Taking Diversity Seriously: Within-Group Heterogeneity in African American Extended Family Support Networks

Abstract

Objective:

This study examined the correlates of involvement in extended family social support networks among African Americans.

Background:

Previous literature has documented the importance of informal social support from extended family members for the African American population. Most research has investigated black-white differences in network involvement or has focused on impoverished African American families. Both approaches conceal important within-group variation in participation among the total African American population.

Method:

This study relied on nationally representative data from the African American sub-sample of the National Survey of American Life (n=3,538). It employed ordinary least squares regression analysis to examine the sociodemographic and family factors that are associated with four key measures of involvement in extended family support networks: receiving and providing extended family support, frequency of family contact, and degree of subjective closeness.

Results:

African Americans routinely interacted with members of their family, displayed a high degree of family closeness, and exchanged support fairly frequently. Findings also revealed significant variation in network involvement by sociodemographic characteristics: women, younger adults, and Southerners were typically most involved; individuals who experienced greater material hardship, were previously incarcerated, or served in the military reported less involvement. Results also showed that family closeness and family contact were particularly salient factors shaping the extent to which network members engaged in support exchanges.

Conclusion:

The magnitude of within-group heterogeneity in network involvement underscores the importance of considering issues of intragroup diversity in the developing literature on African American extended family networks.

Introduction

An established tradition of family research on African Americans documents the importance of informal social support from extended family members for the population as a whole, and especially for vulnerable subgroups such as single mothers and older adults. Extended family members provide valuable forms of support in a variety of circumstances, including offering emotional support to those dealing with mental health challenges (Levine, Taylor, Nguyen, Chatters, & Himle, 2015; Taylor et al., 2015), childcare to single mothers (Jayakody et al., 1993), monetary assistance to family members (Sarkisian & Gerstel, 2004), as well as providing comprehensive caregiving to older relatives (Dilworth-Anderson et al., 2002). Despite their acknowledged importance to African Americans, extended family networks and the supports they provide remain under-researched topics in family studies. The limited scholarship in this area overwhelmingly compares network involvement between Black and white Americans or relies on samples of African Americans from impoverished backgrounds, obscuring important variation in participation within the African American population.

The present study uses nationally representative data from the National Survey of American Life to investigate the extended family support networks of African Americans. In particular, it examines the correlates of the frequency of interacting with family, the degree of subjective family closeness, and both the receipt and provision of informal social support. This study also examines how sociodemographic characteristics and family factors (e.g., contact with family, subjective family closeness) are associated with both receiving and providing to family members. The literature review begins with a discussion of research findings on African American extended family support networks. This is followed by a discussion of the theoretical perspectives framing our analysis—the cultural variant model (Allen, 1978), as well as the family solidarity model and contingency theory. A presentation of the conceptual model guiding our analysis is included in our discussion of the family solidarity model. This section concludes with a description of the focus and goals of the present investigation.

Background and Theoretical Perspectives

Extended Family Social Support Networks of African Americans

Informal social support refers to the assistance provided to or received from individuals outside formal, professional settings (e.g., healthcare institutions or social service agencies). It generally involves emotional, informational, and instrumental support from family members, friends, or informal social groups (Campbell, Wynne-Jones, & Dunne, 2011). Among African Americans, extended family members are a significant source of informal social support. Recent studies indicate that 50%−75% of Black adults report providing and/or receiving some form of informal support from their extended family networks on a regular basis (Lincoln et al., 2013; Silverstein & Waite, 1993; Woodward et al. 2010). Much of the research on African American extended family social support networks examines differences in the levels of support that are given and received relative to white Americans. However, there is a lack of consensus regarding the extent to which African Americans differ from whites in their involvement in extended family social support networks. Some studies suggest that African Americans are more likely than whites to exchange informal social support (e.g., Benin & Keith, 1995; Gertsel & Gallager, 1994; Hogan, Hao & Parish, 1990), whereas other scholarship suggests that they are equally likely to do so (e.g., Silverstein & Waite, 1993). Still other work indicates that Blacks are less likely to receive substantial financial assistance from their relatives than whites (e.g., Sarkisian & Gerstel, 2004; Jayakody, 1998), leading some researchers to assert that African Americans benefit less from their extended family networks than white Americans. Contrarily, recent research by Assari et al. (2018) indicates that compared to whites, the support networks of African Americans are more protective of depressive symptoms. Discrepancies in research findings have largely been attributed to differences in how informal social support has been operationalized across studies, the age and gender of the samples under investigation, the socioeconomic resources available to respondents, and other life circumstances of sample members (e.g., caregiving, helping with mental health challenges [Sarkisian & Gerstel, 2004; Silverstein & Waite, 1993; Taylor, Hernandez, Nicklett, Taylor, & Chatters, 2014]).

While much attention has been paid to black-white differences in informal social support exchanges, less work has focused on within-group heterogeneity in informal social support from extended family. However in a rigorous assessment of racial differences in kin support exchanges, Sarkisian and Gerstel’s (2004) analysis of the National Survey of Families and Households demonstrated that (1) cross-race similarities were more common than differences between groups and (2) within-group heterogeneity on the basis of gender, cultural preferences, and nuclear and extended family composition were more salient predictors of the provision and receipt of support than considerations of race in a simple, dichotomous way (e.g., black-white). These findings suggest that a narrow focus on race differences conceals important within-group heterogeneity in the factors that shape receiving and providing family support. Thus, more research is needed to understand the contextual factors that are associated with family support exchanges among African Americans.

African American Family Theory and Frameworks

The conceptual framework guiding our study is the cultural variant perspective of Black family life as articulated in the seminal work of Allen (1978). Allen (1978) identified three ideological perspectives on the study of Black families—the cultural equivalent, cultural deviant, and cultural variant perspectives—that still resonate in discussions of and research on Black family life today (see Sarkisian, 2007). The cultural equivalent perspective is focused on highlighting characteristics that Black families share with white families. This perspective evaluates Black families from an implicit white, middle-class standard. Any divergences from this standard are minimized, rather than treated as the manifestation of distinctive cultural norms existing within Black families. The cultural deviant perspective, on the other hand, recognizes that Black family life is distinct from white family life, but characterizes these differences as unequivocally negative and pathological. So-called family dysfunction is seen as originating from a history of slavery that has weakened the Black family and disordered family and gender roles (i.e., Black matriarchy). Similar to the cultural equivalent perspective white, middle-class families are considered the normative ideal, and any deviation in family function or structure from this benchmark is taken as evidence of family dysfunction and pathology.

The cultural variant perspective, in contrast, views Black families as constituting a distinctive cultural form, but does not characterize apparent differences as pathological. Rather, this perspective recognizes the importance of the social and cultural environments in which families reside as critical contextual factors that shape family structure and functioning. The cultural variant perspective argues for universality in understanding family forms and function, while acknowledging that situational factors constrain and shape behaviors in characteristic ways. This perspective makes explicit the fact that Black and white families frequently inhabit vastly different sociocultural contexts in U.S. society. Black family life is viewed as an adaptive organization that demonstrates both human agency and planful intention and is capable of modifying its responses to particular and dynamic circumstances. By making racially based social disparities in life circumstances explicit, the cultural variant perspective complicates questions concerning the origins of the problems facing Black families and exposes society’s accountability for prevailing social conditions. Further, cautioning against a radical functionalist stance that views all aspects of Black family life as culturally derived, Allen (1978, p. 125–126) suggests the need for a thorough analysis of Black family life within sociocultural context. Extensions of the cultural variant perspective note that differences in access to economic resources among Black Americans also lead to distinct patterns of family functioning (Billingsley, 1992; Cross, 2021; Dow, 2019; McLoyd et al., 2000; Lacy, 2007, Patillo 1999).

Family Solidarity Model

The family solidarity model is the basis of our conceptual model displayed in Figure 1 and it also guides our analysis. Both the family solidarity framework and Figure 1 predict that positive family behaviors (e.g., frequency of contact) and attitudes (e.g., emotional closeness) are positively related to each other, and furthermore, are positively associated with family support exchanges. The family solidarity theory is based on the well-established work of Bengtson and associates (Bengtson, Giarrusso, Mabry, & Silverstein, 2002; McChesney & Bengtson, 1988), and it has been used with research on African Americans (Taylor et al., 2016) and non-Latino whites (Huo et al., 2019). The family solidarity model maintains that family cohesion is central to relations among relatives, and the receipt and provision of assistance is shaped by familial attitudes, sentiments, and behaviors (Bengtson, Giarrusso, Mabry, & Silverstein, 2002). Dispositions towards family life (e.g., meaning and function of the family), sentiments about family members (e.g., feelings of closeness), and interactions with relatives (e.g., level of contact) are expected to be correlated with each other and also support exchanges (Clarke, Preston, Raskin, & Bengtson, 1999; Parrott & Bengtson, 1999; McChesney & Bengtson, 1988; Taylor, Forsythe-Brown, Lincoln, & Chatters, 2015). Specifically, individuals with a stronger orientation towards the family and those who express more affirmative feelings about family members will have greater contact with relatives and engage in more frequent exchanges of social support. Therefore, we anticipate that subjective family closeness and frequency of contact will be positively related, and both closeness and contact will be positively associated with the receipt and provision of support (Hypothesis 1).

Figure 1.

A Conceptual Model of Extended Family Social Support Networks of African Americans.

Contingency Theory

While the family solidarity model helps identify dimensions of family relations that influence network involvement, contingency theory focuses on individual-level characteristics. This theory posits that individuals’ receipt and provision of support are proportional to and contingent on their needs (Eggebeen & Davey, 1998 Fingerman et al., 2009; Schoeni, 1997). Resource flows are highly dependent on crises or other disruptive life events, and families are most likely to aid members who exhibit greater need for assistance. For instance, Huo et al., (2019) found that parents of adult children were more likely to help their progeny who were struggling financially or with substance abuse. Additionally, adult children have been known to increase support to elderly parents with health issues (e.g., Eggebeen & Davey, 1998; Silverstein, Gans, and Yang, 2006). Given contingency theory’s assertion that resource exchanges between family members are largely need-based, we expect that individuals experiencing greater social or economic hardship (such as lower levels of income and education) will be less likely to provide and more likely to receive support (Hypothesis 2). Prior studies using contingency theory suggest that resources typically flow from the older to younger generation within family systems (e.g., Fingerman et al., 2009). Therefore, in general, we expect that younger adults will report greater receipt of support (Hypothesis 3a). Nonetheless, prior studies using contingency theory also indicate that adult children may increase support to parents in recognition of their advancing age, health declines, and potential needs for assistance (Silverstein et al., 2006). Thus, among older persons, those who are parents may report greater receipt of assistance than older persons without children (Hypothesis 3b). We investigate this possibility by including an age × parental status interaction in our analyses

Beyond social and economic need, findings from recent studies highlight several additional sociodemographic background characteristics that are likely associated with extended family support network involvement (as shown in the conceptual model [Figure 1]). Regardless of age, parenthood is likely related to network involvement. African American parents’ interactions with extended relatives often directly revolve around their children (Billingsley, 1992; Sudarkasa, 1996; Dow, 2019). This may include routine interactions with the extended family network that emphasize family belonging and connections, special occasions, and accomplishments involving children (e.g., birthdays, graduations, sports). For parents with adult children, interactions often center on the adult child and, if relevant, adult children’s spouses (daughter- and son-in laws) and children (their grandchildren). Hence, we anticipate that parents will report greater closeness and contact with extended relatives than adults without children (Hypothesis 4).

Network involvement is also likely to vary by gender. Research on kinscripts (the interplay of family ideology, norms, and behaviors over the life course) indicates that women interact with family members more frequently and conduct the lion’s share of work related to maintaining family ties (Stack & Burton, 1993). From an early age, women are socialized to perform gender-specific roles that are more oriented towards domestic life (e.g., caregiving and housework), and they tend to have broader social networks than men, including ties to extended relatives (Lytton & Romney, 1991; Kawachi & Berkman, 2001). Thus, we expect that women will be more involved in their family networks and will both provide and receive support more frequently than men (Hypothesis 5).

Network involvement may also differ by marital status. Some scholars have described marriage as a “greedy” or privatizing institution that privileges relations between spouses over all others, except potential offspring (e.g., Gerstel & Sarkisian, 2006). This may result in weaker ties to persons besides spouses and children, including extended relatives (Gerstel & Sarkisian, 2006). Conversely, singlehood has been associated with more social connections. (Sarkisian & Gerstel, 2015). Less is known about the social ties of formerly married persons. After marital dissolution, family support relationships that involved one’s former in-laws may be severed, resulting in less frequent family involvement and assistance overall, or they could be associated with increased involvement and support, as relations with family and friends become more salient. Thus, we expect that single persons will have higher levels of network involvement than married persons; this may also be the case for individuals who experience marital dissolution, but only if their reduction in social ties due to loss of a spouse is offset by increased involvement with extended relatives (Hypothesis 6).

While few studies explore geographic differences in family support networks, a small literature points to a clear Southern advantage with respect to network involvement. For example, Sechrist et al., (2007) found that among adult children, Southerners had higher levels of emotional closeness and contact (in person, phone and/or correspondence) with their mothers than did their non-Southern counterparts; regional differences were consistent across race and gender groups. This may occur because the South is distinctive with respect to a broad range of social and political attitudes and behaviors (Hurlbert, 1989). In particular, research emphasizes a distinctive Southern culture in which primary institutions such as the family and religion are more central in daily life relative to other regions (Hurlbert, 1989; Taylor et al., 2004). Consonant with these findings, we expect for Southerners to be subjectively closer to and in frequent contact with their extended relatives and to exchange support most frequently (Hypothesis 7).

Mass incarceration has significantly impacted African American family life and it is likely related to network involvement (Haskins and Lee, 2016; Lee et al. 2015; Miller, 2021; Patterson, Talbert, and Brown, 2021; Wakefield et al., 2016; Wildeman, 2014). Over the last half century, the experience of incarceration has shifted from being a rare event that touched the lives of a small fraction of the U.S. population to a common life course event affecting many African American families, especially those from economically disadvantaged backgrounds (Pettit & Wester, 2004; Lee et al. 2015). Imprisonment is associated with a host of disadvantages that persist even after individuals have served their time; formerly incarcerated persons, especially convicted felons, are marginalized in a variety of ways including being routinely excluded from public services, educational grants, voting, jury service, and many stable, high-paying jobs (Assari et al., 2018; Miller, 2021; Patterson, Talbert, and Brown, 2021; Wakefield et al., 2016; Western, 2006; Uggen, 2010). Upon reentry into society, these adverse circumstances not only negatively impact former prisoners, but also their families. Research on prisoner reentry describes the importance of extended family members in helping formerly incarcerated persons adjust to post-incarceration life. Familial support can be beneficial with regard to transportation needs, financial assistance, finding employment, reducing substance abuse, as well as mitigating post-release depression (see reviews on the role of extended family support on prison reentry by Naser & La Vigne, 2006 and Mowen et al., 2019). Given the significant barriers that former prisoners face, they are often not well positioned to contribute to their families, while at the same time needing significant amounts of financial, social, and psychological support from them (Wakefield et al., 2016). However, imprisonment often negatively impacts the development and maintenance of family ties, which may lead to reduced feelings of closeness and less frequent contact (Miller 2021). As such, we expect formerly incarcerated persons to report less closeness and contact to family than those who have never been imprisoned (Hypothesis 8a). Further, we anticipate that being previously incarcerated will be either not associated or negatively associated with receipt of support (Hypothesis 8b).

While not always explicitly stated, prior theories and empirical findings linking sociodemographic background characteristics (e.g., gender, marital status, region, and incarceration history) to network involvement imply that family closeness and contact attenuate the relationship between these factors and support exchanges. For example, previous work suggests that gender is positively correlated with contact and closeness to extended relatives (Stack & Burton, 1993) and increases in these factors are positively linked with the provision and receipt of support (Bengtson et al., 2002). Likewise, the experience of incarceration is anticipated to be negatively associated with family contact and closeness (Miller, 2021), and lower levels of these factors are inversely related to support exchanges (Bengtson et al., 2002). Hence, in our conceptual model (Figure 1) and analyses, we treat family closeness and contact as factors that are directly associated with the provision and receipt of support and that partially mediate the relationship between sociodemographic background and social support.

The Current Study

The present study builds on the first major investigations of sociodemographic correlates of family support among African Americans that were initiated over 25 years ago using the National Survey of Black Americans (Hatchett & Jackson, 1993; Taylor, 1986). We use data from the National Survey of American Life that includes a comprehensive set of both dependent and independent variables. The current study of informal support among African Americans explores both sociodemographic and family factors (e.g., subjective family closeness, frequency of contact with extended family members) as correlates of receiving and providing social support from extended family.

This study makes several contributions to the literature on African American families. First, it investigates informal social support networks, an area in which there remains a paucity of research, despite the well-known importance of extended family networks among African Americans. Second, it investigates both the receipt and provision of support. Research has noted the lack of work on the provision of support, especially among studies focused on African American extended family networks (Sarkisian & Gerstel, 2004). Third, this analysis includes a measure of material hardship (i.e., difficulty meeting basic living expenses) as an indicator of socio-economic position, in addition to income and education. Material hardship is a novel indicator that has not been used frequently in research on general social support or in research on African American family life. Material hardship has advantages over income and education in that it provides a proximal indicator of the availability of economic resources to meet needs in seven key life domains (e.g., basic expenses, rent/mortgage, utilities, evictions). The use of material hardship in this analysis of social support from extended family adds to a growing literature that examines the impact of material hardship on a variety of social and psychological outcomes (e.g., Gershoff et al., 2007). Fourth, because of this study’s robust sample size, we are able to investigate group differences in traditional demographic correlates (i.e., age, gender), as well as expanded information on marital status (i.e., remarriage, cohabitation) and life circumstances (i.e., military service, incarceration). Currently, there is very little information on marital status differences in support networks among African Americans, and no information exists on how cohabitation or remarriage are related to informal support. Similarly, military service and urbanicity have not been investigated with regards to African American family support networks (because they may be related to network involvement, we explore this possibility in our analyses). This study’s investigation of a full range of extended family network variables, the inclusion of a diverse set of sociodemographic and social background characteristics, and novel assessments of life circumstances (e.g., military service and incarceration history), offers a unique opportunity to investigate social and demographic heterogeneity across diverse dimensions of African American extended family support networks.

Data and Methods

Data

The National Survey of American Life: Coping with Stress in the 21st Century (NSAL) was collected by the Program for Research on Black Americans at the University of Michigan’s Institute for Social Research (Jackson et al., 2004). The field work for the study was completed by the Institute for Social Research’s Survey Research Center, in cooperation with the Program for Research on Black Americans. The NSAL sample has a national multi-stage probability design. Most of the interviews were conducted face-to-face (86%) in respondents’ homes, and the remaining (14%) were telephone interviews. Respondents were compensated for their time. The data collection was conducted from 2001 to 2003. A total of 6,082 interviews were conducted with persons aged 18 or older. This study utilizes the African American sub-sample (n=3,570). The size of the analytic sample for our analysis is 3,538.

The African American sample is the core sample of the NSAL. The core sample consists of 64 primary sampling units (PSUs). Fifty-six of these primary areas overlap substantially with existing the Survey Research Center’s national sample primary areas. The remaining eight primary areas were chosen from the South in order for the sample to represent African Americans in the proportion in which they are distributed nationally. The overall response rate was 72.3%. Final response rates for the NSAL two-phase sample designs were computed using the American Association of Public Opinion Research (AAAPOR) guidelines (for Response Rate 3) (AAPOR, 2006) (see Heeringa et al., 2004 and Jackson et al., 2004 for a more detailed discussion of the NSAL sample).

Measures

Family network variables.

The present analysis investigates four measures of family network involvement (i.e., subjective family closeness, frequency of interacting with extended family members, frequency of receiving support, and frequency of providing support). Degree of subjective family closeness is measured by the question: “How close do you feel towards your family members? Would you say (4) very close, (3) fairly close, (2) not too close, or (1) not close at all?” Frequency of contact with family members is measured by the question: “How often do you see, write or talk on the telephone with family or relatives who do not live with you? Would you say (7) nearly every day, (6) at least once a week, (5) a few times a month, (4) at least once a month, (3) a few times a year, (2) hardly ever or (1) never?” The variable, family contact had 32 missing cases and family closeness had 33 missing cases (the same 32 missing cases and an additional one).

Frequency of receiving support is measured by the item: “How often do people in your family -- including children, grandparents, aunts, uncles, in-laws and so on -- help you out? Finally, frequency of giving support is measured by the question: “How often do you help out people in your family -- including children, grandparents, aunts, uncles, in-laws and so on? Response formats for these two questions used a 4-point Likert scale with a response range of never = 1 to very often = 4. The frequency of receiving support had 203 missing cases including 19 respondents who volunteered that they did not have any family members, and 173 who volunteered that they did not receive help because they did not need help from their extended family. The 19 respondents who volunteered that they did not have any family members are not asked any of the family questions and thus are included in the missing cases for the other family variables. The variable frequency of providing support had 77 missing cases including 45 who volunteered that they did not provide support because their family members did not need it.

Sociodemographic variables.

Several sociodemographic variables are included in this analysis as independent variables (i.e., age, gender, marital status, region, urbanicity, parental status, education, material hardship, family income, military service, and incarceration history). Age is a continuous measure, and gender is coded as male or female (reference). Marital status is a categorical variable indicating whether the respondent is married (reference), remarried, cohabiting, separated, divorced, widowed, or never married. Region is a categorical variable: Northeast, North Central, South (reference) and West. Urbanicity is categorized as urban (reference) or rural. Parental status is a binary variable indicating whether an individual has at least one child (no=0, 1=yes). There were no missing data for gender, age, region, or urbanicity, 17 missing cases for marital status, and 17 missing cases for parental status. Education is coded in years; the log of income is used in the regression analysis. The staff of the Program for Research on Black Americans imputed 74 missing cases for education (1.2% of the total NSAL sample) and 773 cases for income (12.7% of the total NSAL sample). Material hardship is a seven-item index assessing whether respondents could meet basic expenses, pay full rent or mortgage, pay full utilities, had utilities disconnected, had telephone disconnected, were evicted for non-payment, and could not afford leisure activities in the past 12 months. A higher score on this index indicates higher levels of economic hardship (Cronbach’s alpha =.76). Material hardship had 42 missing cases. Military service is a binary variable indicating whether the respondent has ever served in the military (no=0, yes=1). Similarly, incarceration history indicates whether a person has been incarcerated including prison, jail, juvenile detention, and reform school (no=0, yes=1). There were 43 missing cases for military service and 51 missing cases for incarceration history.

Analysis Strategy

Ordinary least squares regression was utilized to identify the correlates of family support networks. We report both standardized and unstandardized regression coefficients. All analyses were conducted in SAS 9.4, and all analyses utilize sample weights to obtain results that are generalizable to the African American population. Weights in the NSAL data account for unequal probabilities of selection, non-response, and post-stratification such that respondents are weighted to their numbers and proportions in the full population. SAS uses the Taylor expansion technique for calculating the complex-design based estimates of variance. This corrects standard error estimates in analysis using complex sample designs (i.e., clustering and stratification). Based upon research indicating the importance of adult children in support networks (e.g., Fingerman et al., 2009; Schoeni, 1997) and previous research on the receipt of informal support from adult offspring among African Americans (Taylor, 1986), an interaction between age and parental status was tested for each of the regression models. The age × parent interaction was only retained in models where it achieved significance.

Results

The distribution of the study variables is presented in Table 1. On the whole, the demographic characteristics of sample members are representative of the U.S. population of African American adults. Roughly half of the sample was female, and the mean age was 42. Approximately 4 out of 10 respondents were married, remarried, or cohabiting, 82% reported that they were parents, 9 out of 10 respondents resided in urban areas, and more than half of the sample resided in the South. On average, respondents had an income of $37,000, 12 years of schooling, and a mean material hardship score of .89. Thirteen percent served in the military and 17% were previously incarcerated. With regards to family network variables, 72% of respondents reported being very close to their families and 50% indicated that they interacted with their extended family members nearly every day. In terms of family support, 47% reported that they provided support to their family members very often, and 33% reported that they received support from their families very often.

Table 1.

Demographic characteristics of the sample and distribution of study variables, National Survey of American Life (NSAL)

	Percent	N	Mean	S.D.	Range
Gender
Female	55.97	2299
Male	44.03	1271
Age		3570	42.32	14.50	18–93
Family income		3570	36832.66	33068.07	0–520000
Education		3570	12.43	2.23	0–17
Marital status
First marriage	24.00	707
Remarried	8.91	253
Cohabiting	8.74	260
Separated	7.16	286
Divorced	11.75	524
Widowed	7.90	353
Never married	31.55	1170
Region
Northeast	15.69	411
North Central	18.81	595
South	56.24	2330
West	9.25	234
Material hardship		3528	0.89	1.31	0–7
Military service
Served in military	13.28	417
Did not serve in military	86.72	3110
Incarceration
Previously incarcerated	16.70	531
Never incarcerated	83.30	2988
Parental status
Parent	81.57	2964
Non-Parent	18.43	588
Urbanicity
Urban	93.89	3320
Rural	6.11	250
Subjective Family closeness		3537	3.64	0.58	1–4
Not close at all	1.50	56
Not too close	4.99	176
Fairly close	21.46	746
Very close	72.05	2557
Family contact		3538	6.07	1.17	1–7
Never	1.31	43
Hardly ever	2.23	81
A few times a year	2.93	82
At least once a month	3.85	132
A few times a month	11.06	351
At least once a week	28.34	985
Nearly every day	50.29	1864
Give support		3492	3.27	0.73	1–4
Never	2.60	101
Not too often	15.59	560
Fairly often	34.33	1186
Very often	47.48	1645
Receive support		3366	2.82	0.92	1–4
Never	11.87	393
Not too often	27.38	939
Fairly often	34.33	936
Very often	32.96	1098

Notes: Percentages are weighted and frequencies are unweighted. Percentages and sample sizes are presented for categorical variables. Means and standard deviations are presented for continuous variables. N=number of sample members. S.D.=standard deviation. No.=number.

Table 2 presents the results of the regression analysis for subjective family closeness and frequency of contact. Region, marital status, material hardship, parental status, military service, and incarceration history were significantly associated with subjective family closeness. Southerners reported higher levels of family closeness than respondents who reside in the West. Separated respondents were less close to their family members than married respondents. Respondents with higher levels of material hardship, those who had served in the military, and previously incarcerated respondents reported lower levels of familial closeness than their counterparts. The interaction between age and parental status was also significant. The nature of this interaction is displayed in Table 4. It shows that age was unrelated to family closeness for respondents who were parents, whereas age was negatively associated with familial closeness for non-parents. For instance, among African Americans without children, 94% of those aged 18–25 were subjectively close to their families. However, this percentage declined to 76% among those who were 65 and older.

Table 2.

Ordinary least squares regression models identifying correlates of subjective family closeness and frequency of family contact among African Americans, National Survey of American Life (NSAL)

	Model (1)		Model (2)		Model (3)
	Subjective Family Closeness		Family Contact		Family Contact
	ß	b(SE)	ß	b(SE)	ß	b(SE)
Age	−0.03	−0.00 (0.00)	−0.09*	−0.01 (0.00)*	−0.13**	−0.01 (0.00)**
Female (vs. male)	−0.05	−0.05 (0.03)	0.30***	0.30 (0.07)***	0.34***	0.34 (0.06)***
Education	0.01	0.00 (0.01)	0.04	0.02 (0.01)	0.03	0.01 (0.01)
Household income	0.02	0.02 (0.02)	0.05	0.05 (0.04)	0.03	0.04 (0.04)
Region (vs. South)
Northeast	−0.03	−0.03 (0.04)	−0.27**	−0.27 (0.09)**	−0.25**	−0.25 (0.09)**
North Central	−0.05	−0.05(0.04)	−0.00	−0.00 (0.09)	0.03	0.03 (0.08)
West	−0.16*	−0.16 (0.06)*	−0.32***	−0.32 (0.07)***	−0.22**	−0.22 (0.07)**
Marital status (vs. first marriage)
Remarried	−0.08	−0.08 (0.06)	−0.04	−0.04 (0.08)	0.01	0.01 (0.09)
Cohabiting	−0.03	−0.03 (0.05)	0.03	0.03 (0.09)	0.06	0.06 (0.09)
Separated	−0.12*	−0.12 (0.06)*	−0.07	−0.07 (0.13)	0.02	0.02 (0.11)
Divorced	−0.09	−0.09 (0.05)	−0.03	−0.03 (0.11)	0.03	0.03 (0.10)
Widowed	0.01	0.01 (0.05)	0.13	0.13 (0.17)	0.10	0.10 (0.16)
Never married	−0.01	−0.01 (0.04)	−0.10	−0.10 (0.08)	−0.09	−0.09 (0.07)
Material hardship	−0.06***	−0.05 (0.01)***	−0.08**	−0.06 (0.02)**	−0.04	−0.03 (0.02)
Served in military (vs. did not serve)	−0.13**	−0.13 (0.04)**	−0.23*	−0.23 (0.11)*	−0.13	−0.13 (0.09)
Previously incarcerated (vs. never incarcerated)	−0.09*	−0.09 (0.04)*	−0.26*	−0.26 (0.10)*	−0.18*	−0.18 (0.09)*
Family Contact	--	--	--	--	--	--
Subjective Family Closeness	--	--	--	--	0.43***	0.74 (0.06)***
Parent (vs. non-parent)	0.13*	−0.19 (0.09)*	0.33***	0.33 (0.07)***	0.29***	0.29 (0.07)***
Age × parent status	0.11*	0.01 (0.00)*	--	--	--	--
Urbanicity
Urban	−0.04	−0.04 (0.07)	0.44***	0.44 (0.07)***	0.47***	0.47 (0.07)***
Constant	3.67***	3.51 (0.20)***	5.41***	4.96 (0.37)***	5.32***	2.48 (0.42)***
F	10.02***		44.18***		47.55***
R-squared	0.05		0.07		0.20
Observations	3437		3438		3436

^*p < .05;

^**p< .01;

^***p < .001

Notes: Data are weighted to account for unequal probabilities of selection, non-response, and post-stratification. ß= Standardized coefficient. b=Unstandardized coefficient. SE=Standard error.

Table 4.

Multivariate cross-tabular analysis of the relationships between family closeness, providing support, receiving support and age, controlling for whether respondents had children.

Age	Family Closeness (N=3437)		Provide Support (N=3393)		Receive Support (N=3267)
	Have Children	No Children	Have Children	No Children	Have Children	No Children
18–25	91.78	94.05	98.91	98.74	95.80	95.37
26–34	92.64	91.11	96.86	99.36	92.93	95.22
35–54	93.70	93.71	97.65	96.47	85.73	82.10
55–64	95.03	84.90	96.74	90.93	81.97	60.11
65-older	96.53	75.84	96.43	84.67	83.42	71.65

Notes: Cross-tabular analyses for Family Closeness, Provide Support, and Receive Support all control for gender, education, household income, region, marital status, material hardship, military service, incarceration history, and urbanicity.

Models 2 and 3 estimate the extent to which family closeness and sociodemographic characteristics are related to frequency of contact with extended family members. In Model 2, gender, age, region, material hardship, military service, incarceration history, parental status, and urbanicity were significantly associated with frequency of family contact. In particular, women interacted more frequently with extended family members than men, older African Americans interacted less frequently than their younger counterparts, parents interacted more frequently than non-parents, urban respondents interacted more frequently than their rural counterparts, and persons who reside in the South interacted more frequently with family members than those residing in the Northeast and West. Additionally, persons who had higher levels of material hardship, those who were previously incarcerated, and those who served in the military had contact with relatives less frequently than those with lower levels of material hardship, and those who were never incarcerated, or served in the armed forces. Model 3 added subjective family closeness to the regression analysis for family contact. Material hardship, military service, incarceration history, and residing in the West, were no longer related to family contact once subjective family closeness was taken into consideration; the other relationships remained statistically significant. Subjective family closeness was significantly associated with frequency of family contact; individuals who were subjectively closer to their extended families had higher levels of contact. An examination of the standardized coefficients in Model 3 reveals that urbanicity (ß =.47), subjective family closeness (ß =.43) and gender (ß =.34) were the largest correlates of family contact.

Table 3 presents findings from regression analyses identifying correlates of providing support and receiving support. Analyses for frequency of providing support are presented in Models 1 and 2. In Model 1, age, gender, military service, and region were significantly related to frequency of providing support. Older respondents provided support less frequently than younger respondents and women provided support more frequently than men. Respondents who resided in the Northeast provided support less frequently than Southerners, and those who served in the military provided less support than those who did not serve. The interaction between age and parental status was also associated with the provision of support. When subjective family closeness and frequency of contact were included in Model 2, we found that military service was no longer significantly related to this outcome, and family closeness and contact were positively associated with providing support. Specifically, respondents who indicated that their families were subjectively closer and who interacted with their families on a more frequent basis reported that they provided support more frequently than their counterparts. The interaction term for parental status and age indicates that among parents, there was no relationship between age and providing support, whereas among non-parents, age was negatively related to providing support. This is evident in the percentages reported in Table 4; for respondents without children, 99% of those aged 18–25 provided support to their family members, whereas only 85% of respondents aged 65 and over provided family support.

Table 3.

Ordinary least squares regression models identifying correlates of frequency of giving and receiving family support among African Americans, National Survey of American Life (NSAL)

	Model (1)		Model (2)		Model (3)		Model (4)
	Providing Support		Providing Support		Receiving Support		Receiving Support
	ß	b(SE)	ß	b(SE)	ß	b(SE)	ß	b(SE)
Age	−0.16***	−0.01 (0.00)***	−0.14**	−0.01 (0.00)**	−0.34***	−0.02 (0.00)***	−0.32***	−0.02 (0.00)***
Female (vs. male)	0.11**	0.11 (0.03)**	0.09*	0.09 (0.04)*	0.12**	0.12 (0.04)**	0.10*	0.10 (0.04)*
Education	0.02	0.01 (0.01)	0.02	0.01 (0.01)	−0.02	−0.01 (0.01)	−0.03	−0.01 (0.01)
Household income	0.03	0.04 (0.02)	0.02	0.03 (0.02)	0.02	0.02 (0.02)	0.00	0.01 (0.02)
Region (vs. South)
Northeast	−0.12*	−0.12 (0.04)*	−0.08*	−0.08 (0.04)*	−0.11**	−0.11 (0.04)**	−0.06	−0.06 (0.05)
North Central	−0.06	−0.06 (0.04)	−0.05	−0.05 (0.04)	−0.19**	−0.19 (0.06)**	−0.16***	−0.16 (0.04)***
West	−0.14	−0.14 (0.10)	−0.08	−0.08 (0.09)	−0.27***	−0.27 (0.06)***	−0.16*	−0.16 (0.06)*
Marital status (vs. first marriage)
Remarried	0.03	0.03 (0.07)	0.04	0.04 (0.07)	−0.23*	−0.23 (0.09)*	−0.19*	−0.19 (0.09)*
Cohabiting	−0.04	−0.04 (0.09)	−0.04	−0.04 (0.08)	−0.04	−0.04 (0.09)	−0.04	−0.04 (0.08)
Separated	−0.01	−0.01 (0.09)	0.02	0.02 (0.09)	−0.07	−0.07 (0.11)	−0.01	−0.01 (0.10)
Divorced	−0.00	−0.00 (0.07)	0.02	0.02 (0.06)	−0.01	−0.01 (0.08)	0.03	0.03 (0.07)
Widowed	0.07	0.07 (0.08)	0.05	0.05 (0.08)	0.15	0.15 (0.11)	0.12	0.12 (0.11)
Never married	0.06	0.06 (0.05)	0.07	0.07 (0.05)	0.09	0.09 (0.06)	0.11	0.11 (0.06)
Material hardship	−0.03	−0.02 (0.01)	−0.01	−0.01 (0.01)	−0.03	−0.02 (0.01)	0.00	0.00 (0.01)
Served in military (vs. never served)	−0.16*	−0.16 (0.06)*	−0.11	−0.11 (0.06)	−0.11	−0.11 (0.08)	−0.02	−0.02 (0.07)
Previously incarcerated (vs. never incarcerated)	0.03	0.03 (0.05)	0.07	0.07 (0.05)	−0.11*	−0.11 (0.05)*	−0.04	−0.04 (0.06)
Family contact	--	--	0.12***	0.10 (0.02)***	--	--	0.13***	0.11 (0.02)***
Subjective Family closeness	--	--	0.11***	0.19 (0.04)***	--	--	0.25***	0.43 (0.03)***
Parent (vs. non-parent)	0.19	−0.22 (0.13)	0.13*	−0.21 (0.13)	0.08	−0.67 (0.11)***	−0.00	−0.62 (0.11)***
Age × parent status	0.14**	0.01 (0.00)**	0.11*	0.01 (0.00)*	0.26***	0.02 (0.00)***	0.21***	0.01 (0.00)***
Urbanicity
Urban	0.11	0.11 (0.09)	0.07	0.07 (0.09)	−0.10	−0.10 (0.06)	−0.13	−0.13 (0.06)
Constant	2.96***	2.94 (0.25)***	3.02***	1.79 (0.25)***	2.82***	3.74 (0.29)***	2.87***	1.65 (0.30)***
F	5.95***		13.64***		14.57***		133.13***
R-squared	0.03		0.09		0.07		0.19
Observations	3394		3393		3270		3267

^*p < .05;

^**p< .01;

^***p < .001

Notes: Data are weighted to account for unequal probabilities of selection, non-response, and post-stratification. ß= Standardized coefficient. b=Unstandardized coefficient SE=Standard error.

Several sociodemographic factors were significantly associated with frequency of receiving support (Table 3, Models 3 and 4). Model 3 shows that age, gender, region, marital status, incarceration history, parental status, and the interaction between age and parental status were significantly associated with frequency of receiving support. Women received support more often than men and respondents who were remarried received assistance less frequently than those who were currently in their first marriage. Southerners received support more frequently than respondents who resided in the Northeast, North Central, and West regions. The magnitude of these associations decreased modestly when subjective family closeness and family contact were considered in Model 4, suggesting that these family factors partially mediate the relationship between sociodemographic background and the receipt of support.

The relationship between incarceration history and receipt of support and Northeast region and receipt of support were no longer significant in Model 4. Respondents who indicated stronger feelings of closeness to family and those who interacted with their families on a more frequent basis reported that they received support more frequently than their counterparts. An examination of the standardized regression coefficients in Model 4 indicates that the age (ß =−.32), family closeness (ß =.25) and the interaction between age and parental status (ß =.21) were particularly large and noteworthy correlates of receipt of support.

The significant interaction between parental status and age indicated that the overall negative relationship between age and receiving family support was stronger for non-parents. This is displayed in Table 4. Among respondents aged 18–25, 96% of parents and 95% of those without children, received support. However, among respondents aged 55–64, 82% of parents received support while 60% of non-parents received support. For the oldest age group of 65 years and older, 83% of parents and 72% of non-parents received support from family members.

In addition, to the results presented here, we also conducted supplemental analyses for possible interactions. Based on previous research and theory on the important role that women play in family support networks, we tested interactions between gender and family contact and gender and family closeness. Neither of these interactions were significant in any of the regression models, so they were not included in the analysis.

Discussion

Overall Findings

This analysis provided a comprehensive investigation of the correlates of several dimensions of African American extended family support networks. We had the benefit of a large national sample which allowed for the investigation of a full range of sociodemographic variables. Concordant with the cultural variant perspective (Allen, 1978), our findings indicate a considerable degree of sociodemographic variation in African American family networks. Further, they confirm expectations from the family solidarity framework, which emphasizes the importance of family connections for social support. Overall, African Americans interacted with their family members on a frequent basis, displayed a high degree of family closeness and contact, and provided and received assistance from extended kin fairly often. These findings are consistent with previous research indicating the importance of familial support relationships among African Americans (Lincoln et al., 2013; McAdoo, 1978; O’Brien, 2012; Sarkisian, 2007; Stack, 1974; Taylor, 1986).

Familial Factors

Overall, our results are congruent with the family solidarity model as articulated in Figure 1 (Bengtson, Giarrusso, Mabry, & Silverstein, 2002) and provide support for our first hypothesis that indicators of greater family cohesion (i.e., family closeness and contact) are positively associated with each other and both giving and receiving family support. Subjective family closeness was positively and significantly related to frequency of contact, as well as the provision and receipt of support. Furthermore, the standardized beta coefficients estimating the strength of the association between family closeness and frequency of contact (ß =.43) and receiving support (ß =.25) reveal that this factor was strongly associated with each of these outcomes.

These findings coincide with prior work on both African Americans (Chatters, et al., 2002; Cross et al., 2018a; Cross et al. 2018b; Hatchett & Jackson, 1993; Taylor, 1986) and whites (Silverstein, Parrot & Bengtson, 1995; Parrott & Bengtson, 1999), indicating that subjective closeness and other measures of family affection are positively related to exchanging informal support. Frequency of contact with extended family members was also positively associated with both receiving and providing assistance, which aligns with prior work on African Americans (Chatters, et al., 2002; Hatchett & Jackson, 1993; Taylor, 1986). Collectively, these findings reveal the importance of family contact for the mobilization and disbursement of social support. Routine contact with relatives provides ongoing opportunities to assess and communicate family members’ needs, alert and mobilize members of the kin network, and provide access to resources.

Sociodemographic Factors

With regard to indicators of social and economic need, there were no differences in family network involvement by household income or education. Further, individuals experiencing material hardship and those who had been previously incarcerated had lower levels of family closeness. While our findings fully confirm the family solidarity framework, they do not support expectations from contingency theory (Hypothesis 2) that support exchanges are largely need-based. It bears noting that contingency theory has been primarily used to explain intergenerational support between parents and children (e.g., Fingerman et al., 2020). Although our findings provide little support for contingency theory’s suggestion that network involvement is largely a function of social or economic need, age was negatively associated with family contact, which is consistent with contingency theory’s emphasis on resources typically flowing from older to younger generations within families (Hypothesis 3a).

Parental status is also related to family contact: adults with children interacted with their family members more frequently than did non-parents. This finding aligns with research suggesting that parenthood has several important influences on involvement with extended family members (e.g., Dow, 2019; Sudarksa, 1996), and it is concordant with our expectation of greater network involvement among parents than adults without children (Hypothesis 4). Our analyses also indicated that parental status moderated the negative relationships between age and both the provision and receipt of support, which supports our expectation that among older adults, those who are parents receive more assistance than those without children (Hypothesis 3b). Specifically, age was negatively associated with both receiving and providing support, but these relationships were much stronger for people who were non-parents than for parents. These findings are congruent with research on the critical role that adult children play in the support networks of older adults among both African Americans (Jackson et al., 2008; Taylor, 1986) and whites (Suitor et al., 2011). For instance, research has consistently found that regardless of race and ethnicity, adult children are the primary caregivers for older parents (Jackson et al., 2008). Our findings also highlight the precarious position of childless older adults in terms of involvement in family networks. Prior research on the size of identified helper networks indicates that older African Americans who are childless have smaller networks comprised of siblings and friends, whereas older persons with children had larger networks that were largely comprised of immediate family (Chatters, Taylor & Jackson, 1985, 1986).

The interaction between age and parental status on subjective family closeness was somewhat different. Among parents, age was unrelated to family closeness, whereas among non-parents age was negatively related to familial closeness. This finding, in conjunction with the other age × parental status interactions, provide evidence of the importance of adult children in the extended family networks of African Americans. However, it is important to note that while some differences exist, the vast majority of African Americans feel close to their family members.

Gender was significantly associated with frequency of contact, as well as frequency of receiving and providing support to extended family members. Also, as indicated by size of the standardized coefficient (ß =.34), the relationship between gender and family contact is particularly strong. This finding is consonant with our fifth research hypothesis and with previous research which finds that women have higher levels of contact with family members and are primarily tasked with maintaining family ties, including remembering and organizing family events (birthdays) (Stack & Burton, 1993). Although gender differences are present, it is important to emphasize that men are involved in their extended families. While women are more involved in family support and family contact than men, these groups do not differ in their reported feelings of family closeness. Overall, these findings are inconsistent with deficit models of Black family life that assert that Black men have abandoned familial responsibilities due to dysfunctional family life, concentrated poverty, and/or high unemployment (see Sarkisian, 2007 for a discussion of this issue). Rather, study findings show that Black men express strong feelings of closeness to family members, which corresponds with recent work indicating that Black men are more likely than white men to live near or with extended family members and interact with extended family more frequently (Sarkisian, 2007).

Counter to our expectation from Hypothesis 6, there were no differences in network involvement between married and single adults. In addition, contrary to Hypothesis 6, we found that respondents who were remarried received assistance from family less frequently than those who were currently in their first marriage. This finding could indicate a decline in family support relationships after experiencing the loss of a spouse due to divorce, separation, or widowhood. Our data do not allow a more in-depth examination of the reasons for remarriage but suggests that the circumstances surrounding this event may have consequences for subsequent relationships with former in-laws and overall level of family support that remarried persons receive.

We note geographic variation in network involvement as well. Concordant with Hypothesis 7, our results indicated that Southerners had higher levels of family closeness, family contact, and family support than persons in other regions. As previously mentioned, very little empirical work has examined geographic differences in extended family support networks, but this result is consistent with the handful of studies in this area indicating that Southerners tend to have higher levels of family involvement (e.g., Sechrist et al., 2007).

There was one significant finding related to urbanicity. African Americans who resided in urban areas had more frequent contact with family members than those in rural areas. This finding aligns with research on family support which notes that proximity is strongly correlated with the frequency of contact (Hatchett & Jackson, 1993). Also, when comparing the magnitude of associations, we observe that urbanicity had the largest standardized coefficient (ß =.47), indicating that the relationship between urbanicity and family contact is particularly strong and noteworthy. However, despite having less frequent contact with family among rural African Americans, they are not less likely than their urban counterparts to receive or provide support.

Military service was negatively associated with subjective family closeness. It was also negatively related to family contact and giving family support, but these relationships were no longer statistically significant after we controlled for subjective family closeness. Several potential reasons for this finding include: 1) African American military members and veterans have higher rates of residential mobility in the United States and globally which may make them more distant from relatives, 2) military personnel have more conservative voting patterns than the African American population as a whole, which may estrange them from more liberal African American family members (Pew Research Center, 2017), and 3) veterans have higher rates of psychiatric disorders including substance abuse and post-traumatic stress disorder (Kessler et al., 2014), which may cause isolation from extended family (Nguyen et al., 2020).

Similar to our finding related to military service, family closeness also appears to mediate the relationship between material hardship and incarceration history and family contact. In other words, African Americans with greater material hardship and those who had been previously incarcerated had lower levels of family contact than their counterparts, but these findings were no longer statistically significant after family closeness was taken into account (Table 2, Models 2). The cross-sectional nature of the NSAL data limit our ability to explicate these relations, but we offer a potential explanation for each finding. Support relationships are typically governed by norms of reciprocity. If a family member is not able to reciprocate support and has exhausted family resources, the quality of family relationships may deteriorate, resulting in lower levels of closeness to family members, and presumably, less contact. Thus, when feelings of closeness are considered, we find that less well-off respondents have rates of contact that are similar to those not experiencing material hardship.

As it relates to incarceration history, we note that prior imprisonment was inversely related to family closeness, contact, and receipt of support. These findings support our expectation that imprisonment is negatively associated with family closeness and contact (Hypothesis 8a), and with our expectation that formerly incarcerated persons receive support less frequently than those who have never been imprisoned (Hypothesis 8b). Emerging research discusses the fundamental ways that incarceration impacts the development, form, and maintenance of family relationships (Miller, 2021). Family relationships develop over time and within the context of mutual interactions that allow for the development of trust, reciprocity, and a shared sense of role expectations and family history. Families impacted by prolonged incarceration lack these opportunities and may experience uncertainty about family roles, expectations, and behaviors, both during and after the incarceration of a family member. The physical separation and social and psychological absence of incarceration negatively impacts the development of family ties, particularly parent-child and spouse/partner bonds. Additionally, although families provide the majority of support to those who have been previously incarcerated, these individuals may still receive less support than people who have never been imprisoned and who have not experienced sustained separation from relatives.

Study Implications

Study findings are important for understanding how sociodemographic factors are associated with family support and have implications for research and practice. Research wise, these findings underscore a consistent pattern of gender differences in family involvement whereby women are more deeply involved in family emotional labor and maintaining contact (kin-keeping), as well as fulfilling family roles associated with providing/receiving support. Black women who have historically high levels of labor force participation (relative to white women) perform double duty in their roles as workers and their involvement in family care behaviors, with the attendant stresses, rewards, and work-family conflicts.

Similarly, study findings provided a clearer understanding of the powerful joint influence of age and parental status in affecting family contact and closeness and the provision and receipt of family support. This raises particularly important implications for a subset of the African American population—older adults without children—who are less likely to have their emotional and instrumental support needs identified and addressed. Further, childless older adults are at elevated risk of experiencing social isolation, loneliness, and their negative physical and mental health effects (Taylor, 2020), that may be particularly harmful for older men (Taylor & Taylor, 2018). African American extended family networks are uniquely positioned to assist older siblings, aunts, uncles, and cousins who are without children. As part of these efforts, extended family members could assist their childless older relatives (e.g., aunts and uncles) with planning for retirement, caregiving, advanced care and other issues.

Finally, this study highlighted the centrality of giving and receiving family support for African Americans. Despite the significance of support exchanges, this investigation showed that there are specific life circumstances experienced by family members (i.e., being formerly incarcerated and military service), that disrupt family relations in ways that may make it difficult to access these resources. Practices and policies that support mass incarceration have made the incarceration of a family member an unfortunately common circumstance that has significant impacts on African American families. Forty percent of the over 2 million people incarcerated are Black, reflecting the outsized impact of incarceration on Black families (Miller, 2021). Current estimates indicate that African Americans are 2.2 times more likely than Whites to be arrested (Federal Bureau of Investigation, 2013), and when arrested, they are six times more likely to be incarcerated (Bonczar, 2003). Our findings demonstrate that incarceration impacts family closeness, contact, and support behaviors. In terms of practical implications, families that have been impacted by incarceration may benefit from family counseling, and family reintegration interventions that acknowledge and renegotiate spouse/partner and parental roles.

A relatively recent development in working with families impacted by incarceration include programs that offer support to children and their parent and alternative caregivers (e.g., grandparents). For example, Mother Intercession in Flint, Michigan (www.unlocktheirfuture.org/) provides services to children and parent/alternative caregivers to combat the negative outcomes associated with parental incarceration, and it offers programming for parents who are incarcerated (e.g., parenting classes, supervised visits for parent-child, mentoring programs) (Miller et al., 2013). These holistic family programs are important in reestablishing and maintaining connections between parents and children, as well as members of the larger family system who have functioned as caregivers to minor children.

This study has several limitations that must be acknowledged. First, the analyses were conducted on data collected in 2001–2003, which may limit the generalizability of the findings to contemporary African American families. Nevertheless, the NSAL dataset remains the most recent national survey with comprehensive information about African American informal social support exchanges. Second and relatedly, our measure of family contact does not include the most up-to-date advances in communication technology including Skype, Facebook, FaceTime, email, texting and GroupMe, which may lead to more conservative estimates of frequency of family contact. Third, our analysis used global measures of the receipt and provision of support from family, rather than specific types of support (e.g., financial support, help during an illness). While this approach allows us to assess overall levels of support within the African American population and the factors associated with general support exchanges, it does not permit a comparison of distinct forms of support (e.g., emotional vs. financial support), representing an opportunity for future research in this area. Fourth, the NSAL cross-sectional design does not allow for an examination of the ongoing nature of family relationships and support transactions. Further, it limits our ability to determine whether findings showing age differences in family contact and giving and receiving support reflect an age or cohort effect. It is possible that as individuals age, they become less involved in their family networks. It is also plausible that younger generations of African Americans have greater opportunities (via new technologies) to interact with family members and to exchange more support than prior generations. However, our age findings align with previous work published in the 1980s (Taylor, 1986), which suggests that these are actual age differences and not cohort effects. Longitudinal assessments of reports of family contact and giving and receiving support would allow us to establish whether African Americans report less involvement in their family networks as they grow older or whether younger cohorts report higher levels of involvement than their predecessors.

Despite these limitations, this study provides a comprehensive assessment of the correlates of informal family support networks among African Americans. Strengths of this analysis include the large, nationally representative sample of African Americans and the inclusion of a broad array of sociodemographic factors and family network measures (i.e., subjective closeness, contact). Despite the fact that the National Survey of American Life was collected in 2001–2003, it remains a relevant and valuable source of information on the Black population in the U.S. The NSAL is the only data set that has a national probability sample of African Americans that is based upon the population distribution of the African American population and not based on an oversampling of a subset of African Americans. Although there are numerous national studies that include an African American subsample, only a handful are national probability samples of the African American population. As such, the NSAL is an ideal data set for fulfilling a primary objective of our investigation to counter monolithic depictions of African Americans by exploring sociodemographic variability in family support networks. Additionally, the study focused on family support networks in relation to diverse groups within the African American population that are rarely examined, including persons experiencing material hardship, military veterans, and those who were formerly incarcerated.

It has been over 40 years since Allen (1978) critiqued cultural deviant and cultural equivalent perspectives on African American families. His articulation of the cultural variant perspective regarded African American families as adaptive organizations with human agency, planful intention, and the ability to modify their responses to particular and dynamic circumstances. An important caveat of the cultural variant perspective was its emphasis on understanding the opportunities and constraints that are embedded in the sociocultural contexts within which African American families are situated. To date, family research has yet to fully realize and explore contemporary sociocultural contexts as a means to deepen knowledge of African American family life. The present study examined within-group variability (i.e., sociodemographic, family contact, and closeness) in support behaviors within extended family networks. Our findings confirmed that African American families are not monolithic and, in fact, demonstrate considerable variability in life circumstances (i.e., being formerly incarcerated and military service), family support behaviors, and their sociodemographic correlates. Strategic research efforts that are squarely focused on understanding within-group variation within the African American population hold promise for enhancing our understanding of how sociocultural contexts are manifested in family life.