Childhood Family Structure and Romantic Relationships During the Transition to Adulthood

Abstract

We use the National Longitudinal Study of Adolescent Health (Add Health) to examine whether childhood family structure experiences influence the development of romantic relationships during adolescence and whether adolescent relationships, in turn, help to shape long-term relationship trajectories. Young people who live in “nontraditional” families during their childhood are more likely than their peers to engage in romantic relationships during adolescence. Family-related mechanisms are significant mediators of this association. Individuals who were raised in stepparent and single-parent families are also more likely to cohabit during adulthood, and those who were raised in single-parent families are less likely to have ever married. Childhood family structure is not associated with serious relationship conflict during adolescence or adulthood, however. Moreover, although adolescent relationship experiences have long-term effects on relationship trajectories, they do not significantly mediate the associations between childhood family structure and relationship outcomes in adulthood.

Keywords

family structure conflict marriage cohabitation romantic relationships

Major shifts in the types of households found in the United States have led to an increase in the diversity of living arrangements experienced by American children, including an increase in single-parent, stepfamily, and cohabiting households. For example, in 1970, 85% of children lived in a home with two parents present, however, this percentage declined to only 70% by 2008 (U.S. Census Bureau, 2009). Although the majority of children today continue to reside with both biological parents, children are increasingly likely to spend a portion of their childhood in an alternative family form (Bumpass & Lu, 2000). As a result, there has been increasing concern about the consequences of children’s family structure and the experience of family change for the well-being of young people as they grow and develop.

Compared with children in “nontraditional” families (i.e., stepparent, single-parent, and nonparent families), children who grow up in two–biological parent households tend to fare better on a wide variety of well-being indicators (Moore et al., 2007; Teachman, 2002). For example, children from single-parent and stepparent households, on average, are more likely to use alcohol and drugs, to drop out of high school, to leave home at a young age, and to have early experiences with sexual activity (e.g., Carlson & Corcoran, 2001; Goldscheider & Goldscheider, 1998; McLanahan & Sandefur, 1994). Additionally, growing up in a stepfamily is associated with more emotional problems (Zill, Morrison, & Coiro, 1993), lower levels of academic achievement, and higher levels of school-related behavior problems during adolescence (Tillman, 2007, 2008).

Children’s living arrangements also tend to be associated with engagement in their own romantic relationships during adolescence and early adulthood (Amato, 1996; Wolfinger, 2005). Although most individuals begin to engage in romantic relationships during their teen years, adolescents from stepfamilies and single-parent families are significantly more likely to do so than are those from two–biological parent households (Cherlin, Kiernan, & Chase-Lansdale, 1995; Sassler, Cunningham, & Lichter, 2009). Although involvement in romantic relationships is not necessarily concerning in itself, more research is needed to examine how relationship experiences unfold. It may be that childhood family experiences set the stage for relationship development at an early age, and those early romantic relationships, in turn, shape long-term relationship trajectories.

Additional research, for example, has suggested that being raised in a single-parent family is associated with higher levels of early cohabitation (Cherlin et al., 1995; Sassler et al., 2009), and being raised in a stepfamily is associated with an increased likelihood of early marriage (Wolfinger, 2005). Unfortunately, research has not yet adequately addressed the question of whether childhood family structure is associated with the quality of these early romantic relationships. Furthermore, we do not know whether experience with adolescent romantic relationships mediates the association between childhood family structure and relationship involvement and/or quality during adulthood.

The long-term effects of early relationship formation are of particular concern to many researchers, social workers, and family-related practitioners. Individuals who engage in relationships during adolescence are learning how to form and maintain romantic unions. As such, they may be learning important, beneficial relationship skills that will help them throughout adulthood. Yet they also are more likely than their nondating peers to be exposed to the potentially negative aspects of romantic relationships, such as intimate-partner conflict and violence (Halpern, Oslak, Young, Martin, & Kupper, 2001), which have been linked to relationship trajectories that include adult unions marked by dangerous and violent interpersonal dynamics (Roscoe & Benaske, 1985; Smith, White, & Holland, 2003). This study uses rich, longitudinal data from the National Longitudinal Study of Adolescent Health (Add Health) to further explicate these issues.

Childhood Family Structure and Adolescent Relationships

This study draws on a life course perspective, which holds that events which occur during one stage of life are partly shaped by the events which occurred in the preceding stages. Thus, development is cumulative and a lifelong process (Elder, 1998). According to this perspective, family structure experiences can be viewed as a trajectory of experiences that happen over a lifetime, from birth through adulthood. Family structure experiences during childhood are also expected to help shape the trajectories that individuals travel in the other spheres of their lives.

As the diversity of family forms has increased, a growing number of children are spending significant portions of their lives in single-parent, stepparent, and cohabiting-parent families, and children are increasingly transitioning into and out of different living arrangements. Research suggests that there is an intergenerational link between parents’ relationship histories (and the resulting family structure experiences faced by children) and children’s own union formation behaviors (Amato, 1996; Wolfinger, 2005). For example, those who have grown up in “nontraditional” families face a greater likelihood of engaging in early romantic unions of their own (Cherlin et al., 1995; Sassler et al., 2009). Living in a nontraditional family structure, particularly residing in a single-mother family for boys, is also a risk factor for experiencing violence during adolescent relationships (Cavanagh, Crissey, & Raley, 2008; Halpern et al., 2001). Although the findings of these studies suggest that the association between family structure and violence differs by gender, the issue of relationship conflict clearly warrants greater attention.

We focus here on two mechanisms through which childhood family structure is thought to contribute to the later relationship trajectories of young people: parent-based social support and parental control and monitoring. Children and adolescents receive social support from people within their surrounding environments, particularly their parents and families. Adolescents’ perceptions of parental and familial support are strong indicators of psychosocial development (Blyth & Traeger, 1988; Greenberg, Siegel, & Leitch, 1983), and tend to vary by family structure. In particular, children of divorced (Amato, 2005), remarried, and cohabiting families (Goldscheider & Goldscheider, 1998) report lower levels of support from parents and other close family members. Adolescents who lack familial support often look to friends and romantic partners (Aquilino, 1991; Goldscheider & Goldscheider, 1998) to provide them with emotional support and love. Thus, adolescents from nontraditional families may be more likely than their peers to engage in romantic relationships and to expect those relationships to provide them with high levels of emotional support. These expectations may also lead adolescents from nontraditional families to become more dissatisfied with their romantic partners. At the same time, their need for support and love may increase the likelihood that these adolescents will remain involved in relationships of poor quality, even if they are marked by conflict or violence.

Parental control and monitoring may also contribute to the link between childhood family structure and adolescent romantic relationships. Compared with two-parent families, single-parent families are less able to monitor their children because there are fewer adults in the household (Thomson, McLanahan, & Curtin, 1992) and the adults generally have to balance multiple tasks by themselves (i.e., going to work, raising children, etc.). Stepparent and cohabiting families may also provide less monitoring/control than two–biological parent families because relationships between nonbiological parent-figures and children often lack clear roles and norms that would establish the parent-figures’ legitimacy and responsibility (Cherlin & Furstenberg, 1994; Thomson et al., 1992). Moreover, biological parents within stepfamilies tend to provide somewhat less supervision than their counterparts in two–biological parent families, as their attention may be split between their children and a relatively new romantic partner. Lower levels of parental supervision and control may allow adolescents in single-parent and stepparent households more opportunities to engage in romantic relationships and sexual activity. Involvement in relationships at earlier ages may also lead to an increased number of partners during adolescence, which is associated with an increased likelihood of experiencing physical aggression (Bergman, 1992). Additionally, insufficient monitoring/control has been shown to predict antisocial behavior, which in turn, increases the risk of aggression toward a romantic partner (Conger, Cui, Bryant, & Elder, 2000).

Childhood Family Structure and Adult Relationships

Some evidence also suggests that childhood family structure may have long-term ramifications for union formation patterns in adulthood. For example, compared with individuals who grow up in two–biological parent households, those who experience parental divorce are more likely to enter into cohabiting unions (Sassler & Goldscheider, 2004), to cohabit prior to marriage (Teachman, 2004), and to cohabit at younger ages (Thornton, 1991). Individuals who grow up in a stable single-mother household (Ryan, Franzetta, Schelar, & Manlove, 2009) and other nontraditional families (Landale, Schoen, & Daniels, 2010) are also at a higher risk of cohabiting in early adulthood. In addition, individuals who grow up in nontraditional families, especially women, tend to move in with a romantic partner more rapidly (Sassler, Addo, & Hartmann, 2010; Teachman, 2003) and to enter into marital relations earlier in life than their peers (Axinn & Thornton, 1996; Ryan et al., 2009; Sassler et al., 2009).

Childhood family structure may also be related to the quality of adult relationships, as individuals from divorced families are more likely than their peers to experience romantic relationships of their own that are unstable and distressed (Amato & Booth, 1997; McLeod, 1991; Ross & Mirowsky, 1999). This association may be explained by the higher than average levels of parental conflict experienced by young people within families that have gone through a divorce. Some studies indicate that children exposed to parental conflict are more likely to experience lower quality relationships in adulthood (Amato & Booth, 2001), to form families early, and to experience union dissolution of their own (Musick & Meier, 2010).

Overall, however, there is little research that directly examines the connections between childhood family structure, patterns of relationship formation, and the quality of adult relationships. Moreover, although numerous studies have focused on the experience of relationship conflict (Brownridge, 2008; Brownridge & Halli, 2002; Yllo & Straus, 1981), no studies to our knowledge have examined childhood family structure as a primary risk factor underlying the level of conflict found within adult unions.

The Mediating Role of Adolescent Relationships

Despite the common tendency to consider adolescent romance as an inconsequential experience in the life course, recent studies indicate that adolescent romantic relationships can have more complicated and enduring implications than anticipated (Carver, Joyner, & Udry, 2003). Of particular interest to this study, findings suggest that adolescent romantic relationships are likely to be an important factor in the transition into adult romantic relationships (Furman & Shaffer, 2003). Individuals who dated during adolescence are not only likely to have more partners during young adulthood but are also more likely to cohabit or marry early compared with their peers who did not experience adolescent relationships (Meier & Allen, 2009; Seiffge-Krenke, 2003). It may be, therefore, that part of the reason why childhood family structure has long-term consequences for adult union formation processes is that family structure influences the start of individuals’ relationship trajectories during the adolescent years.

Although the literature does not clearly reveal an association between adolescent relationship quality and the quality of relationships in adulthood, one might speculate that such an association exists. If a young person’s earliest interpersonal relationships are marked by arguments or violence, he or she may not develop the communication skills necessary to maintain stable, high-quality relationships. On the other hand, adolescents who experienced stable relationships may transfer their successful relationship practices, such as open communication, patience, and healthy emotional coping methods, into their relationships during adulthood. If this is the case, childhood family structure may have long-term consequences for adult relationship quality because of its earlier influence on adolescent relationship quality.

Hypotheses

Given the literature reviewed above, we make the following four hypotheses. Compared with their peers who were raised in a two–biological parent family, individuals from “nontraditional” families (i.e., stepparent, single-parent, and nonparent families) are (Hypothesis 1) more likely to engage in romantic relationships. In particular, they are more likely to date during adolescence, to have ever cohabited by adulthood, and to have ever married by adulthood. They are also (Hypothesis 2) more likely to have adolescent and adult relationships marked by conflict. We also expect that (Hypothesis 3) the association between family structure and adolescent relationship experiences is mediated by family support and parental supervision/control. Last, because the life-course perspective posits that experiences during adolescence should help to shape trajectories in subsequent life stages, we hypothesize that (Hypothesis 4a) there is a direct association between adolescent romantic relationship outcomes and adult relationship outcomes and that (Hypothesis 4b) the associations between childhood family structure and adult relationship outcomes are at least partially mediated by adolescent romantic relationship experiences.

Data and Method

Data for this research come from the Add Health, a nationally representative study of adolescents in Grades 7 through 12 in the United States in 1995. Add Health involves multiple components and several waves of data collection. This research uses contractual data collected from in-home interviews during Wave I (1994-1995), Wave II (1996), and Wave IV (2008), and selected data from the Wave I Parental Questionnaire and School Administrator Questionnaire. This study included respondents who completed Waves I, II, and IV of the in-home interview, whose parents completed the Parental Questionnaire at Wave I, and who had a valid sampling weight (N = 9,196). In all, 567 respondents were dropped from the analytic sample due to a lack of valid information for all the variables included in the multivariate analyses (N = 8,629). We also excluded respondents reporting same-sex partnerships (N = 236) and those who were older than 20 years at Wave II (N = 30). This yielded a final sample size of 8,363 respondents.

Dependent Variables

Adolescent romantic relationship

The first dependent variable is a dichotomous indicator of whether the respondent had been in a recent heterosexual romantic relationship at the time of Wave II. This measure was constructed from a question that directly asked respondents about their involvement in romantic relationships within the past 18 months, and from responses to additional questions regarding behaviors that are indicative of a romantic relationship, including hand holding, kissing, and telling another person that they like or love them. Following the convention set by others using Add Health, if respondents responded yes to any of these questions, even if they did not consider their relationship to be “romantic,” they were coded as having been in a relationship (Carver et al., 2003).

Early adult union formation

The second and third dependent variables, also dichotomous indicators, measure whether the respondent had ever been in a heterosexual cohabiting relationship (by Wave IV) and whether they had ever been married (by Wave IV).

Relationship conflict

Adolescent conflict was measured at Wave II with a series of questions based on a short form of the Conflict Tactics Scale (Strauss, Gelles, & Steinmetz, 1980). Respondents were asked if, within their three most recent relationships, a partner had ever done the following: “call you names, insult you, or treat you disrespectfully in front of others,” “swear at you,” “threaten you with violence,” “push or shove you,” and “throw something at you.” Because few adolescents reported these behaviors, those who answered yes to any of the questions were coded as 1, all others as 0 (Halpern et al., 2001).

Similarly, at Wave IV, respondents were asked about their relationship experiences with one current partner. If the respondent reported multiple current partners, priority was given in the following order: to the marriage partner, cohabiting partner, pregnancy partner, dating partner. If the respondent reported no current partner, questions were asked about the most recent partner. Specifically, respondents were asked how often their partner had ever done any of the following: “threatened you with violence, pushed or shoved you, or thrown something at you that could hurt,” “slapped, hit, or kicked you,” and caused “an injury, such as a sprain, bruise, or cut because of a fight.” The respondents were also asked how often they had committed these acts against their partner. Respondents who answered that these behaviors were committed at least once (by either their partner or themselves) were coded as 1 and all others were coded as 0.

Independent Variables

Childhood family structure

The primary independent variable is childhood family structure, constructed from the household roster at Wave I. Family structure is measured with five dummy variables: two–biological parent (reference category), married stepparent, single-mother, single-father, and nonparent families (i.e., grandparents, other relatives, etc.).

Family-related mechanisms

Measures of family-related mechanisms that might help to explain an association between childhood family structure and adolescent relationship outcomes are taken from the in-home interview at Wave I. The mechanisms included here are family social support, parental control, and parental supervision. Family social support is a five-item index (α = .76) that measures emotional support received from parents and other close family members. Respondents were asked to report, on a scale of 1 (not at all) to 5 (very much), how much they felt that (a) their parents care about them, (b) people in their family understand them, (c) they want to leave home (reverse coded), (d) they and their family have fun together, and (e) their family pays attention to them. Parental control is a count variable ranging from 0 to 7, indicating whether the adolescent makes his or her own decisions about (a) time to be home on the weekends, (b) friends, (c) clothes, (d) how much TV to watch, (e) which TV shows to watch, (f) time to go to bed on week nights, and (g) what to eat. Parental supervision is a count variable ranging from 0 to 4, indicating whether at least one parent-figure is present in the home most or all the time when the adolescent (a) goes to school, (b) comes home from school, (c) eats the evening meal, and (d) goes to bed.

Control Variables

We control for a host of covariates that have been found to be associated with both childhood family structure and romantic relationships in previous literature, including age, gender, race/ethnicity, parent socioeconomic status (SES) (e.g., Cavanagh et al., 2008), and respondent SES (e.g., Guzzo, 2006). The demographic characteristics of age, gender, and race/ethnicity are taken from the Wave I in-home interview. Age is measured in years. Gender is measured such that 1 = male and 0 = female. Race/ethnicity is self-identified and is measured with four dummy indicators—non-Hispanic White (reference category), non-Hispanic Black, Hispanic (of any race), and Asian. Because of small sample sizes, individuals of other races were dropped from the analyses.

Measures of total family income and parental education are taken from the Wave I Parental Questionnaire. Five dummy variables were created to measure total family income—$15,000 or less (reference category), $16,000 to $34,000, $35,000 to $59,000, $60,000 or more, and missing income data. Highest educational attainment achieved by a parent (either the mother or father) is measured with five dummy indicators—less than a high school diploma (reference category), high school diploma or GED, some college, college degree or more, and missing education data.

Respondents’ own income and education are taken from the Wave IV in-home interview. Respondent income is measured with five dummy variables that capture total household income—less than $20,000 (reference category), $20,000 to $49,999, $50,000 to $74,999, $75,000 or more, and missing income data. Respondent education is measured with six dummy indicators that capture educational attainment to date—less than high school (reference category), high school diploma/GED, vocational schooling, some college, college degree, and postbaccalaureate.

Analytic Strategy

Binary logistic regression is used to estimate the effects of childhood family structure on the likelihood of ever being in three types of relationships: adolescent romantic, adult cohabiting, and adult marital relationships. In these models, we also use Sobel–Goodman tests to explore the potential mediating effects of family-related mechanisms on the association between childhood family structure and adolescent relationships, and whether engagement in an adolescent relationship helps to explain the effect of childhood family structure on engagement in adult unions. For each of the three types of relationships, logistic regression is also used to estimate the effects of childhood family structure on the likelihood of being in a conflict-ridden relationship, as opposed to a conflict-free relationship. Finally, we assess whether adolescent relationship experiences help to explain any enduring effect of childhood family structure on the experience of conflict in adult unions.

All analyses account for the multistage, stratified, school-based, cluster sampling design of Add Health by using the robust estimator of variance procedure in STATA. We also control for differential sampling probabilities among individuals by using the Add Health grand sample weights in all estimation procedures (Chantala & Tabor, 1999).

Descriptive Results

Table 1 presents the distribution of childhood family structure and other background characteristics. At Wave I, when the average respondent was 15.5 years old, the majority (58%) of the sample lived in a two–biological parent family. A substantial percentage lived in a single-parent family (20% with a mother and 3% with a father) or a stepparent family (16%), and an additional 3% lived in a nonparent family. The sample is evenly divided between men and women, and is predominantly non-Hispanic White. As indicated by parental education and income, the sample is largely from a middle-class background. The average respondent reported mid-to-high levels of parental support, supervision, and control. At Wave IV, when respondents were in their mid-20s to early 30s, the majority of the sample had at least some college education and earned a mid-to-high level of income.

Table 1.

Weighted Percentages, Means, and Standard Deviations of Sociodemographic and Relationship Variables (N = 8,363).

	Percentage	M	SD
Respondent characteristics at Wave I
Family structure
Two–biological parent family	58.20	—	—
Married stepparent family	16.30	—	—
Single-mother family	19.73	—	—
Single-father family	2.85	—	—
Nonparent family	2.92	—	—
Gender (male)	48.84	—	—
Age in years (at Wave I)		15.47	1.57
Race/ethnicity
White	69.32	—	—
Black	14.90	—	—
Hispanic	11.86	—	—
Asian	3.92	—	—
Parental characteristics at Wave I
Parents’ education
Less than high school	10.96	—	—
High school degree	30.78	—	—
Some college	20.52	—	—
College	34.21	—	—
Missing education data	3.53	—	—
Parents’ income ($)
≤15,000	13.45	—	—
16,000-34,000	20.78	—	—
35,000-59,000	25.43	—	—
60,000+	21.67	—	—
Missing income data	18.67	—	—
Respondent characteristics at Wave IV
Respondent education
Less than high school	8.33	—	—
High school diploma	16.10	—	—
Vocational school	9.53	—	—
Some college	33.57	—	—
College	20.55	—	—
Postbaccalaureate	11.92	—	—
Respondent income ($)
<20,000	10.97	—	—
20,000-49,999	31.26	—	—
50,000-74,999	23.15	—	—
75,000+	28.29	—	—
Missing income data	6.33	—	—
Mediating factors for adolescent analyses at Wave I
Family social support (range 0-5)	—	4.01	0.68
Parental supervision (range 0-4)	—	2.93	0.98
Parental control (range 0-7)	—	2.00	1.56
Outcomes for relationships
Ever in adolescent relationship	67.06	—	—
Ever cohabited	50.41	—	—
Ever married	49.26	—	—
Outcomes for conflict
Adolescent outcomes (N = 5,604)
Relationship marked by conflict	18.76	—	—
Relationship with no conflict	81.24	—	—
Cohabiting outcomes (N = 2,464)
Relationship marked by conflict	33.50	—	—
Relationship with no conflict	66.50	—	—
Marital outcomes (N = 2,593)
Relationship marked by conflict	22.65	—	—
Relationship with no conflict	77.35	—	—

Table 1 also indicates that the vast majority of respondents in the sample reported some experience with romantic relationships over the course of their adolescence and adult years. At Wave II, nearly 67% of the sample reported having been in a recent adolescent relationship. By Wave IV, 50% of the sample reported ever cohabiting and 49% reported having been married. Moreover, a substantial percentage of respondents reported serious conflict. At Wave II, about one fifth (19%) of those in adolescent relationships had experienced conflict at some point within their last three relationships. Among respondents who reported ever cohabiting or marrying by Wave IV, one third (33%) of those whose current or most recent relationship was cohabiting had experienced conflict in that relationship. Twenty-three percent of respondents whose current or most recent relationship was a marriage reported conflict within that marriage.

Multivariate Results

Adolescent Romantic Relationships

The baseline model (Model 1) presented in Table 2 shows that childhood family structure is significantly associated with the odds of engaging in a romantic relationship during adolescence. Compared with individuals who lived in two–biological parent families, those who lived in stepparent, single-mother, single-father, and nonparent families have 52%, 26%, 94%, and 109% higher odds, respectively, of engaging in an adolescent relationship. Controlling for sociodemographic characteristics does little to mediate these associations. As seen in Model 2, however, gender, race/ethnicity, and age have significant direct effects on adolescent romantic engagement, with boys, racial/ethnic minorities, and younger adolescents less likely to report romantic relationships.

Table 2.

Odds Ratios for the Effects of Childhood Family Structure on Engagement in Adolescent Relationship (N = 8,363).

	Adolescent Relationship at Wave II: Odds Ratio (SE)
	Model 1	Model 2	Model 3
Family structure at Wave I
(Two–biological parent family)
Married stepparent family	1.52*** (0.14)	1.55*** (0.15)	1.43*** (0.15)
Single-mother family	1.26** (0.11)	1.45*** (0.14)	1.32** (0.14)
Single-father family	1.94*** (0.39)	1.96** (0.43)	1.67* (0.39)
Nonparent family	2.09*** (0.43)	2.20*** (0.47)	2.11*** (0.06)
Male		0.73*** (0.05)	0.75*** (0.06)
Race/ethnicity
(White)
Black		0.70*** (0.06)	0.71*** (0.07)
Hispanic		0.78** (0.08)	0.79* (0.08)
Asian		0.45*** (0.06)	0.44*** (0.07)
Age of respondent (in years)		1.33*** (0.03)	1.26*** (0.03)
Parents’ education at Wave I
(Less than high school)
High school degree		1.02 (0.11)	0.99 (0.11)
Some college		1.03 (0.12)	1.01 (0.12)
College		0.95 (0.11)	0.93 (0.11)
Parents’ education missing		0.59** (0.11)	0.57** (0.11)
Parents’ income (in $) at Wave I
(≤15,000)
16,000-34,000		1.08 (0.14)	1.07 (0.14)
35,000-59,000		1.17 (0.14)	1.15 (0.14)
60,000+		1.36* (0.17)	1.32* (0.17)
Missing income data		1.14 (0.14)	1.11 (0.13)
Family support at Wave I			0.74*** (0.04)
Parental supervision at Wave I			0.93 (0.03)
Parental control at Wave I			0.94** (0.02)
Log likelihood	−10,336,536	−9,872,484	−9,781,992

p ≤ .05. **p ≤ .01. ***p ≤ .001.

Model 3 considers the mediating role of family-related mechanisms in the association between childhood family structure and adolescent romantic relationships. Results indicate that family-related social support and parental control are both significant independent predictors of adolescent romantic relationships. Specifically, for each unit increase in family support and parental control, the odds of being in a romantic relationship during adolescence decrease by 26% and 6%, respectively. With respect to mediating effects, the Sobel–Goodman test indicates that these family-related mechanisms were significant mediators (p < .05) and accounted for 44% of the total direct effect of family structure on engagement in an adolescent relationship. Yet, given that childhood family structure remains significantly associated with relationship formation, even in the face of controls for family support and parental control, additional factors that are not captured in this model will be important to consider in future research.

Table 3 shows findings regarding the relationship between childhood family structure and the experience of serious conflict among those who reported a recent adolescent relationship (N = 5,604). Contrary to our expectations, Model 1 indicates that, with one exception, adolescents in nontraditional families are no more likely to report serious relationship conflict than are their peers in two–biological parent homes. Those from single-mother families, however, face 25% higher odds of having a conflicted relationship as opposed to a conflict-free relationship.

Table 3.

Odds Ratios for the Effects of Childhood Family Structure on Serious Conflict in Adolescent Relationship (N = 5,604).

	Conflict in Adolescent Relationship at Wave II: Odds Ratio (SE)
	Model 1	Model 2	Model 3
Family structure at Wave I
(Two–biological parent family)
Married stepparent family	1.04 (0.13)	1.00 (0.13)	0.89 (0.12)
Single-mother family	1.25* (0.15)	1.05 (0.14)	0.90 (0.13)
Single-father family	1.27 (0.29)	1.14 (0.26)	0.93 (0.22)
Nonparent family	1.59 (0.43)	1.16 (0.32)	1.14 (0.31)
Male		1.00 (0.10)	1.04 (0.11)
Race/ethnicity
(White)
Black		1.40** (0.18)	1.42** (0.19)
Hispanic		0.93 (0.13)	0.95 (0.13)
Asian		1.27 (0.29)	1.19 (0.26)
Age of respondent (in years)		1.14*** (0.04)	1.09* (0.04)
Parents’ education at Wave I
(Less than high school)
High school degree		0.99 (0.14)	0.95 (0.14)
Some college		0.81 (0.13)	0.79 (0.13)
College		0.84 (0.14)	0.81 (0.13)
Parents’ education missing		1.05 (0.28)	0.99 (0.26)
Parents’ income (in $) at Wave I
(≤15,000)
16,000-34,000		0.84 (0.13)	0.81 (0.12)
35,000-59,000		0.86 (0.17)	0.83 (0.16)
60,000+		0.83 (0.15)	0.78 (0.14)
Missing income data		0.88 (0.15)	0.84 (0.15)
Family support at Wave I			0.74*** (0.05)
Parental supervision at Wave I			0.85*** (0.04)
Parental control at Wave I			0.97 (0.03)
Log likelihood	−5,305,390	−5,241,585	−5,174,879

p ≤ .05. **p ≤ .01. ***p ≤ .001.

The inclusion of sociodemographic characteristics in Model 2, particularly the measures of race/ethnicity and age, reduces to nonsignificance the association between living in a single-mother family and relationship conflict. This suggests that the increased risk of conflict associated with single-mother families results largely from the sociodemographic characteristics of the adolescents within those families.

Although no significant family structure differences in conflict remain, we include our measures of family-related mechanisms in Model 3 to examine their direct effects on adolescent relationship quality. We find that, among adolescents who have had a recent relationship, an increase in family support and parental supervision is associated with a significant decline in the likelihood of having experienced a conflicted relationship (by 26% and 15%, respectively).

Adult Unions

Table 4 shows the results of several nested binary logistic regression models predicting two outcomes at Wave IV: having been in a cohabiting relationship and in a marital union.

Table 4.

Odds Ratios for the Effects of Childhood Family Structure on Engagement in Cohabiting and Marital Relationships (N = 8,363).

	Cohabiting Relationship at Wave IV			Marital Relationship at Wave IV
	Model 1	Model 2	Model 3	Model 1	Model 2	Model 3
Family structure at Wave I
(Two–biological parent family)
Married stepparent family	1.80*** (0.14)	1.58*** (0.13)	1.53*** (0.12)	1.02 (0.08)	0.96 (0.08)	0.94 (0.08)
Single-mother family	1.90*** (0.16)	1.49*** (0.14)	1.44*** (0.13)	0.74*** (0.06)	0.79** (0.07)	0.78** (0.07)
Single-father family	1.99*** (0.33)	1.71*** (0.28)	1.63** (0.27)	0.71* (0.12)	0.62** (0.12)	0.60** (0.11)
Non-parent family	1.76** (0.35)	1.27 (0.25)	1.19 (0.23)	0.97 (0.20)	0.99 (0.21)	0.95 (0.20)
Male		1.07 (0.06)	1.11 (0.06)		0.54 (0.04)	0.55 (0.04)
Race/ethnicity
(White)
Black		1.22* (0.12)	1.26* (0.12)		0.35*** (0.04)	0.36*** (0.04)
Hispanic		0.70** (0.08)	0.71** (0.09)		0.76* (0.09)	0.77* (0.09)
Asian		0.94 (0.17)	1.01 (0.18)		0.54*** (0.09)	0.56*** (0.09)
Age of respondent (in years)		0.93*** (0.02)	0.91*** (0.02)		1.32*** (0.04)	1.30*** (0.04)
Parents’ education at Wave I
(Less than high school)
High school degree		1.00 (0.12)	1.00 (0.12)		0.94 (0.12)	0.94 (0.12)
Some college		0.94 (0.11)	0.93 (0.11)		1.02 (0.14)	1.01 (0.14)
College		0.94 (0.11)	0.93 (0.11)		0.86 (0.12)	0.86 (0.12)
Parents’ education missing		1.08 (0.20)	1.14 (0.22)		0.98 (0.17)	1.01 (0.18)
Parents’ income (in $) at Wave I
(≤15,000)
16,000-34,000		0.98 (0.11)	0.98 (0.11)		0.84 (0.09)	0.84 (0.09)
35,000-59,000		0.92 (0.11)	0.90 (0.11)		0.72** (0.09)	0.71** (0.09)
60,000+		1.11 (0.14)	1.08 (0.14)		0.60*** (0.08)	0.59*** (0.08)
Missing income data		0.88 (0.11)	0.87 (0.11)		0.90 (0.10)	0.89 (0.10)
Respondent education at Wave IV
(Less than high school)
High school diploma		0.69* (0.11)	0.70* (0.11)		0.88 (0.14)	0.89 (0.14)
Vocational school		0.74* (0.11)	0.75* (0.11)		1.14 (0.18)	1.15 (0.18)
Some college		0.56*** (0.08)	0.57*** (0.08)		0.86 (0.12)	0.88 (0.13)
College		0.37*** (0.06)	0.38*** (0.06)		0.55*** (0.09)	0.57*** (0.09)
Postbaccalaureate		0.31*** (0.05)	0.32*** (0.06)		0.67* (0.11)	0.70* (0.12)
Respondent income (in $) at Wave IV
(≤$20,000)
20,000-49,999		0.71*** (0.09)	0.71** (0.09)		1.59*** (0.18)	1.59*** (0.18)
50,000-74,999		0.49*** (0.06)	0.48*** (0.06)		3.14*** (0.38)	3.12*** (0.39)
75,000+		0.53*** (0.06)	0.51*** (0.06)		3.24*** (0.40)	3.17*** (0.40)
Missing income data		0.75 (0.12)	0.77 (0.13)		0.69 (0.13)	0.70 (0.13)
Adolescent relationship at Wave II			1.55*** (0.10)			1.29*** (0.08)
Log likelihood	−11,195,149	−10,727,754	−10,654,815	−11,350,560	−10,179,683	−10,156,140

p ≤ .05. **p ≤ .01. ***p ≤ .001.

Cohabiting Unions

Model 1 of the analysis on cohabiting unions indicates that, compared with adolescents who lived in two–biological parent families, those from stepparent, single-mother, single-father, and nonparent families face odds of cohabiting that are 80%, 90%, 99%, and 76% higher, respectively. Except for those from nonparent families, the association between family structure and cohabitation remains significant even after controlling for sociodemographic characteristics in Model 2. In general, though, Blacks are more likely and Hispanics are less likely than their non-Hispanic White counterparts to report cohabitation experiences. Additionally, we see that older, more educated, and higher income respondents are less likely to have ever cohabited.

The results shown in Model 3 reveal that involvement in an adolescent relationship is a significant independent predictor of adult cohabitation. Compared with those who had not engaged in a romantic relationship by Wave II, young people who had done so face 55% higher odds of cohabiting during adulthood. Contrary to our hypothesis, however, experience with adolescent romantic relationships helps to account for only a small portion of the increased propensity to cohabit found among those from nontraditional families. Although statistically significant (p < .05), a Sobel–Goodman test for mediation indicates that being in an adolescent relationship accounts for only 4% of the total family structure effect on involvement in cohabiting unions.

Marital Unions

The baseline model (Model 1) for the analysis of marital unions indicates that individuals who lived in single-mother and single-father families during their childhood have significantly lower odds of marrying (26% and 29% lower odds, respectively) than do those who lived in two–biological parent families. These associations remain with the inclusion of the sociodemographic characteristics in Model 2. In terms of direct effects, this model indicates that respondents who are male, racial/ethnic minorities, younger, college-educated (as compared with having less than a high school diploma), and those who have parents with higher levels of income are less likely to have been married than are their peers. High levels of personal income, however, are associated with higher odds of ever marrying.

As with cohabitation, having experienced an adolescent romantic relationship is significantly related to the likelihood of marriage (see Model 3). Those who were in a romantic relationship during their adolescence have 29% higher odds of marrying than do their counterparts who did not date during the teen years. Despite the direct relationship between adolescent relationship and the odds of marrying, a Sobel–Goodman test indicates that adolescent relationship experiences do not significantly mediate the remaining association between childhood family structure and marriage.

Table 5 presents the associations between childhood family structure and the experience of serious conflict among respondents within current or recent adult cohabiting (N = 2,464) and marital unions (N = 2,593). In this table, we present two models: the unadjusted model which includes only childhood family structure and an adjusted model which includes measures of sociodemographic characteristics and adolescent relationship experiences.

Table 5.

Odds Ratios for the Effects of Childhood Family Structure on Serious Conflict in Cohabiting and Marital Relationships.

	Cohabiting Relationships at Wave IV (N = 2,464)		Marital Relationships at Wave IV (N = 2,593)
	Model 1	Model 2	Model 1	Model 2
Family structure at Wave I
(Two–biological parent family)
Married stepparent family	1.36* (0.21)	1.23 (0.21)	1.06 (0.19)	1.00 (0.18)
Single-mother family	1.27 (0.18)	0.97 (0.15)	1.50* (0.28)	1.25 (0.26)
Single-father family	1.05 (0.38)	0.85 (0.33)	1.04 (0.41)	0.93 (0.37)
Nonparent family	1.88* (0.53)	1.37 (0.40)	1.17 (0.42)	0.97 (0.41)
Adolescent relationship at Wave II
(Adolescent relationship with no conflict)
Adolescent relationship with conflict		1.04 (0.19)		1.48* (0.24)
No adolescent relationship		0.74* (0.11)		0.84 (0.24)
Log likelihood	−3,260,213	−3,176,462	−2,607,035	−2,470,066

p ≤ .05. **p ≤ .01. ***p ≤ .001.

Conflict in Cohabiting Unions

The unadjusted model shows that respondents who grew up in stepparent and nonparent families have higher odds of experiencing a conflicted cohabiting union (36% and 88%, respectively) than do those who grew up with two biological parents. These associations, however, are reduced to nonsignificance when we account for parent SES (results not shown in the table). Thus, there is little evidence to suggest that young people who grow up in “nontraditional” families are at greater risk for problematic or dangerous cohabiting unions in adulthood. In addition, there appear to be no significant differences in this outcome between adolescent daters who experienced conflict and adolescent daters who did not. Not dating during the teen years, however, appears to have a marginally significant protective effect on the likelihood of experiencing cohabitation conflict. The only other variables examined here that significantly predict conflict in cohabiting relationships are respondent race/ethnicity and education (not shown in the table). Specifically, Blacks have 55% higher odds of experiencing conflict in cohabiting unions than their White peers. Compared with respondents with less than a high school diploma, those with a college degree and a postbaccalaureate degree face 47% and 67% lower odds, respectively, of experiencing serious conflict in a cohabiting relationship (results available on request).

Conflict in Marital Unions

Adults from “nontraditional” families are also no more likely to experience marital conflict than their peers of similar racial/ethnic and SES characteristics. Individuals who lived in single-mother families have 50% higher odds of experiencing conflict in a current or recent marriage than their peers from two–biological parent families (see Model 1). However, this association is reduced to nonsignificance when respondent sociodemographic characteristics are taken into account (results not shown, available on request). In particular, respondents who are male and Black or Hispanic face higher odds of experiencing marital conflict than do women and non-Hispanic White respondents. Higher levels of personal income and educational attainment are associated with lower odds of experiencing marital conflict. Interestingly, we see in Model 2 that the experience of serious conflict in an adolescent relationship is a significant predictor of conflict in a marital union. Compared with their sociodemographically similar peers, individuals who experienced a conflicted adolescent relationship face 48% higher odds of engaging in a conflicted marital union.

Discussion

This research makes several important contributions to our knowledge about relationship formation during the transition to adulthood. First, this research helps to further explicate the mechanisms linking childhood family structure to relationship formation during adolescence. Although the majority of respondents from all types of family structures have engaged in an adolescent romantic relationship, adolescents from “nontraditional” families (i.e., stepparent, single-parent, and nonparent families) have a significantly greater likelihood of having done so than adolescents from two–biological parent families. Our findings indicate that lower levels of family support and parental control help to explain a significant portion (44%) of this increased likelihood. Low levels of support and control may lead these adolescents to search for external sources of social support in the form of boyfriends/girlfriends and may allow them more freedom to make their own decisions regarding where and with whom they spend time.

In addition to a higher likelihood of engaging in adolescent relationships, young people raised in stepparent and single-parent families face a significantly greater likelihood of cohabiting as adults than do their counterparts from two–biological parent families. At the same time, those raised in single-parent families are less likely to have married by their mid-20s to early 30s. In contrast to our expectations, we find that controlling for involvement in an adolescent relationship does not have a strong mediating effect on the association between childhood family structure and adult union formation. In fact, doing so only explains 4% of the total family structure effect on the likelihood of cohabiting and explains none of the total family structure effect on the likelihood of marriage. Thus, it appears that there is something about childhood family structure experiences, above and beyond their influence over people’s initial forays into romantic relationships, which continues to affect their likelihood of entering into coresidential unions during adulthood. To further explicate the mechanisms at work here, future research should explore in greater detail the family and peer contexts in which adolescents from differing family structures are raised.

A second important contribution of this study is its focus on serious relationship conflict, including behaviors that are verbally and physically abusive. Overall, we find that a relatively high percentage of adolescents (19% of those with recent relationships) and adults (33% of those in current/recent cohabitations and 23% of those in current/recent marriages) have experienced serious relationship conflict. Our findings suggest that, in general, parental supervision and perceptions of family support are important factors influencing adolescents’ likelihood of engaging in dangerous relationships. As found in at least one previous study (Cavanagh et al., 2008), however, childhood family structure is not a significant independent predictor of conflict within adolescent relationships once sociodemographic characteristics are held constant.

Contrary to our hypotheses, we also find few long-term effects of childhood family structure on the likelihood of experiencing conflict-ridden relationships during adulthood. Although childhood family structure may influence the likelihood with which young people engage in coresidential adult relationships, it does not seem to influence the likelihood that those unions, when entered into, will be marked by overt conflict or violence. What does appear to influence the likelihood that individuals’ marital unions will include conflict is their experience with adolescent relationships, particularly if those earlier relationships included serious conflict.

In sum, these findings are generally supportive of predictions made using a life-course perspective and confirm the results of earlier studies that show that both childhood family structure and experiences in early-life relationships have enduring long-term effects on relationship trajectories during adulthood (Roscoe & Benaske, 1985). Yet the specific mechanisms linking childhood experiences to adult outcomes still have not been fully explicated. We expected that the association between childhood family structure and adult relationship outcomes would be accounted for by intervening adolescent relationship experiences. Yet intervening adolescent relationships only explain a small portion of the association between childhood family structure and engagement in adult cohabitation and do not appear to be important mechanisms underlying the associations between family structure and marriage or family structure and adult relationship quality. Thus, the focus of future work should be to examine the other important ways in which childhood family structure influences the experiences of young people, leading to distinct relationship and union formation trajectories.

Limitations

Although these findings offer important contributions to the existing literature, there are several limitations that should be acknowledged. First, we use a cross-sectional measure of childhood family structure. As such, we have not captured any family structure transitions that respondents may have experienced prior to Wave I. The literature suggests that both the number and type of family structure transitions that children experience may have lasting effects on their future outcomes (e.g., Cavanagh & Hutson, 2006; Fomby & Cherlin, 2007; Tillman, 2007). Future research on the predictors of relationship formation and quality would benefit from a more detailed examination of the family structure histories that individuals experience throughout the course of their lifetime. Another limitation is the overly conservative nature of our measure of relationship conflict in adult unions. Because we are only able to measure conflict within one current or most recent cohabiting/marital union, we cannot capture serious relationship conflict that occurred in previous relationships or in a current relationship about which no questions were asked.

Finally, the adolescent interviews in the Add Health lack measures to examine the mediating role of parental socialization and respondents’ experience with family-related violence during childhood (either in the form of child abuse or exposure to parental spousal/partner abuse). Research indicates that children observe and acquire relationship skills, norms, and expectations from their parents, and that these may vary by the parent(s)’ marital/relationship status (Amato & DeBoer, 2001). Moreover, observing parental partner conflict or experiencing child abuse has also been associated with poor relationship quality in later relationships (Amato & Booth, 2001). The long-term association between adolescent relationship conflict and later marital conflict may be explained, therefore, by the greater likelihood of experiencing family-related violence during childhood. Future data collection efforts should aim to include this kind of information so that research can better examine the family-related factors that influence the development of conflicted and/or violent relationship trajectories.

Conclusion

In sum, the findings of this study suggest that we must continue to focus our efforts on understanding the experiences that affect relationship outcomes among adolescents and young adults. As with many other types of developmental trajectories, it is important to examine the influence of early life experiences, such as family structure, on social outcomes in adulthood. Our study provides evidence that union formation behavior is influenced by the family context in which one grew up. Individuals from nontraditional families are more likely to engage in adolescent relationships and to cohabit as adults, and are less likely to engage in marital unions than their peers from two–biological parent families. Contrary to expectations, childhood family structure does not predict serious conflict in adolescent or adult cohabiting or marital relationships.

Thus, if young people from nontraditional families are at an increased risk of engaging in problematic behaviors, as some previous research has suggested (McLanahan & Sandefur, 1994; Tillman, 2007, 2008), these behaviors do not appear to be manifesting most frequently within romantic relationships in the form of overt physical and verbal conflict. Although this finding is reassuring, we recognize that further research is needed to examine other potential relationship problems that may be linked to the experience of nontraditional family structures. In addition to childhood family structure, contextual factors, such as school environment, peer networks, and neighborhood characteristics should also be included in future studies that examine romantic relationship trajectories and the likelihood of engaging in a “problematic” or dangerous relationship during adolescence and adulthood.

Footnotes

Acknowledgements

Special acknowledgment is due to Ronald R. Rindfuss and Barbara Entwisle for assistance in the original design.

Authors’ Note

This research uses data from Add Health, a program project directed by Kathleen Mullan Harris and designed by J. Richard Udry, Peter S. Bearman, and Kathleen Mullan Harris at the University of North Carolina at Chapel Hill, and funded by Grant P01-HD31921 from the Eunice Kennedy Shriver National Institute of Child Health and Human Development, with cooperative funding from 23 other federal agencies and foundations. No direct support was received from Grant P01-HD31921 for this analysis. This article benefited from the helpful comments of Karin L. Brewster and Isaac W. Eberstein. Information on how to obtain the Add Health data files is available on the Add Health website ().

Declaration of Conflicting Interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The author(s) received no financial support for the research, authorship, and/or publication of this article.

References

Amato

P. R.

(1996). Explaining the intergenerational transmission of divorce. Journal of Marriage and the Family, 58, 628-640.

Amato

P. R.

(2005). The impact of family formation change on the cognitive, social, and emotional well-being of the next generation. Future of Children, 15, 75-96.

Amato

P. R.

Booth

(1997). A generation at risk: Growing up in an era of family upheaval. Cambridge MA: Harvard University Press.

Amato

P. R.

Booth

(2001). The legacy of parents’ marital discord: Consequences for children’s marital quality. Journal of Personality and Social Psychology, 81, 627-638.

Amato

P. R.

DeBoer

D. D.

(2001). The transmission of marital instability across generations: Relationship skills or commitment to marriage? Journal of Marriage and Family, 63, 1038-1051.

Aquilino

W. S.

(1991). Family structure and home-leaving: A further specification of the relationship. Journal of Marriage and the Family, 53, 999-1010.

Axinn

W. G.

Thornton

(1996). The influence of parents’ marital dissolutions on children’s attitudes toward family formation. Demography, 33, 66-81.

Bergman

(1992). Dating violence among high school students. Sociology of Work, 37, 21-27.

Blyth

D. A.

Traeger

(1988). Adolescent self-esteem and perceived relationships with parents and peers. In Salzinger

Antrobus

Hammer

(Eds.), Social networks of children, adolescents, and college students (pp. 23-56). Hillsdale, NJ: Lawrence Erlbaum.

10.

Brownridge

D. A.

(2008). The elevated risk for violence against cohabiting women: A comparison of three nationally representative surveys of Canada. Violence Against Women, 14, 809-832.

11.

Brownridge

D. A.

Halli

S. S.

(2002). Understanding male partner violence against cohabiting and married women: An empirical investigation with a synthesized model. Journal of Family Violence, 17, 341-361.

12.

Bumpass

L. L.

(2000). Trends in cohabitation and implications for children’s family contexts in the United States. Population Studies, 54, 29-41.

13.

Carlson

M. J.

Corcoran

M. E.

(2001). Family structure and children’s behavioral outcomes. Journal of Marriage and Family, 63, 779-792.

14.

Carver

Joyner

Udry

J. R.

(2003). National estimates of adolescent romantic relationships. In Florsheim

(Ed.), Adolescent romantic relations and sexual behavior: Theory, research, and practical implications (pp. 23-56). Mahwah, NJ: Lawrence Erlbaum.

15.

Cavanagh

S. E.

Crissey

S. R.

Raley

R. K.

(2008). Family structure history and adolescent romance. Journal of Marriage and Family, 70, 698-714.

16.

Cavanagh

S. E.

Huston

A. C.

(2006). Family instability and children’s early problem behaviors. Social Forces, 85, 551-581.

17.

Chantala

Tabor

J. W.

(1999, June). Strategies to perform a design-based analysis using the Add Health data (National Longitudinal Study of Adolescent Health). Chapel Hill: Carolina Population Center, University of North Carolina at Chapel Hill. Retrieved from http://www.cpc.unc.edu/projects/addhealth/data/guides/weight1.pdf

18.

Cherlin

A. J.

Furstenberg

F. F.

(1994). Stepfamilies in the United States: A reconsideration. Annual Review of Sociology, 20, 359-381.

19.

Cherlin

A. J.

Kiernan

Chase-Lansdale

P. L.

(1995). Parental divorce in childhood and demographic outcomes in young adulthood. Demography, 32, 299-318.

20.

Conger

R. D.

Cui

Bryant

C. M.

Elder

G. H.

(2000). Competence in early adult romantic relationships: A developmental perspective on family influences. Journal of Personality and Social Psychology, 79, 224-237.

21.

Elder

G. H.

Jr. (1998). The life course as developmental theory. Child Development, 69, 1-12.

22.

Fomby

Cherlin

A. J.

(2007). Family instability and child well-being. American Sociological Review, 72, 181-204.

23.

Furman

Shaffer

(2003). The role of romantic relationships in adolescent development. In Florsheim

(Ed.), Adolescent romance, relations and sexual behavior: Theory, research, and practical implications (pp. 3-22). Mahwah, NJ: Lawrence Erlbaum.

24.

Goldscheider

F. K.

Goldscheider

(1998). The effect of childhood family structure on leaving and returning home. Journal of Marriage and the Family, 60, 745-756.

25.

Greenberg

M. T.

Siegel

J. M.

Leitch

C. J.

(1983). The nature and importance of attachment relationships to parents and peers during adolescence. Journal of Youth and Adolescence, 12, 373-386.

26.

Guzzo

K. B.

(2006). The relationship between life course events and union formation. Social Science Research, 35, 384-408.

27.

Halpern

C. T.

Oslak

S. G.

Young

M. L.

Martin

S. L.

Kupper

L. L.

(2001). Partner violence among adolescents in opposite-sex romantic relationships: Findings from the National Longitudinal Study of Adolescent Health. American Journal of Public Health, 19, 1679-1685.

28.

Landale

N. S.

Schoen

Daniels

(2010). Early family formation among White, Black, and Mexican American women. Journal of Family Issues, 31, 445-474.

29.

McLanahan

S. S.

Sandefur

(1994). Growing up with a single-parent: What hurts, what helps. Cambridge, MA: Harvard University Press.

30.

McLeod

J. D.

(1991). Childhood parental loss and adult depression. Journal of Health and Social Behavior, 32, 205-220.

31.

Meier

Allen

(2009). Romantic relationships from adolescence to young adulthood: Evidence from the National Longitudinal Study of Adolescent Health. Sociological Quarterly, 50, 308-335.

32.

Moore

K. A.

Bronte-Tinkew

Jekielek

Guzman

Ryan

Redd

Matthews

(2007). Developing measures of healthy marriages and relationships. In Hofferth

S. L.

Casper

L. M.

(Eds.), Handbook of measurement issues in family research (pp. 101-121). Mahwah, NJ: Lawrence Erlbaum.

33.

Musick

Meier

(2010). Are both parents always better than one? Parental conflict and young adult well-being. Social Science Research, 39, 814-830.

34.

Roscoe

Benaske

(1985). Courtship violence experienced by abused wives: Similarities in patterns of abuse. Family Relations, 43, 419-424.

35.

Ross

C. E.

Mirowsky

(1999). Parental divorce, life course disruption, and adult depression. Journal of Marriage and the Family, 61, 1034-1045.

36.

Ryan

Franzetta

Schelar

Manlove

(2009). Family structure history: Links to relationship formation behaviors in young adulthood. Journal of Marriage and Family, 71, 935-953.

37.

Sassler

Addo

Hartmann

(2010). The tempo of relationship progression among low-income couples. Social Science Research, 39, 831-844.

38.

Sassler

Cunningham

Lichter

D. T.

(2009). Intergenerational patterns of union formation and relationship quality. Journal of Family Issues, 30, 757-786.

39.

Sassler

Goldscheider

F. K.

(2004). Revisiting Jane Austen’s theory of marriage timing: Union formation among American men in the late 20th century. Journal of Family Issues, 25, 139-166.

40.

Seiffge-Krenke

(2003). Testing theories of romantic development from adolescence to young adulthood: Evidence of developmental sequence. International Journal of Behavioral Development, 27, 519-531.

41.

Smith

P. H.

White

J. W.

Holland

L. J.

(2003). A longitudinal perspective on dating violence among adolescent and college-age women. American Journal of Public Health, 93, 1104-1109.

42.

Strauss

Gelles

Steinmetz

(1980). Behind closed doors: Violence in the family. Garden City, NY: Anchor/Doubleday.

43.

Teachman

J. D.

(2002). Childhood living arrangements and the intergenerational transmission of divorce. Journal of Marriage and Family, 64, 717-729.

44.

Teachman

J. D.

(2003). Premarital sex, premarital cohabitation, and the risk of subsequent marital dissolution among women. Journal of Marriage and Family, 65, 444-455.

45.

Teachman

J. D.

(2004). The childhood living arrangements of children and the characteristics of their marriages. Journal of Family Issues, 25, 86-111.

46.

Thomson

McLanahan

S. S.

Curtin

R. B.

(1992). Family structure, gender, and parental socialization. Journal of Marriage and the Family, 54, 368-378.

47.

Thornton

(1991). Influence of marital history of parents on the marital and cohabitational experiences of children. American Journal of Sociology, 96, 868-894.

48.

Tillman

K. H.

(2007). Family structure pathways and academic disadvantage among adolescents in stepfamilies. Sociological Inquiry, 77, 383-424.

49.

Tillman

K. H.

(2008). Non-traditional siblings and the academic outcomes of adolescents. Social Science Research, 37, 88-108.

50.

U.S. Census Bureau. (2009). Living arrangements of children under 18 years old: 1960 to present. Retrieved from http://www.census.gov/population/www/socdemo/hh-fam.html

51.

Wolfinger

N. H.

(2005). Understanding the divorce cycle: The children of divorce in their own marriages. New York, NY: Cambridge University Press.

52.

Yllo

Straus

M. A.

(1981). Interpersonal violence among married and cohabiting couples. Family Relations, 30, 339-347.

53.

Zill

Morrison

Coiro

(1993). Long-term effects of parental divorce on parent-child relationships, adjustment, and achievement in young adulthood. Journal of Family Psychology, 7, 91-103.