PreprintPDF Available

Racial/Ethnic Differences in the Association Between Family Structure and Children's Education

November 2019

Authors:

Harvard University

Abstract: While an extensive literature has shown that children raised by both biological parents fare better academically than children raised in any other family structure, there has been little research to explain an important finding: living apart from a biological parent is less negatively consequential for racial/ethnic minority children than white children. To address this gap, I test two explanations that have been posited to account for racial/ethnic differences in the association between family structure and children’s educational attainment: socioeconomic stress and extended family embeddedness. I assess whether racial/ethnic variation in these two mechanisms explain group differences in the association between family structure and on-time high school completion and college enrollment for white, black, and Hispanic children. Results indicate that both socioeconomic stress and extended family embeddedness attenuate the effect of family structure on these two measures of educational attainment, though the former to a much greater extent. Differences in socioeconomic resources accounted for up to nearly 50% of the gap in these outcomes, and extended family embeddedness explained roughly 15-20%. These findings lend support for the socioeconomic stress hypothesis, which posits that the negative effect of familial disruption may be less independently impactful for racial/ethnic groups facing many socioeconomic disadvantages to begin with. Results are less consistent with the hypothesis that racial/ethnic minority children’s deeper embeddedness in their extended family network protects against the negative effects of familial disruption.

Content uploaded by Christina Cross

Content may be subject to copyright.

Racial/Ethnic Differences in the Association Between Family Structure and Children’s

Education

Christina J. Cross

Department of Sociology

Center for Population and Development Studies

Harvard University

33 Kirkland Street

Cambridge, MA 02138

crossc@fas.harvard.edu

ABSTRACT

While an extensive literature has shown that children raised by both biological parents fare better

academically than children raised in any other family structure, there has been little research to

explain an important finding: living apart from a biological parent is less negatively

consequential for racial/ethnic minority children than white children. To address this gap, I test

two explanations that have been posited to account for racial/ethnic differences in the association

between family structure and children’s educational attainment: socioeconomic stress and

extended family embeddedness. I assess whether racial/ethnic variation in these two mechanisms

explain group differences in the association between family structure and on-time high school

completion and college enrollment for white, black, and Hispanic children. Results indicate that

both socioeconomic stress and extended family embeddedness attenuate the effect of family

structure on these two measures of educational attainment, though the former to a much greater

extent. Differences in socioeconomic resources accounted for up to nearly 50% of the gap in

these outcomes, and extended family embeddedness explained roughly 15-20%. These findings

lend support for the socioeconomic stress hypothesis, which posits that the negative effect of

familial disruption may be less independently impactful for racial/ethnic groups facing many

socioeconomic disadvantages to begin with. Results are less consistent with the hypothesis that

racial/ethnic minority children’s deeper embeddedness in their extended family network protects

against the negative effects of familial disruption.

Keywords: Family Structure, Race/Ethnicity, Extended Family, Educational Attainment, Family

Diversity, Child Wellbeing

Introduction

Decades of research have shown that children fare better academically when they

are raised in a two-biological-parent family (i.e., a household with a child and his or her

birth mother and biological father; McLanahan & Percheski, 2008). Children who live

apart from a parent during childhood are less likely to finish high school, attend college,

or earn a bachelor’s degree (Amato & Keith, 1991; Hetherington, Camara, &

Featherman, 1983; Manning & Lamb, 2003; McLanahan & Sandefur, 1994). A growing

body of research has demonstrated, however, that the effects of family structure are not

equal across all groups. Studies have found that this negative association between

familial disruption and children’s educational attainment is significantly weaker for

minority children than white children (e.g., Amato, 2000; Shaw, Winslow, & Flanagan,

1999; Sun & Li, 2007). While they have yet to be formally tested, two main mechanisms

have been posited to explain racial/ethnic differences in the effect of family structure on

children’s academic outcomes: socioeconomic stress and extended family embeddedness

(McLoyd, Cauce, Takeuchi, & Wilson, 2000).

Using nationally representative, longitudinal data this study investigates the extent

to which racial/ethnic variation in these two mechanisms mediates the association

between family structure and two important measures of educational attainment: on-time

high school completion and college enrollment. This paper builds on previous work in

several key ways. First, unlike prior research that typically uses cross-sectional or short-

term measures of family structure, I construct children’s full histories of family structure

(from birth to age 17), including both nuclear and extended family household

configurations. My analyses also include lifetime measures of a wide range of family

factors including parents’ employment and health statuses and family wealth. This allows

me to better articulate how cumulative childhood family experiences influence long-term

outcomes for children. Second, this study moves beyond black-white comparisons that

have traditionally characterized research in this area to incorporate Hispanics, who now

represent the largest minority group in the U.S. (Passel & Cohn, 2008 [inconsistencies in

racial categorizations over time in the dataset and small cell sizes do not permit inclusion

of other racial/ethnic groups]). Given group differences in levels of exposure to

socioeconomic stress, participation in extended family networks, and rates of two-parent

households, including Hispanic families provides a unique opportunity to compare

families of color to one another and to non-Hispanic whites and to assess whether

patterns of association for Hispanic children are similar to those of black or white

children, or altogether distinctive. Third, as it relates to measures of extended family

embeddedness, instead of using coresidence with extended relatives as the sole proxy for

extended family support, I also include indicators of the levels of practical and emotional

support that parents receive from family members. Including additional measures of

support allows me to better approximate the total amount of extended family support that

is available to parents, both within and across households.

Understanding whether racial/ethnic differences in socioeconomic and extended

family resources help explain the differential association between family structure and

children’s educational attainment has implications for research and policy. To begin, a

focus on group differences in the effects of family structure advances a more diverse

portrait of American families, one that more adequately reflects the longstanding and

growing racial and ethnic diversity of the United States, which has been lacking in extant

family research (Gerstel, 2011; McLoyd et al., 2000). It also provides an important

corrective to work that assumes the primacy of the white middle class two-biological-

parent family model (Coontz, 2004; Roschelle, 1999). If minority children are less

vulnerable to the adverse effects of parental separation than white children, then

valorizing the two-biological-parent family over other family forms is unwarranted and

may prove to be unfruitful in reducing racial/ethnic disparities in child outcomes. Further,

contemporary policy discussions about family structure center on questions concerning

the role of the two-parent family in promoting child wellbeing. (U.S. House of

Representatives, 1996). If racial/ethnic differences in socioeconomic resources mediate

the relationship between family structure and educational attainment, then what deserves

policy attention is not minority families’ deviation from the two-biological-parent family

model, but rather the social structures that produce and maintain racialized

socioeconomic inequities. Moreover, many welfare reform policies assume that extended

families will act as a social safety net to their families during hard times (Cherlin &

Seltzer, 2014). However, recent studies have called into question the viability of the

extended family safety net (Garrett-Peters & Burton, 2016; McDonald & Armstrong,

2001). Examining the extent to which embeddedness in one’s extended family network

serves as a buffer against the negative effects of familial disruption can better inform

policies that encourage reliance on extended family networks to enhance child wellbeing.

Background and Theoretical Perspectives

Family Structure and Children’s Educational Attainment

Children raised in two-biological-parent families tend to fare better academically

than children raised in any other family form (Brown, 2010). In their landmark study on

single-parent families and child wellbeing, McLanahan and Sandefur (1994) found that

youth living with both biological parents were less likely to drop out of high school and

more likely to enroll in college, compared to youth living with single or divorced parents

(though the negative effect of family structure was somewhat smaller for college

enrollment than high school graduation). Subsequent studies have produced similar

findings related to these outcomes (e.g., Deleire & Kalil, 2002; Ginther & Pollak, 2004;

Hill, Yeung, & Duncan, 2001).

Theoretical explanations that account for why children growing up outside of a

two-biological-parent family are at an elevated risk of experiencing academic problems

can be classified into three categories: (1) economic resources (2) parental socialization,

and (3) stress and family instability (Amato, 2005; Brown, 2010; Fomby & Cherlin,

2007; Wu, 1996). Economic explanations maintain that some family structures,

particularly single-parent families, tend to be economically disadvantaged, having fewer

economic resources such as income and wealth available to facilitate effective family

functioning. Economic deprivation undermines parents’ ability to provide the material

goods and services needed to cultivate children’s academic growth (Amato, 2005).

Parental socialization explanations focus on child-rearing, foremost parenting practices.

The argument here is that there is typically a weaker parental authority structure in

single-, step-, and cohabiting families than in two-biological-parent families, and there is

less monitoring and supervision, which in turn reduces the quality of parenting that

children receive (Amato 2005, Brown 2010; Sandberg and Hofferth, 2001). Family

instability explanations assert that the number of family transitions a child experiences

during childhood, independent of family type, has negative consequences for children’s

academic performance. The idea behind this is that children fare better when family

structure is stable; transitions lead to stress due to people moving in and out of the

household, potentially disrupting family routines and leading to new role configurations.

This familial disruption may also be accompanied by residential and school changes, that

potentially further undermine children’s educational outcomes (Fomby and Cherlin,

2007; Teachman, 2008; Wu 1996).

In addition to these explanations, researchers have noted that selection may also

account for the association between family structure and children’s academic outcomes.

This perspective suggests that it is not family structure or instability per se that influence

child wellbeing. Rather, children perform better in two-biological-parent families because

adults who form and maintain such households may be more well-adjusted and

socioeconomically advantaged in the first place (Brown, 2010; Hofferth, 2006).

Racial/Ethnic Disparities in the Effects of Family Structure on Educational Outcomes

While numerous studies have examined the relationship between family structure

and children’s educational attainment, only a few have investigated whether the effect of

family structure varies across racial/ethnic groups. Among those focused on racial/ethnic

differences, scholars generally find that the negative effects of living outside of a two-

biological-parent family are weaker for minority children than white children. For

example, in a rigorous meta-analysis of research on the long-term effects of divorce,

Amato & Keith (1991) found that the magnitude of the association between parental

divorce and educational attainment was nearly twice that for white children than for black

children. Amato (2001) found similar racial differences in his follow-up meta-analysis a

decade later. In another study focused on high school academic achievement, Smith

(1997) observed a stronger effect of parental separation on children’s grades for white

youth, relative to their black peers. Further, McLanahan and Sandefur (1994) found that

the proportionate increase in risk of dropping out of high school due to familial disruption

was significantly greater for whites than for blacks or Hispanics. However, for high

school idleness, they found that the effect of familial disruption was stronger for blacks

than whites. This latter finding suggests that the differential association between family

structure and educational outcomes may depend on the outcome of interest under

investigation. Two major mechanisms have been hypothesized to explain why minority

children may be less vulnerable to the negative consequences of living outside of a two-

biological-parent family than white children: socioeconomic stress and extended family

embeddedness.

Socioeconomic stress hypothesis. In the U.S., race and ethnicity are fundamental

axes of inequality along which socioeconomic resources are distributed (Bobo, Kluegel,

& Smith, 1997; Bonilla-Silva, 2006; Conley, 2009). Due to historic and contemporary

structural racism, minority families on average have less access to socioeconomic

resources such as income, wealth, and quality neighborhoods than white families

(Charles, 2006; Krysan, Couper, Farley, & Foreman, 2009; Mazumder, 2008; Proctor,

Semega & Kollar, 2016; Wilson, 1987). Because of this, minority children are more

likely to be exposed to socioeconomically stressful environments, and some scholars

have suggested that their exposure to sustained socioeconomic deprivation means that the

additional stress incurred by living apart from a parent is only marginally impactful,

above and beyond existing social disadvantages (McLoyd et al., 2000; Smith, 1997). This

is referred to as the socioeconomic stress hypothesis.

This hypothesis is derived from family stress theory, which focuses on a family’s

ability to cope with stressful situations and crises. It maintains that a families’ ability to

respond and adapt to stressors depends on the nature and number of stressful events, the

family resources available to facilitate coping, and family members’ perception of the

event (McCubbin, 1993; McCubbin & McCubbin, 1989; Friedman 1998; Fomby et al.

2010). Given the cumulative number of socioeconomic stressors that minority children

face, and the extent to which the pervasiveness of these stressors is more normative, one

might expect that the independent effect of an event like parental divorce to be smaller

for this group. On the other hand, for children of a racial/ethnic group who on average

benefit from greater access to socioeconomic resources (i.e., white children), one might

expect that the effect of familial disruption may be more pronounced.

Extended family embeddedness hypothesis. Several studies suggest that minority

children are more deeply embedded in extended family networks compared to white

children (Burton, 1992; Hunter, 1997; Margolis, Fosco, & Stormshak, 2014; Sarkisian &

Gerstel, 2012; Stack 1974; Taylor, 1986). Indeed, research on the extended family

networks of Black Americans finds that this group is engaged in ongoing and reciprocal

exchanges of practical, emotional, and financial support, and black families are more

involved in practical support (e.g., help with household chores, transportation, and child

care) than white families (Jayakody, Chatters, & Taylor, 1993; Sarkisian & Gerstel, 2004;

Stack & Burton, 1993; Taylor, Chae, Lincoln, & Chatters, 2015). They also tend to live

in closer proximity to extended relatives than whites (Farber, Miller-Cribbs, & Reitmeier,

2005). Although limited, extant research on Hispanic extended family networks suggest

that they are more integrated and may offer higher levels of support than white extended

family networks (Baca Zinn & Wells, 2000; Kamo, 2000). Greater involvement in these

extended family networks may reduce of some the negative psychosocial effects

associated with parental conflict or separation (Lamborn & Nguyen, 2004; Pittman, 2007;

McLoyd et al., 2000; Smith, 1997). This is known as the extended family embeddedness

hypothesis.

This hypothesis is grounded in social capital theory, which asserts that social

capital within the family, that is supportive relations between parents and children, is

crucial for children’s intellectual growth (Coleman, 1988). Social capital within the

family gives children access to other forms of parental capital that help facilitate

children’s educational performance (i.e., economic, cultural, and human capital)

(Coleman, 1988). The absence of a parent from the household may diminish the

availability of these resources to children, and thus undermine their academic success.

Extended family networks, however, may increase the number of nurturant and

supportive adults available in the child’s life to provide such resources (McLoyd, et al.

2000). Thus, to the extent that extended family networks contribute additional resources

such as monitoring and supervision and familial involvement in education, one would

expect that embeddedness in these networks may weaken or even countervail the

negative effects of parental absence. However, if demands for support exchanges increase

stress and diminish household resources, one would anticipate that greater embeddedness

in extended family networks would adversely affect children’s educational attainment

(Pilkauskas, Campbell, & Wimer, 2017; Dominguez & Watkins, 2003).

The Current Study

While plausible, the relative importance of these mechanisms in explaining racial

differences in the effects of family structure has yet to be examined (but see Fomby et al.,

2010 for an investigation of racial/ethnic differences in the effects of family instability on

children’s risk behaviors and see Sun & Li, 2007 for a test of the socioeconomic stress

hypothesis on parental divorce/separation during late adolescence). This study seeks to

overcome this limitation by using nationally representative longitudinal data to assess

whether racial/ethnic differences in socioeconomic stress and extended family

embeddedness account for group differences in the association between family structure

and children’s educational attainment. I begin by investigating whether children’s

lifetime experience of family structure influences their on-time high school completion

and college enrollment. I then determine if any observed associations between family

structure and these two measures of educational attainment differ by race/ethnicity.

Finally, I evaluate the extent to which factors associated with socioeconomic stress and

extended family embeddedness explain any group differences in associations by family

structure.

Method

Data

This study draws on data from the Panel Study of Income Dynamics (1985-2015)

and its two youth-centered supplements, the Child Development Supplement (CDS)

(1997-2007) and the Transition into Adulthood Supplement (TAS) (2005-2015). The

PSID began in 1968 as a nationally-representative sample of nearly 5,000 U.S.

households. Original sample members and their descendants were followed annually until

1997 and have been followed biennially since then. To maintain population

representativeness, in 1997, a sample refresher added approximately 500 households

headed by immigrants who had entered the United States since 1968. At each wave, the

household head or the spouse or cohabiting partner of the head reports on household

composition, and household members’ employment, income, educational attainment, and

health status. In 2015, the study collected information on nearly 25,000 individuals in

approximately 9,000 households.

While the PSID has always collected some information about children, in 1997,

the PSID supplemented its main data collection with the CDS, which provides additional

information on up to two children aged 0 to 12 years per household. Information is

obtained through interviews with primary and secondary caregivers and with older

children, and through assessments and interviewer observations. Children and their

caregivers were re-interviewed in 2002 and 2007, or until children reached age 18. In

2005, the Transition to Adulthood Supplement (TAS) was developed to continue

following the original CDS cohort when they aged out of CDS or left high school. TAS

has continued to follow respondents biennially as they complete their educations and

enter the labor force. In 2015, TAS respondents were between 18 and 28 years old. My

analytic sample includes children who were present in at least 50% of the waves in which

they could have been observed in the PSID core data and lived with mothers with a

known union history, who responded to at least one wave of the Transition to Adulthood

Supplement between 2005 and 2015, and who provided information on their educational

status and attainment (N=2,589). This includes 86% of the age-eligible original CDS

sample. Young adults who were not included in this analytic sample were typically

younger, Hispanic, and from families with lower household incomes and wealth at birth.

Weighted statistics are representative of young adults born between 1985 and 1997 who

were residing in the United States in 1997.

Measures

Dependent variables. Two dependent variables are used as indicators of

children’s educational attainment: on-time high school completion and college-

enrollment. On-time high school completion is a binary variable based on whether the

child (1) completed high school, and (2) graduated without repeating a grade in school.

This information is gathered from CDS and TAS. In CDS, individuals are asked whether

they ever repeated a grade in school, and in TAS, respondents are asked whether they

completed high school, and in which month and year they did so. If participants

responded affirmatively to having repeated a grade, they are assigned a value of 0,

regardless of completion status. If participants did not repeat a grade, but also did not

finish high school, they are also assigned a value of 0. Only respondents who did not

repeat a grade and finished high school are assigned a value of 1. I have chosen this

measure over any-time high school completion because high school graduation rates are

universally high in the U.S. (Chapman, Laird, Ifill, KewalRamani, 2011), but on-time

completion rates are significantly lower, and youth who do not graduate high school on-

time are more likely to never graduate, not pursue higher education, and be unemployed

(National Center for Education Statistics, 2014). College enrollment is also a binary

variable. Respondents are coded as 1 if they reported attending a two-year or four-year

college in at least one wave of the TAS in which they were observed. They are coded as 0

if they did not finish high school and/or if they did not report attending college in any of

the TAS waves in which they were observed. About one-quarter of college enrollees did

not report an initial enrollment date, so the analysis considers ever having been enrolled,

rather than enrollment immediately after high school.

Independent variable. Children’s lifetime experience of family structure (from

birth to age 17) is based on mother’s union status: (1) the proportion of childhood spent

living with two biological parents, (2) the proportion of childhood spent living with a

mother and a non-biological parent, and (3) the proportion of childhood spent with a

single mother. Following the example of Carlson & Corcoran (2001), this categorization

of single-mother families includes never married, divorced, widowed, and separated

mothers. Although these categories of single-parenthood are not identical, I combined

them due to sample-size limitations and to be consistent with prior research. Additionally,

a child’s non-biological parent may include either the married or cohabiting partner of the

child’s mother.

I used the PSID’s marriage and birth history files to construct the family structure

variable. Based on the child’s reported birth month and year and his or her mother’s

reported start and end date for each of her unions (also reported by month and year), I

determined the union status of each child’s mother when he or she was born, and in each

subsequent year of his or her life until age 17. This approach is an improvement upon

measurements that rely solely on PSID relationship status variables and/or household

roster information, as mother’s union status information is not available in non-interview

years after 1997 or in years in which she did not participate in the survey. If a child’s

mother was unobserved in one year, but she reports information on her union status in a

subsequent year of the marriage history file, I used the reported union status start and end

dates to recover information about her union status for missing years, retrospectively. To

obtain the proportion of childhood spent in each family structure type, I totaled the

number of waves a child is observed living in a given family structure and divided this

value by the total number of waves a child is observed before age 18.

Socioeconomic stress. To test the socioeconomic stress hypothesis, I used several

measures that have been previously identified as indicators of children’s socioeconomic

circumstances: (Conley, 2009; Fomby et al. 2010): parents’ education, self-reported

measures of neighborhood quality, safety, and homeownership, mother’s age at child’s

birth, and lifetime measures of family wealth and parents’ employment and health status.

Mothers’ education is specified as the highest level of education completed by the child’s

mother: less than high school, high school, some college, and bachelor’s degree or higher

(reference). Education was used instead of family income in this analysis due to high

levels of collinearity between the two variables and because income is likely to be

endogenous (affected by mother’s union status). For neighborhood quality, I used the

child’s primary caregiver’s (PCG) self- reported rating of their neighborhood as a place

to raise children (1=excellent/very good, 2=good, 3=fair/poor). Neighborhood safety was

measured based on the PCG’s report of how safe they perceived their neighborhood to be

to walk around alone after dark (0=neighborhood is somewhat or extremely dangerous,

1=neighborhood is completely or fairly safe). Homeownership is a binary variable that is

based on the PCG’s report of the percentage of their neighbors who owned their home

(0=less than 50%, 1=more than 50%). Mother’s age at birth is a categorical variable: 19

and under (reference), 20-29, and 30+. Average family wealth and parents’ employment

and health statuses are all lifetime measures that I created by summing the values for

each of these variables across all the waves that a child was observed and dividing them

by the total number of waves that a child was observed. For family wealth, I included the

inverse hyperbolic sine of wealth excluding home equity to account for the highly

skewed distribution of wealth in the sample. Mothers’ employment and health status are

both dichotomous variables indicating whether the child’s mother is employed and in

good health (0=no, 1=yes).

Extended family embeddedness. I included five indicators to test the extended

family embeddedness hypothesis: (1) coresidence with extended relatives, (2) geographic

proximity to extended relatives, (3) frequency of contact with extended relatives, and (4)

the PCG’s reported satisfaction with the level of practical support and (5) emotional

support he or she received from extended family members. Coresidence with extended

relatives was determined based on information from the household roster and from the

Family Identification Mapping System (FIMS). A child was identified as living with an

extended relative if he or she resided with an adult relative (by birth, adoption, or

marriage) who is not his or her parent or sibling. This includes grandparents, aunts,

uncles, and adult cousins; it does not include nonrelatives such as friends or boarders or

underage extended family members. Following Fomby et al. (2010), I distinguish

between short-term and long-term extended family coresidence with a three-category

variable: 1=never lived with an extended relative (reference), 2=lived with an extended

relative for less than a quarter of childhood, and 3=lived with an extended relative for at

least a quarter of childhood. Geographic proximity to extended relatives is measured

based on the PCG’s report of the number of family members (besides the child’s other

parent if they are not present in the household) who live in the same neighborhood. Due

to the highly positively skewed distribution of responses, I constructed a categorical

variable that collapsed responses into quartiles. Frequency of contact with extended

relatives is based on how often the PCG reports that the nuclear family “gets together

with relatives or friends”: less than once a month (reference), one to three times a month,

or once a week or more. Measures of practical and emotional support from extended

family is based on the PCG’s report of how satisfied he or she is with the level of support

that he or she is receiving from family members besides the child’s other parent.

Responses are based on a 7-point Likert scale with 1 indicating complete dissatisfaction

and 7 indicating complete satisfaction. Higher values on these scales signify higher

levels of satisfaction with practical and emotional support received by the primary

caregiver.

Moderator. Race/ethnicity is coded into four categories based on the PCG’s report

of the child’s race: (1) non-Hispanic white (reference); (2) non-Hispanic black; (3)

Hispanic; and (4) other race.

Control variables. In addition to these variables, I also control for several

demographic characteristics of the child: gender, region where the child grew up, and the

average number of children living in the child’s household. Gender is measured as male

(reference) or female. Region (South vs non-South) measures the proportion of all

observed waves that a child lived in a state in the US South. Average number of children

in the household was constructed in a similar fashion. At each wave that a child was

observed, he or she was assigned a value indicating the total number of children living in

his or her household. These values were then totaled and averaged across the total

number of waves that a child was observed.

All covariates except child’s race, gender, age, region, family income, and

number of children present in the household included missing data on at least some

observations. To restore missing cases and improve the generalizability of my findings, I

used multiple imputation with chained equations in Stata 15. The imputation model

included all dependent, independent, and control variables with missing values, as well as

child’s race, gender, and age, a family-level identifier shared by all related individuals in

the sample, and stratum and clustering indicators. My multivariate regressions used

multiply imputed datasets where covariates include imputed values but dependent

variables are not imputed (von Hippel 2007).

Analytic Strategy

I use logistic regression analysis to examine the relationship between children’s

lifetime experience of family structure and their educational attainment. I begin with

baseline models that estimate the strength of the association between the main effects of

family structure and on-time high school completion and college enrollment, taking into

account the aforementioned control variables (Model 1), before including interaction

terms for white, black, and Hispanic children (Model 2). I then separately add in

indicators of the socioeconomic stress (Model 3) and extended family embeddedness

(Model 4) hypotheses, before jointly including all covariates (Model 5) to assess the

extent to which these mechanisms help explain group differences in the association of

family structure between family structure and educational attainment. Next, I conduct

Wald tests to determine whether the family structure-race/ethnicity interactions are

jointly significant. Then, I calculate predicted probabilities to show how, if at all, the

effects of family structure on high school completion and college enrollment vary by

race/ethnicity. To appropriately compare changes in coefficients across logistic

regression models and to quantify the amount of variance in the two dependent variables

that is explained by socioeconomic stress and extended family embeddedness, I calculate

the average partial effects of these two sets of factors using the Karlson, Hohm and

Breen (KHB) method for decomposing total effects (Kohler, Karlson and Holm 2011).

Results

Descriptive Results

Table 1 displays the distribution of the dependent variables and sample characteristics by

race/ethnicity. Statistically significant group differences (black vs. Hispanic, black vs.

white, and Hispanic vs. white) are indicated with footnotes. Starting with outcome

variables, I found that there were significant differences by race/ethnicity in educational

attainment. White youths had the highest rates of on-time high school completion (85%)

and college enrollment (81%) among the three groups. Hispanic youths followed with

74% for both outcomes, and black youths with 61% and 70%, respectively.

Consistent with findings from cross-sectional research (U.S. Department of

Health and Human Services, Health Resources and Services Administration, 2014), the

longitudinal estimates of exposure to various family structures indicate that black

children less of their childhood living in a two-biological-parent and more time living

with a single mother, compared to white and Hispanic children. Black children also spent

more time living with their mothers and her partner who is not their biological parent

(both married and cohabiting), relative to their white peers, whereas Hispanic youths and

white youths spent similar amounts of time with their mother’s non-biological partner.

Turning to socioeconomic stress, black and Hispanic children had significantly

less access to socioeconomic resources during childhood than their white peers. For

instance, the average family wealth for white children was roughly 1.5 to two times

higher than that of Hispanic and black children, respectively. Similarly, white parents

were approximately 1.5 times more likely to rate their neighborhood as “excellent” or

“very good”, relative to black and Hispanic parents. Black and Hispanic children were,

however, more likely to live with or in close proximity to extended relatives, and their

parents expressed higher levels of satisfaction with the amount of practical support

received from extended family members. Eighty percent of white children never lived in

an extended family, compared to approximately 40% of black children and 60% of

Hispanic children. Further, white parents reported significantly lower levels of

satisfaction with the amount of practical support received from extended family members

than black and Hispanic parents.

In sum, descriptive results generally support findings from prior research

indicating racial/ethnic differences in levels of educational attainment and family

structure. They also show that on average, black and Hispanic children are confronted

with more socioeconomic hardship than their white peers. Moreover, they lend some

evidence to support the assertion that minority children are more deeply embedded in

their extended family networks than white children. The multivariate analyses in the

following section assess the extent to which racial/ethnic differences in these family

resources account for group differences in the effects of family structure.

Multivariate Results

On-time high school completion. In the baseline model (Model 1) for Table 2, the

variables indicating the main effects of family structure show that more time spent living

with a single-mother or with a mother and a non-biological parent are associated with a

lower likelihood of finishing high-school on-time, compared with more time spent with

two biological parents. The main effects of race/ethnicity are statistically significant,

indicating that black and Hispanic youth are less likely to complete high school on-time,

relative to white youth. The interaction term between single-mother family and

race/ethnicity (Model 2) is highly significant and positive for black youth (OR 3.049),

signifying a weaker negative association between family structure and on-time high

school completion for black youth, compared to white youth. The association between

family structure and this outcome was not different between white and Hispanic youth.

Model 3 accounts for indicators of socioeconomic stress. Children who spent

more time not living in a two-biological-parent family remained less likely to graduate

high school on-time (though the coefficient for time spent living with mother and a non-

biological parent is only marginally significant). The interaction term for single-mother

family and black children is no longer statistically significant, signifying that

socioeconomic stressors help explain the differential association between family structure

and on-time high-school completion between blacks and whites. Indicators of

socioeconomic stress, namely parents’ education, age, and health status are positively

related to high school completion and appear to be explaining much of the variation in

this outcome. Children whose parents earned a bachelor’s degree or higher and those

whose mother was older at childbirth and whose mother spent more time in good health

during their childhood were more likely to graduate on-time.

Model 4 includes indicators of extended family embeddedness. Here, we see no

change in the significance of the association between family structure and on-time high

completion or the family structure-race/ethnicity interaction term for black youth and

high school completion. This suggests that indicators of extended family embeddedness

explained little of the main interaction effects of family structure on high school

graduation between blacks and whites. When both sets of factors are taken into account

in the full model (model 5), the main effects for family structure are no longer

statistically significant, and neither is the interaction term for blacks. Thus, combined,

indicators of socioeconomic stress and extended family embeddedness help explain the

relationship between family structure and high school completion, as well as the

differential effect of family structure for black youth. Wald tests indicate that the family

structure-race/ethnicity interactions are statistically significant in models 2 through 5

(Table 2), providing evidence that the association between family structure and on-time

high school completion differs for at least one group (in this case, black children).

To better understand the interaction terms for race/ethnicity and family structure, I

calculated and graphed the predicted probabilities for each family structure-race/ethnicity

combination from Table 2 model 2 that had statistically significant differences by

race/ethnicity, setting all other values to their mean or modal values. Figure 1 displays

these values for on-time high school completion. This graph shows that as black children

spend longer periods of time in a single-mother family, the change in slope for the effect

of living in a single-mother family declines less sharply than the change in slope for

whites. This pattern indicates that more time spent in this family structure is less

negatively consequential for black youth. This pattern is consistent for Hispanic children,

though as demonstrated in Table 2, this differential effect is not statistically significant at

p<.05. Summarizing Table 2 and Figure 1, the results show that living in a single-mother

family for the majority of childhood is less detrimental for black children’s on-time high

school completion, compared to white children.

College enrollment. In the baseline model for college enrollment (Table 3), the

main effect of family structure indicates that children who spent some time in a single-

mother family or with a mother and a non-biological parent are less likely to enroll in

college, relative to those who spent a greater proportion of childhood living in a two-

biological-parent family. The main family structure-race/ethnicity interaction terms for

blacks and Hispanics, however, are not significant (Model 2). This suggests that the

effect of family structure on college enrollment does not differ by race/ethnicity.

Model 3 adds in factors related to socioeconomic stress. Unlike the baseline

model, the main effect of family structure is no longer statistically significant, indicating

that socioeconomic factors help explain the association between the main effect of family

structure and college enrollment. In particular, parents’ education, family wealth, and

mother’s age at childbirth are positively related to college enrollment. On the contrary,

when indicators of extended family embeddedness are taken into account (Model 4), the

main effect of family structure remains similar to that in the baseline model. This

suggests that the extended family embeddedness hypothesis does little to explain the

association between family structure and college enrollment. Extended family

embeddedness does appear to be independently associated with college enrollment,

however. Specifically, children who spent some time living with an extended relative had

a roughly 40% lower odds of attending college than those who never lived with an

extended relative during childhood.

In the full model, when I account for factors related to both socioeconomic stress

and extended family embeddedness, the main effects of family structure are

nonsignificant. Similar to Model 3, the coefficients for parents’ education, family wealth,

and mother’s age are highly and positively related to college enrollment. Coresidence

with extended family members is no longer related to college-enrollment. This finding

suggests that socioeconomic factors are more predictive of college enrollment than

family structure itself or extended family embeddedness. Wald tests show that the family

structure-race/ethnicity interactions are not statistically significant in each model (see

Table 3), providing further evidence that the association between family structure and

college enrollment does not differ by race/ethnicity.

Table 4 displays the average partial effect of family structure in the baseline and

full models estimated using the Karlson, Hohm and Breen (KHB) method for

decomposing total effects (Kohler, Karlson and Holm 2011). The average partial effect

indicates the difference in the predicted probability that a child experienced an outcome

of interest (in this case, on-time high school completion and college enrollment), given

the time he or she spent in a single-mother family or with a mother and a non-biological

parent, relative to time spent in a two-biological-parent family. The percentage change in

the average partial effect between the baseline and full models demonstrates the

difference in how much of the variance is explained by family structure for each outcome

before and after taking into account indicators of socioeconomic stress and extended

family embeddedness and controlling for sociodemographic characteristics.

Average partial effects from the KHB decomposition method reveal that

indicators of both socioeconomic stress and extended family embeddedness significantly

attenuated the relationship between family structure and children’s educational

attainment, though the latter did so to a much lesser degree. The reduced model estimated

that on average, the probability of on-time high school completion for each additional

year a child spent living with his or her mother and a non-biological parent was 17.6

percentage points lower than for each additional year spent in a two-biological-parent

family. In the full model, once indicators of socioeconomic stress are taken into account,

the average partial effect decreased to 10.2 percentage points. Thus, the difference in the

size of average partial effects in the reduced and full models indicates that 42% of the gap

in on-time high school completion between children who spend an additional year living

with a mother and a non-biological parent compared to with two biological parents is

explained by socioeconomic factors included in the full model. Extended family

embeddedness, on the other hand, explained 15% of the gap. When we compare

differences between the amount of time spent living with a single mother compared to

two biological parents, we see that socioeconomic factors explained 22% and extended

family embeddedness explained 13% of the gap in on-time high school completion. The

results for college enrollment echo those of on-time high school completion. For each

additional year a child spent with his or her mother and a non-biological parent compared

to with both biological parents, socioeconomic factors accounted for 48% of the variance

in college enrollment, whereas extended family embeddedness accounted for 19% of the

variance in this outcome. Socioeconomic factors explained 36% and extended family

embeddedness explained 13% of the gap in college enrollment between children who

spent more time living with a single mother, compared to those who spent more time

living with both parents.

Sensitivity Analysis

I assess the robustness of my findings in several ways. First, research suggests

that children who experience multiple transitions in family structure face worse

educational outcomes than children raised in stable families, independent of the type of

stable family structure (Fomby & Cherlin, 2007). In a supplemental analysis, I controlled

for whether a child experienced one or more changes in family structure during

childhood. Children who spent more time outside of a two-biological-parent family were

still less likely to complete high school on-time and enroll in college, and black children

in single-mother families remained marginally more likely to finish high school on-time,

relative to white children who spent similar periods of time in this family structure.

Second, I considered an alternative approach to measuring family structure that

captures both mother’s relationship type and transitions (but not biological parental

status): (1) stable two-parent family, (2) stable one-parent family, (3) two-parent to one-

parent family (4), one-parent to two-parent family, and (5) two or more family

transitions. Consistent with my main findings, children who grew up in a stable two-

parent family were more likely to graduate high school on-time, compared to children

raised in any other family type. Black children raised in a stable one-parent family were

more likely to complete high school on-time than white children raised in a stable one-

parent family.

Third, my analytic sample includes children who were present in at least half of

the waves in which they could have been observed in the PSID. I also ran my analysis on

children who were present in all waves of the study. Results from this more restricted

sample are representative of the results that I obtained when I included individuals who

were missing in up to 50% of waves. Further, my analytic sample also contains 326

sibling pairs. Given that observations between siblings are non-independent, I ran all

models clustering on family unit identifiers rather than on the indicators of the PSID’s

multistage sampling design and observed similar results.

Finally, research has suggested that there is negative selection into extended

families; that is, children who live with extended relatives may do so because their

parents lack the resources to live independently (Chase-Lansdale, Brooks-Gunn, &

Zamsky, 1994; Pilkauskas & Dunifon, 2016). If this is the case, then the observed

negative association between extended family coresidence and on-time high school

completion and college enrollment may be attributable, in part to, to the fact that children

raised in these households are less well-off to begin with.

To account for factors related to the selection into an extended family, I used

propensity score matching (PSM) methods to estimate the average treatment effect of

extended family coresidence. This was accomplished by comparing the on-time high

school completion and college enrollment of children who spent time in an extended

family during childhood with children who never spent time in an extended family but

are otherwise similar in terms of family background characteristics. Measures in the

matching model included the extended family coresidence status (treatment), child’s

gender, race, sex, and region, number of children in the household at birth, mother’s

education, and mother’s age, union, and health status at birth. In a sample balanced on

these characteristics, children who lived in an extended family for less than or a quarter

or more of their childhood were not significantly less likely to finish high school on-time

or enroll in college, compared to those who never lived in an extended family. These

findings suggest that the negative association between extended family coresidence and

children’s educational attainment is largely explained by the negative selection into this

living arrangement. Findings from PSM methods may differ from those of the logistic

regression analyses, which regress toward the mean, if the treatment group (i.e., those

likely to experience extended family coresidence) and the control group do not have

substantial overlap on characteristics that distinguish the two groups. All in all, however,

the results of the sensitivity checks considered here lend confidence to the robustness of

my main findings.

Discussion

This study used nationally representative, longitudinal data to examine the

association between children’s lifetime experience of family structure and their

educational attainment, the extent to which this association varied by race/ethnicity, and

the relative importance of two hypothesized mechanisms in explaining any observed

group differences. More specifically, I assessed the relationship between childhood

family structure and on-time high school completion and college enrollment and tested

whether racial/ethnic differences in socioeconomic stress and extended family

embeddedness mediate these relationships for white, black, and Hispanic children.

Results from this study confirm findings from prior research demonstrating that children

raised outside of a two-biological-parent family are at an elevated risk of experiencing

lower educational attainment than those raised in this family type. They are also

consistent with previous studies showing that the consequences of living in a single-

mother family are less negatively impactful for black children, relative to white children

who spend similar amounts of time in this family structure (e.g., McLanahan & Sandefur,

1994). This suggests that the experience of single-parenthood for white children has

limited generalizability to black children, as the two groups may adjust differently to life

in a single-parent family.

I did not, however, observe racial differences in the link between family

structure and college enrollment. This indicates that the differential association between

family structure and educational attainment, or child wellbeing more generally, may

depend on the outcome of interest. Indeed, prior research has identified stronger family

structure effects for more proximate outcomes such as behavioral and psychological

wellbeing, than more distal ones such as educational attainment (Amato, 2005), and only

in some instances have researchers observed racial/differences in these outcomes

(Dunifon & Kowaleski-Jones, 2002; Manning & Brown, 2006; McLanahan & Sandefur,

1994). Overall, findings on racial differences suggest that our examination of the impact

of family structure on children’s educational attainment, and wellbeing broadly speaking,

is best approached from a pluralistic perspective, one that does not privilege one family

type over another, since family structure does not always have the same social

consequences for all groups.

There were no observed differences between whites and Hispanics in the effects

of family structure on either measure of educational attainment. While this finding is

consistent with Fomby et al. (2010)’s work on racial differences in the effect of family

instability on adolescent’s risk behavior, it does not preclude the possibility of group

differences for other dimensions of child wellbeing. Given the limited number of studies

that focus on Hispanic children, more work is needed to investigate potential differential

effects of family structure on child wellbeing for this group.

What accounts for racial/ethnic variation in the association between family

structure and educational attainment? Findings from this study suggest that both

socioeconomic stress and extended family embeddedness attenuate the effect of family

structure on on-time high school completion and college enrollment. Socioeconomic

stress, however, appears to have greater relative influence. Differences in socioeconomic

resources accounted for 22-48% of the gap in on-time high school completion and

college enrollment between amount of time spent in a two-biological-parent family and

those living with a mother and nonbiological parent or single mother. Extended family

embeddedness factors helped explain roughly 15-20% of the variance in these outcomes.

These findings lend support for the socioeconomic stress hypothesis, which posits that

the negative effect of familial disruption may be less independently impactful for children

of racial/ethnic groups facing a large number of socioeconomic disadvantages to begin

with. In contrast, they provide little evidence to support the hypothesis that racial/ethnic

minority children’s deeper embeddedness in their extended family network protects

against the negative effects of familial disruption.

In interpreting these findings, it is important to acknowledge that exposure to

socioeconomic hardship is harmful to children, irrespective of race/ethnicity and that

black children are not uniquely impervious to the negative consequences of familial

disruption. Rather, they may have developed adaptive strategies derived from their

structural position in society that aids them in coping with this stressor. Indeed, research

suggests that while exposure to racially and economically problematic life circumstances

is detrimental to child wellbeing, black children may develop competencies that help

them respond to major life events such as parental separation (Smith and Carlson, 1997).

Future work should explore racial/ethnic variation in children’s responses to familial

disruption and its link to child outcomes.

Given the key role that socioeconomic factors play in explaining the relationship

between family structure and children’s educational attainment, these results call into

question whether welfare funds should be diverted to encourage the formation and

maintenance of two-parent nuclear families and discourage nonmarital childbearing (U.S.

House of Representatives, 1996), rather than being spent directly on alleviating financial

hardships for disadvantaged groups. They also suggest that expectations for the extended

family to compensate for family hardship may be misplaced. Extended family

embeddedness does not appear to fully protect against the negative impact of living

outside of a two-biological-parent family. Therefore, efforts aimed at improving child

wellbeing for disadvantaged groups may be more effective by focusing on reducing

socioeconomic hardships for low-income families and adopting reasonable expectations

for the extended family to meet the needs of its disadvantaged members.

Results should be interpreted within the context of the study’s limitations. First,

while I included a robust set of indicators of socioeconomic stress and extended family

embeddedness, the factors used in this study are not exhaustive. I would have liked to

include direct measures of the amount and frequency of practical and emotional support

that mothers received from extended family members, as well as measures of relationship

quality among extended family members, but was unable to do so due to limitations of

the data. This would allow me to more accurately assess the levels of extended family

support available to mothers in various family types, as well the amount of extended

family cohesion they experience. Second, the relatively small sample size for Hispanic

children diminishes my statistical power, making it difficult to observe statistically

significant differences and to generalize to the broader U.S. population for this group.

Third, although my use of the birth and marriage history files allows me to

retrospectively identify mothers’ union status between waves and during skipped waves, I

miss children whose mothers attrited. As noted earlier, individuals who attrited were

more likely to be economically disadvantaged. To the extent that children of

disadvantaged mothers are more likely to have lower levels of educational attainment,

attrition is likely to have underestimated the influence of socioeconomic stress on

children’s on-time completion and college enrollment. Lastly, while the two hypotheses

that I tested help partially account for racial/ethnic differences in the association between

family structure and children’s educational attainment, neither of them fully explain this

phenomenon. Future research should investigate how other factors related to children’s

familial and community context may account for these differences.

Despite its limitations, this study takes important steps towards understanding

why black children are less affected by family structure with regard to educational

attainment than their white peers. Using nationally representative longitudinal data, this

research shows that group differences in access to socioeconomic resources are an

important mechanism for explaining this gap. Combined, study findings highlight the

importance of examining racial/ethnic differences in the consequences of family structure

for children’s educational attainment and their wellbeing more generally.

References

Amato, P. R. (2000). The consequences of divorce for adults and children. Journal of

Marriage and Family, 62, 1269-1287.

Amato, P. R. (2001). Children of divorce in the 1990s: An update of the Amato and Keith

(1991) meta-analysis. Journal of Family Psychology, 15, 355-370.

Amato, P. R. (2005). The impact of family formation change on the cognitive, social, and

emotional well-being of the next generation. The Future of Children, 15, 75-96.

Amato, P. R., & Keith, B. (1991). Parental divorce and adult well-being: A meta-

analysis. Journal of Marriage and the Family, 53, 43-58.

Baca Zinn, M. & Wells, B. (2000). Diversity within Latino families: New lessons for

family social science. In D. H. Demo, K. R. Allen, & M. A. Fine (Eds.).

Handbook of Family Diversity (262–73). New York, NY: Oxford University

Press.

Brown, S. L. (2010). Marriage and child well-being: Research and policy perspectives.

Journal of Marriage and Family, 72, 1059-1077.

Bobo, L., Kluegel, J. R., & Smith, R. A. (1997). Laissez faire racism: The crystallization

of a ‘kinder, gentler’ anti-black ideology. In S.A. Tuch & J. Martin (Eds.). Racial

attitudes in the 1990s: continuity and change (15-44). Greenwood, CT: Praeger.

Bonilla-Silva, E. (2006). Racism without racists: Color-blind racism and the persistence

of racial inequality in America. Lanham, Maryland: Rowman and Littlefield

Publishers.

Bourdieu, P. & Passerson, J. (1977). Reproduction in education, society and culture.

Thousand Oaks, CA: Sage Publications.

Burton, L.M. (1992). Black grandparents rearing children of drug-addicted parents:

Stressors, outcomes, and social service needs. Gerontologist, 32, 744-751.

Bradley, R. H., Corwyn, R. F., Burchinal, M., McAdoo, H. P., & Coll, C. G. (2001a).

The home environments of children in the United States Part I: Relations with

behavioral development through age thirteen. Child Development 72, 6, 1868-

1886.

Bradley, R. H., Corwyn, R. F., McAdoo, H. P., & Coll, C. G. (2001b.) The home

environments of children in the United States Part I: Variations by age, ethnicity,

and poverty status. Child Development 72, 6, 1844-67.

Carlson, M. J. & Corcoran, M. E. (2001). Family structure and children's behavioral and

cognitive outcomes. Journal of Marriage and Family, 63, 779-792.

Chapman, C., Laird, J., Ifill, N., & KewalRamani, A. (2011). Trends in high school

dropout and completion rates in the United States: 1972-2009. National Center

for Education Statistics, Compendium Report, 1-53.

Charles, C. Z. (2006) Won’t you be my neighbor? Race, class and residence in Los

Angeles. New York, NY: Russell Sage Foundation.

Chase-Lansdale, P. L., Brooks-Gunn, J., & Zamsky, E. S. (1994). Young African-

American multigenerational families in poverty: Quality of mothering and grand-

mothering. Child Development, 65, 373-393.

Cherlin, A. J., & Seltzer, J. A. (2014). Family complexity, the family safety net, and

public policy. The ANNALS of the American Academy of Political and Social

Science, 654, 231-239.

Coleman, J. (1988). Social capital in the creation of human capital. American Journal of

Sociology, 94, S95-S120.

Conley, D. (2009). Being black, living in the red (2nd ed.). Berkeley, CA: University of

California Press.

Deleire, T & Kalil, A. (2002). Good things come in threes: Single-parent

multigenerational family structure and adolescent adjustment. Demography, 39,

393-413.

Dominguez, S. & Watkins, C. (2003). Creating networks for survival and mobility:

Social capital among African-American and Latin-American low-income

mothers. Social Problems, 50, 111-135.

Dunifon, R., & Kowaleski-Jones, L. (2002). Who’s in the house? Race differences in

cohabitation, single-parenthood, and child development. Child Development, 73,

1249 – 1264.

Dunifon, R., Ziol-Guest, K., & Kopko, K. (2014). Grandparent coresidence and family

well-being implications for research and policy. The ANNALS of the American

Academy of Political and Social Science, 654, 110-126.

Farber, N., Miller-Cribbs, J., & Reitmeier, M. Kin networks in the South: A comparison

of low-income, rural African-American and white women. Rural Social Work and

Community Practice, 10, 52–63.

Fomby, P. & Cherlin, A. J. (2007). Family instability and child wellbeing. American

Sociological Review, 72, 181-204.

Fomby, P., Mollborn, S., & Sennott, C. (2010). Race/ethnic differences in effects of

family instability on adolescents’ risk behavior. Journal of Marriage and Family,

72, 234-253.

Friedman, M. M. (1998). Family nursing (4th Ed.). Stamford, CO: Appleton & Lange.

Garrett-Peters, R. & Burton, L.M. (2016). Tenuous ties: The nature and costs of kin

support among low-income rural black mothers. Women, Gender, and Families of

Color, 4, 4-35.

Gerstel, N. (2011). Rethinking families and community: The color, class, and centrality

of extended kin ties. Sociological Forum, 26, 1–20.

Ginther, D. K. & Pollak, R. A. (2004) Family structure and children's educational

outcomes: Blended families, stylized facts, and descriptive regressions.

Demography, 41, 671-696.

Hill, M. S., Yeung, W. J., & Duncan, G. J. (2001). Childhood family structure and young

adult behaviors. Journal of Population Economics, 14, 271-299.

Hofferth, S. L. (2006). Residential father family type and child well-being: Investment

versus selection. Demography, 43, 53–77.

Hetherington, E., Camara, K., & Featherman, D. (1983). Achievement and intellectual

functioning of children in one-parent households. In J. Spence (Ed.). Achievement

and achievement motives: Psychological and sociological approaches (pp. 205-

284). San Francisco: W. H. Freeman and Company.

Hunter, A. G. (1997). Counting on grandmothers: Black mothers’ and fathers’ reliance on

grandmothers for parenting support. Journal of Family Issues, 18, 251-269.

Jayakody, R., Chatters, L.M., & Taylor, R.J. (1993). Family support to single and married

African American mothers: The provision of financial, emotional, and child care

assistance. Journal of Marriage and the Family 55, 261–276.

Karlson, K. B., & Holm, A. (2011). Decomposing primary and secondary effects: a new

decomposition method. Research in Social Stratification and Mobility, 29, 221-

237.

Kohler, U., Karlson, K. B., & Holm, A. (2011). Comparing coefficients of nested

nonlinear probability models. The Stata Journal, 11, 420-438.

Krysan, Maria, Mick P. Couper, Reynolds Farley, and Tyrone A. Forman. (2009). Does

race matter in neighborhood preferences? Results from a video experiment.

American Journal of Sociology, 115, 527-559.

Lamborn, S. G., & Nguyen, D. T. (2004). African American adolescents' perceptions of

family interactions: Kinship support, parent-child Relationships, and teen

adjustment. Journal of Youth and Adolescence, 33, 547-558.

Lareau, Annette. 2000. Home Advantage: Social Class and Parental Intervention in

Elementary Education. Lanham, MD: Rowman and Littlefield.

Lareau, Annette. 2011. Unequal Childhoods Class, Race, and Family Life (2nd Edition).

Berkeley, CA: University of California Press.

Panel Study of Income Dynamics. 2010. Child development supplement user's guide

for CDS-II. Ann Arbor, MI: Institute for Social Research.

Pilkauskas, N.V., Campbell, C., & Wimer, C. (2017). Giving unto others: Private

financial transfers and hardship among families with children. Journal of

Marriage and Family, 79, 709-722.

Pilkauskas, N. V., & Dunifon, R. E. (2016). Understanding grandfamilies: Characteristics

of grandparents, nonresident parents, and children. Journal of Marriage and

Family, 78, 623-633.

Pittman, L. D. (2007). Grandmothers' involvement among young adolescents growing up

in poverty. Journal of Research on Adolescence, 17, 89-115.

Proctor, B. D., Semega, J.L, & Kollar, M.A. (2016). Income and poverty in the United

States: 2015. Washington, DC: United States Census Bureau.

Manning, W. D., & Brown, S. L. (2006). Children’s economic well-being in marriage

and cohabiting parent families. Journal of Marriage and Family, 68, 345 – 362.

Manning, W. D., & Lamb, K. A. (2003). Adolescent well-being in cohabitating, married,

and single-parent families. Journal of Marriage and Family, 65, 876-893.

Margolis, K. L., Fosco, G. M., & Stormshak, E. A. (2014). Circle of care: Extending

beyond primary caregivers to examine collaborative caretaking in adolescent

development. Journal of Family Issues, 37, 1179-1202.

Mazumder, B. (2008). Upward intergenerational economic mobility in the United States.

Economic Mobility Project. Washington, DC: Pew Charitable Trusts.

Mc Cubbin, M.A. (1993). Family stress theory and the development of nursing

knowledge about family adaptation. In S.L. Feetham, S.B. Meister, J.M.Bell, &

C.L. Gillis (Eds.). The Nursing Family (46-58). New Bury Park, CA: Sage

Publications.

McCubbin, M. A., & McCubbin, H. I. (1989). Families coping with illness: The

resiliency model of family stress, adjustment and adaptation. In, C.B. Danielson,

B. Hamel-Bissel, & P.Winstead-Fry (Eds.). Families, health & illness:

Perspectives on coping and intervention (21-64). St. Louis, MO: Mosby.

McDonald, K. B., & Armstrong, E. M. (2001). De-romanticizing black intergenerational

support: The questionable expectations of welfare reform. Journal of Marriage

and Family, 63, 213-223.

McLanahan, S. & Percheski, C. (2008). Family structure and the reproduction of

inequalities. Annual Review of Sociology, 34, 257-276.

McLanahan, S. & Sandefur, G. (1994). Growing up with a single-parent: What hurts,

what helps. Cambridge, MA: Harvard University Press.

McLoyd V. C., Cauce A. M., Takeuchi, D., & Wilson, L. (2000). Marital processes and

parental socialization in families of color: A decade review of research. Journal of

Marriage and the Family, 62, 1070–1093.

National Center for Education Statistics. (2014). On-time high school graduation: The

opportunity picture. Washington, D.C.: U.S. Department of Education. Retrieved

from: http://opportunityindex.org/wp-content/uploads/2014/12/HS-Graduation-

2014.pdf.

Passel, J. S., & Cohn, D. (2008). U.S. population projections: 2005 – 2050. Washington,

DC: Pew Research Center.

Roschelle, A. R. 1997. No more kin: Exploring race, class, and gender in family

networks. Thousand Oaks, CA: Sage.

Sandberg, J. F., & Hoffereth, S. F. (2001). Changes in children's time with parents:

United States, 1981-1997. Demography, 38, 423-436.

Sarkisian, N. & Gerstel, N. (2004). Kin support among blacks and whites: Race and

family organization. American Sociological Review, 69, 812-837.

Sarkisian, N. & Gerstel, N. (2012). Nuclear family values, extended family lives: The

power of race, class, and gender. New York, NY: Routledge.

Shaw, D. S., Winslow, E. B., & Flanagan, C. (1999). A prospective study in the effects of

marital status and family relations on young children’s adjustment among African

American and European American families. Child Development, 70, 742–755.

Smith, T. E. (1997). Differences between black and white students in the effect of

parental separation on school grades. Journal of Divorce and Remarriage, 27, 25-

42.

Smith, C. & Carlson B.E. (1997). Stress, coping, and cesilience in youth. Social Service

Review, 71, 231-256.

Stack, C. B. (1974). All our kin: strategies for survival in a black community.

New York, NY: Harper & Row.

Stack, C. B. & Burton, L. M. (1993). Kinscripts. Journal of Comparative Family Studies,

24, 157-170.

Sun, Y. & Li, Y. (2007). Racial and ethnic differences in experiencing parents' marital

disruption during late adolescence. Journal of Marriage and Family, 69, 742-762.

Taylor, R. J. (1986). Receipt of support from family among black Americans:

Demographic and familial differences. Journal of Marriage and Family, 48, 67-

77.

Taylor, R. J., Chae, D.H., Lincoln, K.D., & Chatters, L.M. (2015). Extended family and

friendship support networks are both protective and risk factors for major

depressive disorder and depressive symptoms among African-Americans and

Black Caribbeans. Journal of Nervous and Mental Disease, 203, 132-140.

Teachman, J. D. (2008). The living arrangements of children and their educational well-

being. Journal of Family Issues, 29, 734-761.

U.S. Department of Health and Human Services, Health Resources and Services

Administration, Maternal and Child Health Bureau. (2014). Child health USA

2014. Rockville, Maryland: U.S. Department of Health and Human Services.

U.S. House of Representatives, Committee on Ways and Means. 1996. Personal

Responsibility and Work Opportunity Reconciliation Act of 1996: Conference

Report to Accompany H.R. 3734.Section 401, Purpose (p.9), July 30.

Von Hippel, P. T. (2007). Regression with missing Ys: An improved strategy for

analyzing multiply imputed data. Sociological Methodology, 37, 83-117.

Winship, C. & Mare, R. D. (1984). Regression with ordinal variables. American

Sociological Review, 49, 512-525.

Wilson, W. J. (1987). The truly disadvantaged: The inner city, the underclass, and public

policy. Chicago, IL: The University of Chicago Press.

Wu, L. (1996). Effects of family instability, income and income instability on the risk of

a premarital birth. American Journal of Sociology, 68, 386-406.

Table 1. Sample characteristics by race/ethnicity, Panel Study of Income Dynamics, Child Development

Supplement, Transition into Adulthood Study, (1985-2015)

Variables

Full

Sample

Black

White

Hispanic

Other

Proportion of childhood with two biological parentsabc

0.68

0.33

0.78

0.73

0.62

Proportion of childhood with mother and non-biological partnerab

0.09

0.12

0.09

0.08

0.10

Proportion of childhood with single motherabc

0.22

0.55

0.13

0.19

0.28

On-time high school completionabc

0.79

0.63

0.85

0.74

0.83

College enrollmentab

0.77

0.61

0.81

0.74

0.76

Femalea

0.48

0.42

0.49

0.48

0.53

Child's ageab

24.6

25.1

24.6

24.0

(SD)

0.09

0.16

0.12

0.31

0.34

Proportion of childhood spent living in the Southabc

0.33

0.59

0.30

0.22

0.24

No. of childrenabc

2.33

2.52

2.18

2.80

2.39

(SD)

0.02

0.06

0.02

0.09

0.13

Parents' education

<HSabc

0.13

0.15

0.05

0.47

0.17

0.34

0.38

0.32

0.25

0.27

Some collegebc

0.29

0.33

0.27

0.15

0.25

BA or higherabc

0.25

0.14

0.37

0.13

0.31

Family wealth without home equityabc

7.24

4.54

8.33

5.97

6.30

(SD)

0.14

0.23

0.17

0.41

0.71

Proportion of childhood mom is employedabc

0.65

0.60

0.69

0.53

0.57

Neighborhood quality

Excellent/very goodac

0.56

0.44

0.71

0.43

0.49

Goodab

0.26

0.30

0.21

0.32

0.25

Fair/poorac

0.18

0.26

0.08

0.25

0.26

>50% of neighbors own their homeabc

0.68

0.43

0.80

0.59

0.45

Neighorhood is completely or fairly safeac

0.88

0.74

0.94

0.74

0.90

Mother's age at child's birth

19 and belowa

0.12

0.17

0.07

0.13

0.11

20-29c

0.55

0.54

0.56

0.51

30 and aboveac

0.33

0.27

0.39

0.29

0.36

Proportion of childhood mom is in good healthac

0.90

0.84

0.93

0.85

0.82

Extended family coresidence

Never lived with an extended relativeac

0.60

0.39

0.80

0.62

0.61

Lived with an extended relative for <1/4 of childhoodab

0.18

0.24

0.13

0.14

0.21

Lived with an extended relative for >=1/4 of childhoodac

0.22

0.37

0.07

0.24

0.18

No. of extended relatives living in R's neighborhood by quartile

Bottom 25th percentileac

0.45

0.35

0.53

0.45

0.53

25th-50th percentile

0.07

0.04

0.08

50th-75th percentileac

0.23

0.27

0.19

0.27

0.16

75th-99th percentileac

0.25

0.31

0.20

0.25

0.23

Frequency of contact with extended relatives and friends

Less than once a month

0.24

0.28

0.19

0.30

0.25

One to three times a montha

0.16

0.13

0.18

0.15

0.19

Greater than or equal to once a weekbc

0.60

0.59

0.62

0.55

PCG satisfaction with the level of practical support received from

extended family membersabc

5.13

5.32

4.89

5.83

5.53

(SD)

0.04

0.10

0.05

0.12

0.17

PCG satisfaction with the level of emotional support received from

extended family membersbc

5.47

5.28

5.41

5.76

5.86

(SD)

0.04

0.10

0.05

0.14

0.15

No. of Observations

2,589

1,105

1,163

183

137

Notes: Values are percentages unless otherwise noted. Values are weighted to account for the complex multistage

clustered design of the PSID. Total for the time-varying characteristics is the child-year average. aBlack-White difference

significant at p<0.05. bBlack-Hispanic difference significant at p<0.05. cHispanic-White difference significant at p<.05.

No.=Number. SD=Standard deviation. PCG=Primary caregiver.

Table 2: Odds ratios from logistic regression models predicting on-time high school completion, Panel Study of Income Dynamics,

Child Development Supplement (CDS), Transition into Adulthood Supplement (TAS), 1985-2015

Model 1

Model 2

Model 3

Model 4

Model 5

Proportion of childhood with single mother

0.466***

0.231***

0.428**

0.294***

0.462*

(-3.141)

(-3.858)

(-2.055)

(-3.220)

(-1.837)

Single mother X Black

3.049**

2.254

2.896**

2.239

(2.219)

(1.557)

(2.119)

(1.547)

Single mother X Hispanic

1.750

1.473

1.673

1.388

(0.667)

(0.449)

(0.619)

(0.376)

Proportion of childhood with mother and non-biological partner

0.233***

0.271***

0.446*

0.312***

0.459*

(-4.309)

(-3.118)

(-1.752)

(-2.795)

(-1.695)

Mother and non-biological partner X Black

3.822*

2.889

5.264**

3.596

(1.723)

(1.325)

(2.157)

(1.621)

Mother and non-biological partner X Hispanic

0.726

0.374

0.698

0.362

(-0.267)

(-0.763)

(-0.289)

(-0.795)

Race/ethnicity (vs. White)

Black

0.473***

0.280***

0.412***

0.310***

0.424***

(-3.701)

(-4.614)

(-3.114)

(-4.319)

(-3.074)

Hispanic

0.600**

0.578*

0.927

0.688

1.016

(-2.179)

(-1.797)

(-0.209)

(-1.174)

-0.044

Female

1.676***

1.652***

1.728***

1.675***

1.738***

(3.774)

(3.686)

(3.856)

(3.687)

(3.836)

Child's age

0.938***

0.943***

0.935***

0.946***

0.937***

(-3.163)

(-2.924)

(-3.055)

(-2.687)

(-2.927)

Proportion of childhood spent living in the South

0.790

0.807

0.839

0.831

0.855

(-1.564)

(-1.414)

(-1.142)

(-1.236)

(-1.027)

No. of children

0.798***

0.783***

0.854**

0.792***

0.856**

(-3.034)

(-3.329)

(-2.018)

(-3.131)

(-1.996)

Parents' education (vs. BA or higher)

Less than high school

0.495**

0.509**

(-2.522)

(-2.416)

High school

0.366***

0.384***

(-4.737)

(-4.488)

Some college

0.522***

0.520***

(-2.958)

(-2.955)

Family wealth without home equity

1.001

1.000

(0.095)

(-0.022)

Proportion of childhood mom is employed

1.099

1.060

(0.393)

(0.245)

Neighborhood quality (vs. excellent/very good)

Good

0.796

0.784

(-1.243)

(-1.320)

Poor

0.781

0.784

(-1.146)

(-1.121)

>50% of neighbors own their homeabc

1.153

1.189

(0.713)

(0.872)

Neighorhood is completely or fairly safe

1.357

1.295

(1.456)

(1.244)

Mother's age at child's birth

20-29

2.201***

2.109***

(3.598)

(3.325)

30 and above

2.108***

2.000***

(3.008)

(2.737)

Proportion of childhood mom is in good health

3.526***

3.501***

(3.691)

(3.707)

Extended family coresidence (vs. never lived with an extended relative)

Lived with an extended relative for <1/4 of childhood

0.705*

0.839

(-1.822)

(-0.911)

Lived with an extended relative for >=1/4 of childhood

0.613**

0.820

(-2.456)

(-0.972)

No. of extended relatives living in R's neighborhood by quartile (vs. below

25th percentile)

25th-50th percentile

1.025

1.172

(0.074)

(0.456)

50th-75th percentile

0.689*

0.715

(-1.848)

(-1.614)

75th-99th percentile

1.071

1.067

(0.329)

(0.300)

Frequency of contact with extended relatives (vs. less than once a month)

One to three times a month

1.458*

1.425

(1.721)

(1.590)

Greater than or equal to once a week

1.387**

1.280

(2.006)

(1.503)

PCG satisfaction with practical support received from extended family

members

0.954

0.970

(-0.707)

(-0.444)

PCG satisfaction with practical support received from extended family

members

1.088

1.048

(1.286)

(0.680)

Wald Test p-value

0.003***

0.002***

0.001***

Constant

50.663***

51.588***

8.333***

31.266***

6.847**

(7.009)

(7.187)

(2.886)

(5.140)

(2.380)

Observations

2,589

t-statistics in parentheses

*** p<0.01, ** p<0.05, * p<0.1

Notes: "Other" race category omitted. Analyses use sample weights to account for the complex multistage clustered design of the PSID.

The Wald test provides evidence for whether the family structure-race/ethnicity interactions are jointly significant. No.=Number.

OR=Odds ratios.

Table 3: Odds ratios from logistic regression models predicting college enrollment, Panel Study of Income Dynamics, Child

Development Supplement (CDS), and Transition into Adulthood Supplement (TAS), 1985-2015

Model 1

Model 2

Model 3

Model 4

Model 5

Proportion of childhood with single mother

0.429***

0.348***

0.915

0.450**

1.007

(-3.302)

(-2.853)

(-0.213)

(-2.048)

(0.016)

Single mother X Black

1.663

0.928

1.631

0.906

(0.974)

(-0.131)

(0.921)

(-0.173)

Single mother X Hispanic

0.510

0.264

0.470

0.239

(-0.803)

(-1.416)

(-0.845)

(-1.498)

Proportion of childhood with mother and non-biological partner

0.228***

0.208***

0.435*

0.241***

0.445*

(-4.497)

(-3.845)

(-1.762)

(-3.492)

(-1.727)

Mother and non-biological partner X Black

1.811

1.044

2.489

1.215

(0.680)

(0.043)

(1.061)

(0.198)

Mother and non-biological partner X Hispanic

6.623*

3.404

7.594*

3.812

(1.800)

(1.028)

(1.806)

(1.084)

Race/ethnicity (vs. White)

Black

0.740

0.552*

0.975

0.638

1.052

(-1.363)

(-1.841)

(-0.073)

(-1.389)

(0.146)

Hispanic

1.009

0.991

1.966*

1.180

2.160*

(0.035)

(-0.028)

(1.739)

(0.487)

(1.924)

Female

1.737***

1.739***

1.841***

1.803***

1.867***

(3.884)

(3.878)

(4.077)

(3.985)

(4.056)

Child's age

1.020

1.023

1.022

1.025

1.019

(0.989)

(1.127)

(0.941)

(1.224)

(0.844)

Proportion of childhood spent living in the South

0.834

0.862

0.902

0.861

0.889

(-1.090)

(-0.893)

(-0.597)

(-0.903)

(-0.683)

No. of children

0.782***

0.777***

0.888

0.778***

0.882

(-3.207)

(-3.331)

(-1.363)

(-3.292)

(-1.447)

Proportion of childhood mom is in good health

1.410

1.407

(0.898)

(0.897)

Parents' education (vs. BA or higher)

Less than high school

0.270***

0.280***

(-3.903)

(-3.823)

High school

0.214***

0.223***

(-6.413)

(-6.221)

Some college

0.373***

0.374***

(-3.948)

(-3.952)

Family wealth without home equity

1.046***

1.047***

(2.965)

(3.056)

Proportion of childhood mom is employed

1.087

1.050

(0.303)

(0.176)

Neighborhood quality (vs. excellent/very good)

Good

0.890

0.888

(-0.605)

(-0.613)

Poor

0.837

0.826

(-0.731)

(-0.781)

>50% of neighbors own their homeabc

1.323

1.364

(1.322)

(1.456)

Neighorhood is completely or fairly safe

1.381

1.405

(1.411)

(1.477)

Mother's age at child's birth

20-29

1.820**

1.780**

(2.463)

(2.325)

30 and above

1.870**

1.756**

(2.230)

(1.988)

Extended family coresidence (vs. never lived with an extended relative)

Lived with an extended relative for <1/4 of childhood

0.572***

0.738

(-2.891)

(-1.545)

Lived with an extended relative for >=1/4 of childhood

0.581**

0.834

(-2.550)

(-0.798)

No. of extended relatives living in R's neighborhood by quartile (vs. below 25th

percentile)

25th-50th percentile

0.963

1.254

(-0.102)

(0.570)

50th-75th percentile

0.767

0.846

(-1.140)

(-0.705)

75th-99th percentile

1.155

1.253

(0.690)

(1.056)

Frequency of contact with extended relatives (vs. less than once a month)

One to three times a month

1.227

1.135

(0.851)

(0.513)

Greater than or equal to once a week

1.134

1.015

(0.736)

(0.083)

PCG satisfaction with practical support received from extended family members

0.928

0.940

(-1.108)

(-0.844)

PCG satisfaction with practical support received from extended family members

1.085

1.026

(1.211)

(0.344)

Wald Test p-value

0.209

0.501

0.191

0.451

Constant

5.891***

5.820***

1.540

4.951**

1.921

(3.226)

(3.287)

(0.546)

(2.473)

(0.747)

Observations

2,436

t-statistics in parentheses

*** p<0.01, ** p<0.05, * p<0.1

Notes: "Other" race category omitted. Analyses use sample weights to account for the complex multistage clustered design of the PSID. The

Wald test provides evidence for whether the family structure-race/ethnicity interactions are jointly significant. No.=Number. OR=Odds

ratios.

Table 4. Average partial effects of socioeconomic stress and extended family embeddedness on children's

educational attainment, Panel Study of Income Dynamics, Child Development Supplement (CDS), and Transition

into Adulthood Supplement (TAS), 1985-2015

Socioeconomic stress

Extended family embeddedness

Family structure

Mother and non-

biological partner

Single

mother

Mother and non-

biological partner

Single

mother

On-time high school completion

Reduced model

-0.176***

-0.221***

-0.177***

-0.220***

(0.024)

(0.039)

(0.025)

(0.039)

Full model

-0.102***

-0.173***

-0.151***

-0.192***

(0.027)

(0.040)

(0.026)

(0.041)

Percentage change

42.0%

21.7%

14.7%

12.7%

Observations

2,589

Socioeconomic stress

Extended family embeddedness

Family structure

Mother and non-

biological partner

Single

mother

Mother and non-

biological partner

Single

mother

College enrollment

Reduced model

-0.155***

-0.169***

-0.157***

-0.170***

(0.025)

(0.039)

(0.026)

(0.040)

Full model

-0.082***

-0.109***

-0.128***

-0.148***

(0.027)

(0.041)

(0.027)

(0.040)

Percentage change

48.4%

35.5%

18.5%

12.9%

Observations

2,436

Standard errors in parentheses

*** p<0.01, ** p<0.05, * p<0.1

Notes: Table 4 describes the average partial effect of family structure on on-time high school completion and college

enrollment in the reduced (baseline) and full models, estimated using the Karlson, Hohm and Breen (KHB) method for

decomposing total effects (Kohler, Karlson and Holm 2011).

ResearchGate has not been able to resolve any citations for this publication.

Giving Unto Others: Private Financial Transfers and Hardship Among Families With Children: Private Financial Transfers and Hardship

Article

Full-text available

Jun 2017
J MARRIAGE FAM

Prior research shows that financial assistance from family and friends is an important source of support for families with children. Research on financial transfers has largely focused on the recipients of transfers, however. In this study, using longitudinal data from the Fragile Families and Child Wellbeing Study (n∼16,000 person-waves), the authors examine the association between the provision of financial assistance to family and friends and material hardship. The results from pooled regression and fixed effects models indicate that providing financial transfers is associated with an increased risk of hardship. The most economically disadvantaged groups, single mothers, those in the bottom income tertile, and Black mothers are the most likely to experience hardship after giving a transfer. These findings have important implications for understanding why families may have difficulty meeting basic and essential needs and how social networks may exacerbate the challenges of escaping poverty and establishing economic self-sufficiency

Understanding Grandfamilies: Characteristics of Grandparents, Nonresident Parents, and Children

Article

Full-text available

Mar 2016
J MARRIAGE FAM

Using data from the Year 9 Fragile Families and Child Wellbeing Study (N ∼ 3,182), we investigated the characteristics grandfamilies (grandparents raising their grandchildren with no parent present, N = 84) and compared them to other key groups, including children's nonresident parents and other economically disadvantaged families with children. Results show that grandparents raising their grandchildren were generally better off in terms of educational attainment, marital status, and economic well-being than the child's parents. Grandparents raising their grandchildren also had characteristics very similar to other disadvantaged mothers. Academic and socioemotional well-being were poorer among children in grandfamilies compared with those living with their mothers, but parenting practices were very similar. These findings suggest that although children in grandfamilies may be at a disadvantage academically and socioemotionally, grandparent caregivers are in many ways similar to other fragile-family mothers. Overall, this study enhances our knowledge of an important yet understudied family type.

The Truly Disadvantaged: The Inner City, The Underclass, and Public Policy.

Article

Sep 1989

Social Capital in the Creation of Human Capital

Article

Jan 2000

J.S. Coleman

Family stress theory and the development of nursing knowledge about family adaptation

Article

Jan 1993

M.A. McCubbin

Families coping with illness: the resiliency model of family stress, adjustment, and adaptation

Article

Jan 1993

Tenuous Ties: The Nature and Costs of Kin Support among Low-income Rural Black Mothers

Article

Apr 2016

Since Stack's (1974) landmark ethnography of kin support in a close-knit group of poor black mothers in the Midwest, there has been ample research on social support among low-income black families. While this body of work has largely painted a picture of the cohesive and supportive nature of families in black communities, recent research has highlighted the limited nature of kin support, especially the support available to low-income black mothers. Much of this work, however, has focused primarily on urban black mothers and paid less attention to the conditions that poor rural black mothers face when seeking and giving family support. Using longitudinal ethnographic data from a sample of 16 low-income black mothers in the rural South, we draw on social exchange, negotiated-order, and social capital perspectives to scrutinize the nature and costs of kin support in family networks marked by limited resources, instability, and chronic need. Our findings reveal the centrality of problematic resources and unpredictable family networks as conditions that diminish mothers' autonomy and compromise important "side bets" as mothers seek out, manage, and repay support. Implications of this study for theories of social support and social capital are also discussed.

Won't You Be My Neighbor? Race, Class and Residence in Los Angeles

Article

Jan 2006

Camille Charles

Los Angeles is a city of delicate racial and ethnic balance. As evidenced by the 1965 Watts violence, the 1992 Rodney King riots, and this year's award-winning film Crash, the city's myriad racial groups coexist uneasily together, often on the brink of confrontation. In fact, Los Angeles is highly segregated, with racial and ethnic groups clustered in homogeneous neighborhoods. These residential groupings have profound effects on the economic well-being and quality of life of residents, dictating which jobs they can access, which social networks they can tap in to, and which schools they attend. In Won't You Be My Neighbor? sociologist Camille Zubrinsky Charles explores how modern racial attitudes shape and are shaped by the places in which people live. Using in-depth survey data and information from focus groups with members of L.A.'s largest racial and ethnic groups, Won't You Be My Neighbor? explores why Los Angeles remains a segregated city. Charles finds that people of all backgrounds prefer both racial integration and a critical mass of same-race neighbors. When asked to reveal their preferred level of racial integration, people of all races show a clear and consistent order of preference, with whites considered the most highly desired neighbors and blacks the least desirable. This is even true among recent immigrants who have little experience with American race relations. Charles finds that these preferences, which are driven primarily by racial prejudice and minority-group fears of white hostility, taken together with financial considerations, strongly affect people's decisions about where they live. Still, Charles offers reasons for optimism: over time and with increased exposure to other racial and ethnic groups, people show an increased willingness to live with neighbors of other races. In a racially and ethnically diverse city, segregated neighborhoods can foster distrust, reinforce stereotypes, and agitate inter-group tensions. Won't You Be My Neighbor? zeroes in on segregated neighborhoods to provide a compelling examination of the way contemporary racial attitudes shape, and are shaped by, the places where we live.

Nuclear Family Values, Extended Family Lives: The Power of Race, Class, and Gender

Article

Apr 2012

Natalia Sarkisian

Nuclear Family Values, Extended Family Lives shows how the current emphasis on the nuclear family - with its exclusion of the extended family - is narrow, even deleterious, and misses much of family life. This omission is tied to gender, race, and class.

Unequal Childhoods: Class, Race, and Family Life

Book

Sep 2011

Annette Lareau

Class does make a difference in the lives and futures of American children. Drawing on in-depth observations of black and white middle-class, working-class, and poor families, Unequal Childhoods explores this fact, offering a picture of childhood today. Here are the frenetic families managing their children's hectic schedules of "leisure" activities; and here are families with plenty of time but little economic security. Lareau shows how middle-class parents, whether black or white, engage in a process of "concerted cultivation" designed to draw out children's talents and skills, while working-class and poor families rely on "the accomplishment of natural growth," in which a child's development unfolds spontaneously—as long as basic comfort, food, and shelter are provided. Each of these approaches to childrearing brings its own benefits and its own drawbacks. In identifying and analyzing differences between the two, Lareau demonstrates the power, and limits, of social class in shaping the lives of America's children. The first edition of Unequal Childhoods was an instant classic, portraying in riveting detail the unexpected ways in which social class influences parenting in white and African American families. A decade later, Annette Lareau has revisited the same families and interviewed the original subjects to examine the impact of social class in the transition to adulthood.

Racial/Ethnic Differences in the Association Between Family Structure and Children's Education

Abstract

Recommended publications

Dropping Out Among Language Minority Youth

Racial/Ethnic Differences in the Association Between Family Structure and Children's Education

The Color, Class, and Context of Family Structure and Its Association with Children's Educational Pe...

Extended family households among children in the United States: Differences by race/ethnicity and so...

Race and ethnic differences in step‐ versus biological parent support to adult children