, 1994). Although the latter findings are
academic ability), household (e.g., ratio of adults to chil- compatible with the welfare culture theory, they are open
dren), and child characteristics. to other interpretations, among them that persistent reli-
ance on welfare reflects severe deprivation of tangible
Academic Achievement
This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.

resources that is unmeasured by family income (M. A.

This document is copyrighted by the American Psychological Association or one of its allied publishers.

Poor and Iow-SES children, on average, perform signifi- Hill & O'Neill, 1994; Mayer, 1997).
cantly less well than nonpoor and middle-class children
on numerous indicators of academic achievement, includ- Duration, Timing, and Neighborhood Context of
Poverty
ing achievement test scores, grade retentions, course fail-
ures, placement in special education, high school gradua- School achievement, like cognitive functioning during the
tion rate, high school dropout rate, and completed years preschool years, typically declines with increases in the
of schooling (e.g., Coleman et al., 1966; Conger, Con- duration of poverty (Korenman et al., 1995; Pagani, Bou-
ger, & Elder, 1997; Entwisle & Alexander, 1990; Have- lerice, & Tremblay, 1997; Sherman, 1994; J. Smith et al.,
man & Wolfe, 1995; M. S. Hill & Duncan, 1987; Pat- 1997; Zill et al., 1995). The chance that children will be
terson, Kupersmidt, & Vaden, 1990; White, 1982). Meta- retained in grades or placed in special education in-
analyses suggest that among traditional indicators of creases by 2 - 3 % for every year the children live in pov-
SES, family income is the highest single correlate of erty, controlling for low birth weight and parents' educa-
academic achievement, followed by parental occupation tion (Sherman, 1994; Zill et al., 1995). Persistent, but
and parental education (White, 1982), and that measures not transitory, poverty between ages 8 and 12 predicted
of SES that combine income and occupation, education higher rates of grade retention among Canadian children,
and occupation, or all three components are only slightly even with controls for level of classroom inattention at
more highly correlated with academic achievement than age 6, maternal education, and other family variables
is income alone. In an analysis that provided some control (Pagani et al., 1997).
for genetically based family characteristics, Duncan et Korenman et al.'s (1995) study of the effects of
al. (in press) found that differences in family income persistent poverty, based on data from children in the
experienced by siblings during the period between birth NLSY, is especially noteworthy because it assessed the
and age 15 predicted sibling differences in adult educa- economic status of the children prenatally, at birth, and
tional attainment, with income disparities between birth throughout childhood (over a 13-year period) and in-
and age 5 being especially strong predictors. cluded a range of important control variables. Deficits in
Public welfare as a source of income for poor fami- verbal, mathematical, and reading skills associated with
lies has drawn criticism from some policy analysts on long-term poverty were two to three times larger than
the grounds that it causes intergenerational dependence those associated with current poverty status, controlling
by undermining children's self-reliance and motivation for differences in maternal education, maternal IQ, family
(principally through exposure to parents who exemplify structure, maternal behaviors during pregnancy (i.e.,
these undesirable qualities) and, in turn, their educational smoking, alcohol consumption), infants' health status
and economic attainment ("welfare culture theory"). (e.g., low birth weight), children's nutritional status, and
This sentiment partly underlies the recently implemented age of mother at first birth. The effects of income-needs
policy mandating a five-year lifetime limit on receipt of ratios found in this study were unusually large, perhaps
Aid to Families With Dependent Children (Greenberg, because measurement of the latter covered the prenatal
1996). It is known that the majority of daughters from and early preschool years, when material deprivation can
families highly dependent on welfare do not become be especially harmful to children's development.
highly dependent on welfare, although they have a greater Indeed, recently reported data suggest that the pre-
likelihood of sharing this fate than do daughters from school years are a period of elevated vulnerability to the
nonrecipient families (Duncan, Hill, & Hoffman, 1988). impact of poverty, at least in terms of years of schooling.
However, studies of the relation between sources of in- Ironically, it is precisely this period during which chil-
come (i.e., public welfare vs. other sources) and chil- dren are at highest risk of being poor (Bronfenbrenner
dren's achievement are sparse and report conflicting et al., 1996). Using data from the PSID, Duncan et al.

192 February 1998 • American Psychologist

 (in press) found that poverty during the first five years in the NLSY. Likewise, among eighth graders in the Na-
of life was far more detrimental to completed years of tional Education Longitudinal Study, demographic and
schooling than poverty during middle childhood and ado- home support variables (e.g., literacy resources in the
lescence. The differential impact of income by childhood home, discussion of school matters with families, paren-
stage was especially strong for African Americans as tal education, mothers' educational expectations) reduced
compared with Whites. Differences in the timing of pov- the effect of poverty on reading achievement by more
erty within the preschool period appear to have no effects than half (Lee & Croninger, 1994). These findings are
on achievement scores (J. Smith et al., 1997) or class- consonant with evidence that variation in parental
room placement (Pagani et al., 1997). involvement contributes to disparities in achievement
The socioeconomic mix of children's neighbor- within and across income groups (e.g., Clark, 1983;
hoods also is related to completed years of schooling. Comer, 1980; Eccles & Harold, 1993; Epstein, 1990) and
Adolescents who grow up in affluent neighborhoods or evidence that preschool intervention enhances achieve-
neighborhoods with a higher percentage of affluent fami- ment in first grade partly by increasing parental involve-
lies complete more years of school and have lower school ment (Reynolds, 1991).
dropout rates than adolescents from similar families who Entwisle et al. (1997) found evidence that home
grow up in poor neighborhoods or neighborhoods with resources (e.g., books, computers, trips) have their great-
This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.

proportionately fewer affluent families (Brooks-Gunn, est impact on children's academic growth during the
This document is copyrighted by the American Psychological Association or one of its allied publishers.

Duncan, et al., 1993; Corcoran et al., 1992; Datcher, summer, when schools are closed, and, furthermore, that
1982; Duncan, 1994), although race and gender moderate poor families are less likely than nonpoor families to
some of these relations. The percentage of affluent neigh- command the resources needed to promote the continued
bors has greater predictive power than the percentage of cognitive growth of low-SES children during this period.
low-income neighbors, a finding that appears to endorse Low-SES and high-SES grade school children in En-
the notion that neighorhoods affect children's develop- twisle et al.'s study made comparable gains in reading
ment through positive collective socialization (e.g., role and math when schools were open (winters). However,
models who enforce societal norms) and institutional re- in keeping with previous findings (Heyns, 1978), when
sources (Brooks-Gunn, Duncan, et al., 1993; Duncan, schools were closed (summers), low-SES children lost
1994). ground, whereas high-SES children continued to improve
Especially provocative is evidence that the impact their academic skills (Entwisle et al., 1997).
of neighborhood resources on children's achievement is Noneconomic theories of educational attainment
seasonal. In Entwisle, Alexander, and Olson's (1997) lon- suggest that family income positively affects child out-
gitudinal study of elementary school children, they found comes not so much for what it can buy (i.e., income as
that, controlling for family resources, neighborhood re- a resource that parents invest in their children)but for
sources positively affected children's academic growth what it represents. Central to some of these theories is
during summers, when schools were closed, but had neg- parents' status as role models. Parents with greater labor
ligible or inverse effects during winters, when schools market earnings, for instance, are thought to define suc-
were in session. Summer gains were higher among chil- cess at higher levels than parents with lesser earnings or
dren living in neighborhoods where poverty rates were welfare-dependent parents, stimulating children's motiva-
lower. When children are not in school, higher SES neigh- tion for achievement, which in turn promotes greater
borhoods apparently are better able than poor neighbor- school success. While it is appealing, there appears to be
hoods to supplant cognition-enhancing resources fur- less empirical support for the role-modeling hypothesis
nished by schools during the academic year (Entwisle et than the economic resource hypothesis, although direct
al., 1997). contrasts of the explanatory power of these two hypothe-
ses are rare (M. S. Hill & Duncan, 1987).
Mediators of Family-Level Income and
Socioeconomic Status Effects Teachers" behavior and school characteristics.
Characteristics of schools and classrooms that define
Parental and home factors, In their synthesis their social character (e.g., teachers' attitudes, school val-
of results from studies of preschool, primary, and grade ues, school-classroom climate) are significant sources
school children, Hess and Holloway (1984) identified of variation in students' achievement (Brookover, Beady,
several variables linking socioeconomic factors to school Flood, Schweitzer, & Wisenbaker, 1979; Rutter, 1983).
achievement, including verbal interactions between moth- They also appear to contribute to the achievement gap
ers and children, expectations of parents for achievement, between economic groups through interpersonal pro-
positive affective relations between parents and children, cesses. Research indicates that even in kindergarten and
and discipline and control strategies. Recent studies fo- first grade, teachers tend to perceive poor and low-SES
cusing specifically on poverty have identified similar me- students less positively (e.g., as having less maturity and
diators. For example, the amount of emotional support self-regulatory skills) and to have lower achievement ex-
and cognitive stimulation in children's home environ- pectations for them than for nonpoor children, largely
ments was found by Korenman et al. (1995) to account on the basis of noncognitive considerations (e.g., speech
for one third to one half of the disadvantages in verbal, patterns and dress). Teachers of lower SES pupils, as
reading, and math skills among persistently poor children compared with those of higher SES pupils, also are prone

February 1998 • American Psychologist 193

 to perceive the school and classroom climate less posi- range of financial and material resources usually avail-
tively (Alexander, Entwisle, & Thompson, 1987; Rist, able to schools, the precise level of these resources ap-
1970). pears to be of little consequence for achievement out-
Teachers who hold such perceptions provide poor comes (Rutter, 1983). However, this does not mean that
children with less positive attention, fewer learning op- resources are not important. Many schools serving poor
portunities (sometimes enabled by seating arrangements children are stunningly inadequate, lacking even basic
and assignments to groups), and less reinforcement of financial and material resources. Deficits in instructional
instances of good performance. As a consequence of dif- tools, teachers' morale, school climate, and other factors
ferential expectations and treatment, objectively deter- of the severity seen in these schools undoubtedly dampen
mined differences between the achievement of poor chil- educational outcomes (Kozol, 1991).
dren and nonpoor children may emerge or become exac- Physical health status, Less attention has been
erbated (e.g., test scores and grades) and, in turn, may given to the role of health-related factors in producing
further buttress differences in teachers' achievement ex- economic group differences in school achievement. How-
pectations (Alexander et al., 1987; Gouldner, 1978; Rist, ever, well-documented linkages between poverty and
1970). Teachers who grew up in middle-class homes, health and between health and school achievement pro-
rather than in lower-class homes, are more given to social vide a strong case for hypothesizing that poverty ad-
This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.

class biases, and especially racial biases, in their achieve- versely affects children's school achievement partly by
This document is copyrighted by the American Psychological Association or one of its allied publishers.

ment expectancies and perceptions of first graders' matu- diminishing their physical health status and, in turn, by
rity. These biases may stem from their unfamiliarity with engendering cognitive deficits (Crooks, 1995). For exam-
poor and minority individuals and their culture and a ple, children who were low birth weight and who suffered
resulting tendency to misconstrue "misleading cues" various perinatal illnesses as infants experience greater
(e.g., speech patterns, dress, deportment) as fundamental school failure and more school-loss days (McGauhey,
failures in these children (Alexander et al., 1987). Starfield, Alexander, & Ensminger, 1991). In addition, as
Duncan et al. (in press) lacked data to explain why noted previously, lead contamination is associated with
poverty during early childhood has more detrimental ef- cognitive deficits and lower school achievement (Nee-
fects on children's completed years of school than pov- dleman et al., 1990). Although this research does not
erty during middle childhood and adolescence, especially specifically address children in poverty, the fact that poor
among African Americans. However, as they noted, there children have higher rates of prematurity and higher lev-
is a compelling reason to believe that this timing effect els of lead in their blood suggests a connection (Crooks,
may reflect the influence of school readiness and, in turn, 1995). Logic also dictates that because school readiness
teachers' affective responses and expectancies, both of and cognitive skills during the preschool years predict
which predict later school achievement (Alexander et al., later school achievement (e.g., achievement test scores,
1987; Brooks-Gunn, Guo, & Faarstenberg, 1993; Rist, grade failure, school dropout; Brooks-Gunn, Guo, & Fur-
1970). One of the few tests of causal models of preschool stenberg, 1993), mediators of the impact of poverty and
intervention effects lends support to this speculation. low SES on early cognitive functioning (e.g., perinatal
Reynolds (1991) found that academic-readiness skills at health status, cognitive stimulation in the home environ-
the beginning and the end of kindergarten, boosted by ment, lead poisoning) also are likely to be implicated
preschool intervention, produced substantial positive ef- in economic group differences in school achievement.
fects on academic performance during first and second Similar arguments can be made about socioemotional
grade directly and indirectly through increased achieve- outcomes, given the relations among, for example, pov-
ment motivation and persistence in academic tasks, with erty, prematurity, prenatal exposure to alcohol and illegal
the effects of prior achievement increasing over time. drugs, and behavioral problems (McLoyd, Ceballo, &
That the effects of academic readiness persisted suggests Mangelsdorf, 1996).
the difficulty poor children experience in overcoming
Protective Processes
weak academic skills on entry to public school--an ob-
duracy that may be the product of labeling and low Parenting that is strict and highly directive (i.e., well-
teacher expectations (e.g., Rist, 1970) or the cumulative defined house rules, clear sanctions for breaking rules,
nature of academic skills. This difficulty may be even close supervision), combined with high levels of warmth,
greater for poor African American children because they helps poor, inner-city children resist forces in their extra-
often labor under the added burden of negative racial familial environments that in ordinary circumstances
stereotypes about their cognitive ability (Murray & Jack- contribute to low levels of achievement (e.g., peer pres-
son, 1982/1983; Steele, 1992). sure against achievement and poor-quality schooling).
Schools also contribute to the achievement gap be- These parenting behaviors distinguish poor, inner-city
tween poor and affluent children through differences children exhibiting high academic achievement from
in instructional practices and educational resources. their low-achieving counterparts exposed to similar stres-
Schools that stratify classes by ability levels, as compared sors (Baldwin, Baldwin, & Cole, 1990; Clark, 1983; Jar-
with those that do not, produce greater social class differ- rett, 1995).
ences in achievement between students in the same The protective effects of the cognitive stimulation
schools (Lee & Croninger, 1994). Within the general afforded by preschool education have been amply dem-

194 February 1998 • American Psychologist

 onstrated (Barnett, 1995; Haskins, 1989). Poor children anxiety, sadness-depression, dependency; Achenbach,
who attend large-scale (e.g., Head Start) and small, model Verhulst, Edelbrock, Baron, & Akkerhuis, 1987). Social
preschool intervention programs have superior academic- class differences in rates of extemalizing problems grad-
readiness skills or show greater gains in these skills than ually increase during the preschool and early school years
their nonenrolled counterparts (Currie & Thomas, 1995; (Stevenson, Richman, & Graham, 1985; for a review of
McKey, Condelli, Ganson, Barrett, McConkey, & Plantz, these studies, see McLoyd et al., 1996). Low SES also
1985; Reynolds, 1991). These programs also lower rates is a risk factor for chronic delinquency and early onset
of grade retention and assignment to special education, of antisocial behavior and is a correlate of numerous risk
enhance children's performance on standardized achieve- factors that have been linked to delinquency, including
ment tests, and increase rates of high school graduation large family size, family discord, and parental mental
(e.g., Fuerst & Fuerst, 1993; Hebbeler, 1985; Reynolds, illness (for a review of these studies, see Yoshikawa,
1994). These effects have been replicated in experimental 1994). Although income and SES are linked to children's
studies with random assignment (Royce et al., 1983) and socioemotional functioning, they generally are less pow-
in quasi-experimental studies that carefully control for erful in accounting for differences in children's socio-
preexisting differences in children's cognitive function- emotional functioning than differences in their cognitive
ing and background characteristics such as maternal edu- development and school achievement (e.g., Dubow & Ip-
This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.

cation and father presence (Lee, Brooks-Gunn, & Schnur, polito, 1994; Duncan et al., 1994; Pagani et al., 1997).
This document is copyrighted by the American Psychological Association or one of its allied publishers.

1988; Lee, Brooks-Gunn, Schnur, & Liaw, 1990). Partici-

pation in formal afterschool care programs that provide
Duration, Timing, and Neighborhood Context of
Poverty
cognitive stimulation also is associated with improved
academic achievement among low-income children Internalizing problems (Duncan et al., 1994; Hanson,
(Posner & Vandell, 1994). McLanahan, & Thomson, 1997; McLeod & Shanahan,
Unfortunately, cognitive and achievement gains pro- 1993; Pagani et al., 1997) and externalizing symptoms
duced by preschool education programs are typically lost (Duncan et al., 1994; Hanson et al., 1997; Korenman et
by the third or fourth grade, if not before (Lee et al., al., 1995) become more prevalent the longer children have
1990; McKey et al., 1985), although gains produced by been living in poverty. Neither internalizing nor exter-
model programs and programs of longer duration (i.e., nalizing symptoms have been linked to timing of poverty
more than two years) and greater intensity tend to persist within the first four years of life (Duncan et al., 1994),
longer (Campbell & Ramey, 1995; Ramey, Ramey, but it remains to be seen whether timing of poverty as
Gaines, & Blair, 1995; Royce et al., 1983). There is strong assessed across the entire period of childhood and adoles-
reason to believe that this diminution of effects is partly cence influences children's socioemotional functioning,
due to poor-quality schooling subsequent to the interven- Many theories assume that there is more crime in
tion (Lee & Loeb, 1995). Moreover, in a trend reminis- poor neighborhoods than affluent neighborhoods because
cent of effects found for academic readiness, at posttest living in a poor neighborhood increases adolescents'
and follow-up, poor children who participate in large- chances of committing antisocial and delinquent acts
scale preschool education perform markedly below na- (through peer influences, negative adult role models out-
tional norms and below the performance of nonpoor chil- side the family, etc.). Data on this point are not only
dren on standardized achievement tests (Hebbeler, 1985; contradictory but also surprisingly sparse (Jencks &
Reynolds, 1994). In sum, although preschool intervention Mayer, 1990). In a rare study of neighborhood effects on
offers some protection from the negative effects of pov- preschoolers' socioemotional functioning, Duncan et al.
erty, over the long run, it does not bestow levels of cogni- (1994) found that having more low-income neighbors
tive and academic competence comparable to those seen predicted higher levels of externalizing problem behav-
among nonpoor children in the general population. iors among five-year-olds (e.g., temper tantrums, destruc-
tion of objects), controlling for family income, poverty
Socioemotional Functioning status, and other family variables. These authors specu-
Studies based on samples drawn from primary care facili- lated that parents in poorer neighborhoods may be less
ties and pediatric practice settings and samples of nonre- prone to reduce aggression and acting out in children
ferred children in home and school settings have reported because the need for children to defend themselves
a higher prevalence of emotional and behavioral prob- through such behaviors is greater in poorer neighbor-
lems among poor and low-SES children and adolescents hoods. Peer group influences and lower quality schools
than their middle-class counterparts (e.g., Adams, Hill- and child-care environments also may underlie this
man, & Gaydos, 1994; Butler, Starfield, & Stenmark, neighborhood effect (e.g., less adult supervision as re-
1984), although a few epidemiological studies of very flected in higher child-staff ratios may result in less
young children (three years or younger) have not (e.g., inhibition of acting out and aggressive behaviors).
Earls, 1980). Parents' reports, teachers' reports, and self-
reports reveal these social class effects, with effects gen- Mediators of Family-Level Income and
erally being more pronounced for externalizing problems
Socioeconomic Status Effects
(e.g., disobedience, fighting, difficulty getting along with Parental dysphoria and harsh, inconsistent
others, impulsivity) than internalizing problems (e.g., parenting. A strong pattern of findings from recent

February 1998 • American Psychologist 195

 studies suggests that poverty and economic stress elevate loss is more debilitating to families and children than is
socioemotional problems in children partly by increasing chronic hardship or poverty. Second, whereas Elder
parents' tendency to discipline children in a punitive and (1974) found that economic loss adversely affected only
inconsistent manner and to ignore children's dependency the behavior of fathers, and only those fathers who were
needs (Conger et al., 1992; Conger, Ge, Elder, Lorenz, & emotionally unstable prior to the Depression, research
Simons, 1994; Dodge, Pettit, & Bates, 1994; Lempers, with contemporary families experiencing either eco-
Clark-Lempers, & Simons, 1989; McLeod & Shanahan, nomic loss or chronic poverty has demonstrated more
1993; McLoyd, Jayaratne, Ceballo, & Borquez, 1994), pervasive effects extending to the emotional functioning
although some research contradicts this pattern (Hanson and parenting behavior of both mothers and fathers (e.g.,
et al., 1997). Likewise, overwhelming evidence exists Conger et al., 1994; Lempers et al., 1989; McLoyd, 1993;
that these parenting behaviors stem partly from an over- McLoyd et al., 1994). These processes apparently operate
abundance of negative life events and conditions that within contemporary families without regard to parents'
confront poor adults (McLoyd, 1990). emotional histories, although studies using a prospective
In McLeod and Shanahan's (1993) study of four- research design similar to Elder's would provide more
to eight-year-old children in the NLSY, mothers' low lev- definitive evidence concerning this issue. Finally, whereas
els of involvement with their children and frequent use Elder's research pointed to irritability and anger as criti-
This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.

of physical punishment explained the effect of current cal emotional responses that produce harmful effects in
This document is copyrighted by the American Psychological Association or one of its allied publishers.

poverty on children's internalizing and externalizing children, studies of contemporary families also have
symptoms. However, length of time spent in poverty nei- pointed to depressive symptoms and dysphoria as promi-
ther increased the frequency of physical punishment nor nent mediating states, leading to both withdrawal from
decreased mothers' emotional responsiveness, perhaps and punitiveness toward children (Duncan et al., 1994;
indicating that family interactions stabilize as families McLeod & Shanahan, 1993; McLoyd et al., 1994).
adapt to economic deprivation. In a similar vein, eco- Besides direct tests of mediating influences, studies
nomic pressure created by the Iowa farm crisis during of bivariate relations provide an abundance of indirect,
the 1980s led to more self-reported anxiety, depression, converging evidence of the role of parents' emotional
and delinquent behavior in children by increasing par- responses and child-rearing behavior in linking economic
ents' tendencies to treat children in a hostile and coercive hardship to children's development. For example, it is
fashion. Parental hostility was precipitated by marital well-documented that (a) poor adults have more mental
conflict and parent-child conflicts over money (Conger health problems than nonpoor adults; (b) mothers with
et al., 1994). Maternal depression mediated the effects more mental health problems exhibit fewer positive be-
of poverty on both internalizing and externalizing symp- haviors and more hostile, dominating, and coercive be-
toms in Duncan et al.'s (1994) study of five-year-olds in havior in disciplinary encounters; (c) poor adults experi-
the IHDP sample. ence more negative life events and chronic conditions
These studies of contemporary families are in- than nonpoor adults; (d) negative life events increase
formed by and generally consistent with Elder's (1974; punitive, harsh, and inconsistent parenting behavior; and
Elder, Nguyen, & Caspi, 1985) pathbreaking studies of (e) child-rearing behaviors more prevalent among poor
the effects of paternal income and job loss during the parents are predictive of various socioemotional prob-
Great Depression of the 1930s. In that research, eco- lems in children (for extensive reviews of these bodies
nomic loss was found to have few direct effects on chil- of literature, see Dix, 1991; Downey & Coyne, 1990;
dren's socioemotional functioning. Rather, most adverse Loeber & Dishion, 1983; McLoyd, 1990; Yoshikawa,
effects were produced indirectly through negative effects 1994).
on fathers' psychological functioning and parenting. Fa- Discrete and chronic s t r e s s o r s , Studies have not
thers who sustained heavy financial losses became more directly assessed exposure to negative events and condi-
irritable, tense, and explosive, which in turn increased tions as a mediator of the effects of family-level income
their tendency to be punitive, arbitrary, and inconsistent or SES on children's socioemotional functioning, but it
in disciplining their children. These behaviors predicted is highly likely that such mediating processes exist. Poor
temper tantrums, irritability, and negativism in young and low-SES children experience more negative or unde-
children, especially boys, and moodiness, hypersensitiv- sirable life events and adverse conditions, and this over-
ity, and feelings of inadequacy in female adolescents. abundance can place demands on them that exceed their
Although drawing heavily from Elder's work (Elder, coping resources (Pryor-Brown, Cowen, Hightower, &
1974; Elder et al., 1985), studies of the effects of eco- Lotyczewski, 1986; Sterling, Cowen, Weissberg, Lotyc-
nomic hardship on contemporary families diverge from zewski, & Boike, 1985). Undesirability of life events is
and extend Elder's research and conceptualizations in a consistent predictor of socioemotional maladjustment
important ways. First, persistent poverty, not transitory (e.g., depression, aggression, somatic complaints; Attar,
poverty, is consistently associated with more harmful ef- Guerra, & Tolan, 1994; Pryor-Brown, Powell, & Earls,
fects on children's socioemotional functioning (Duncan 1989; Sterling et al., 1985).
et al., 1994; Korenman et al., 1995; McLeod & Shanahan, Adversity that is chronic also exacts a toll on chil-
1993; Pagani et al., 1997), a finding that contradicts Elder dren's mental health. Shaw and Emery (1988), for exam-
and Caspi's (1988) contention that substantial income ple, found that low-income school-age children exposed

196 February 1998 • American Psychologist

 to a greater number of chronic family stressors (i.e., pa- Protective Processes
rental conflict, maternal depression, overcrowding) had
more internalizing symptoms, more clinically elevated Whereas parenting behavior is an important mediator of
child behavior problems, and lower self-worth. For poor the effects of poverty on children's socioemotional func-
and low-SES children, chronic stressors often include tioning, it also may be crucial in buffering poor children
neighborhood and housing conditions. Children are not from the negative effects of discrete and chronic stressors
oblivious to their community environments, and, like exogenous to parents. In Cowen, Wyman, Work, and Par-
adults, they make judgments about the quality of their ker's (1990) sample of urban children exposed to a high
neighborhoods that are consistent with their neighbor- number of chronic adversities and negative life events
hoods' objective characteristics (Homel & Burns, 1987; (poverty, family turmoil, illness-death, violence), factors
Zill, 1984). Children living in poor, high-crime neighbor- that distinguished stress-resilient children (as indexed by
hoods report experiencing more stressors than those liv- high ratings from parents and current or former teachers
ing in more advantaged neighborhoods (Attar et al., on peer relations, likeability, leadership qualities, and
1994). Street violence is a problem in many poor, urban school performance) from stress-affected children in-
communities (Garbarino, Kostelny, & Dubrow, 1991; cluded nonseparation of child and primary caregiver dur-
Osofsky, 1995), and male adolescents are especially sub- ing infancy; positive parent-child relations during the
This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.

ject to be victims of and witnesses to it (Fitzpatrick & preschool and elementary school years; a strong sense of
This document is copyrighted by the American Psychological Association or one of its allied publishers.

Boldizar, 1993). Witnessing and being victimized by parenting efficacy by the primary caregivers; and parental
community and domestic violence are associated with use of reasoned, age-appropriate, consistent disciplinary
aggression (Attar et al., 1994) and with high levels of practices. Similar parenting variables have predicted
posttraumatic stress symptoms (e.g., fear, anxiety, depres- stress resilience in other samples (Masten, Morison, Pelli-
sion, hostility; Kotlowitz, 1991; Richters & Martinez, grini, & Tellegen, 1990; Rutter, 1990; Werner & Smith,
1993; Zinsmeister, 1990), especially among female ado- 1982). Also noteworthy is that stress-resilient children
lescents (Fitzpatrick & Boldizar, 1993). identify more strongly with their parents and perceive
Poor children are more likely than nonpoor children their parents as more Supportive and less harsh than do
to live in housing located on commercial-industrial highly stressed children with severe learning and behav-
streets, and there is evidence that this residential setting ioral problems (Masten et al., 1990; Wemer & Smith,
can be a source of internalizing symptoms (e.g., worry, 1982).
fear, loneliness). Commercial-industrial streets may pro- Family members and extrafamilial adults living out-
duce these effects because they are less friendly environ- side the home may indirectly contribute to socioemo-
ments, offer fewer opportunities for social exchange tional resilience in poor children by providing emotional
among children, and increase parents' insistence that and parenting support to the mother and bolstering paren-
children stay in or close to home during free time (Du- tal control (Cowen et al., 1990; Dornbusch et al., 1985).
brow & Garbarino, 1989; Homel & Burns, 1989). Poor Because parental depression is a risk factor for difficul-
children's self-esteem and sense of hope also may be ties in parenting, factors that protect against matemal
eroded by pernicious housing conditions (e.g., lack of depression (e.g., companionship, availability of a confi-
heat, inadequate plumbing, insect and rodent infestation), dant, and assistance in child rearing) are likely to enhance
by circumstances that publicly mark their membership in parenting, facilitate positive parent-child relations, and
a stigmatized group (e.g., using food stamps, living in in turn foster adaptive resilience in the children (for a
public housing, living in neighborhoods branded as review of these studies, see McLoyd, 1990). Nonparental
"bad"), and by ostracism from nonpoor individuals in adults also may foster resilience in poor children by mod-
schools and other settings evoked by these public markers eling positive behavior and providing support directly to
of stigma (Brantlinger, 1991; Kotlowitz, 1991; MacLeod, the children. Positive adult role models figured promi-
1987; Wiltfang & Scarbecz, 1990). nently in the lives of poor, inner-city African American
Homelessness is associated with more externalizing "superkids" in Williams and Kornblum's (1985) ethno-
(but not internalizing) problems in poor children and ado- graphic study. Similarly, all of the high-risk, resilient
lescents, probably because it precipitates and follows on children in Wemer and Smith's (1982) study received
the heels of negative life events out of the children's guidance and support from a mentor.
control and impedes mothers' ability to provide nurturant Several educationally oriented preschool interven-
and consistent parenting (Hausman & Hammen, 1993). tions have produced immediate (Lee et al., 1990; McKey
Masten, Miliotis, Graham-Bermann, Ramirez, and Nee- et al., 1985) and longer term gains in self-esteem and
mann (1993) found that low-income children living in social competence (Lally, Mangione, & Honig, 1988) and
an emergency shelter had experienced significantly more reductions in aggression, acting out, and antisocial be-
discrete and chronic stressors than low-income children havior (Johnson, 1988; Lally et al., 1988; Seitz, Rosen-
living at home. Delinquent and aggressive behavior was baum, & Apfel, 1985). Moreover, some have proved to
elevated in homeless children, but parental distress, cu- be effective in reducing the number of probation cases,
mulative risk history, and recent negative life events, the number of months on probation or parole, the number
rather than housing status, were the primary predictors of lifetime (juvenile and adult) criminal arrests (including
of child behavior problems. drug-related, property, and personal-violence crimes),

February 1998 • American Psychologist 197

 and the number of frequent offenders (Lally et al., 1988; ment of school-readiness skills, setting in motion a pat-
Schweinhart, Barnes, & Weikart, 1993). Most of these tern of underachievement.
effect sizes are large enough to be socially meaningful, Higher rates of perinatal complications, reduced ac-
with the latter one being especially important given that cess to resources that buffer the negative effects of peri-
most crimes are committed by frequent offenders (J. Q. natal complications, increased exposure to lead, and less
Wilson & Herrnstein, 1985). home-based cognitive stimulation partly account for di-
Interventions that produce such effects are typically minished cognitive functioning in poor children. These
family-focused, whereby preschool education is accom- factors, along with lower teacher expectancies and poorer
panied by various core services to parents and other fam- academic-readiness skills, also appear to contribute to
ily members (e.g., parenting education, assistance with lower levels of school achievement among poor children.
housing and legal problems). They appear to attenuate Harsh, inconsistent parenting and elevated exposure to
aggressive and delinquent behavior by reducing multiple acute and chronic stressors appear to be key mediators of
risk factors associated with these outcomes (e.g., cogni- the link between poverty and children's socioemotional
tive deficits, impaired socioemotional functioning in early functioning.
life, low parental involvement with children, hostile or
rejecting parenting, low levels of schooling and earnings
PRACTICE AND POLICY IMPLICATIONS
This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.

during late adolescence). With rare exception (Schwein- Research on the effects of poverty suggests that policies
This document is copyrighted by the American Psychological Association or one of its allied publishers.

hart et al., 1993), these hypothesized pathways have not that raise the incomes of poor families will enhance chil-
been directly tested, but they are consistent with an exten- dren's development, especially their cognitive function-
sive body of research on the precursors of delinquent and ing and educational attainments. Nonlinear effects of in-
externalizing behaviors (Yoshikawa, 1994, 1995; Zigler, come on cognitive and academic functioning are conso-
Taussig, & Black, 1992). nant with the view that "reducing poverty is associated
with improved outcomes for children, whereas increasing
Summary affluence is not" (Garbarino, 1992, p. 233). They bolster
the assertion that targeting primarily poor families, rather
Family-level poverty, low SES, and residence in less eco- than middle and upper income families, for income sub-
nomically advantaged neighborhoods each independently sidies and tax relief, overall, will have a more salutary
predicts lower scores on tests of intelligence and cogni- impact on children's development. Barring income incre-
tive functioning, lower levels of school achievement, and ments for all poor families, Duncan et al.'s (in press)
increased levels of socioemotional problems, controlling research on the differential effects of the timing of pov-
for various parent and family characteristics. Increments erty argues for welfare policies that give highest priority
in family income bear a nonlinear relation to both cogni- to the elimination of deep and persistent poverty during
tive functioning and completed years of schooling, ex- children's early years of life. It also strengthens the case
erring much larger impacts on lower income children than for implementation of Head Start for poor children from
higher income children. Family income generally is more birth to age three (Zigler, 1994), whereas evidence that
powerful in accounting for differences in children's cog- more intensive preschool intervention produces more en-
nitive development and school achievement than differ- during effects argues for making Head Start a full-day
ences in socioemotional functioning. Neighborhood in- rather than a half-day program, five days per week, year-
come consistently has less potent effects on children's round (currently, centers operate on a nine-month school
development than does family income. Social class ef- year; fewer than 15% are full-day programs, and 35%
fects are more pronounced for externalizing behavior are open less than five days per week; Hofferth, 1994;
problems than internalizing symptoms, with class effects Zigler & Styfco, 1994). Such an expansion also would
on rates of externalizing problems increasing markedly respond to child-care needs created by the work require-
during the preschool and early school years. ments of the 1996 federal welfare reform law and help
Persistent poverty is repeatedly found to have more redress the paucity of regulated, high-quality child care
detrimental effects on IQ, school achievement, and so- (Greenberg, 1996).
cioemotional functioning than does occasional or transi- The 1996 welfare reform law mandates a five-year
tory poverty. Relatively little research has assessed the lifetime limit on welfare assistance in the form of cash
deveaopmental significance of the timing of poverty dur- aid, work slots, or noncash aid, although several states
ing childhood. The few extant studies suggest that timing plan to impose even shorter time limits (Greenberg, 1996;
of poverty is unrelated to IQ and socioemotional func- Pear, 1997). Cutoffs of welfare assistance can trigger
tioning but matters a great deal in shaping children's several changes that can have harmful effects on children
educational attainment. Poverty experienced during the (e.g., fewer material resources, increased levels of paren-
first five years of a child' s life attenuates completed years tal distress, punitiveness toward children, declines in the
of schooling more so than does poverty during middle quality of home environments as families double up).
childhood and adolescence, an effect that is more pro- Extrapolating from Duncan et al.'s (in press) research,
nounced among African Americans than Whites. One cutoffs that coincide with children's preschool years and
plausible interpretation of this finding is that poverty ex- that result in even deeper poverty are likely to be espe-
perienced during early childhood thwarts the develop- cially detrimental to children's educational attainment.

198 February 1998 • American Psychologist

 Policies that increase the availability and the quality a broad base of bipartisan and popular support. That such
of prenatal care to poor women, especially when prenatal interventions are the core essence of America's antipov-
care is combined with nutritional services, and policies erty policies reflects Americans' preference for indirect
that reduce children's exposure to lead can be expected rather than direct approaches to poverty reduction (Has-
to translate into enhanced cognitive functioning in poor kins, 1989; Oyemade, 1985), their aversion to social in-
children. Evidence of the mediating influence of chil- surance programs similar to those that exist in European
dren's home learning experiences is in accord with the countries (e.g., universal child allowances, universal
assumption of "two-generation" programs, namely, that health care, free or inexpensive child care), and their
increasing levels of education among poor mothers will conviction that governmental policies should promote
indirectly and positively affect children's cognitive func- equality of opportunity rather than equality of condition
tioning through provision of more stimulating home envi- (Hailer, Hollinger, & Raubal, 1990). Evidence that family
ronments (S. Smith & Zaslow, 1995). income is a stronger predictor of IQ and school achieve-
Studies documenting the adverse and cumulative ef- ment than traditional indicators of social class, such as
fects of risk factors such as stressful life events and low parental education, invites a number of inferences, one
social support on parental functioning, home learning of which is that programs that directly raise incomes of
environments, and children's development argue for in- poor families (e.g., increasing welfare benefits, tax cred-
This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.

tervention and prevention efforts that are ecological, in- its, minimum wage) may be comparatively more effective
This document is copyrighted by the American Psychological Association or one of its allied publishers.

tegrative, and family-centered in their approach and that in fostering children's development (in addition to being
encompass delivery of a broad range of concrete services easier to design and administer) than programs that seek
intended to ease the stressors attendant to poverty (Ill- to modify family characteristics. Productive debate of
back, 1994; Schorr, 1989). They also justify efforts to this issue requires cost-benefit analyses in which expen-
shore up the support-services component of Head Start ditures on income-transfer programs and service delivery
(Takanishi & DeLeon, 1994) and buttress the assumption programs are compared and "judged by the benefits they
of family support programs that children's development produce relative to their costs" (Duncan et al., in press).
can be enhanced indirectly by family support services Setting aside the issue of comparative benefits, it is
that (a) reduce the occurrence and buffer the negative clear that the prevalence of poverty in America has not
effects of stressful life events and (b) increase parents' been substantially reduced by antipoverty policies whose
ability to provide children with cognitive stimulation and core strategy involves educating or ministering to the
emotional support (Weissbourd, 1987). Such services, acute needs of poor children and families in the absence
however, are no substitute for direct educational services of job creation and the presence of massive numbers of
to preschoolers (Ramey et al., 1995). The relation be- jobs that do not pay parents living wages. At best, service
tween cognitive development and home learning environ- delivery programs have only blunted the force of poverty
ments underscores the potential value of offering parents and rendered modest improvements in children's environ-
information about children's developmental needs and mental circumstances and developmental outcomes
encouraging parents to transport cognitively stimulating (McLoyd, 1997). The stressors and the resource deficits
activities from classrooms to home settings (McLoyd, of chronic poverty are simply too multifarious and conta-
1997). gious. As Zigler and Styfco (1994) recently lamented,
Besides home-based cognitive stimulation, parental "Neither Head Start nor any preschool program can inoc-
involvement in children's schooling, teachers' expecta- ulate children against the ravages of poverty. Early inter-
tions and affective orientations, and the material re- vention simply cannot overpower the effects of poor liv-
sources and instructional policies of schools contribute ing conditions, inadequate nutrition and health care, neg-
to economic group differences in children's school ative role models, and substandard schools" (p. 129).
achievement. Each of these pathways suggests strategies The limited effectiveness of service delivery programs is
for raising the level of scholastic achievement among rooted in the fact that many of the causes of poverty and
poor and low-SES children. Instructing teachers about the difficult life conditions confronting poor families are
normative behaviors (e.g., speech patterns) among lower- structural in nature and, as such, are largely impervious
class and minority children and affirming the legitimacy to child- and family-level interventions (e.g., historic and
of those behaviors may be prerequisites to raising teach- contemporary racism in the labor market, lending institu-
ers' expectancies of these children. Interventions of this tions, and housing; poor-quality schools; low wages paid
nature in the early grades seem warranted, because nega- by traditionally "female" jobs; unavailability of af-
tive teacher evaluations and affective responses intrude fordable, high-quality child care). In essence, service de-
very early in the educational experiences of poor and livery programs arguably represent a case of using sec-
African American children and mark the onset of class- ondary strategies to deal with primary problems (Hal-
and race-differentiated achievement trajectories (Alexan- pern, 1988, 1991). Nonetheless, to argue that these pro-
der et al., 1987; Rist, 1970). grams are insufficient to radically reduce poverty is not
Interventions that provide poor Americans with edu- to suggest that they be eliminated. Rather, in the spirit
cation and support services (e.g., Head Start, home-vis- of Zigler and Styfco' s admonition, it is a call to acknowl-
iting programs, education and job-training programs) but edge that child- and family-level interventions and service
not with more cash income and material support enjoy delivery programs can "counter some of the injuries of

February 1998 • American Psychologist 199

 inequality, b u t . . , c a n n o t d e s t r o y i n e q u a l i t y i t s e l f " (de Butler, J. A., Starfield, B., & Stenmark, S. (1984). Child health policy.
In H. Stevenson & A. Siegel (Eds.), Child development research and
L o n e , 1979, p. 68). In so doing, these p r o g r a m s are re- social policy (pp. 110-188). Chicago: University of Chicago Press.
lieved of a responsibility they cannot discharge--pro- Campbell, E, & Ramey, C. (1995). Cognitive and school outcomes for
d u c i n g g r e a t e r equality. high-risk African-American students at middle adolescence: Positive
effects of early intervention. American Educational Research Jour-
REFERENCES nal 32, 743-772.
Center for the Study of Social Policy. (1986). The "flip-side" of Black
Achenbach, T., Verhulst, E, Edelbrock, C., Baron, G. D., & Akkerhuis, families headed by women: The economic status of Black men. In
G. (1987). Epidemiological comparisons of American and Dutch R. Staples (Ed.), The Black family: Essays and studies (pp. 232-
children: II. Behavioral/emotional problems reported by teachers for 238). Belmont, CA: Wadsworth.
ages 6 to 11. Journal of the American Academy of Child and Adoles- Chase-Lansdale, P. L., & Gordon, R. A. (1996). Economic hardship and
cent Psychiatry, 26, 326-332. the development of 5- and 6-year-olds: Neighborhood and regional
Adams, C., Hillman, N., & Gaydos, G. (1994). Behavioral difficulties perspectives. Child Development, 67, 3338-3367.
in toddlers: Impact of sociocultural and biological risk factors. Jour- Citro, C. E, & Michael, R. T. (Eds.). (1995). Measuring poverty: A new
nal of Clinical Child Psychology, 23, 373-381. approach. Washington, DC: National Academy Press.
Alexander, K., Entwisle, D., & Thompson, M. (1987). School perfor- Clark, R. (1983). Family life and school achievement: Why poor Black
mance, status relations, and the structure of sentiment: Bringing the children succeed or fail. Chicago: University of Chicago Press.
teacher back in. American Sociological Review, 52, 665-682. Clarke-Stewart, K. A., & Apfel, N. (1978). Evaluating parental effects
This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.

Attar, B., Guerra, N., & Tolan, P. (1994). Neighborhood disadvantage, on child development. In L. S. Shulman (Ed.), Review of research
This document is copyrighted by the American Psychological Association or one of its allied publishers.

stressful life events, and adjustment in urban elementary-school chil- in education (Vol. 6, pp. 47-119). Itasca, IL: Peacock.
dren. Journal of Clinical Child Psychology, 23, 391-400. Coleman, J. S., Campbell, E., Hobson, C., McPartland, J., Mood, A.,
Baldwin, A.L., Baldwin, C., & Cole, R.E. (1990). Stress-resistant Weinfeld, E, & York, R. (1966). Equality of educational opportunity.
families and stress-resistant children. In J. Rolf, A. S. Masten, D. Washington, DC: U.S. Government Printing Office.
Cicchetti, K. Nuechterlein, & S. Weintraub (Eds.), Risk and protec- Comer, J.P. (1980). School power: Implications of an intervention
tive factors in the development of psychopathology (pp. 257-280). project. New York: Free Press.
Cambridge, England: Cambridge University Press. Conger, R.D., Conger, K.J., & Elder, G. (1997). Family economic
Barnett, W. S. (1995). Long-term effects of early childhood programs hardship and adolescent academic performance: Mediating and mod-
on cognitive and school outcomes. The Future of Children, 5, 25- erating processes. In G. Duncan & J. Brooks-Gunn (Eds.), Conse-
50. quences of growing up poor (pp. 288-310). New York: Russell Sage
Bartfeld, J., & Meyer, D. (1994). Are there really deadbeat dads? The Foundation.
relationship between ability to pay, enforcement, and compliance in Conger, R. D., Conger, K., Elder, G., Lorenz, E, Simons, R., & Whit-
nonmarital child support cases. Social Service Review, 68, 219-235. beck, L. (1992). A family process model of economic hardship and
Bee, H., Barnard, K., Eyres, S., Gray, C., Hammond, M., Spietz, A., adjustment of early adolescent boys. Child Development, 63, 526-
Snyder, C., & Clark, B. (1982). Prediction of IQ and language skill 541.
from perinatal status, child performance, family characteristics, and Conger, R., Ge, X., Elder, G., Lorenz, E, & Simons, R. (1994). Eco-
mother-infant interaction. Child Development, 53, 1134-1156. nomic stress, coercive family process, and developmental problems
Bloom, B. S. (1964). Stability and change in human characteristics. of adolescents. Child Development, 65, 541-561.
New York: Wiley. Corcoran, M., Gordon, R., Laren, D., & Solon, G. (1992). The associa-
Bradley, R. H., Caldweil, B., Rock, S. L., Ramey, C., Barnard, K., Gray, tion between men's economic status and their family and community
C., Hammond, M., Mitchell, S., Gotffried, A., Siegel, L., & Johnson, origins. Journal of Human Resources, 27, 575-601.
D. (1989). Home environment and cognitive development in the first Cowen, E. L., Wyman, P. A., Work, W. C., & Parker, G. R. (1990). The
3 years of life: A collaborative study involving six sites and three Rochester Child Resilience Project: Overview and summary of first
ethnic groups in North America. Developmental Psychology, 25, year findings. Development and Psychopathology, 2, 193-212.
217-235. Crooks, D. (1995). American children at risk: Poverty and its conse-
Bradley, R., Whiteside, L., Mundfrom, D., Casey, P., Kelleher, K., & quences for children's health, growth, and school achievement. Year-
Pope, S. (1994). Early indications of resilience and their relation to book of Physical Anthropology, 38, 57-86.
experiences in the home environments of low birthweight, premature Currie, J., & Thomas, D. (1995). Does Head Start make a difference?
children living in poverty. Child Development, 65, 346-360. American Economic Review, 85, 341-364.
Brantlinger, E. (1991). Social class distinctions in adolescents' reports
Danziger, S., & Danziger, S. (1993). Child poverty and public policy:
of problems and punishment in school. Behavioral Disorders, 17,
Toward a comprehensive antipoverty agenda. Daedalus: America's
36-46. Childhood, 122, 57-84.
Bronfenbrenner, U. (1986). Ecology of the family as a context for
Datcher, L. (1982). Effects of community and family background on
human development: Research perspectives. Developmental Psychol-
achievement. Review of Economics and Statistics, 64, 32-41.
ogy, 22, 723-742.
Bronfenbrenner, U., McClelland, P., Wethington, E., Moen, P., & Ceci, de Lone, R. (1979). Small futures: Children, inequality, and the limits
S. (1996). The state of Americans: This generation and the next. of liberal reform. New York: Harcourt Brace Jovanovich.
New York: Free Press. Deutsch, C. (1973). Social class and child development. In B. Cald-
Brookover, W., Beady, C., Flood, P., Schweitzer, J., & Wisenbaker, J. well & H. Ricciuti (Eds.), Review of child development research
(1979). School social systems and student achievement: Schools can (Vol. 3, pp. 233-282). Chicago: University of Chicago Press.
make a difference. New York: Praeger. Dix, T. (1991). The affective organization of parenting: Adaptive and
Brooks-Gunn, J., Duncan, G., Klebanov, P., &Sealand, N. (1993). Do maladaptative processes. Psychological Bulletin, 110, 3-25.
neighborhoods influence child and adolescent development? Ameri- Dodge, K., Pettit, G., & Bates, J. (1994). Socialization mediators of
can Journal of Sociology, 99, 353-395. the relation between socioeconomic status and child conduct prob-
Brooks-Gunn, J., Guo, G., & Furstenberg, E (1993). Who drops out lems. Child Development, 65, 649-665.
and who continues beyond high school? A 20-year follow-up of Dornbusch, S., Carlsmith, J., Bushwall, S., Ritter, P., Leiderman, H.,
Black urban youth. Journal of Research on Adolescence, 3, 271- Hastorf, A., & Gross, R. (1985). Single parents, extended households,
294. and the control of adolescents. Child Development, 56, 326-341.
Brooks-Gunn, J., Klebanov, P., & Liaw, E (1995). The learning, physi- Downey, G., & Coyne, J. (1990). Children of depressed parents: An
cal, and emotional environment of the home in the context of poverty: integrative review. Psychological Bulletin, 108, 50-76.
The Infant Health and Development Program. Children and Youth Dubow, E., & Ippolito, M. E (1994). Effects of poverty and quality of
Services Review, 17, 231-250. the home environment on changes in the academic and behavioral

200 F e b r u a r y 1998 • A m e r i c a n P s y c h o l o g i s t
 adjustment of elementary school-age children. Journal of Clinical Garbarino, J. (1992). The meaning of poverty in the world of children.
Child Psychology, 23, 401-412. American Behavioral Scientist, 35, 220-237.
Dubrow, N. E, & Garbarino, J. (1989). Living in the war zone: Mothers Garbarino, J., Kostelny, K., & Dubrow, N. (1991). What children can
and young children in a public housing development. Child Welfare, tell us about living in danger. American Psychologist, 46, 376-383.
68, 3-20. Garfinkel, I., & McLanahan, S. (1986). Single mothers and their chil-
Duncan, G. (1984). Years ofpoverty, years ofplenty. Ann Arbor: Univer- dren: A new American dilemma. Washington, DC: Urban Institute
sity of Michigan, Institute for Social Research. Press.
Duncan, G. (1991). The economic environment of childhood. In A. Garrett, P., Ng'andu, N., & Ferron, J. (1994). Poverty experiences of
Huston (Ed.), Children in poverty: Child development and public young children and the quality of their home environments. Child
policy (pp. 23-50). New York: Cambridge University Press. Development, 65, 331-345.
Duncan, G. (1994). Families and neighbors as sources of disadvantage Gottschalk, P. (1992). Is the intergenerational correlation in welfare
in the schooling decisions of White and Black adolescents. American participation across generations spurious? Unpublished manuscript,
Journal of Education, 103, 20-53. Boston College.
Duncan, G., & Brooks-Gunn, J. (Eds.). (1997), Consequences of grow- Gouldner, H. (1978). Teachers' pets, troublemakers, and nobodies:
ing up poor. New York: Russell Sage Foundation. Black children in elementary school. Westport, CT: Greenwood
Duncan, G., Brooks-Gunn, J., & Klebanov, E (1994). Economic depri- Press.
vation and early childhood development. Child Development, 65, Greenberg, M. (1996). No duty, no floor: The real meaning of "ending
296-318. entitlements." Washington, DC: Center for Law and Social Policy.
Duncan, G., Hill, M., & Hoffman, S. D. 0988, January 29). Welfare Hailer, M., Hollinger, E, & Raubai, O. (1990). Leviathan or welfare
This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.

dependence within and across generations. Science, 239, 467-471. state? Attitudes toward the role of government in six advanced West-
This document is copyrighted by the American Psychological Association or one of its allied publishers.

Duncan, G., & Rodgers, W. (1988). Longitudinal aspects of childhood ern nations. In J. Becker, J. Davis, P. Ester, & P. Mohler (Eds,),
poverty. Journal of Marriage and the Family, 50, 1007-1021. Attitudes to inequality and the role of government (pp. 33-62).
Duncan, G., & Rodgers, W. (1991). Has children's poverty become Rijswijk, The Netherlands: Sociaal en Cultureel Planbureau.
more persistent? American Sociological Review, 56, 538-550. Halpern, R. (1988). Parent support and education for low-income fami-
Duncan, G., & Yeung, W. (1995). Extent and consequences of welfare lies: Historical and current perspectives. Children and Youth Services
dependence among America's children. Children and Youth Services Review, 10, 283-303.
Review, 17, 157-182. Halpem, R. (1991). Supportive services for families in poverty: Dilem-
Duncan, G., Yeung, W., Brooks-Gunn, J., & Smith, J. (in press). How mas of reform. Social Service Review, 65, 343-364.
much does childhood poverty affect the life chances of children? Hanson, T., McLanahan, S., & Thomson, E. (1997). Economic re-
American Sociological Review. sources, parental practices, and children's well-being. In G. Dun-
can & J. Brooks-Gunn (Eds.), Consequences of growing up poor
Earls, E (1980). Prevalence of behavior problems in 3-year-old children:
(pp. 190-238). New York: Russell Sage Foundation.
A cross-national replication. Archives of General Psychiatry, 37,
Haskins, R. (1989). Beyond metaphor: The efficacy of early childhood
1153-1157.
education. American Psychologist, 44, 274-282.
Eccles, J., & Harold, R. (1993). Parent-school involvement during the Hauser, R., & Cart, D. (1995). Measuring poverty and socioeconomic
early adolescent years. Teachers College Record, 94, 568-587.
status in studies of health and well-being (Working Paper No. 94-
Eggebeen, D., & Lichter, D. (1991). Race, family structure, and chang-
24). Madison: University of Wisconsin, Center for Demography and
ing poverty among American children. American Sociological Re- Ecology.
view, 56, 801-817. Hausman, B., & Hammen, C. (1993). Parenting in homeless families:
Elder, G. (1974). Children of the Great Depression. Chicago: University The double crisis. American Journal of Orthopsychiatry, 63, 358-
of Chicago Press. 369.
Elder, G. (1997). The life course and human development. In W. Damon Haveman, R. H. (1987). Poverty policy andpoverty research. Madison:
(Series Ed.) & R. Lerner (Vol. Ed.), Handbook of child psychology: University of Wisconsin Press.
Vol. 1. Theoretical models of human development (5th ed., pp. 939- Haveman, R., & Wolfe, B. (1995). The determinants of children's at-
991). New York: Wiley. tainments: A review of methods and findings. Journal of Economic
Elder, G., & Caspi, A. (1988). Economic stress in lives: Developmental Literature, 33, 1829-1878.
perspectives. Journal of Social Issues, 44, 25-45. Hawley, T, & Disney, E. (1992). Crack's children: The consequences
Elder, G., Nguyen, T., & Caspi, A. (1985). Linking family hardship to of maternal cocaine abuse. Social Policy Report, 6(4), 1-22.
children's lives. Child Development, 56, 361-375. Hebbeler, K. (1985). An old and a new question on the effects of
Entwisle, D., & Alexander, K. (1990). Beginning school math compe- early education for children from low income families. Educational
tence: Minority and majority comparisons. Child Development, 61, Evaluation and Policy Analysis, 7, 207-216.
454-471. Hess, R., & Holloway, S. (1984). Family and school as educational
Entwisle, D., Alexander, K., & Olson, L. (1997). Children, schools, institutions. In R, D. Parke (Ed.), Review of child development re-
and inequality. Boulder, CO: Westview Press. search (Vol. 7, pp. 179-222). Chicago: University of Chicago Press.
Epstein, J. (1990). School and family connection: Theory, research, and Heyns, B. (1978). Summer learning and the effects of schooling. New
implications for interpreting sociologies of education and family. York: Academic Press.
Marriage and Family Review, 15, 99-126. Hill, M. A., & O'Neill, J. (1994). Family endowments and the achieve-
Escalona, S. K. (1984). Social and other environmental influences on ment of young children with special reference to the underclass.
the cognitive and personality development of low birthweight infants. Journal of Human Resources, 29, 1064-1100.
American Journal of Mental Deficiency, 88, 508-512. Hill, M.S., & Duncan, G. (1987). Parental family income and the
Fitzpatrick, K., & Boldizar, J. (1993). The prevalence and consequences socioeconomic attainment of children. Social Science Research, 16,
of exposure to violence among African-American youth. Journal of 39-73.
the American Academy of Child and Adolescent Psychiatry, 32, 424- Hofferth, S. (1994). Who enrolls in Head Start? A demographic analysis
430. of Head Start-eligible children. Early Childhood Research Quarterly,
Frank, R., Strobino, D., Salkever, D., & Jackson, C. (1992). Updated 9, 243-268.
estimates of the impact of prenatal care on birthweight outcomes by Hoffman, L.W. (1985). The changing genetics/socialization balance.
race. Journal of Human Resources, 27, 631-642. Journal of Social Issues, 41, 127-148.
Fuerst, J., & Fuerst, D. (1993). Chicago experience with an early child- Homel, R., & Bums, A. (1987). Is this a good place to grow up in?
hood program: The special case of the Child Parent Center Program. Neighborhood quality and children's evaluations. Landscape and Ur-
Urban Education, 28, 69-96. ban Planning, 14, 101-116.
Fusfeld, D., & Bates, T. (1984). The political economy of the urban Homel, R., & Bums, A. (1989). Environmental quality and the wellbe-
ghetto. Carbondale: Southern Illinois University Press. ing of children. Social Indicators Research, 21, 133-158.

F e b r u a r y 1998 • A m e r i c a n P s y c h o l o g i s t 201
 House, J. (1981). Social structure and personality. In M. Rosenberg & Martin, E, & Midgley, E. (1994). Immigration to the United States:
R. Turner (Eds.), Social psychology: Sociological perspectives (pp. Journey to an uncertain destination. Population Bulletin, 49(2), 2 -
525-561). New York: Basic Books. 45.
Huston, A. (1991). Children in poverty: Developmental and policy is- Massey, D. (1994). America's apartheid and the urban underclass. Social
sues. In A. Huston (Ed.), Children in poverty: Child development Service Review, 68, 471-487.
andpublic policy (pp. 1- 22). New York: Cambridge University Press. Masten, A., Miliotis, S., Graham-Bermann, S., Ramirez, M., & Nee-
Huston, A., McLoyd, V. C., & Garcia Coil, C. (1994). Children and mann, J. (1993). Children in homeless families: Risks to mental
poverty: Issues in contemporary research. Child Development, 65, health and development. Journal of Consulting and Clinical Psychol-
275-282. ogy, 61, 335-343.
Huston, A., McLoyd, V.C., & Garcia Coil, C. (1997). Poverty and Masten, A., Morison, E, Pelligrini, D., & Tellegen, A. (1990). Compe-
behavior: The case for multiple methods and levels of analysis. Devel- tence under stress: Risk and protective factors. In J. Rolf, A. S. Mas-
opmental Review, 17, 376-393. ten, D. Cicchetti, K. Nuechterlein, & S. Weintraub (Eds.), Risk and
Illback, R. (1994). Poverty and the crisis in children's services: The protective factors in the development of psychopathology (pp. 236-
need for services integration. Journal of Clinical Child Psychology, 256). Cambridge, England: Cambridge University Press.
23, 413-424. Mayer, S. (1997). What money can't buy. Cambridge, MA: Harvard
James, S. D. (1985). The impact of cybernation technology on Black University Press.
automobile workers in the U.S. Ann Arbor: UMI Research Press. McGauhey, E, Starfield, B., Alexander, C., & Ensminger, M. (1991).
Jargowsky, P, (1994). Ghetto poverty among Blacks in the 1980s. Jour- Social environment and vulnerability of low birth weight children:
nal of Policy Analysis and Management, 13, 288-310. A social-epidemiological perspective. Pediatrics, 88, 943-953.
This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.

Jarrett, R. (1995). Growing up poor: The family experiences of socially McKey, R., Condelli, L., Ganson, H., Barrett, B., McConkey, C., &
This document is copyrighted by the American Psychological Association or one of its allied publishers.

mobile youth in low-income African American neighborhoods. Jour- Plantz, M. (1985). The impact of Head Start on children, families,
nal of Adolescent Research, 10, 111 - 135. and communities: Final report of the Head Start Evaluation, Synthe-
Jencks, C., & Mayer, S. (1990). The social consequences of growing sis and Utilization Project (DHHS Publication No. OHDS 90-
up in a poor neighborhood: A review. In M. McGeary & L. Lynn 31193). Washington, DC: U.S. Government Printing Office.
(Eds.), Inner city poverty in the United States. Washington, DC: McLeod, J., & Shanahan, M. (1993). Poverty, parenting, and children's
National Academy Press. mental health. American Sociological Review, 58, 351-366.
Johnson, D.L. (1988). Primary prevention of behavior problems in McLoyd, V.C. (1990). The impact of economic hardship on Black
young children: The Houston Parent-Child Development Center. In families and children: Psychological distress, parenting, and socio-
R. Price, E. Cowen, R. Lorion, & J. Ramos-McKay (Eds.), 14 ounces emotional development. Child Development, 61, 311-346.
of prevention: A casebook for practitioners (pp. 44-52). Washington, McLoyd, V. C. (1993). Employment among African American mothers
DC: American Psychological Association. in dual-earner families: Antecedents and consequences for family life
Knox, V. W., & Bane, M. (1994). Child support and schooling. In I. and child development. In J. Frankel (Ed,), The employed mother
Garfinkel & S. McLanahan (Eds.), Child support and child well- and the family context (pp. 180-226). New York: Springer-Verlag.
being (pp. 285-316). Washington, DC: Urban Institute Press. McLoyd, V. (1997). Children in poverty: Development, public policy,
Korenman, S., Miller, J., & Sjaastad, J. (1995). Long-term poverty and and practice. In W. Damon (Series Ed.), I. Sigel, & A. Renninger
child development in the United States: Results from the NLSY. (Vol. Eds.), Handbook of child psychology: Vol. 4. Child psychology
Children and Youth Services Review, 17, 127-155. in practice (5th ed., pp. 135-208). New York: Wiley.
Kotlowitz, A. (1991). There are no children here. New York: Doubleday. McLoyd, V. C., Ceballo, R., & Mangelsdorf, S. (1996). The effects of
Kozol, J. (1991). Savage inequalities: Children in America's schools. poverty on children's socioemotional development. In J. Noshpitz
New York: Crown. (Series Ed.) & N. Alessi (Vol. Ed.), Handbook of child and adolescent
Laily, J. R., Mangione, P., & Honig, A. (1988). The Syracuse University psychiatry: Vol. 4. Varieties of development (pp. 189-206). New
Family Development Research Program: Long-range impact of an York: Wiley.
early intervention with low-income children and their families. In I. McLoyd, V. C., Jayaratne, T., Ceballo, R., & Borquez, J. (1994). Unem-
Sigel (Series Ed.) & D. Powell (Vol. Ed.), Advances in applied ployment and work interruption among African American single
developmental psychology: Vol. 3. Parent education as early child- mothers: Effects on parenting and adolescent socioemotional func-
hood intervention: Emerging directions in theory, research, and prac- tioning. Child Development, 65, 562-589.
tice (pp. 79-104). Norwood, NJ: Ablex. Meisels, S. J., & Plunkett, J. W. (1988). Developmental consequences
Lee, V., Brooks-Gunn, J., & Schnur, E. (1988). Does Head Start work? of preterm birth: Are there long-term effects? In E B. Baltes, D. L.
A 1-year follow-up comparison of disadvantaged children attending Featherman, & R. M. Lerner (Eds.), Life-span development and be-
Head Start, no preschool, and other preschool programs. Develop- havior (Vol. 9, pp. 87-128). Hillsdale, NJ: Erlbaum.
mental Psychology, 24, 210-222. Mueller, C., & Parcel, T. (1981). Measures of socioeconomic status:
Lee, V., Brooks-Gunn, J., Schnur, E., & Liaw, E (1990). Are Head Alternatives and recommendations. Child Development, 52, 13-30.
Start effects sustained? A longitudinal follow-up comparison of dis- Murray, C. B., & Jackson, J. S. (1982/1983). The conditioned failure
advantaged children attending Head Start, no preschool, and other model of Black educational underachievement. Humboldt Journal of
preschool programs. Child Development, 61, 495-507. Social Relations, 10, 276-300.
Lee, V., & Croninger, R. (1994). The relative importance of home and Needleman, H. L., Schell, A., Bellinger, D., Leviton, A., & Allred, E.
school in the development of literacy skills for middle-grade students. (1990). The long-term effects of low doses of lead in childhood: An
American Journal of Education, 102, 286-329. 11-year follow-up report. New England Journal of Medicine, 322,
Lee, V., & Loeb, S. (1995). Where do Head Start attendees end up? 83-88.
One reason why preschool effects fade out. Educational Evaluation Olds, D., & Kitzman, H. (1993). Review of research on home visiting
and Policy Analysis, 17, 62-82. for pregnant women and parents of young children. The Future of
Lempers, J., Clark-Lempers, D., & Simons, R. (1989). Economic hard- Children, 3, 53-92.
ship, parenting, and distress in adolescence. Child Development, 60, Osofsky, J. (1995). The effects of exposure to violence on young chil-
25 -49. dren. American Psychologist, 50, 782-788.
Liaw, E, & Brooks-Gunn, J. (1994). Cumulative familial risks and Oyemade, U. J. (1985). The rationale for Head Start as a vehicle for
low-birthweight children's cognitive and behavioral development. the upward mobility of minority families: A minority perspective.
Journal of Clinical Child Psychology, 23, 360-372. American Journal of Orthopsychiatry, 55, 591-602.
Loeber, R., & Dishion, T. (1983). Early predictors of male delinquency: Pagani, L., Boulerice, B., & Tremblay, R. (1997). The influence of
A review. Psychological Bulletin, 94, 68-99. poverty on children's classroom placement and behavior problems.
MacLeod, J. (1987). Ain't no makin' it: Leveled aspirations in a low- In G. Duncan & J. Brooks-Gunn (Eds.), Consequences of growing
income neighborhood. Boulder, CO: Westview Press. up poor (pp. 311-339). New York: Russell Sage Foundation.

202 F e b r u a r y 1998 • A m e r i c a n P s y c h o l o g i s t
 Patterson, C., Kupersmidt, J., & Vaden, N. (1990). Income level, gender, Scarr, S. (1993). Biological and cultural diversity: The legacy of Darwin
ethnicity, and household composition as predictors of children's for development. Child Development, 64, 1333-1353.
school-based competence. Child Development, 61, 485-494. Schorr, L. (1989). Within our reach: Breaking the cycle of disadvantage.
Pear, R. (1997, February 23). Rewards and penalties vary in states' New York: Doubleday.
welfare programs. The New York ~mes, p. 26. Schweinhart, L. J., Barnes, H., & Weikart, D. (1993). Significant bene-
Portes, A., & Zhou, M. (1993). The new second generation: Segmented fits: The High/Scope Perry Preschool Study through age 27. Mono-
assimilation and its variants. Annals of the American Academy of graphs of the High~Scope Educational Research Foundation, 10.
Political and Social Sciences, 530, 74-96. Seitz, V., Rosenbanm, L., & Apfel, N. (1985). Effects of family support
Posner, J. K., & Vandell, D. (1994). Low-income children's after-school intervention: A ten-year follow-up. Child Development, 56, 376-
care: Are there beneficial effects of after-school programs? Child 391.
Development, 65, 440-456. Shaw, D., & Emery, R. (1988). Chronic family adversity and school-
Pryor-Brown, L., Cowen, E., Hightower, A., & Lotyczewski, B. (1986). age children's adjustment. Journal of the American Academy of Child
Demographic differences among children in judging and experienc- and Adolescent Psychiatry, 27, 200-206.
ing specific stressful life events. Journal of Special Education, 20, Sherman, A. (1994). Wasting America "sfuture: The Children 'S Defense
339-346. Fund report on the costs of child poverty. Boston: Beacon Press.
Pryor-Brown, L., Powell, J., & Earls, E (1989). Stressful life events Shinn, M., & Gillespie, C. (1994). The roles of housing and poverty
and psychiatric symptoms in Black adolescent females. Journal of in the origins of homelessness. American Behavioral Scientist, 37,
Adolescent Research, 4, 140-151. 505-521.
Ramey, C., Ramey, S., Gaines, K., & Blair, C. (1995). Two-generation Siegel, L. (1982). Reproductive, perinatal, and environmental factors
This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.

early intervention programs: A child development perspective. In as predictors of the cognitive and language development of preterm
This document is copyrighted by the American Psychological Association or one of its allied publishers.

I. Sigel (Series Ed.) & S. Smith (Vol. Ed.), Advances in applied and full-term infants. Child Development, 53, 963-973.
developmental psychology: Vol. 9. Two-generation programs for Smeeding, T. M. (1992, January/February). Why the U.S. antipoverty
families in poverty: A new intervention strategy (pp. 199-228). Nor- system doesn't work very well. Challenge, 35, 30-35.
wood, NJ: Ablex.
Smith, J., Brooks-Gunn, J., & Klebanov, P. (1997). Consequences of
Randolph, S., & Adams-Taylor, S. (in press). The health status of chil-
living in poverty for young children's cognitive and verbal ability and
dren and adolescents in urban environments. In G. King & W. Davis
early school achievement. In G. Duncan & J. Brooks-Gunn (Eds.),
(Eds.), The health of Black America: Social causes and conse-
Consequences of growing up poor (pp. 132-189). New York: Russell
quences. New York: Oxford University Press. Sage Foundation.
Reddy, M. A. (Ed.). (1993). Statistical record of Hispanic Americans.
Detroit, MI: Gale Research. Smith, S., & Zaslow, M. (1995). Rationale and policy context for two-
generation interventions. In I. Sigel (Series Ed.) & S. Smith (Vol.
Reynolds, A. (1991). Early schooling of children at risk. American
Educational Research Journal, 28, 392-422. Ed.), Advances in applied developmental psychology: Vol. 9. Two-
generation programs for families in poverty: A new intervention
Reynolds, A. (1994). Effects of a preschool plus follow-on intervention
strategy (pp. 1-35). Norwood, NJ: Ablex.
for children at risk. Developmental Psychology, 30, 787-804.
Richters, J. E., & Martinez, P. (1993). The NIMH Community Violence Steele, C. (1992, April). Race and the schooling of Black Americans.
Project: Vol. 1. Children as victims of and witnesses to violence. Atlantic Monthly, 68-78.
Psychiatry, 56, 7-21. Sterling, S., Cowen, E.L., Weissberg, R.P., Lotyczewski, B.S., &
Rist, R. (1970). Student social class and teacher expectations: The self- Boike, M. (1985). Recent stressful life events and young children's
fulfilling prophecy in ghetto education. Harvard Educational Review, school adjustment. American Journal of Community Psychology, 13,
40, 411-451. 87-98.
Rodgers, J., & Rodgers, J.L. (1993). Chronic poverty in the United Stevenson, J., Richman, N., & Graham, P. (1985). Behaviour problems
States. Journal of Human Resources, 28, 25-54. and language abilities at three years and behavioural deviance at
Rowe, D., & Rodgers, J. (1997). Poverty and behavior: Are environmen- eight years. Journal of Child Psychology and Psychiatry, 26, 215-
tal measures nature and nurture? Developmental Review, 17, 358- 230.
375. Sugland, B., Zaslow, M., Smith, J., Brooks-Gunn, J., Coates, D., Blu-
Royce, J., Darlington, R., & Murray, H. (1983). Pooled analyses: Find- menthal, C., Moore, K., Griffin, T., & Bradley, R. (1995). The Early
ings across studies. In Consortium for Longitudinal Studies (Ed.), Childhood HOME Inventory and HOME--Short Form in differing
As the twig is bent." Lasting effects of preschool programs (pp. 411- racial/ethnic groups: Are there differences in underlying structure,
459). Hillsdale, NJ: Erlbaum. internal consistency of subscales, and patterns of prediction? Journal
Ruggles, P. (1990). Drawing the line: Alternative poverty measures of Family Issues, 16, 632-663.
and their implications for public policy. Washington, DC: Urban Takanishi, R., & DeLeon, P. (1994). A Head Start for the 21st century.
Institute Press. American Psychologist, 49, 120-122.
Rumbaut, R. (1994). The crucible within: Ethnic identity, self-esteem, Tienda, M. (1991). Poor people and poor places: Deciphering neighbor-
and segmented assimilation among children of immigrants. Interna- hood effects on poverty outcomes. In J. Huber (Ed.), Macro-micro
tional Migration Review, 28, 748-794. linkages in sociology (pp. 244-262). Newbury Park, CA: Sage.
Rutter, M. (1983). School effects on pupil progress: Research findings • U.S. Bureau of the Census. (1994). Statistical abstract of the United
and policy implications. Child Development, 54, 1-29. States: 1994. Washington, DC: U.S. Government Printing Office.
Rutter, M. (1990). Psychosocial resilience and protective mechanisms. U.S. Bureau of the Census. (1996). Statistical abstract of the United
In J. Rolf, A. S. Masten, D. Cicchetti, K. Nuechterlein, & S. Wein- States: 1996. Washington, DC: U.S. Government Printing Office.
traub (Eds.), Risk and protective factors in the development of psy- Vaughan, D. (1993). Exploring the use of the public's views to set
chopathology (pp. 181-215). Cambridge, England: Cambridge Uni- income poverty thresholds and adjust them over time. Social Security
versity Press. Bulletin, 56, 22-46.
Salkind, N., & Haskins, R. (1982). Negative income tax: The impact Weissbourd, B. (1987). A brief history of family support programs. In
on children from low-income families. Journal of Family Issues, 3, S. Kagan, D. Powell, B. Weissbourd, & E. Zigler (Eds.), America's
165-180. family support programs (pp. 38-56). New Haven, CT: Yale Univer-
Sameroff, A. (1986). Environmental context of child development. sity Press.
Journal of Pediatrics, 109, 192-200. Werner, E., & Smith, R. (1982). Vulnerable but invincible: A study of
Sameroff, A., Seifer, R., Barocas, R., Zax, M., & Greenspan, S. (1987). resilient children. New York: McGraw-Hill.
Intelligence quotient scores of 4-year-old children: Social-environ- White, K. (1982). The relation between socioeconomic status and aca-
mental risk factors. Pediatrics, 79, 343-350. demic achievement. Psychological Bulletin, 91, 461-481.
Scarf, S. (1981). Race, social class, and individual differences in IQ. Williams, T., & Kornblum, W. (1985). Growing up poor. Lexington,
Hillsdale, NJ: Erlbaum. MA: Lexington Books.

February 1998 • American Psychologist 203

 Wilson, J. Q., & Herrnstein, R.J. (1985). Crime and human nature. Zigler, E. (1994). Reshaping early childhood intervention to be a more
New York: Simon & Schuster. effective weapon against poverty. American Journal of Community
Wilson, W.J. (1987). The truly disadvantaged: The inner city, the Psychology, 22, 37-47.
underclass, and public policy. Chicago: University of Chicago Press. Zigler, E., & Berman, W. (1983). Discerning the future of early child-
Wilson, W. J. (1996). When work disappears: The world of the new hood intervention. American Psychologist, 33, 894-906.
urban poor. New York: Knopf. Zigler, E., & Styfco, S. (1994). Head Start: Criticisms in a constructive
Wilson, W. J., & Neckerman, K. (1986). Poverty and family structure: context. American Psychologist, 49, 127-132.
The widening gap between evidence and public policy issues. In S. Zigler, E., Taussig, C., & Black, K. (1992). Early childhood interven-
Danziger & D. Weinberg (Eds.), Fighting poverty: What works and tion: A promising preventative for juvenile delinquency. American
what doesn't (pp. 232-259). Cambridge, MA: Harvard University Psychologist, 47, 997-1006.
Press.
Wiltfang, G., & Scarbecz, M. (1990). Social class and adolescents' Zill, N. (1984). American children: Happy, healthy and insecure. New
self-esteem: Another look. Social Psychology Quarterly, 53, 174- York: Doubleday Anchor.
183. Zill, N., Moore, K., Smith, E., Stief, T., & Coiro, M. (1995). The life
Yoshikawa, H. (1994). Prevention as cumulative protection: Effects of circumstances and development of children in welfare families: A
early family support and education on chronic delinquency and its profile based on national survey data. In P. L. Chase-Lansdale & J.
risks. Psychological Bulletin, 115, 28-54. Brooks-Gunn (Eds.), Escape from poverty: What makes a difference
Yoshikawa, H. (1995). Long-term effects of early childhood programs for children? (pp. 38-59). New York: Cambridge University Press.
on social outcomes and delinquency. The Future of Children, 5, 51 - Zinsmeister, K. (1990, June). Growing up scared. Atlantic Monthly,
75. 49-66.
This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.
This document is copyrighted by the American Psychological Association or one of its allied publishers.

204 F e b r u a r y 1998 • A m e r i c a n P s y c h o l o g i s t