Skip to main content
PMC home page
NIHPA Author Manuscripts logoLink to NIHPA Author Manuscripts
. Author manuscript; available in PMC: 2017 Jan 1.
Published in final edited form as: J Fam Issues. 2014 Jan 9;37(1):29–52. doi: 10.1177/0192513X13518211

Birds of a Feather Have Babies Together?: Family Structure Homogamy and Union Stability among Cohabiting Parents

Robin S Högnäs 1,*, Jason R Thomas 2
PMCID: PMC4666318  NIHMSID: NIHMS555674  PMID: 26640313

Abstract

The association between childhood family structure and offspring wellbeing is well-documented. Recent research shows that adult children of divorced parents will likely marry someone whose parents’ divorced (i.e., family structure homogamy) and are subsequently likely to divorce themselves. This literature has focused primarily on marital unions, despite the rise in cohabitation and nonmarital childbearing. Research suggests that marriage and cohabitation are different types of unions and have different implications for the wellbeing of children. Therefore, we extend the literature by examining the role of family structure homogamy in matching patterns and union stability among unmarried, cohabiting couples. Data from the Fragile Families and Child Wellbeing Study suggest that unmarried, cohabiting mothers and fathers are both more likely to be from nonintact childhood family structures and are significantly more likely to dissolve their unions compared to married parents who both tend to be from intact childhood family structures.

Keywords: Family Structure Homogamy, Fragile Families, Union Stability

Childhood family structure—whether or not children grow up with both biological parents—is associated with children’s outcomes well into adulthood. For example, research consistently shows that children of divorced parents often experience their own divorce (Amato, 1996; Wolfinger, 2003, 2011) and more recent evidence suggests that nonmarital childbearing is transmitted from one generation to the next (Högnäs & Carlson, 2012). When romantic partners share nonintact childhood family structures (i.e., both partners’ parents broke up, often referred to as family structure homogamy; Wolfinger, 2003), the likelihood of union dissolution increases relative to those in which one or neither partner’s parents broke up (McGue & Lykken, 1992; Amato, 1996; Wolfinger, 2003).

Prior research has emphasized marital stability, yet little is known about the association between family structure homogamy and cohabiting unions, which are increasingly prevalent; and the formation and dissolution of cohabiting and marital unions differ in important ways (Schoen and Weinick, 1993; Brines and Joyner, 1999). The rise in nonmarital childbearing (i.e., approximately 41% of children are born outside of marriage; Martin, 2011) has meant that many children are born within cohabiting unions, which tend to be much more fragile than marital unions (Beck et al., 2010; McLanahan, 2011). Therefore, we examine the extent to which family structure homogamy is associated with childbearing and the stability of parental unions within the first 5 years of a child’s life. If sharing a childhood family structure influences the childbearing patterns and relationship stability of unmarried, cohabiting parents—or exacerbates the fragility of these unions net of confounding factors (e.g., education, relationship quality)— family structure homogamy among unmarried, cohabiting couples may contribute significantly to the transmission of parental instability across generations.

Using data from the Fragile Families and Child Wellbeing Study (N = 3,069), we begin by analyzing the matching patterns of parents with respect to childhood family structure (i.e., whether or not biological parents were together at age 15) for different types of relationships. We then estimate the association between family structure homogamy and parents’ likelihood of breaking up following the birth of their common child. First, this study extends the literature by replicating previous studies showing that parents from similar family structures are more likely to partner, and that married parents who are both from nonintact family structures are more likely to separate. Second, we extend prior research by examining whether family structure homogamy is associated with whom one bears a child and the stability of parental unions among unmarried, cohabiting couples.

BACKGROUND

The life course perspective draws our attention to the “linked lives” of people in social life and the fact that the experiences of one generation affect the experiences of older and younger generations (Bengtson & Allen, 1993; Elder, 1994, 1998). In particular, the timing and context in which adults’ transition into parenthood, and the stability of parental relationships, influence the experiences of children. Amato’s (1996) model of the pathways connecting the experiences of two generations suggests that parental divorce operates through life course and socioeconomic characteristics (e.g., age at marriage, educational attainment, income, mothers’ employment), child attitudes toward divorce, and interpersonal behavior problems to influence adult offspring’s risk of divorce. Amato further suggests that these pathways are applicable to parents and adult offspring who form nonmarital, cohabiting unions. We borrow from Amato’s framework to describe the links between childhood family structure and both partner selection and union dissolution among cohabiting and married couples.

Couple Matching

An extensive literature describes the partner characteristics that people value and the pressures and constraints faced in forming romantic relationships. In general, people tend to form relationships with others who are similar demographically (Kalmijn, 1998), although differences between cohabiting and married couples have been noted (Schoen & Weinick, 1993; Blackwell & Lichter, 2004). Using data from the National Survey of Families and Households, for example, Schoen and Weinick (1993) find that cohabitation is a “looser bond” than marriage, meaning partners tend to be more concerned with “short-term and achieved characteristics” (e.g., education) than the “long-term” or enduring characteristics (e.g., race and religion) that often concern married couples. Blackwell and Lichter (2000) find similar matching differences among married and cohabiting couples in the 1990 decennial census. While these differences are important and potentially consequential, overall, couples tend to be homogamous (Smock, 2000; Blackwell & Lichter, 2004; Goldstein & Harknett, 2006).

Despite the well-established literature on the matching patterns of couples, particularly married couples, little is known about the extent to which people from similar or different childhood family structures form romantic relationships and bear children (Wolfinger, 2003). Nonintact childhood family structures may influence the formation of romantic unions in two ways. First, indirect effects of family structure background on the socioeconomic characteristics (Amato & Keith, 1991) and the life course of offspring may impact the composition of partnership markets. Offspring of divorced parents are more likely to marry at an earlier age (McLanahan & Bumpass, 1988; Thornton, 1991) and attain less education than children whose parents remain married (McLanhan & Sandefur, 1994; Sun & Li, 2001; Frisco, Muller, & Frank, 2007). School, the age at which one leaves school, and the age at first marriage help structure partnership markets and selection (Mare, 1991). Therefore, divorce and parental separation may operate through educational attainment and age at first marriage indirectly to induce family structure homogamy; and conditioning on these characteristics may attenuate the degree of family structure homogamy.

Schoen and Weinick (1993) find a higher propensity for cohabiting versus married couples to match on educational attainment, but a lower propensity to match on age. These offsetting tendencies make it difficult to predict whether family structure homogamy is more prevalent among cohabiting or married couples. 1 Even so, growing up in a nonintact family is associated with lower educational attainment, which is associated with an increased likelihood of cohabitation. Therefore, one might predict that cohabiting couples share nonintact family structure backgrounds more frequently than married couples. The increasing prevalence of cohabitation (Bumpass & Lu, 2000), however, may mean that family structure background is less predictive of the type of union that couples form. On the other hand, research suggests that those reared in nonintact families are more likely to cohabit prior to, or in place of, marriage (Bumpass, Sweet, & Cherlin, 1989; Thornton, 1991).

The second way that childhood family structure may influence partner selection is through attitudes and personal interactions. In reference to Amato’s model (1996), we note that parental divorce is associated with offspring’s attitudes toward divorce and interpersonal behavior problems, both of which potentially influence the selection of a romantic partner directly. Research suggests that children of divorce share similar attitudes about marriage and divorce which facilitate the process of union formation (Wolfinger, 2003). To the extent that parental separation or divorce increases the next generation’s negative attitudes toward marriage, and reduces the likelihood of marriage, offspring who share these experiences may be more likely to cohabit compared to those whose parents remained married.

Adult offspring’s interpersonal behavior problems (associated with parental separation/divorce) introduce challenges to forming and maintaining romantic relationships (Amato, 1996). When parental unions dissolve, children may be placed in precarious social situations, particularly when conflict emerges subsequently. These environments are likely to influence how children relate to others well into adulthood. Wolfinger (2003) hypothesizes that children of divorce harbor feelings of anger or jealousy toward those who did not share their experience, but relate better to those who did. The shared experience of parental divorce potentially reduces the negative impact of behavior problems on union formation, but may increase the risk of these unions dissolving. In addition, those from intact families may have difficulty forming unions with others who have negative attitudes toward marriage, and behavior problems associated with parental divorce or separation. Indeed, research suggests that married couples tend to share an intact childhood family structure,2 although educational attainment partially attenuates the association (Wolfinger, 2003). To our knowledge, family structure homogamy among unmarried, cohabiting couples has not been examined in the extant literature, despite the growing number of children born within these unions.

Family Structure Homogamy and Childbearing

The family structure in which offspring are reared is associated with the timing and context of their childbearing (McLanahan & Sandefur, 1994; Wu 1996; Carlson, VanOrman, & Pilkauskas, 2010). Indeed, childbearing practices have been shown to be transmitted across generations. Barber (2001), for example, finds that children born to young mothers become young parents themselves; and evidence suggests that children of divorce and those born to unmarried parents both are more likely to have children outside of marriage (Wu & Martinson, 1993; Hofferth & Goldscheider, 2010; Högnäs & Carlson, 2012). These studies, however, have focused primarily on the family structure experiences of mothers and/or fathers independently, and have not addressed (to our knowledge) the extent to which unmarried women and men who are from similar family structure backgrounds have children together. Given the likelihood that growing up in a nonintact family increases the risk of nonmarital childbearing, we would expect this risk to be exacerbated for prospective parents who both are from nonintact family structures compared to prospective parents who both are from intact families (or who do not match on this characteristic).

Union Dissolution

An extensive body of research focuses on the factors associated with union dissolution, particularly among married couples. Race, socioeconomic characteristics, premarital cohabitation, and age at first marriage are associated with divorce (DeMaris & Rao, 1992; Smock, 2000; Dush et al., 2003; Cherlin, 2005). Although this work emphasizes marriage, the rise in cohabitation and nonmarital childbearing, has increased interest in nonmarital union stability. Studies show that among unmarried couples with children, infidelity (Edin & Kefalas, 2005; Hill, 2007), substance abuse (Reed, 2007), physical violence and women’s distrust in men (Carlson, McLanahan, & England, 2004), fathers’ multipartnered fertility (Harknett & McLanahan, 2004; Monte, 2007), and economic circumstances (Tach & Edin 2009; Smock & Manning, 1997) are associated with union dissolution. Important individual correlates associated with union stability also are associated with family experiences during childhood.

The focus here is primarily on union dissolution among unmarried, cohabiting parents, however, we return to Amato’s (1996) conceptualization of the pathways through which childhood experiences with divorce influence offspring’s risk of divorce, which is relevant to union dissolution more generally. Amato argues that three pathways are associated with parental divorce and adult offspring’s subsequent divorce: 1) life course and socioeconomic characteristics (e.g., age at marriage, cohabitation, educational attainment, income, wife’s employment); 2) offspring’s attitudes toward divorce; and 3) interpersonal behavior problems. Parental divorce or separation may operate through each of these mediators to reduce the benefits of marriage, eliminate barriers to divorce or separation, or increase the alternatives to marriage (Amato, 1996; Levinger, 1976).

Data from the study of marriage over the life course suggest that among the three pathways, interpersonal behavior problems, a proposed consequence of parental divorce or separation, primarily influence offspring’s risk of divorce3. Couples in which both sets of parents divorced report higher levels of behavior problems and these couples are also at a higher risk of divorce, although these associations are not significant (Amato, 1996). While others do not directly examine behavior problems, they find that family structure homogamy increases the risk of divorce among married couples (McGue & Lykken, 1992; Amato & Rogers, 1997; Wolfinger, 2003). On the other hand, Bumpass, Martin, and Sweet (1991) find that the risk of divorce is higher among couples in which the wife was from an intact family and the husband was not (net of social and demographic characteristics); however, in this case, family structure homogamy is measured by whether spouses’ families were ‘intact at age 16,’ as opposed to parental divorce during childhood. This unexpected finding is interpreted with caution,4 and generally suggests that outcomes of union dissolution may be particularly sensitive to family structure homogamy measurements.

Again, the extant literature has yet to examine directly family structure homogamy and union stability among unmarried, cohabiting parents; however, some studies examine the childhood family structure of one partner (Manning & Smock, 1995; Wolfinger, 2001; Lichter, Qian, & Mellott, 2006), or control for the childhood family structures of both partners (Smock & Manning, 1997). These studies do not, however, focus on the joint family structure background of the couple nor the association between family structure homogamy and union dissolution among unmarried couples. Generally, growing up in a nonintact family may increase the risk of dissolving a cohabiting union (versus continued cohabitation or moving into marriage), although in some cases, the association may not be significant (e.g., Wolfinger, 2001). In addition, some research suggests that family structure background influences union dissolution primarily within lower socioeconomic contexts. Lichter et al. (2006), for example, find that poor women from intact families are 20% less likely than poor women from intact childhood families to end their cohabiting unions, but find no statistically significant difference among nonpoor women.

In this paper, we extend prior literature by first analyzing the matching patterns for unmarried, cohabiting couples by family structure homogamy, something not yet examined in the extant literature (to our knowledge). We further extend the literature by considering the effects of the joint childhood family structures on union dissolution among cohabiting parents, and by testing whether the results differ for married couples. We hypothesize that the risk of union dissolution will be the highest among unmarried, cohabiting couples in which both members are from nonintact families. The extent to which family structure homogamy influences the dissolution of cohabiting unions may exacerbate other risks of dissolution, including fewer barriers associated with breaking up and less commitment to these relationships (Nock, 1995). In addition, our measure of childhood family structure is similar to Bumpass et al.’s (1991) measure; therefore, our results concerning married parents will serve as a useful comparison to their unexpected findings. It is important to note that we do not have information about whether respondents lived in nonintact families due to parental divorce or possibly the death of a parent. Therefore, our results may be biased to the extent that living in a nonintact family at age 15 resulted from parental death as opposed to divorce. On the other hand, because parental mortality may be less common among offspring at age 15, bias introduced by parental death is likely to be small.

METHODS

Data

We use data from the Fragile Families and Child Wellbeing Study (FFCWS), a longitudinal birth cohort study with an oversample of unmarried parents, to examine family structure homogamy among unmarried, cohabiting parents and the stability of these unions. The full FFCWS includes 4,897 births—3,710 to unmarried parents and 1,187 to married parents. The weighted sample represents nonmarital births in U.S. cities with populations more than 200,000. Baseline interviews with mothers and fathers took place in 75 hospitals just after the birth of a child in 20 U.S. cities between 1998 and 2000. Follow-up interviews were conducted at 1, 3, and 5 years following the birth. Response rates were 88% for unmarried mothers and 75% for unmarried fathers at baseline; 85% of mothers were retained in the study by the 5-year interview, and 88% of fathers were interviewed at least once. Our analytic sample is restricted to cases with no missing information on any of the variables in the analyses.5 When possible, we use mother reports for missing information about fathers to increase our sample size.

Variables and Analytic Approach

We examine two primary research questions in this paper. First, are parents from the same childhood family structure (i.e., family structure homogamy) more likely than those from different childhood family structures to form a union and have a child together? Both mothers and fathers were asked whether or not they lived with both biological parents at age 15 at the baseline survey. Those who lived with both biological parents at age 15 are from “intact” childhood family structures, and those who did not are from “nonintact” childhood family structures. These combinations form a two-by-two contingency table which can be examined using log-linear models (e.g. Mare, 1991; Schwartz and Mare, 2005; Schwartz, 2010) to determine the association between mothers’ and fathers’ childhood family structure.

We hypothesize that the strength of the association between mothers’ childhood family structure (M) and fathers’ childhood family structure (F) will vary by the type of relationship. The relationship status of parents at baseline (R) is classified by: married, cohabiting, or nonresident.6 Given that previous research shows that race and education tend to be associated with both childhood family structure and assortative mating (Cherlin, 2005; Kalmijn, 1998), these measures are included. Race/ethnicity of the mother (B) was dichotomized as black, non-Hispanic or other,7 and mothers’ education at baseline (E) was measured using four categories, less than a high school degree, high school degree, some college, and a college degree or more.

Our analysis of family structure homogamy begins with a baseline model which includes all of the two-way interactions except for MF; in other words, the baseline model assumes there is no association between mothers’ and fathers’ childhood family structure. We then test whether model specifications that allow for family structure homogamy fit data better than the baseline model. This would suggest that parents sort based on childhood family structure. The fit of the models is assessed using Bayesian information criterion, BIC (Raftery, 1995),8 a measure that balances model fit with complexity, favoring the more parsimonious specification among models that fit data similarly well.

Our second research question concerns the relative stability of parental unions distinguished by family structure homogamy. We hypothesize that parental unions characterized by family structure homogamy are more likely to dissolve than parental couples with any other combination of childhood family structures. This hypothesis is tested by fitting the following logistic regression model:

  • log(odds of union dissolution) = β0 + β1 married + β2(intact mom, intact dad) + β3(nonintact mom, intact dad) + β4(intact mom, nonintact dad) + β5 controls,

for couples who were married or cohabiting at baseline. Greater differences between β2, β3, and β4 and the reference category (nonintact mom, nonintact dad) indicate stronger support for our hypothesis. Union stability is determined by relationship status at baseline and at 1, 3, and 5 years following the birth of a child (mothers’ reports). Married or cohabiting relationships at baseline which change to nonresident status (e.g., friends) at the follow-up surveys are coded as dissolved.

Childhood family structure may operate indirectly to influence union dissolution (Amato 1996). Indirect pathways through which childhood family structure may operate include life course and socioeconomic variables such has whether or not the mother was young (less than 25 years of age), the mother’s level of education, and couple’s relationship status (married or cohabiting), all of which were measured at the baseline survey. Childhood family structure also may influence union dissolution through the offspring’s attitudes, which we model by including a measure of the mother’s agreement with the statement “parents should stay together for the children even if they don’t get along”. For this measure, response choices were 1 = strongly disagree, 2 = disagree, 3 = agree, and 4 = strongly agree. Interpersonal behavior problems, the third pathway in Amato’s framework, are measured using four measures reported by the mother about the father at baseline: whether the father was fair and willing to compromise over a disagreement, whether he expressed affection or love for the mother, whether he insulted or criticized the mother’s ideas (reverse coded), and whether he encouraged or helped the mother to do things that were important to her (α=.66). Response choices were 1 = never, 2 = sometimes, and 3 = often. Higher scores indicated greater levels of support and less conflict stemming from interpersonal behavior problems.

Mothers’ race was included to control for racial differences in the propensity of divorce or union dissolution (Cherlin, 2005). Finally, we examine whether the association between childhood family structure and union dissolution differs by union status given that marriage and cohabitation are distinct types of relationships (Schoen & Weinick, 1993).

RESULTS

Descriptive Statistics

Table 1 shows the weighted descriptive statistics for our analytic sample. Fifty-three percent of the couples are married at the birth of their child, 27% are cohabiting, and 20% are living apart (at baseline). Among those who are married, 47% are couples composed of both mothers and fathers from intact childhood families, which is more than twenty percentage points larger than any other configuration of childhood family structures. Conversely, couples consisting of both mothers and fathers from nonintact families make up the modal category among cohabiting and nonresident couples (35% and 47%, respectively). As expected, married parental unions are more stable than cohabiting parental unions. Following the focal birth, 2% of marriages dissolve after 1 year, 7% after 3 years, and 16% dissolve after 5 years. The corresponding numbers for cohabiting parents are 20%, 34% and 44%.

Table 1.

Weighted Means of Sample Characteristics by Relationship Status at Baseline

All Married Cohabiting Nonresident
Relationship Status at Baseline .53 .27 .20
FAMILY BACKGROUND
    Mother and Father lived w/ both biological parents at age 15 (yes/no) .33 .47 .21 .11
    Only Father lived w/ both biological parents at age 15 (yes/no) .23 .24 .26 .18
    Only Mother lived w/ both biological parents at age 15 (yes/no) .20 .18 .18 .24
    Neither Mother nor Father lived w/ both biological parents at age 15 (yes/no) .24 .10 .35 .47
UNION INSTABILITY (Married and Cohabiting only)
    Union dissolves by 1 year after birth of child .08 .02 .20 --
    Union dissolves by 3 years after birth of child* .16 .07 .34 --
    Union dissolves by 5 years after birth of child* .25 .16 .44 --
DEMOGRAPHIC AND SOCIOECONOMIC STATUS
    Mother is Black, non-Hispanic .37 .19 .46 .73
    Mother's age 26.9 (.16) 29.2 (.39) 25.20 (.21) 22.90 (.28)
    Mother has less than HS degree .27 .18 .37 .37
    Mother has HS degree .34 .26 .41 .47
    Mother has some college .20 .22 .20 .15
    Mother has college degree or more .18 .34 .01 .01
ATTITUDINAL MEASURES (Mothers')
    Parents should stay together for the children even if they don't get along (1–4) 1.94 (.04) 2.00 (.06) 1.85 (.03) 1.91 (.05)
RELATIONSHIP QUALITY
    Mother feels supported by father (range= 1–3) 2.69 (.02) 2.73 (.03) 2.70 (.01) 2.57 (.02)
N 3,069 715 1,345 1,009
Open in a new tab

Note: Means are weighted by city sampling means. Number of cases (Ns) are not weighted. Jacknife standard errors are in parentheses. All variables are measured at baseline, except for union dissolution which is measured 1 year after the birth of the child.

*

Sample sizes for union dissolution by the 3rd year after the child's birth are 702 and 1,291 for married and cohabiting parents, respectively. The corresponding sample sizes for union dissolution by the 5th year after the child's birth are 680 (married) and 1,285 (cohabiting).

In terms of demographic and socioeconomic characteristics, parents in nonresident relationships at the birth of their child are more likely to be non-Hispanic Black (73%) compared to married (19%) and cohabiting couples (46%). The average age of mothers is 27 and married mothers tend to be older than cohabiting and nonresident mothers (29, 25, and 23, respectively). Most mothers (61%) had a high school degree or less, with variation by relationship status. Fifty-six percent of married mothers have some college or more education compared to less than 25% of cohabiting and nonresident mothers. Mothers’ attitudes about parents staying together for children even when they do not get along ranges from an average of 1.85 among cohabiting mothers to 2.00 for married mothers. In terms of relationship quality, nonresident mothers tend to report slightly lower levels than married and cohabiting mothers (2.57, 2.73, and 2.70, respectively).

Couple Matching

Figure 1 shows the distributions of parental matching—defined by mothers’ and fathers’ childhood family structures—separated by mother’s education (top panel) and mother’s race (bottom panel). Family structure homogamy appears at the different levels education for mothers, with this exception of mothers who have a high school degree. At higher levels of education, mothers and fathers both tend to be from intact families. Conversely, family structure homogamy is more likely to characterize the partnerships of mothers with less than a high school degree. The bottom panel in Figure 1 suggests that family structure homogamy also varies by mother’s race. Over 40% of non-Hispanic, Black mothers are from nonintact families and had the focal child with a father from a nonintact family, whereas more than 40% of Hispanic and White mothers share an intact childhood family structure with their children’s fathers.

Figure 1.

Figure 1

Open in a new tab

Parental Matching by Childhood Family Structure, Education, and Race

Note. Childhood Family Structure I (intact) = Lived with Both Biological Parents at Age 15, NI (non-intact) = Did Not Live with Both Biological Parents at Age 15)

To examine the presence of family structure homogamy net of relationship type, race, and education, we turn to the results of our log-linear models. Table 2 shows five models with corresponding BIC values which assess each model’s fit. Our baseline model includes all two-way interactions with the exception of MF, the interaction between mothers’ and fathers’ childhood family structure.9 Comparing the baseline model (Model 1) to Model 2, which includes the MF interaction term, provides a test of the hypothesis that mothers from nonintact families are more likely to have had the focal child with a father from a nonintact family compared to a father from an intact family (and vice versa). The BIC for the baseline model is −406 and the corresponding value for Model 2, with the MF interaction, is −409, suggesting that the inclusion of family structure homogamy improves the fit of the model (more negative values of BIC indicate improvements in model fit).10 We also examined whether family structure homogamy depends on couples’ relationship status at baseline (Model 3) and mothers’ race (Model 4) and education (Model 5). The BIC values suggest that none of these models are preferred to Model 1, our baseline model (i.e., values remain the same or decrease), and that simply including the MF interaction provides the best fit to these data.

Table 2.

Fit of Log-linear Models of the Association between Mother's and Father's Childhood Family Structure (N = 3,069)

Model Deviance df BIC
1. [MR] [MB] [ME] [FR] [FB] [FE] 108 64 −406
2. [MR] [MB] [ME] [FR] [FB] [FE] [MF] 96 63 −409
3. [MFR] [MB] [ME] [FB] [FE] 96 61 −394
4. [MR] [MFB] [ME] [FR] [FE] 92 62 −406
5. [MR] [MB] [MFE] [FR] [FB] 93 60 −388
Open in a new tab

Note: M=Mother's Childhood Family Structure (2 categories); F=Father's Childhood Family Structure; R=Relationship Status (3 categores); B=Mother is non-Hispanic, Black; E=Mother's education (4 categories). Estimates are unweighted.

Union Stability

We now turn to our results regarding family structure homogamy and union dissolution following the birth of a child. Figure 2 shows the proportion of dissolved unions at 1, 3, and 5 years following the birth, by childhood family structure matching. Couples characterized by family structure homogamy have the highest probability of dissolution at every time point, followed by couples in which only the mother is from a nonintact family. Dissolution is the least likely for couples in which both mothers and fathers are from intact family structures, although even among mothers and fathers who are both from intact family structures, dissolution increases between years 1 and 5.

Figure 2.

Figure 2

Open in a new tab

The Proportion of Dissolved Unions by Childhood Family Structure Matching at 1, 3, and 5 Years after the Birth of a Child

Table 3 presents estimates from nested logistic regression models of union dissolution 1 year after baseline (to retain the highest number of cases) for parents who were married and cohabiting at baseline.11 We present the nested models separately for cohabiting and married couples given that the processes underlying union dissolution for these groups are likely to differ. Models 1 and 1a (for cohabiting and married couples respectively) include childhood family structure matching; Models 2 and 2a add dummy variables for mothers’ race, mother’s education, and whether the mother is less than age 25 (all measured at the baseline survey); Models 3 and 3a add mothers’ attitude toward union stability and relationship quality.

Table 3.

Logistic Regression Model of the Odds of Union Dissolution 1 Year After the Birth of a Child (N = 2,060)

Model 1 Cohabiting
Model 2		 Model 3 Model 1a Married
Model 2a		 Model 3a
Childhood Family Structure
Neither Parent lived w/ both biological parents at age 15 reference category
    Mother and Father lived w/ both biological parents at age 15 .49***
(.33, .71)		 .69†
(.46, 1.03)		 .71†
(.48, 1.07)		 14***
(.05, .42)		 .29*
(.09, .91)		 .28*
(.09, .94)
    Only Father lived w/ both biological parents at age 15 .76
(.55, 1.06)		 .88
(.63, 1.24)		 .90
(.64, 1.27)		 .18**
(.05, .64)		 .24*
(.07, .91)		 .27*
(.07, 1.02)
    Only Mother lived w/ both biological parents at age 15 .97
(.70, 1.34)		 1.10
(.79, 1.54)		 1.16
(.83, 1.64)		 .40†
(.15, 1.10)		 .58
(.20, 1.66)		 .75
(.25, 2.22)
Demographics
    Mother is Black, non-Hispanic 1.94***
(1.48, 2.55)		 1.88***
(1.43, 2.48)		 1.39
(.58, 3.32)		 1.43
(.59, 3.46)
    Mother is less than age 25 at baseline (yes/no) 1.27†
(.97, 1.68)		 1.29†
(.98, 1.70)		 2.18†
(.91, 5.22)		 1.95
(.78, 4.83)
    Mother has HS degree (reference = < HS degree omitted) .86
(.64, 1.16)		 .87
(.65, 1.17)		 .98
(.34, 2.84)		 .72
(.24, 2.18)
    Mother has some college .78
(.56, 1.10)		 .79
(.56, 1.12)		 .51
(.17, 1.56)		 .41
(.13, 1.30)
    Mother has college degree or more 1.03
(.49, 2.17)		 1.04
(.49, 2.20)		 .22†
(.04, 1.22)		 .20
(.04, 1.15)
Mother's Attitude about Stability & Relationship Quality
    Parents should stay together for the children even if they don't get along
range: 1 – 4, high values indicate stronger agreement		 .84†
(.69, 1.02)		 .45*
(.22, .93)
    Mother feels supported by father
range: 1 – 3, high values indicate more support		 .44**
(.30, .64)		 .15**
(.05, .50)
N 1,345 715
Open in a new tab
***

Note: p < .001;

**

p < .01;

*

p < .05;

†

p ≤ .10. 95% confidence intervals shown in parentheses. All covariates are measured at baseline. Estimates are unweighted.

Beginning with Models 1 and 1a, the estimated risk of union dissolution is over 50% lower for cohabiting couples and over 80% lower for married couples in which both the mother and father are from intact families compared to couples in which both of the parents are from nonintact families. The risk of dissolution when only one parent is from a nonintact family structure is also lower than if both parents are from nonintact families. These differences in union dissolution are attenuated when demographic characteristics are added in Model 2 for cohabiting couples, and only unions consisting of mothers and fathers from intact families face a significantly lower risk of union dissolution relative to couples where both parents are from a nonintact family. Among married couples, the risk union dissolution, however, also remains statistically lower when only the father lived with both biological parents at age 15. Very little changes when the attitudinal and relationship quality measures are added in Models 3 and 3a. These variables, however, are associated significantly with the risk of union dissolution for both cohabiting and married couples.12

Focusing on Models 3 and 3a, contrary to our expectations (but perhaps consistent elsewhere, Wolfinger, 2001), the estimated effects of childhood family structure appear to be stronger among married versus cohabiting parents. In the full model, the estimated odds of union dissolution among married couples in which both parents are from intact families are more than 70% lower than for couples in which both spouses are from nonintact families. The reduction in the odds of union dissolution associated with couples in which only the mother is from a nonintact family is low also, although the estimate is marginally significant. The estimated odds of union dissolution among cohabiting couples does not appear to be associated with the childhood family structures of the parents; however, one exception is that the odds are slightly lower when both cohabiting parents are from intact families compared to when both are from nonintact families.

Childhood Family Structure Robustness

Our measure of childhood family structure does not allow us to account for the timing of parental separation (nor for parental repartnering), which may influence the attitudes and behavior of adult offspring. In addition, parents may have been temporarily separated when offspring were age 15. To address this possibility, we compared respondents reports of their childhood family structure at age 15 (i.e., whether their biological parents were or were not together) to their reports of whether or not their parents were together at the time that they were interviewed for the 5-year follow-up survey. More than 80% of respondents reported the same relationship status for their parents when they were age 15 (which was reported at baseline) and when they were asked at time of the 5-year follow-up survey. This consistency suggests that our measure of childhood family structure is a reasonable indicator of family stability.

DISCUSSION

In this paper, we set out to answer two questions about family structure homogamy among unmarried, cohabiting couples. First, are cohabiting couples more likely to have a child with someone from a similar childhood family structure? Results from our log-linear models suggest a positive association between the childhood family structures of mothers and fathers. Our analysis provides evidence that matching on family structure homogamy extends beyond marriage to childbearing, and that the pattern is observed among married and cohabiting parents. The strength of this association does not appear to depend on the type of relationship or the education of the mother, although we do find some weak evidence of an interaction with the mother’s race.

Our second research question concerns the stability of married and cohabiting parents in which both mothers and fathers are from nonintact family structures. Among married couples, it appears that mothers and fathers who are both from nonintact families face the highest likelihood of union dissolution one year after the birth of their child. The likelihood of marital dissolution is not significantly different when there is family structure homogamy compared to only one parent being from a nonintact family. One exception is a significantly lower likelihood of dissolution when only the married father is from a nonintact family. The same is true for couples in which only the married mother is from a nonintact family, but this estimate is not statistically significant. These findings are generally consistent with previous research (Amato, 1996; Wolfinger, 2003).

Why might the likelihood of union dissolution decrease when only married fathers are from intact childhood family structures? Although we are unable to test mechanisms due to missing measures and small sample sizes with the inclusion of characteristics of both mothers and fathers, we draw from existing literature to speculate about the social processes underlying our findings. One possible explanation is that the effects of family structure background on factors like interpersonal behaviors play out differently for men and women, particularly following the birth of a child. In addition, fathers from intact family structures may face fewer barriers with respect to socioeconomic status, which tends to be associated with romantic relationship problems (Tach & Edin, 2009; Smock & Manning, 1997). The growing educational attainment of women (Crissey, 2009) may mean that women from nonintact families and men from intact families match on education increasingly, despite the negative association between parental divorce and separation and offspring’s education (McLanahan & Sandefur, 1994; Sun & Li, 2001; Frisco et al., 2007). Educational attainment may buffer relationships in which mothers are from nonintact and fathers are from intact families.

The findings for cohabiting couples differ in that only mothers from intact families who are living with fathers from the same family background are at a (moderate) significantly lower risk of union dissolution one year after the birth of their child, relative to families consisting of both parents from nonintact families. In addition, cohabiting families in which only the mother is from a nonintact family face lower odds of union dissolution (compared to the reference group), but the difference is small and insignificant. Conversely, cohabiting families in which only the father is from a nonintact family actually face a higher risk of union dissolution, but again the difference is not statistically significant.

Our findings differ with respect to married and cohabiting couples in which only mothers are from intact families—the likelihood of dissolution decreases for married parents, but increases for cohabiting parents. Here again, it may be that the potential effects of growing up in a nonintact family structure differ for men and women, particularly following the birth of a child for fathers. It is conceivable that the impact of family structure backgrounds on romantic partnerships surface once cohabiting men from nonintact families have children as opposed to the period during which men form romantic partnerships. In addition, mismatches in family structure backgrounds among married and cohabiting parents may influence relationship stability differently in part due to unobserved characteristics associated with union formation.

Finally, our results suggest that parents’ childhood family structures are not associated with union stability beyond one year following the birth of a child. This is true for both married and unmarried, cohabiting parents. It may be that longer relationship duration following the birth of a child reduces the potential influence of family structure homogamy on couples. Interpersonal behaviors and attitudes associated with experiencing parental separation or divorce during childhood may change over time. In addition, couples may experience external shocks (e.g., a financial crisis) that influence their relationships above and beyond couple or personal characteristics. Future investigation is necessary to determine whether small sample sizes or a diminishing importance of family structure accounts for no association between family structure homogamy and union stability 5 years following the birth of a child.

We are unaware of research examining the effect of family structure homogamy on the union stability of unmarried, cohabiting couples; therefore, these findings contribute to our understanding of the importance of family structure homogamy for a large, growing proportion of American families. This paper is not, however, without limitations. First, this study is limited to those who live in urban areas with populations of 200,000 or more. Therefore, we cannot generalize our findings to families who live in rural areas. Given high concentrations of poverty and low educational attainment in rural areas, we may miss couples who are both more likely to come from nonintact families and more likely to have children within cohabitation versus marriage. If anything, without rural couples in our sample, we may be underestimating the association between family structure homogamy and union stability among cohabiting parents. In addition, while we are able to consider the childhood family structure of both mothers and fathers, our sample shrinks when we include additional information on both parents. We know, for example, that educational homogamy often characterizes couple relationships (e.g. Mare, 1991; Schwartz and Mare, 2005; Schwartz, 2010); however, in the interest of retaining our sample size, we cannot examine these potentially informative patterns.

Overall, it is unlikely that people actually consider the childhood family structure of a potential romantic partner when deciding whether or not to pursue a relationship, and more importantly, with whom to bear a child. We are, therefore, left with the question of why couples and parents tend to match on this characteristic. It may be simply that personal social networks are generally characterized by homophily (McPherson, Smith-Lovin, & Cook, 2001). That is, men and women are more likely to socialize, form partnerships, and have children with others who are like them across a range of social dimensions (e.g., education, race, religion), including childhood family structure. Equally likely, however, is that the attitudes, personality traits, and relationship skills developed among children from nonintact families make relationships more amenable when these traits are shared (Amato, 1991).

Further exploration of these hypotheses would benefit from large longitudinal samples with information on attitudes and characteristics measured before romantic and parental relationships begin. These measures would allow one to determine whether attitudes are associated with both the formation and dissolution of unions. Better information on the timing of—and exposure to—parental divorce/separation also would be helpful. That is, the extent to which the timing of exposure to parental relationship dissolution (e.g., early childhood, middle childhood, adolescence) differentially impacts the future relationship stability of offspring is worth exploring. It may be that earlier exposure to divorce and parental separation impacts the relationship trajectories of offspring more strongly than later exposure.

Further exploration of how childhood family structure is linked to parental unions is worthwhile, particularly given the deleterious impact of parental instability on children well into adulthood (McLanahan & Sandefur, 1994; Amato, 1991; Sun & Li, 2001; Sun, 2001). A close examination of the mechanisms or pathways which underlie the patterns that we describe here stands to contribute to knowledge about the social conditions associated with the systematic fragility of cohabiting, parental unions. Along the same lines, the literature has emphasized strongly the influence of individual characteristics on union stability; however, more research focused on the joint characteristics, and family background experiences, of mothers and fathers is needed to fully understand why cohabiting parents’ relationships end frequently and often times quickly.

Acknowledgments

This research was carried out using the facilities at the University of Louisville, Pennsylvania State University, and the Center for Demography and Ecology at the University of Wisconsin-Madison. Some of the first author’s time on this paper was supported by an NICHD postdoctoral fellowship (#R24HD047873) and by an ARRA supplement (#R01HD57894).This study was also supported by an additional grant from the NICHD (2R24HD041025-11) and grants from the National Institute of Aging (5P30AG017266 and 5T32AG000129-22.) The authors would like to thank Marcy Carlson, Alberto Palloni, and Christine Schwartz for helpful advice during the early stages of the project and anonymous reviewers for insightful comments which helped to improve the manuscript.

Footnotes

1

Goldstein and Harknett (2006) look at the association between the type of relationship (married, cohabiting, romantically involved, not romantically involved) among couples and their educational attainment, but they focus on the effects of education on relationship type rather than the degree of homogamy by relationship type.

2

In the only other study of family structure homogamy that we know of, Landis (1956) finds no evidence of family structure homogamy among a nonrandom sample of college students.

3

In a logistic regression of offspring divorce on parental divorce of both spouses, Amato (1996) finds the risk of offspring divorce, relative to a couple where neither spouse experienced a parental divorce, increase by 26% (not statistically significant), 59% (statistically significant), and 189% (statistically significant) when the husband’s parents divorced, the wife’s parents divorced, and both spouses, parents’ divorced, respectively. There is a similar gradient when the mediating variables are included, but the differences are not statistically significant.

4

Gender differences in the effects of childhood family structure on union dissolution may result from differences in how sons and daughters are affected by the separation of their parents, but we do not have sufficient information to test associated hypotheses and thus our work is primarily descriptive.

5

Our analytic sample is more likely to include fathers from nonintact families, cohabiting couples, and stable unions compared to the cases excluded from our analysis. While these differences may introduce bias, findings from additional analyses (not shown) using multiply imputed data were qualitatively similar to those presented here.

6

Nonresident fathers are those who are “visiting,” are just friends, or hardly ever or never talk to mothers.

7

We collapsed Hispanic and White, non-Hispanic mothers to maximize cell counts in our contingency table. The differences between these groups were small.

8

BIC = Deviance – df*log(N), where df is the degrees of freedom for the model, and N is the sample size. More negative values of BIC indicate a better model fit (Raftery 1995).

9

We start with this baseline model because childhood family structure is associated with relationship status (see Table 1), as well as race and education (see Figure 1). Previous research also supports two-way associations between race and education; race and relationship status; and education and relationship status. The baseline model fit better than simpler models that lacked the two-way interactions, as indicated by BIC.

10

Raftery (1995) suggests that BIC differences of 0–2, 2–6, 6–10, and 10 or more indicate weak, positive/moderate, strong, and very strong evidence (respectively) that models with smaller (more negative) BICs provide better fits to data.

11

We also examined the odds of union dissolution at years 3 and 5 (results not shown). The results were similar, but differences between couple types were attenuated and not statistically significant in our full models. This was due partially to the significant reduction in the sample size associated with missing values on all variables for mothers and fathers (where fathers’ reports were used).

12

In additional analyses not shown, we find that intact childhood family structures significantly predict positive attitudes towards union stability net of socioeconomic and demographic characteristics, although we find no significant effect on our measure of relationship quality. These results suggest that childhood family structure operates indirectly through attitudes to influence union stability.

Contributor Information

Robin S. Högnäs, University of Louisville.

Jason R. Thomas, Pennsylvania State University

REFERENCES

  1. Amato PR, Keith B. Separation from a parent during childhood and adult socioeconomic attainment. Social Forces. 1991;70(1):187–206. [Google Scholar]
  2. Amato PR. Explaining the intergenerational transmission of divorce. Journal of Marriage and Family. 1996;58:628–640. [Google Scholar]
  3. Amato PR, Rogers SJ. A longitudinal study of marital problems and subsequent divorce. Journal of Marriage and Family. 1997;59:612–624. [Google Scholar]
  4. Barber JS. The intergenerational transmission of age at first birth among married and unmarried men and women. Social Science Research. 2001;30:219–247. [Google Scholar]
  5. Beck AN, Cooper CE, McLanahan SS, Brooks-Gunn J. Relationship transitions and maternal parenting. Journal of Marriage and Family. 2010;72:219–233. doi: 10.1111/j.1741-3737.2010.00695.x. [DOI] [PMC free article] [PubMed] [Google Scholar]
  6. Bengtson VL, Allen KR. The life course perspective applied to families over time. In: Boss P, Doherty WJ, LaRossa R, Schumm WR, Steinmetz SK, editors. Sourcebook of Family Theories and Methods: A Contextual Approach. New York: Springer US; 1993. pp. 469–504. [Google Scholar]
  7. Blackwell DL, Lichter DT. Mate selection among married and cohabiting couples. Journal of Family Issues. 2000;21:275–302. [Google Scholar]
  8. Blackwell DL, Lichter DT. Homogamy among dating, cohabiting, and married couples. Sociological Quarterly. 2004;45:719–737. [Google Scholar]
  9. Brines J, Joyner K. The ties that bind: Principles of cohesion in cohabitation and marriage. American Sociological Review. 1999;64(3):333–355. [Google Scholar]
  10. Bumpass LL, Martin TC, Sweet JA. The impact of family background and early marital factors on marital disruption. Journal of Family Issues. 1991;12:22–42. doi: 10.1177/019251391012001003. [DOI] [PubMed] [Google Scholar]
  11. Bumpass LL, Sweet JA, Cherlin A. The role of cohabitation in declining rates of marriage. Journal of Marriage and the Family. 1989;35:913–927. [Google Scholar]
  12. Bumpass L, Lu H. Trends in cohabitation and implications for children's family contexts in the United States. Population Studies. 2000;54(1):29–41. doi: 10.1080/713779060. [DOI] [PubMed] [Google Scholar]
  13. Carlson M, Corcoran M. Family structure and children's behavioral and cognitive outcomes. Journal of Marriage and the Family. 2004;53:779–792. [Google Scholar]
  14. Carlson M, McLanahan S, England P. Union formation in fragile families. Demography. 2004;41:237–262. doi: 10.1353/dem.2004.0012. [DOI] [PMC free article] [PubMed] [Google Scholar]
  15. Carlson M, VanOrman A, Pilkauskas N. Examining the antecedents of U.S. nonmarital childbearing. Center for Demography and Ecology Working Paper. 2010 doi: 10.1007/s13524-013-0201-9. 2010–12. [DOI] [PMC free article] [PubMed] [Google Scholar]
  16. Cherlin A. American marriage in the early twenty-first century. Future of Children. 2005;15(2):33–55. doi: 10.1353/foc.2005.0015. [DOI] [PubMed] [Google Scholar]
  17. Crissey SR. Current Population Reports U.S. Census Bureau. Washington, DC: 2009. Educational attainment in the United States: 2007. [Google Scholar]
  18. Demaris A, Rao KV. Premarital cohabitation and subsequent marital stability in the United States: A reassessment. Journal of Marriage and the Family. 1992;54:178–190. [Google Scholar]
  19. Dush CMK, Cohan CL, Amato P. The relationship between cohabitation and marital quality and stability: Change across cohort. Journal of Marriage and the Family. 2003;65(3):539–549. [Google Scholar]
  20. Edin K, Kefalas M. Promises I can keep: why poor mothers put motherhood before marriage. Berkeley and Los Angeles, CA: University of California Press; 2005. [Google Scholar]
  21. Elder GH., Jr Time, human agency, and social change: Perspectives on the life course. Social Psychology Quarterly. 1994;57:4–15. [Google Scholar]
  22. Elder GH., Jr The life course as developmental theory. Child Development. 1998;69:1–12. [PubMed] [Google Scholar]
  23. Frisco ML, Muller C, Frank K. Parents' union dissolution and adolescents' school performance: Comparing methodological approaches. Journal of Marriage and Family. 2007;69:721–741. doi: 10.1111/j.1741-3737.2007.00402.x. [DOI] [PMC free article] [PubMed] [Google Scholar]
  24. Goldstein JR, Harknett K. Parenting across racial and class lines: Assortative mating patterns of new parents who are married, cohabiting, dating or no longer romantically involved. Social Forces. 2006;85(1):121–143. [Google Scholar]
  25. Harknett K, McLanahan SS. Racial and ethnic differences in marriage after the birth of a child. American Sociological Review. 2004;69:790–811. [Google Scholar]
  26. Hill HD. Steppin’ out: infidelity and sexual jealousy among unmarried parents. In: England P, Edin K, editors. Unmarried Coupes with Children. New York: Russell Sage Foundation; 2007. pp. 104–132. [Google Scholar]
  27. Hofferth SL, Goldscheider Family structure and the transition to early parenthood. Demography. 2010;47(2):415–437. doi: 10.1353/dem.0.0102. [DOI] [PMC free article] [PubMed] [Google Scholar]
  28. Högnäs RH, Carlson MJ. “Like parent, like child?”: The intergenerational transmission of nonmarital childbearing. Social Science Research. 2012;41:1480–1494. doi: 10.1016/j.ssresearch.2012.05.012. [DOI] [PMC free article] [PubMed] [Google Scholar]
  29. Kalmijn M. Intermarriage and homogamy: Causes, patterns, trends. Annual Review of Sociology. 1998;24:395–421. doi: 10.1146/annurev.soc.24.1.395. [DOI] [PubMed] [Google Scholar]
  30. Landis JT. The pattern of divorce in three generations. Social Forces. 1956;34:213–216. [Google Scholar]
  31. Levinger G. A socio-psychological perspective on marital dissolution. Journal of Social Issues. 1976;52:21–47. [Google Scholar]
  32. Lichter DL, Qian Z, Mellott LM. Marriage or dissolution? union transitions among poor cohabiting women. Demography. 2006;42(2):223–240. doi: 10.1353/dem.2006.0016. [DOI] [PubMed] [Google Scholar]
  33. Manning WD, Smock PJ. Why marry? race and the transition to marriage among cohabitors. Demography. 1995;32(4):509–520. [PubMed] [Google Scholar]
  34. Mare R. Five decades of educational assortative mating. American Sociological Review. 1991;56(1):15–32. [Google Scholar]
  35. Martin JA, Hamilton BE, Ventura SJ, Osterman MJK, Kirmeyer S, Matthews TJ, et al. “Births: Final data for 2009” 60 (1). Hyattsville, MD. National Center for Health Statistics. 2011 [PubMed] [Google Scholar]
  36. McGue M, Lykken DT. Genetic influence on risk of divorce. Psychological Science. 1992;3:368–373. [Google Scholar]
  37. McLanahan S, Bumpass L. Intergenerational consequences of family disruption. American Journal of Sociology. 1988;94(1):130–152. [Google Scholar]
  38. McLanahan S, Sandefur G. Growing up with a single parent: what hurts? what helps? Cambridge, MA: Harvard University Press; 1994. [Google Scholar]
  39. McLanahan S. Family instability and complexity after a nonmarital birth: Outcomes for children in fragile families. In: Carlson MJ, England P, editors. Social Class and Changing Families in an Unequal America. Stanford, CA: Stanford University Press; 2011. [Google Scholar]
  40. McPherson M, Smith-Lovin L, Cook JM. Birds of a feather: Homophily in social networks. Annual Review of Sociology. 2001;27:415–44. [Google Scholar]
  41. Monte LM. Blended but not the Bradys: Navigating unmarried multiple partner fertility. In: England P, Edin K, editors. Unmarried Couples with Children. New York: Russell Sage; 2007. pp. 183–203. [Google Scholar]
  42. Nock SL. A comparison of marriages and cohabiting relationships. Journal of Family Issues. 1995;16(1):53–76. [Google Scholar]
  43. Raftery AE. Bayesian model selection in social research. Sociological Methodology. 1995;25:111–163. [Google Scholar]
  44. Reed J. Anatomy of the breakup: How and why do unmarried couples with children break up? In: England P, Edin K, editors. Unmarried Couples with Children. New York: Russell Sage Foundations; 2007. pp. 133–155. [Google Scholar]
  45. Schwartz CR. Pathways to educational homogamy in marital and cohabiting unions. Demography. 2010;47(3):735–753. doi: 10.1353/dem.0.0124. [DOI] [PMC free article] [PubMed] [Google Scholar]
  46. Schwartz CR, Mare RD. Trends in educational assortative mating from 1940 to 2003. Demography. 2005;42(4):621–646. doi: 10.1353/dem.2005.0036. [DOI] [PubMed] [Google Scholar]
  47. Schoen R, Weinick Partner choice in marriages and cohabitations. Journal of Marriage and Family. 1993;55(2):408–414. [Google Scholar]
  48. Smock P. Cohabitation in the United States: An appraisal of research themes, findings, and implications. Annual Review of Sociology. 2000;26:1–20. [Google Scholar]
  49. Smock P, Manning WD. Cohabiting partners' economic circumstances and marriage. Demography. 1997;34(3):331–341. [PubMed] [Google Scholar]
  50. Sun Y. Family environment and adolescents' well-being before and after parents' marital disruption: a longitudinal analysis. Journal of Marriage and Family. 2001;63:697–713. [Google Scholar]
  51. Sun Y, Li Y. Marital disruption, parental involvement, and children's academic achievement: A prospective analysis. Journal of Family Issues. 2001;22(1):27–62. [Google Scholar]
  52. Tach L, Edin K. The structural and institutional deteminants of union dissolution in fragile families. Paper prepared for presentation at the 2009 annual meeting of the Population Association for America. 2009:1–52. [Google Scholar]
  53. Thornton A. Influence of the marital history of parents on the marital and cohabitational experiences of children. American Journal of Sociology. 1991;96:868–894. [Google Scholar]
  54. Wolfinger NH. The effects of family structure of origin on offspring cohabitation duration. Sociological Inquiry. 2001;71:293–313. [Google Scholar]
  55. Wolfinger NH. Family structure homogamy: The effects of parental divorce on partner selection and marital stability. Social Science Research. 2003;32:80–97. [Google Scholar]
  56. Wolfinger NH. More evidence for trends in the intergenerational transmission of divorce: A completed cohort approach using data from the General Social Survey. Demography. 2011;48(2):581–592. doi: 10.1007/s13524-011-0025-4. [DOI] [PubMed] [Google Scholar]
  57. Wu LL, Martinson BC. Family structure and the risk of a premarital birth. American Sociological Review. 1993;58(2):210–232. [Google Scholar]
  58. Wu LL. Effects of family instability, income, and income instability on the risk of premarital birth. American Sociological Review. 1996;61(3):386–406. [Google Scholar]

RESOURCES