Skip to main content
NCBI home page
NIHPA Author Manuscripts logoLink to NIHPA Author Manuscripts
. Author manuscript; available in PMC: 2013 Aug 1.
Published in final edited form as: J Marriage Fam. 2012 Jul 13;74(4):726–742. doi: 10.1111/j.1741-3737.2012.00985.x

Marriage Expectations Among African American Couples in Early Adulthood: A Dyadic Analysis

Ashley B Barr 1, Ronald L Simons 1
PMCID: PMC3435147  NIHMSID: NIHMS372464  PMID: 22962498

Abstract

Using Family and Community Health Study data consisting of 168 unmarried, primarily African American couples, the current study sought to understand the dyadic interplay among school, work, and partner-specific marriage expectations in early adulthood. Drawing on the economic prospects, adult transitions, and work – family literatures, the authors hypothesized and found ample support that expectations to marry a romantic partner were linked not only to one’s own school and work-related experiences but also to those of a partner. These associations held while controlling for relationship satisfaction, general views of marriage, and other covariates that have been posited to explain racial inequalities in relationship and marriage patterns. Furthermore, the authors found that actor covariates of marital expectations differed from partner covariates, a finding that highlights the advantages of dyadic analyses in helping researchers understand marriage as both a developmental and interpersonal process.

Keywords: African Americans, dyadic/couple data, families and work, union formation, youth/emergent adulthood

Individual marital behavior and general marriage patterns have long been of interest to family researchers, demographers, and public policy experts. In particular, marital behavior and trends among African Americans, perhaps more than any other group in the United States, have been and continue to be a primary source of concern and debate. This concern centers around the degree to which African American families differ from those of other racial groups (Osborne, Manning, & Smock, 2007; Preston, Lim, & Morgan, 1992), the structural and cultural origins of such differences (Furstenberg, 2009; Moynihan, 1965; Ruggles, 1994), and the implications of such differences for families and communities (McLanahan & Percheski, 2008). Although much debate exists with regard to the origins and implications of African American family structure, much research over the past several decades has documented racial inequalities in the frequency, quality, and stability of romantic relationships in the United States. For example, African Americans are less likely to marry and more likely to divorce than Whites (Goldstein & Kenney, 2001; Sweeney & Phillips, 2004). Likewise, African American couples tend to report lower satisfaction and higher rates of violence than couples from other racial groups (Broman, 2005). Among young people, general expectations to marry are lower among African Americans compared to their White counterparts (Crissey, 2005).

For as long as such racial inequalities have been documented, cultural and structural explanations for their persistence have flourished (see Dixon, 2009; Furstenberg, 2009, for an overview of these often competing explanations). In the present study we sought to contextualize young, African American couples’ relationships as embedded within their school- and work-related experiences. More specifically, the present study focused on understanding young couples’ marriage expectations as they relate to recent unemployment history, current school and work status, earnings, weekly hours worked, and job satisfaction. It is important to note that such associations were explored while holding constant relationship characteristics and general views of marriage, potential cultural influences that others have deemed partially responsible for the aforementioned racial inequalities in relationship patterns (Moynihan, 1965).

African Americans are uniquely situated within U.S. educational institutions and labor markets, and the educational and employment patterns of African Americans, especially African American young adults, tend to be marked by even more pluralism and less linearity than those of Whites (Furstenberg, 2010; Settersten & Ray, 2010). Thus, understanding the associations between African American young adults’ school- and work-related experiences and their union-formation expectations may help to shed some light on the racial disparities in the frequency, quality, and stability of romantic relationships in early adulthood and later in life. Furthermore, by exploring the school- and work-related correlates of marital expectations in a dyadic manner, in this study we took into consideration not only the intersections of work and family domains in young adulthood (Sneed, Hamagami, McArdle, Cohen, & Chen, 2007) but also the theoretical and empirical dependency of romantic partners. Research to date has largely lacked this simultaneous consideration of both developmental and relational interdependence and hence has potentially failed to capture the complexity of relationships in young adulthood and how such relationships intersect with other life domains.

Perhaps because of the popular concern regarding the declining significance of marriage, research on marital expectations to date has concentrated primarily on general and generational trends in youth’s expectations to marry. Much of this research, then, has examined marital expectations and their predictors in an abstract sense by asking young people about their general expectations to marry anyone in the future rather than their expectations to marry a given partner (Gassanov, Nicholson, & Koch-Turner, 2008; Manning, Longmore, & Giordano, 2007). Given that marital expectations are likely linked not only to the general value one assigns to marriage but also to the unique dynamics of each romantic relationship, as acknowledged by Manning et al. (2007), the partner-specific, dyadic approach we take herein may provide a more relational perspective on the marriage process.

Not only are partner-specific marital expectations a strong predictor of marriage (Waller & McLanahan, 2005), but they may also offer insight into intimate relationships that is not offered by studies of generalized marital expectations or marital behavior. Most important is that focusing on the development and correlates of partner-specific marital expectations requires that one view marriage as an interpersonal process. Given that this process begins long before the point of marriage, understanding partners’ marital expectations, or lack thereof, may offer insight into whether, how, and when two people decide to marry and why they decide to marry one another. In addition, as Hall (2006) and Byrd (2009) have shown, cultural and individual notions of marriage play an important role in the lives and relationships of unmarried individuals. Given the rising age at first marriage, the study of marital expectations (and other non-behavior – based measures of marital significance) is ever more important to research on young adulthood, because studies of marital behavior are becoming increasingly irrelevant and inapplicable to the young adult population. Finally, given changing expectations about the timing of marriage and the ongoing debate about the meaning of marriage relative to other relationship statuses (e.g., Heuveline & Timberlake, 2004; Sassler, 2010), a greater understanding of the development of marital expectations in a relationship-specific context could offer much more nuanced insight into the changing meaning of intimate relationships for young people (Casper & Bianchi, 2002; Heuveline & Timberlake, 2004; Smock, 2000).

Linking School, Work, and Marriage Expectations

Although past work has consistently shown that school enrollment and higher educational attainment affect marital behavior by delaying entry into marriage (Guzzo, 2006; Koball, 1998), and that stable employment and the ability to earn a family wage, in particular among men, tend to accelerate entry into marriage (Bulcroft & Bulcroft, 1993), little is known about how such factors might affect intentions to marry a specific partner, presumably a necessary precursor to marriage. Hence, researchers know plenty about the intraindividual processes that predict marital behavior, but little is known about the interindividual processes that affect the marriage process and the place of marriage in nonmarital romantic relationships. In assessing the relationships between school- and work-related experiences and young people’s expectations to marry their romantic partners, we find three lines of research to be most relevant. We discuss each of these literatures below.

Economic Potential and Marriage Expectations in Young Adulthood

As Settersten and Ray (2010) articulated, “Young adults have heard the message loud and clear: To get ahead, one needs a college degree” (p. 26). This sentiment suggests that young adults are expected to be enrolled in higher education, because such an education is vital for future economic success. Despite the expectation and the necessity of higher education, the majority of young adults in the United States do not have a 4-year degree, and in 2005, only 15% of African Americans aged 25 through 34 held a bachelor’s degree (Rumbaut & Komaie, 2007). Of those African American young adults with only a high school diploma, over half were disengaged from conventional social institutions in that they were not in school, in the military, or working (Settersten & Ray, 2010).

Research assessing the implications of education, employment, and social disengagement on general mate preferences or marital transitions suggests that financial security is important to many young people in deciding whom and when to marry. For instance, Burton and Tucker’s (2009) qualitative work revealed the role of uncertainty and instability, financial and otherwise, in preventing African American women from marrying. Furthermore, both Duvander (1999) and Smock, Manning, and Porter (2005) demonstrated that a lack of financial security was a significant barrier to marriage for cohabiting couples in Sweden and in the United States, respectively. Such results were echoed by Gibson-Davis, Edin, and McLanahan (2005) and England and Edin (2010) for low-income, unmarried parents. Hence, although economic security has been consistently associated with marital behavior, young adulthood is increasingly unstable in that young people today are taking longer than those in the past to finish their education and establish their careers (Arnett, 2004). Marital expectations in the developmental period under investigation, then, might be linked not to current socioeconomic status or education but to indicators of future potential, captured here by school enrollment, earnings, and recent unemployment. To the extent that such factors do indeed signal future economic potential, they were expected to be related to marital expectations in the following ways:

Hypothesis 1A: Controlling for work status and educational attainment, school enrollment of oneself and one’s partner will be positively associated with the expectation to marry.

Hypothesis 1B: Controlling for current work and school status, the experience of recent unemployment by oneself or one’s partner will be negatively associated marital expectations.

Hypothesis 1C: One’s own and one’s partner’s earnings will be positively associated with the expectation to marry.

Research with primarily White samples has revealed the worker identity to be more salient for men than for women (Rothbard & Edwards, 2003). African American women, however, have long assumed the (co)provider role in their families because of the vulnerable economic contexts in which they are more likely to be situated and the relatively high unemployment and incarceration rates among African American men (Pettit & Western, 2004). Furthermore, Sweeney (2002) showed that the importance of both White and African American women’s earnings in predicting marriage has grown over time. Given this work, we did not hypothesize any gender differences in the predictors of marital expectations, because those of both women and men were expected to be responsive to indicators of their own and their partners’ economic prospects.

Adult Transitions and Marriage Expectations in Young Adulthood

A second line of research has shown certain factors to be important in predicting when young people begin to assume an adult identity. For many young people in the United States, economic independence has been shown to be a primary factor in establishing such an identity (Furstenberg, 2008, 2010). Furthermore, as Sneed et al. (2007) revealed, development across multiple life domains is interdependent, meaning that each marker of adulthood facilitates young people in taking on other adult roles. Given this developmental interdependence, milestones like the completion of schooling or establishing economic independence in young adulthood may be important for young people in making plans for marriage, because “marriage has become the culminating event” of adulthood, embarked upon only after other indicators of maturity have been reached (Furstenberg, 2010, p. 75). Although lacking an examination of partner effects, Guzzo (2006) provided support for this notion by showing that being enrolled in school was negatively associated with a transition to both marriage and cohabitation.

The adult-transitions literature leads to hypotheses similar to those in the economic prospects literature regarding the role of earnings and unemployment in predicting marital expectations; that is, both higher earnings and the lack of unemployment by oneself and one’s partner presumably allow for increased feelings of adult status. From this perspective, the relationships outlined in Hypotheses 1B and 1C were expected to hold. Contrary to Hypothesis 1A, however, to the extent that school enrollment prevents partners from assuming adult identities, school status should be negatively associated with marriage expectations, and work status should be positively associated with such expectations. Stated formally, this hypothesis is as follows:

Hypothesis 2A: One’s own and one’s partner’s school enrollment will be negatively associated, and work status positively associated, with marriage expectations.

From the adult-transitions perspective, it is not only work status and earnings but also work hours that might encourage the development of an adult identity, because increased work hours indicate increased investment in an adult role. In fact, in assessing whether young people expected to marry within the next 5 years, Gassanov et al. (2008) found hours of employment to be positively associated with the general expectation to marry. Hence, we hypothesized the following with regard to work hours:

Hypothesis 2B: One’s own and one’s partner’s hours of employment will be positively associated with marriage expectations.

Marriage Expectations, Job Satisfaction, Spillover, and Time Binds

A third line of research located in the work – family literature suggests two other mechanisms through which work-related experiences, in particular job satisfaction and work hours, may influence family-related experiences. The first of these linking mechanisms is spillover, defined by Edwards and Rothbard (2000) as the “effects of work and family on one another that generate similarities between the two domains” (p. 180). The concept of positive spillover is commonly used to explain the consistent association between job satisfaction and life or family satisfaction (Heller, Watson, & Ilies, 2006). Although most of the research that has examined spillover thus far has used samples of formally defined families (i.e. married couples with or without children), it seems plausible that spillover may occur between the work and relationship domains occupied by unmarried young adults. For instance, job satisfaction may be associated with young people’s expectations to marry through, perhaps, its association with relationship satisfaction.

Hypothesis 3: Job satisfaction of oneself or one’s partner will be positively associated with marital expectations.

Alternatively, there exists a second mechanism in the work – family literature that may link job satisfaction, as well as work hours, to young couple’s marriage expectations. This mechanism stems from the notion that work and family are different spheres that compete for time, resources, and commitment (Greenhaus & Beutell, 1985). Such competition results in what Hochschild (1997) referred to as the time bind, a tension between the pull of work and family. This tension may lead young people to focus, at least temporarily, on one particular domain. Contrary to the spillover hypothesis, then, a satisfying or time-consuming job might entice one to invest in that job rather than a relationship. Given that married women assume more responsibility than do married men for household work (Sayer, England, Bittman, & Bianchi, 2009), responsibility that may limit the amount of time they can contribute to paid work (Becker & Moen, 1999), young women may be especially likely to prioritize a satisfying or demanding job over the prospect of marriage. More formally, these hypotheses are as follows:

Hypothesis 4A: Both one’s own and one’s partner’s job satisfaction and work hours will be negatively associated with marital expectations.

Hypothesis 4B: The actor associations in Hypothesis 4A will be stronger for women than for men.

Other Relevant Factors

In testing the above hypotheses, it was necessary to include statistical covariates that might confound the associations of interest. Because partner-specific marital expectations may be tied up in respondents’ general views of marriage or generalized expectations to marry, we controlled for respondents’ attitudes toward marriage. Although religiosity has previously been associated with marital expectations (Manning et al., 2007), we found that religious involvement had no effect above and beyond that of general views of marriage and thus did not include it as a covariate in the final analyses presented here. Furthermore, because relationship quality has been associated with views of marriage (Simons, Simons, Lei, & Landor, 2011) and with marital behavior (Kenney & McLanahan, 2006), we controlled for relationship satisfaction and several other individual- and dyad-level factors that have been shown to predict either satisfaction or marital expectations. These factors included gender (Amato, Johnson, Booth, & Rogers, 2003), whether or not the partners shared a child (Crohan, 1996), whether or not the couple was interracial (Forry, Leslie, & Letiecq, 2007), whether or not the partners were sexually active “on a regular basis” (Sprecher, 2002), cohabitation status (Waller & McLanahan, 2005), and relationship length (Qu, 2003). Third, so as not to confound current work and school status with educational attainment or age, we controlled for both partners’ age and highest level of education.

Method

Data

We evaluated the above hypotheses using data from Wave 5 of the Family and Community Health Study (FACHS), an ongoing, longitudinal research project that examines the social, psychological, and contextual risk and protective factors associated with African American families’ health and well-being (see Gibbons, Gerrard, Cleveland, Wills, & Brody, 2004, and Simons et al., 2002, for a detailed description of sampling procedures). In brief, a total of 867 African American families from Iowa and Georgia participated in the first wave of data collection in 1997. At the time of recruitment, each family had a child who was in the fifth grade in the public school system. Follow-up interviews with these children and their families were conducted every 2 to 3 years thereafter.

A total of 689 target youth (79.47% of the original sample) completed surveys at Wave 5 of data collection in 2008 and 2009. By this time, the participants no longer resided in only Georgia or Iowa but were dispersed across 23 states. Also during the fifth wave of data collection, the romantic partners of those targets who were involved in an ongoing romantic relationship, defined as anything from dating one person on a regular basis to being married, were invited to participate in the study. In total, 377 respondents (54.72% of the Wave 5 sample) were involved in an ongoing romantic relationship; of these, 237 (62.86%) romantic partners participated in the study. There was little evidence of selection bias when we compared respondents who had participating romantic partners to those who did not. These groups did not differ significantly with regard to age, education, work status, self-esteem, criminal involvement, or religious involvement. Also, as discussed elsewhere (Simons et al., 2011), there was little evidence of selective attrition across the five study waves.

Given their potentially unique characteristics and their minority status, the same-sex couples (n = 7, or 2.95%, all of whom were women) were dropped from the sample. Furthermore, couples who were married (n = 25, or 10.55%) and for whom either partner exceeded 30 years of age (n = 11, or 4.64%) were excluded from the analyses. Couples in which one partner exceeded 30 years of age were excluded both because the present study concerned the developmental period of young adulthood and because the wide age discrepancy between dyad members was not representative of the general dating patterns of young adults in the sample. An additional 26 couples (10.97%) were excluded for either partner’s incomplete data on the primary variables of interest. The present study, then, focused on the subset of respondents who were involved in an ongoing, heterosexual, nonmarital relationship and who provided complete data. The final sample consisted of 336 individuals, or 168 dyads. Although the original FACHS data were restricted to African American targets, the romantic partners of these targets were not required to be African American in order to participate in the study. Hence, all couples in the sample included at least one African American partner, but the vast majority (72%) comprised two.

Dependent Variable

The dependent variable of interest, expectation to marry, was assessed by asking respondents “Do you think you will marry [partner’s name]?” Responses ranged from 1 (definitely yes) to 4 (definitely not). Because the large majority of respondents (86%) fell into the bottom two response categories (either definitely yes or maybe), we dummy coded the variable so that the definitely yes category, indicating a high expectation to marry one’s partner, was coded 1 and all other categories were coded 0.

Key Independent Variables

Respondents were coded as being enrolled in school if they reported that they were currently attending school or would be attending school in the next academic year. Respondents were coded as working if they reported working for pay when asked about their current work situation. School enrollment and working for pay were not mutually exclusive statuses. Respondents were coded as having experienced recent unemployment if they responded affirmatively to the following question: “In the past 12 months, were you ever unemployed when you wanted a job?” Among those who were working, weekly work hours were reported in whole numbers, and weekly income was reported in whole dollars. The hourly earnings measure was constructed as a ratio of weekly income to weekly work hours. Job satisfaction was assessed with one item that asked respondents “How happy are you with your job(s)?” Responses ranged from 1(very unhappy) to 5 (very happy).

In line with other research that has used internal moderators (see, e.g., Frech & Williams, 2007, and Mirowsky, 1999), both the hourly earnings and job satisfaction variables were standardized, and respondents who were not working (and hence had no income, work hours, or job satisfaction) were coded at the mean of 0. Such a technique allowed us to maintain the full sample, independent of work status, in all analyses and is theoretically important because it provides evidence of the degree to which it is the status (i.e., work) or the characteristics of that status (i.e., earnings, hours, and job satisfaction) that predict marital expectations. Using the internal-moderator technique was particularly important in the current study because it permitted couples in which one person worked for pay and the other did not to be retained in the analyses. Standard techniques of limiting the sample to working couples would not have allowed for this heterogeneity.

Control Variables

General relationship satisfaction was assessed with three questions: (a) “How well do you and your romantic partner get along compared to most couples?”; (b) “How happy are you, all things considered, with your relationship?”; and (c) “All in all, how satisfied are you with your relationship?” Items were standardized and averaged to form the index of relationship satisfaction (α = .76). Respondents’ general view of marriage was also assessed with three items that indicated the degree to which they agreed with the following three statements: (a) “Marriage leads to a fuller life,” (b) “Marriage leads to a happier life,” and (c) “Getting married is the most important part of my life.” Response categories ranged from 1 (strongly disagree) to 5 (strongly agree), and we averaged the scores across items to form the index (α = .76). Age and highest level of education at the time of the interview were also reported by respondents. Dyad-level controls included whether or not the couple had a child together (1 = yes), the length of the relationship (in years), whether or not the romantic partner reported his or her race as something other than African American (1 = interracial), cohabitation status (1 = cohabiting), and the target’s report of whether or not the couple had sex “on a regular basis” (1 = yes).

The Actor – Partner Interdependence Model

Most studies of romantic relationships focus on the effect of one’s own characteristics on one’s own behavior/outcomes. This is termed an actor effect, and it typically is estimated for only one member of a dyad (e.g., Guzzo, 2006; Simons et al., 2011). Even when couple-level data are available, researchers have tended to run separate analyses for each member of the dyad (e.g., Fagan, Schmitz, & Lloyd, 2007; Taft et al., 2006). Although such analyses are insightful, they often do not consider the possibility that each person’s outcome is affected not only by his or her own characteristics but also by those of his or her partner (the latter of which is termed a partner effect). To take full advantage of the dyadic data available here, and to overcome the limitations of previous relationship research by accounting for the dependency between romantic partners’ and examining transactional, or partner, effects, we used the Actor – Partner Interdependence Model (APIM) developed by Kenny and his colleagues (Campbell & Kashy, 2002; Kashy & Kenny, 1999; Kenny, Kashy, & Cook, 2006). The APIM includes both actor effects (i.e., the association between one’s own predictor and one’s own outcome) and partner effects (i.e., the association between one’s partner’s predictor and one’s own outcome). In allowing for the simultaneous estimation of actor and partner effects while accounting for shared couple-level context, such a model helps to identify “truly relational phenomena” (Kenny et al., 2006, p. 147).

Plan of Analysis

The APIM can be estimated via structural equation modeling or multilevel modeling (Kenny et al., 2006). Given the large number of control variables and the absence of latent constructs from our analyses, we used the multilevel modeling approach for dyadic data analysis (Campbell & Kashy, 2002) via HLM version 6.0 (Raudenbush & Bryk, 2002). Our data, like all dyadic data, had a hierarchical structure of individuals nested within couples. Given the small number of lower level units per couple, a restriction had to be placed with regard to the random component, leaving only the intercept allowed to vary across dyads (Kenny et al., 2006). It was in this variation of intercepts that the nonindependence in romantic partner’s scores was modeled. Restricting the models to include only fixed effects controlled for unobserved population heterogeneity and did not bias the slope estimates, because the restriction merely “confounds the variance of the slopes with error variance” (Kenny et al., 2006, p. 89).

Given the binary nature of the dependent variable, Bernoulli modeling, a type of hierarchical generalized linear model (HGLM), was used in all of the analyses that follow. The Bernoulli model predicts the log-odds of being in a given category on the dependent variable; in this case, the results show the log odds of expecting to marry one’s partner. For easier interpretation of results, we exponentiated the log odds, thereby transforming them into odds ratios. In addition, all continuous predictors were centered at their grand mean, and all dummy variables were effects coded for easy interpretation of the intercepts.

Results

Descriptive Statistics and Intercorrelations

Table 1 displays the means and standard deviations, and Table 2 provides the intercorrelations for all variables used in subsequent analyses. In Table 2, individual-level variables are separated by respondent gender, and significant gender differences are indicated where appropriate. Of the 168 couples in the study, 41% (n = 68) had a child together, and 33% (n = 55) were cohabiting. On average, the couples had been together for 2.74 years. Interracial couples made up 28% (n = 47) of the sample, and about one half of couples reported having sex “on a regular basis.”

Table 1.

Descriptive Statistics for Dyad-Level and Individual-Level Variables (N = 168 Dyads)

Variables M SD Range
Dyad level
 Child present 0.40 0.49 0 – 1
 Interracial couple 0.28 0.45 0 – 1
 Relationship length (in years) 2.74 1.75 0 – 5
 Cohabiting 0.33 0.47 0 – 1
 Regular sex 0.51 0.50 0 – 1
Individual level Women Gender Difference Men α
M SD Range M SD Range
Age 21.51 2.23 17 – 30 *** 22.67 2.29 19 – 30
Education 12.72 2.17 1 – 16 * 12.18 2.46 1 – 16
Positive view of marriage 3.36 0.87 1 – 5 * 3.57 0.93 1 – 5 .76
Relationship satisfaction −0.01 0.86 3.01 – 1.39 −0.05 0.78 −2.71 – 1.39 .76
Expect to marry partner 0.38 0.49 0 – 1 0.35 0.48 0 – 1
In school 0.64 0.48 0 – 1 *** 0.40 0.49 0 – 1
Working 0.65 0.48 0 – 1 0.62 0.49 0 – 1
Unemployment in past year 0.43 0.50 0 – 1 ** 0.60 0.49 0 – 1
Hourly wagea 9.66 5.24 30 10.74 6.54 0.10 – 46.88
Hours workeda 34.48 10.05 8 – 70 37.57 13.56 3 – 80
Job satisfactiona 3.61 0.93 0 – 5 3.50 1.03 0 – 5
Open in a new tab
a

Means and standard deviations were determined on the basis of data from working respondents before we constructed the internal moderator variable.

*

p ≤ .05.

**

p ≤ .01.

***

p ≤ .001 (two-tailed).

Table 2.

Intercorrelations Among Study Variables for Women (W) and Men (M; N = 168 Dyads)

Variable 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27
1. Age (W) —
2. Age (M) .16* —
3. Education (W) .14* −.22* —
4. Education (M) −.13* .07 .23* —
5. In school (W) −.10 −.14* .35* .13* —
6. In school (M) −.10 −.08 .16* .20* .34* —
7. Working (W) .02 −.07 .04 .12* −.05 .05 —
8. Working (M) −.07 .08 .00 .03 .03 −.05 .12* —
9. Recent unemployment (W) −.09 .02 −.16* −.21* .05 −.06 −.34* .02 —
10. Recent unemployment (M) −.07 −.07 .03 −.08 −.08 .01 −.02 −.40* .06 —
11. Hourly wage (W) .16* .05 .14* .06 −.07 .00 .00 −.05 −.06 .13* —
12. Hourly wage (M) .05 .20* .09 .07 −.05 −.14* −.08 .00 −.01 −.01 .03 —
13. Job satisfaction (W) .05 −.01 .23* .14* .07 .04 .00 .14* −.08 −.07 .16* .06 —
14. Job satisfaction (M) .06 .02 −.04 .05 .05 −.10 −.05 .00 .00 .05 −.01 −.01 .02 —
15. Hours worked (W) .05 .04 −.12 −.18 −.18 −0.18* .00 .01 −.05 .00 .09 −.06 −.01 −.08 —
16. Hours worked (M) .02 −.02 −.01 .08 .05 −.09 .01 .00 .04 −.05 .05 −.22 −.04 .01 .15* —
17. View of marriage (W) .08 .13 −.02 .06 .00 −.02 .11 .00 −.02 −.08 −.17* .09 −.04 −.16* .10 .01 —
18. View of marriage (M) .01 −.11* −.01 −.03 −.05 .04 −.01 .08 −.04 −.05 −.09 .00 −.03 .06 −.02 −.12 .26* —
19. Expect to marry partner (W) −.03 .02 −.03 .08 −.08 .13* .06 .01 −.07 −.13* .15* −.03 −.03 −.12* .14* .11* .17* .15* —
20. Expect to parry partner (M) −.12* −.04 −.07 .07 −.02 .05 −.06 .09 −.07 −.08 −.13* −.02 −.04 −.03 .10 .16* .08 .43* .27* —
21. Relationship satisfaction (W) .05 −.01 .11* .16* −.06 −.05 .07 .02 −.09 −.06 −.10 .11* .03 .01 .05 .02 .04 .08 .37* .16* —
22. Relationship satisfaction (M) .03 −.08 .11* .13* −.07 .07 .03 .02 −.10 −.14* −.09 .02 .06 .05 −.03 .01 −.03 .32* .14* .38* .42* —
23. Child present .06 .16* −.21* −.14* −.27* −.06 −.10 .00 −.01 .16* .05 −.06 −.09 −.06 .06 .01 −.05 .07 .13* .13* −.04 −.06 —
24. Interracial couple .02 −.12* .04 −.06 −.06 −.03 .10 .11 −.06 .00 .19* −.01 .09 −.02 −.05 −.13 −.27* .01 −.02 −.04 .01 .08 .03 —
25. Relationship length .01 .10 −.23* −.05 −.14* −.14* .06 −.03 −.05 .10 .03 .00 .02 −.07 .08 .06 .09 .03 .22* .06 .00 −.17* .32* −.17* —
26. Cohabiting .17* .16* −.01 −.06 −.19* −.14* −.04 .05 −.02 .01 .06 −.04 .04 .06 0.23* .08 .03 .09 .08 .18* .07 .10 .15* .05 .07 —
27. Regular sex .10 −.06 .05 .03 −.19* −.16* −.08 −.04 −.14* .06 .17* −.03 −.02 .18* .09 .13* .00 .11* −.03 .08 .21* .13* −.01 .17* .03 .06 —
Open in a new tab

Note: W = women; M = men.

*

p < .05.

The men and women in the study averaged 22.67 and 21.51 years of age, respectively, a significant gender difference. On average, both men and women had at least a high school diploma, but women reported about one half year more education than men, a significant gender difference. Women also reported a less favorable view of marriage than did men, but neither current relationship satisfaction nor expectations to marry significantly differed by gender. Women in the sample were more likely than men to be in school (64% of women, 41% of men), but both men and women were equally likely to be working (65% of women, 62% of men). Among those who were working, women reported, on average, a lower hourly wage ($9.66 for women, $10.74 for men) and fewer weekly work hours (34.48 for women, 37.57 for men) than their male counterparts, but these differences were not statistically significant. Last, working men and women reported similar levels of job satisfaction, but men were significantly more likely than women to have experienced recent unemployment (44% of women, 60% of men).

Several bivariate relations in the correlation matrix (see Table 2) were noteworthy. First, partners’ marital expectations were significantly and positively correlated (r = .27). For binary outcomes, this pairwise intraclass correlation coefficient (McMahon, Pouget, & Tortu, 2006) indicated that, as expected, partners’ marital expectations were not independent of one another. Nonetheless, the strength of the correlation was only weak to moderate. Such lack of perfect agreement between partners’ marital expectations was also found by Sassler and McNally (2003) for cohabiting couples. In addition, for both women and men, marital expectations were significantly and positively correlated with general view of marriage and relationship satisfaction. These significant associations, among others, reinforced the necessity of controlling for such factors in the analyses that follow.

HGLM Results

Table 3 displays the HGLM results in a series of Bernoulli models. As mentioned above, we discuss our findings using the presented odds ratios. Regression coefficients and corresponding standard errors are available on request. Model 1 includes only the dyad- and individual-level control variables. Of the dyad-level (Level 2) controls, only one was significantly predictive of couples’ expectations to marry: Couples who had been together longer showed increased odds of expecting to marry. This association between relationship length and marital expectations maintained its significance across all subsequent models.

Table 3.

Bernoulli Regression Results Predicting Partner-Specific Expectation to Marry (N = 168 Dyads)

Model
Predictor 1 2 3 4 5 6 7
Intercept 0.54*** 0.51** 0.56** 0.54** 0.56** 0.52** .49**
Level 2 predictors
 Child present 1.33 1.37 1.29 1.41 1.45* 1.46* 1.49*
 Interracial couple 1.01 1.01 1.01 1.01 1.02 1.14 1.14
 Relationship length (years) 1.24* 1.26* 1.29* 1.28* 1.29* 1.27* 1.32*
 Cohabiting 1.28 1.34 1.32 1.37 1.38 1.27 1.31
 Regular sex 0.85 0.90 0.88 0.88 0.89 0.81 0.75
Level 1 predictors
 Female 1.17 1.27 1.25 1.34 1.33 1.38 1.50*
 Age
  Actor 0.95 0.96 0.95 0.94 0.94 0.92 0.92
  Partner 0.88 0.90 0.90 0.88 0.88 0.90 0.88
 Years of education
  Actor 1.02 1.02 1.03 1.01 1.02 1.00 0.99
  Partner 1.01 0.96 0.97 0.94 0.95 0.95 0.95
 Positive view of marriage
  Actor 2.10*** 2.15*** 2.16*** 2.16*** 2.16*** 2.45*** 2.77***
  Partner 1.11 1.10 1.11 1.10 1.10 1.18 1.20
 Relationship satisfaction
  Actor 3.22*** 3.59*** 3.65*** 3.63*** 3.67*** 4.11*** 4.46***
  Partner 0.97 0.95 0.92 0.92 0.92 0.88 0.90
 In school
  Actor 0.95 0.93 0.94 0.94 1.04 1.06
  Partner 1.53* 1.49* 1.55* 1.55* 1.54* 1.53*
  Actor × female 0.83
  Partner × female 1.27
 Working
  Actor 1.18 1.18 1.17 1.17 1.17 1.23
  Partner 0.90 0.88 0.78 0.77 0.74 0.69
  Actor × female 0.88
  Partner × female 1.17
 Recently unemployed
  Actor 0.93 0.93 0.94 0.98
  Partner 0.69* 0.69* 0.65* 0.58*
  Actor × female 1.02
  Partner × female 0.84
 Hours
  Actor 2.04** 1.85**
  Partner 1.57 1.60
  Actor × female 0.82
  Partner × female 1.03
 Wage
  Actor 1.73** 2.16**
  Partner 0.73 0.65
  Actor × female 1.75*
  Partner × female 1.47
 Job satisfaction
  Actor 0.85 0.77
  Partner 0.85 0.86
  Actor × Female 0.93
  Partner × Female 1.04
Open in a new tab

Note: All table values are odds ratios.

*

p ≤ .05.

**

p ≤ .01.

***

p ≤ .001.

Of the individual-level (Level 1) predictors, actor view of marriage and actor relationship satisfaction proved statistically significant. Controlling for all else in the model, both actor view of marriage and actor relationship satisfaction were positively associated with marital expectations. In other words, individuals who held more favorable general attitudes toward marriage and those who were more satisfied in their relationships were significantly more likely to expect to marry their romantic partner. Of note is that the coefficients for these latter two variables (view of marriage and relationship satisfaction) increased substantially in magnitude as school- and work-related characteristics were taken into consideration, a point we return to in the Discussion section.

Model 2 is the first model to incorporate school- and work-related variables into the analysis. Work and school status variables for both actor and partner were entered into the model simultaneously, and only partner school status proved statistically significant. Respondents with a partner in school were 53% more likely to expect to marry that partner. To assess whether or not the association between partner’s school status and marital expectations differed by gender, Model 3 includes the gender interactions. As shown by the lack of significance for all interaction terms, none of the work/school status effects differed significantly for men and women. Such findings show partial support for Hypothesis 1A and no support for Hypothesis 2A in that partner’s school enrollment was associated with increased, rather than decreased, odds of expecting marriage. Furthermore, women’s and men’s marital expectations were similarly predicted by partner’s school enrollment.

Model 4 tested the hypothesis concerning the association between recent unemployment and marriage expectations. The results suggest that a partner’s but not actor’s recent unemployment was significantly and negatively predictive of the expectation to marry. In fact, controlling for all else in the model, including one’s own experience of unemployment, having a partner who was recently unemployed was associated with a 31% decrease in the odds of expecting to marry that partner. Such findings provide partial support for Hypothesis 1B. As shown in Model 5, this partner effect did not differ significantly by gender.

We saw in Model 2 that work status alone was not predictive of marital expectations; Model 6 tested the degree to which the characteristics of that status might be associated with marital expectations. Although partner hours and partner wage proved to be nonsignificant, actor hours and actor wage were both significantly and positively associated with respondents’ expectations to marry their romantic partner. All else being equal, a 1-SD increase in work hours was associated with a twofold increase in the odds of expecting marriage, whereas a 1-SD increase in hourly wage was similarly associated with a 73% increase in the odds of expecting marriage. These findings provide partial support for Hypotheses 1C and 2A, but no support for Hypothesis 4A, in that respondents’ own working hours and earnings were positively related to marriage expectations. Furthermore, as indicated by the significant Actor Wage × Female interaction in Model 7, the association between actor wage and marriage expectations was much stronger for young women than for young men.

Also shown in Models 6 and 7 are the associations between job satisfaction and marital expectations. Such models provide no support for the notions of positive spillover (Hypothesis 3), because neither actor nor partner job satisfaction was significantly associated with marital expectations. Given the hypothesized association between job satisfaction and relationship satisfaction, and hence the possibility that relationship satisfaction might be mediating the job satisfaction – marital expectations link, we also ran these models without the relationship satisfaction control (results not shown). In the models excluding relationship satisfaction, the directionality and magnitude of the job satisfaction effects were consistent with those presented in Model 6. Furthermore, contrary to Hypothesis 4B, Model 7 revealed that these null associations between job satisfaction and marital expectations did not differ by gender.

Discussion

Work – family research to date has been primarily concerned with already-established, formally recognized families. Given the rising age of first marriage (Cherlin, 2010), however, education-and work-related experiences may become increasingly important in influencing the very formation of such families or in influencing couples not traditionally defined as families, such as those who are unmarried (Powell, Bolzendahl, Geist, & Steelman, 2010). Little research has examined how work-related factors may influence the experiences and expectations of unmarried couples, especially from a dyadic perspective. As Wickrama, Florensia, and Bryant (2011) pointed out, even less research has examined African American families in particular. The research conducted so far, then, lacks an understanding of marriage as both a developmental and dyadic process. The present study revealed the connectedness of both partners’ school and work experiences and union formation expectations among predominantly African American couples in young adulthood, a life stage that is gaining salience among life course researchers (Arnett, 2004). Consistent with other work in this life stage, the findings presented here offer support for viewing developmental domains (e.g., relationships, work, schooling) in early adulthood as interdependent (Sneed et al., 2007).

Furthermore, the multivariate results brought to light the importance of taking into account not only developmental interdependence but also relational interdependence, in that young people’s marital expectations were affected by not only their own experiences but also those of their partner. In particular, we found marital expectations to be positively associated with a partner’s school enrollment and negatively associated with a partner’s recent unemployment. The former effect seems to counter the finding by Guzzo (2006), Gassanov et al. (2008), and Koball (1998) that school enrollment was negatively associated with union formation. Such differential findings, however, may be attributable to the lack of partner effects examined in these studies and to their focus on transitions rather than expectations. Although school enrollment may impede a union transition in the present, it, along with the partner’s lack of recent unemployment, may enhance one’s expectations for a union transition in the future by signaling future economic well-being. Because the expectation is that young people should be enrolled in school and working toward a secure future, this may be especially true during the developmental period studied here.

The fact that partners’ actual earnings, job satisfaction, and work hours were unrelated to marital expectations is consistent with the interpretation that marital expectations, unlike marital behavior, in early adulthood are more closely associated with a partner’s future potential than his or her current marriageability. This was the case for both men and women in the sample, given that neither the partner school or partner unemployment effects differed by gender. The lack of gender differences found for partner effects combined with the significant gender difference for the effect of actor earnings supports other researchers (Burton & Tucker, 2009; Hill, 2005; Wickrama et al., 2011) in critiquing the male breadwinner – female caregiver model of family relations as historically Eurocentric and inapplicable to the experiences of many African Americans. Our results suggest that both women’s and men’s marital expectations were tied up in their partners’ economic prospects and that, among workers, one’s own earnings were more strongly associated with marital expectations for women than for men. Hence, the findings presented here grant the important economic role that African American women play in building their families and, in a broader sense, support arguments made by Oppenheimer (1997) against the notion that women’s economic independence makes marriage unattractive.

The hypotheses drawn from the work – family literature received little support in the current study. As previously noted, actor work hours were positively predictive of marital expectations, thus supporting the adult-transitions perspective drawn from the life course literature rather than the competing-roles perspective drawn from the work – family literature. Furthermore, job satisfaction proved to not be associated with marital expectations, although both actor and partner effects were negative in direction and not insubstantial in size. For job satisfaction, then, such results provide no support for the positive spillover hypothesis and only trend-based support for the notion of competing roles in early adulthood. Given the direction and size of the job satisfaction coefficients, a sample with a greater percentage of working young people might yield stronger support for the notion that young people who are satisfied in their jobs may recognize the potential competition for time posed by marriage and hence may be choosing (at least temporarily) not to take seriously their romantic relationships or the prospect of marriage.

Although both relationship satisfaction and general view of marriage proved to be consistent predictors of marriage expectations across all models, it is worth noting that the size of the coefficients grew steadily as school- and work-related variables were entered into the regression equations. Such a pattern suggests that as external (i.e., school and employment) factors are controlled for, the predictive power of internal (i.e., intrarelationship and intraindividual) characteristics increases. In other words, as external factors become more conducive to marriage, couples may be freer to make union formation decisions that are based on their own values and the quality of their relationships. Such a pattern of results supports qualitative work that has revealed structural factors to be prominent in people’s union formation plans and behaviors (Burton & Tucker, 2009; Edin & Kefalas, 2005; Edin & Kissane, 2010; Edin & Reed, 2005; England & Edin, 2010) and corroborates such arguments, like that of Cooke and Baxter (2010), about the importance of broader contexts in regulating individual effects. Future studies of work and family must consider this possibility more extensively and, as Dixon (2009) argued, must take seriously the importance of both internal (e.g., relationship) and external (e.g., school or work) factors in predicting relationship quality and formation, in particular among African Americans given their unique structural location in the U.S. labor market.

Limitations, Implications, and Future Directions

The findings of the present study must be considered in light of several limitations. First and foremost, although the project from which the current data were drawn is longitudinal, the couple-level data used in the current analyses were available at only the latest wave of data collection. Thus, the present study was cross-sectional and could not establish causality between school- and work-related factors and marital expectations. Future research could benefit from longitudinal couple data to parse out the directionality of effects and could also explore whether the marital expectations examined here predict marital transitions. Both are important in understanding marriage as a developmental and dyadic process.

A second limitation of the current study concerns the generalizability of its findings. The present sample included primarily African American couples and was limited to the target youth who lived in Georgia and Iowa at the time FACHS was initiated. Although the respondents are now dispersed across the United States, such a sample necessarily restricts the study’s generalizability, and thus future studies are needed to replicate and expand on its findings on a larger scale. Nonetheless, this lack of generalizability is somewhat offset by the present focus on an often-ignored population of young people in a theoretically significant developmental period of the life course. Besides capturing the heterogeneity among African American young adults and their intimate relationships, the current sample shifts the focus in the work – family literature away from White, already-established families to young African American couples. Given the documented racial inequalities not only in the frequency, quality, and stability of romantic relationships but also in educational attainment, (un)employment patterns, and work characteristics, a shift in focus is not only warranted but also critical for work – family scholars. In addition, the current sample allows for recognition of the heterogeneity among African American individuals and couples, because those represented here ranged considerably in their attitudes toward marriage, relationship satisfaction, and marital expectations. As Chambers and Kravitz (2011) argued, such heterogeneity is often lacking in research on African Americans.

Last, some of the measures we used were relatively simplistic, and others were lacking because of data limitations. For example, the data available to measure the concept of spillover were limited to a one-item indicator of job satisfaction. Although the job satisfaction – family satisfaction link is one of the most basic indicators of positive spillover, perhaps other, or more complex, indicators, like job security, job strain, or work – family conflict, might have yielded different results. Likewise, the FACHS data lack measures of family background for romantic partners, and hence we were unable to control for things like parents’ marital history or socioeconomic status. It is possible that such family background characteristics predict both marital expectations and school and work-related experiences, thus making the association between them spurious. Our inclusion of respondents’ general views toward marriage as well as their educational attainment as controls, both of which are associated with family background (Conger, Cui, Bryant, & Elder, 2000), makes the possibility of complete spuriousness less likely, however. Nonetheless, it is important to include additional job characteristics as well as family background indicators in future work.

Despite the cross-sectional nature, limited generalizability, and restrictive measures, this study was the first to examine the intersections of schooling, work, and marital expectations with a heterogeneous sample of African American couples in young adulthood. This examination revealed that marital expectations are differentially predicted by actor and partner characteristics. Although one’s own earnings and work hours, potential indicators of adult status, were positively associated with the expectation to marry, it was a partner’s school enrollment and lack of recent unemployment that enhanced marital expectations. Partner effects, then, support the economic prospects perspective, but actor effects seem to support the adult-transitions perspective. In general, though, both perspectives reinforce the role of socioeconomic factors in determining the centrality of marriage in African American young adults’ lives and the “marriageability” of romantic partners. Thus, the economic foundations of marital behavior (Sweeney, 2002) appear to be laid long before such behavior can be measured.

To date, most researchers have failed to assess marital expectations from a dyadic perspective or have tended to restrict their samples to either cohabiting couples or couples expecting a child. As the significant actor and partner effects in the present study reveal, a relational approach is crucial to understanding the partner-specific marital expectations of young people today and hence to understanding the development and progression of romantic relationships. Furthermore, the present findings suggest that researchers should endeavor to gather more heterogeneous samples of couples to better understand the plurality of romantic relationships that exists within young adulthood. The current findings are clear in suggesting interdependence of multiple developmental domains in this life stage and in demonstrating the interdependence of romantic partners across such domains. Future researchers would be wise to take this interdependence seriously in order to better understand marriage and relationship processes across the life course, as well as the inequalities that may arise throughout such processes.

Acknowledgments

This research was supported by the National Institute of Mental Health (Grants MH48165 and MH62669) and the Centers for Disease Control and Prevention (Grant 029136-02). Additional funding for this project was provided by the National Institute on Drug Abuse, the National Institute on Alcohol Abuse and Alcoholism, and the Iowa Agriculture and Home Economics Experiment Station (Project No. 3320).

References

  1. Amato PR, Johnson DR, Booth A, Rogers SJ. Continuity and change in marital quality between 1980 and 2000. Journal of Marriage and Family. 2003;65:1–22. [Google Scholar]
  2. Arnett JJ. Emerging adulthood: The winding road from the late teens through the twenties. New York: Oxford University Press; 2004. [Google Scholar]
  3. Becker PE, Moen P. Scaling back: Dual-earner couples’ work – family strategies. Journal of Marriage and the Family. 1999;61:995–1007. [Google Scholar]
  4. Broman CL. Marital quality in Black and White marriages. Journal of Family Issues. 2005;26:431–441. [Google Scholar]
  5. Bulcroft RA, Bulcroft KA. Race differences in attitudinal and motivational factors in the decision to marry. Journal of Marriage and the Family. 1993;55:338–355. [Google Scholar]
  6. Burton LM, Tucker MB. Romantic unions in an era of uncertainty: A post-Moynihan perspective on African American women and marriage. Annals of the American Academy of Political and Social Science. 2009;621:132–148. [Google Scholar]
  7. Byrd SE. The social construction of marital commitment. Journal of Marriage and Family. 2009;71:318–336. [Google Scholar]
  8. Campbell L, Kashy DA. Estimating actor, partner, and interaction effects for dyadic data using PROC MIXED and HLM: A user-friendly guide. Personal Relationships. 2002;9:327–342. [Google Scholar]
  9. Casper LM, Bianchi SM. Continuity and change in the American family. Thousand Oaks, CA: Sage; 2002. [Google Scholar]
  10. Chambers AL, Kravitz A. Understanding the disproportionaly low marriage rate among African Americans: An amalgam of sociological and psychological constraints. Family Relations. 2011;60:648–660. [Google Scholar]
  11. Cherlin AJ. Demographic trends in the United States: A review of research in the 2000s. Journal of Marriage and Family. 2010;72:403–419. doi: 10.1111/j.1741-3737.2010.00710.x. [DOI] [PMC free article] [PubMed] [Google Scholar]
  12. Conger RD, Cui M, Bryant CM, Elder GH. Competence in early adult romantic relationships: A developmental perspective on family influences. Journal of Personality and Social Psychology. 2000;79:224–237. doi: 10.1037//0022-3514.79.2.224. [DOI] [PubMed] [Google Scholar]
  13. Cooke LP, Baxter J. “Families” in international context: Comparing institutional effects across Western societies. Journal of Marriage and Family. 2010;72:516–536. [Google Scholar]
  14. Crissey SR. Race/ethnic differences in the marital expectations of adolescents: The role of romantic relationships. Journal of Marriage and Family. 2005;67:697–709. [Google Scholar]
  15. Crohan SE. Marital quality and conflict across the transition to parenthood in African American and White couples. Journal of Marriage and the Family. 1996;58:933–944. [Google Scholar]
  16. Dixon P. Marriage among African Americans: What does the research reveal? Journal of African American Studies. 2009;13:29–46. [Google Scholar]
  17. Duvander AZE. The transition from cohabitation to marriage: A longitudinal study of the propensity to marry in Sweden in the early 1990s. Journal of Family Issues. 1999;20:698–717. [Google Scholar]
  18. Edin K, Kefalas M. Promises I can keep: Why poor women put motherhood before marriage. Los Angeles: University of California Press; 2005. [Google Scholar]
  19. Edin K, Kissane RJ. Poverty and the American family: A decade in review. Journal of Marriage and Family. 2010;72:460–479. [Google Scholar]
  20. Edin K, Reed JM. Why don’t they just get married? Barriers to marriage among the disadvantaged. The Future of Children. 2005;15:117–137. doi: 10.1353/foc.2005.0017. [DOI] [PubMed] [Google Scholar]
  21. Edwards JR, Rothbard NP. Mechanisms linking work and family: Clarifying the relationship between work and family constructs. Academy of Management Review. 2000;25:178–199. [Google Scholar]
  22. England P, Edin K. Unmarried couples with children: Why don’t they marry? How can policy-makers promote more stable relationships? In: Risman BJ, editor. Families as they really are. New York: Norton; 2010. pp. 307–312. [Google Scholar]
  23. Fagan J, Schmitz MF, Lloyd JJ. The relationship between adolescent and young fathers’ capital and marital plans of couples expecting a baby. Family Relations. 2007;56:231–243. [Google Scholar]
  24. Forry ND, Leslie LA, Letiecq BL. Marital quality in interracial relationships. Journal of Family Issues. 2007;28:1538–1552. [Google Scholar]
  25. Frech A, Williams K. Depression and the psychological benefits of entering marriage. Journal of Health and Social Behavior. 2007;48:149–163. doi: 10.1177/002214650704800204. [DOI] [PubMed] [Google Scholar]
  26. Furstenberg FF. The intersections of social class and the transition to adulthood. New Directions for Child and Adolescent Development. 2008;2008(119):1–10. doi: 10.1002/cd.205. [DOI] [PubMed] [Google Scholar]
  27. Furstenberg FF. If Moynihan had only known: Race, class, and family change in the late twentieth century. Annals of the American Academy of Political and Social Science. 2009;621:94–110. [Google Scholar]
  28. Furstenberg FF. On a new schedule: Transitions to adulthood and family change. The Future of Children. 2010;20:67–87. doi: 10.1353/foc.0.0038. [DOI] [PubMed] [Google Scholar]
  29. Gassanov MA, Nicholson LM, Koch-Turner A. Expectations to marry among American youth: The effects of unwed fertility, economic activity, and cohabitation. Youth & Society. 2008;40:265–288. [Google Scholar]
  30. Gibbons FX, Gerrard M, Cleveland MJ, Wills TA, Brody G. Perceived discrimination and substance use in African American parents and their children: A panel study. Journal of Personality and Social Psychology. 2004;86:517–529. doi: 10.1037/0022-3514.86.4.517. [DOI] [PubMed] [Google Scholar]
  31. Gibson-Davis CM, Edin K, McLanahan S. High hopes but even higher expectations: The retreat from marriage among low-income couples. Journal of Marriage and Family. 2005;67:1301–1312. [Google Scholar]
  32. Goldstein JR, Kenney CT. Marriage delayed or marriage forgone? New cohort forecasts of first marriage for U.S. women. American Sociological Review. 2001;66:506–519. [Google Scholar]
  33. Greenhaus JH, Beutell NJ. Sources of conflict between work and family roles. Academy of Management Review. 1985;10:76–88. [Google Scholar]
  34. Guzzo KB. The relationship between life course events and union formation. Social Science Research. 2006;35:384–408. [Google Scholar]
  35. Hall SS. Marital meaning: Exploring young adults’ belief systems about marriage. Journal of Family Issues. 2006;27:1437–1458. [Google Scholar]
  36. Heller D, Watson D, Ilies R. The dynamic process of life satisfaction. Journal of Personality. 2006;74:1421–1450. doi: 10.1111/j.1467-6494.2006.00415.x. [DOI] [PubMed] [Google Scholar]
  37. Heuveline P, Timberlake JM. The role of cohabitation in family formation: The United States in comparative perspective. Journal of Marriage and Family. 2004;66:1214–1230. doi: 10.1111/j.0022-2445.2004.00088.x. [DOI] [PMC free article] [PubMed] [Google Scholar]
  38. Hill SA. Black intimacies: A gender perspective on families and relationships. Walnut Creek, CA: AltaMira Press; 2005. [Google Scholar]
  39. Hochschild A. The time bind: When home becomes work and work becomes home. New York: Metropolitan/Holt; 1997. [Google Scholar]
  40. Kashy DA, Kenny DA. The analysis of data from dyads and groups. In: Reis HT, Judd CM, editors. Handbook of research methods in social psychology. New York: Cambridge University Press; 1999. pp. 451–477. [Google Scholar]
  41. Kenney CT, McLanahan SS. Why are cohabiting relationships more violent than marriages. Demography. 2006;43:127–140. doi: 10.1353/dem.2006.0007. [DOI] [PubMed] [Google Scholar]
  42. Kenny DA, Kashy DA, Cook WL. Dyadic data analysis. New York: Guilford Press; 2006. [Google Scholar]
  43. Koball H. Have African American men become less committed to marriage? Explaining the twentieth century racial cross over in men’s marriage timing. Demography. 1998;35:251–258. [PubMed] [Google Scholar]
  44. Manning WD, Longmore MA, Giordano PC. The changing institution of marriage: Adolescents’ expectations to cohabit and to marry. Journal of Marriage and Family. 2007;69:559–575. [Google Scholar]
  45. McLanahan SS, Percheski C. Family structure and the reproduction of inequalities. Annual Review of Sociology. 2008;34:257–276. [Google Scholar]
  46. McMahon JM, Pouget ER, Tortu S. A guide for multilevel modeling of dyadic data with binary outcomes using SAS PROC NLMIXED. Computational Statistics & Data Analysis. 2006;50:3663–3680. doi: 10.1016/j.csda.2005.08.008. [DOI] [PMC free article] [PubMed] [Google Scholar]
  47. Mirowsky J. Analyzing associations between mental health and social circumstances. In: Aneshensel CS, Phelan JC, editors. Handbook of the sociology of mental health. New York: Plenum Press; 1999. pp. 105–123. [Google Scholar]
  48. Moynihan DP. The Negro family: The case for national action. Washington, DC: U.S. Department of Labor, Office of Political Planning Research; 1965. [Google Scholar]
  49. Oppenheimer VK. Women’s employment and the gain to marriage: The specialization and trading model. Annual Review of Sociology. 1997;23:431–453. doi: 10.1146/annurev.soc.23.1.431. [DOI] [PubMed] [Google Scholar]
  50. Osborne C, Manning WD, Smock PJ. Married and cohabiting parents’ relationship stability: A focus on race and ethnicity. Journal of Marriage and Family. 2007;69:1345–1366. [Google Scholar]
  51. Pettit B, Western B. Mass imprisonment and the life course: Race and class inequality in U.S. incarceration. American Sociological Review. 2004;69:151–169. [Google Scholar]
  52. Powell B, Bolzendahl C, Geist C, Steelman LC. Counted out: Same-sex relations and Americans’ definitions of family. New York: Russell Sage Foundation; 2010. [Google Scholar]
  53. Preston SH, Lim S, Morgan SP. African American marriage in 1910: Beneath the surface of census data. Demography. 1992;29:1–15. [PubMed] [Google Scholar]
  54. Qu L. Expectations of marriage among cohabiting couples. Family Trends. 2003;64:36–39. [Google Scholar]
  55. Raudenbush SW, Bryk AS. Hierarchical linear models: Applications and data analysis methods. 2. Thousand Oaks, CA: Sage; 2002. [Google Scholar]
  56. Rothbard NP, Edwards JR. Investment in work and family roles: A test of identity and utilitarian motives. Personnel Psychology. 2003;56:699–729. [Google Scholar]
  57. Ruggles S. The origins of African American family structure. American Sociological Review. 1994;59:136–151. [Google Scholar]
  58. Rumbaut RG, Komaie G. Young adults in the United States: A mid-decade profile. 2007 Retrieved from http://www.transad.pop.upenn.edu/downloads/Rumbaut%20Komaie%20-%20Young%20Adults%20in%20US.pdf.
  59. Sassler S. Partnering across the life course: Sex, relationships, and mate selection. Journal of Marriage and Family. 2010;72:557–575. doi: 10.1111/j.1741-3737.2010.00718.x. [DOI] [PMC free article] [PubMed] [Google Scholar]
  60. Sassler S, McNally J. Cohabiting couples’ economic circumstances and union transitions: A re-examination using multiple imputation techniques. Social Science Research. 2003;32:553–578. [Google Scholar]
  61. Sayer LC, England P, Bittman M, Bianchi SM. How long is the second (plus first) shift? Gender differences in paid, unpaid, and total work time in Australia and the United States. Journal of Comparative Family Studies. 2009;40:523–545. [Google Scholar]
  62. Settersten RA, Ray B. What’s going on with young people today? The long and twisting path to adulthood. The Future of Children. 2010;20:19–41. doi: 10.1353/foc.0.0044. [DOI] [PubMed] [Google Scholar]
  63. Simons RL, Lin KH, Gordon LC, Brody GH, Murry V, Conger R. Community differences in the association between parenting practices and child conduct problems. Journal of Marriage and Family. 2002;64:331–345. [Google Scholar]
  64. Simons RL, Simons LG, Lei MK, Landor A. Relational schemas, hostile romantic relationships, and beliefs about marriage among young African American adults. Journal of Social and Personal Relationships. 2011;29:77–101. doi: 10.1177/0265407511406897. [DOI] [PMC free article] [PubMed] [Google Scholar]
  65. Smock PJ. Cohabitation in the United States: An appraisal of research themes, findings, and implications. Annual Review of Sociology. 2000;26:1–20. [Google Scholar]
  66. Smock PJ, Manning WD, Porter M. “Everything’s there except money”: How money shapes decisions to marry among cohabitors. Journal of Marriage and Family. 2005;67:680–696. [Google Scholar]
  67. Sneed JR, Hamagami F, McArdle JJ, Cohen P, Chen H. The dynamic interdependence of developmental domains across emerging adulthood. Journal of Youth and Adolescence. 2007;36:351–362. doi: 10.1007/s10964-006-9081-2. [DOI] [PubMed] [Google Scholar]
  68. Sprecher S. Sexual satisfaction in premarital relationships: Associations with satisfaction, love, commitment, and stability. Journal of Sex Research. 2002;39:190–196. doi: 10.1080/00224490209552141. [DOI] [PubMed] [Google Scholar]
  69. Sweeney MM. Two decades of family change: The shifting economic foundations of marriage. American Sociological Review. 2002;67:132–147. [Google Scholar]
  70. Sweeney MM, Phillips JA. Understanding racial differences in marital disruption: Recent trends and explanations. Journal of Marriage and Family. 2004;66:639–650. [Google Scholar]
  71. Taft CT, O’Farrell TJ, Torres SE, Panuzio J, Monson CM, Murphy M, Murphy CM. Examining the correlates of psychological aggression among a community sample of couples. Journal of Family Psychology. 2006;20:581–588. doi: 10.1037/0893-3200.20.4.581. [DOI] [PubMed] [Google Scholar]
  72. Waller MR, McLanahan SS. “His” and “her” marriage expectations: Determinants and consequences. Journal of Marriage and Family. 2005;67:53–67. [Google Scholar]
  73. Wickrama KAS, Florensia SF, Bryant CM. Work, marriage, and health behaviors of African American working couples. Personal Relationships. 2011;18:519–531. [Google Scholar]

RESOURCES