Journal of Family Issues - eCommons@Cornell - Cornell University

2 downloads 0 Views 178KB Size Report
May 12, 2009 - those who spent their childhoods living with biological parents in married- ... marriages initiated in the 1990s are expected to end in divorce ...
Journal of Family Issues http://jfi.sagepub.com

Intergenerational Patterns of Union Formation and Relationship Quality Sharon Sassler, Anna Cunningham and Daniel T. Lichter Journal of Family Issues 2009; 30; 757 originally published online Feb 12, 2009; DOI: 10.1177/0192513X09331580 The online version of this article can be found at: http://jfi.sagepub.com/cgi/content/abstract/30/6/757

Published by: http://www.sagepublications.com

Additional services and information for Journal of Family Issues can be found at: Email Alerts: http://jfi.sagepub.com/cgi/alerts Subscriptions: http://jfi.sagepub.com/subscriptions Reprints: http://www.sagepub.com/journalsReprints.nav Permissions: http://www.sagepub.com/journalsPermissions.nav Citations http://jfi.sagepub.com/cgi/content/refs/30/6/757

Downloaded from http://jfi.sagepub.com at CORNELL UNIV on May 12, 2009

Intergenerational Patterns of Union Formation and Relationship Quality

Journal of Family Issues Volume 30 Number 6 june 2009 757-786 © 2009 Sage Publications 10.1177/0192513X09331580 http://jfi.sagepub.com hosted at http://online.sagepub.com

Sharon Sassler Cornell University, Ithaca, New York

Anna Cunningham The Ohio State University, Columbus

Daniel T. Lichter Cornell University, Ithaca, New York The authors examine whether young adults who experienced their parents’ divorce and new relationships have different relationship trajectories than those who spent their childhoods living with biological parents in marriedcouple families. The analysis is based on longitudinal reports from more than 1,500 children from Wave 1 of the 1987-1988 National Survey of Families and Households who were ages 18 to 34 at Wave 3 (in 2001-2002). The results suggest that parents’ intimate relationships serve as templates for their children. Children of divorce had elevated rates of cohabitation as adults, relative to marriage. But union outcomes were not uniform for all children who experienced parental divorce. Those whose parents cohabited following divorce exhibited elevated odds of cohabiting themselves, compared to young adults whose parents remarried without first cohabiting or remained in stable marriages. Parental cohabitation also undermines relationship quality and stability among married or dating young adults. Keywords:   divorce; cohabitation; marriage; intergenerational reproduction; marital quality

C

hanges in union formation and dissolution have transformed American family life over the last three decades (Cherlin, 2004; Lichter & Qian,

Authors’ Note: An earlier version of this article was presented at the International Conference on Children and Divorce, University of East Anglia, Norwich, UK, July 27, 2006. The authors acknowledge the support of Cornell University’s Bronfenbrenner Life Course Center and the helpful comments and suggestions of two anonymous reviewers and Regina Bures. Please address correspondence to Sharon Sassler, Department of Policy Analysis & Management, Cornell University, 134 Martha Van Rensselaer Hall, Ithaca, NY 14853; e-mail: [email protected]. 757 Downloaded from http://jfi.sagepub.com at CORNELL UNIV on May 12, 2009

758   Journal of Family Issues

2004). Young adults today represent the first full generation of Americans who experienced—as growing children—high rates of marital instability and remarriage of their parents during the so-called divorce revolution and beyond. About 40% of American children who grew up in the 1970s and 1980s experienced the breakup of their parents’ marriages (Bumpass, 1984). Today’s young adults also were exposed during childhood to newly emerging alternatives to traditional nuclear family life and were often challenged economically and psychosocially by unstable and sometimes chaotic family living arrangements (Fomby & Cherlin, 2007; Heuveline & Timberlake, 2004; Manning & Lichter, 1996; Raley & Wildsmith, 2004). Many lived with their cohabiting parents and partners, while still more faced the prospect of splitting their time and emotional energy between two new stepfamilies—including stepparents, stepsiblings, and half siblings—when parents remarried following a divorce (Goldscheider & Sassler, 2006; Graefe & Lichter, 1999). America’s young adults are now forming intimate relationships of their own—dating, cohabiting, and getting married—and many have high hopes for a stable and healthy family life (Manning, Longmore, & Giordano, 2007; Thornton & Young-DeMarco, 2001). Are children of divorce more likely to reproduce the disruptive family experiences of their parents? Or have their childhood experiences sensitized them to the difficulties of finding a compatible partner and made them more cautious but perhaps more successful in finding a lifelong mate? In this article, we examine marriage and relationship data on adult children from the third wave of the National Survey of Families and Households (NSFH), which is linked intergenerationally to the marital histories that their parents reported in earlier waves. Specifically, we address the question of whether young adults who experienced firsthand their parents’ divorce and new relationships are less (or more) likely to marry, cohabit, or form steady romantic attachments than those who spent their childhoods living with both biological (or adoptive) parents in traditional married-couple families. We also assess the impact of parental union transitions on young adults’ assessments of the quality of their current relationships. Unlike previous studies, we distinguish among various types of postdivorce parental relationships such as cohabitation and remarriage. Our study contributes to a growing body of research on the intergenerational reproduction of marriage and marital quality (Amato & Cheadle, 2005; Teachman, 2002; Webster, Orbuch, & House, 1995).

Downloaded from http://jfi.sagepub.com at CORNELL UNIV on May 12, 2009

Sassler et  al. / Patterns of Union Formation   759

Intergenerational Family Patterns Compared with other developed nations, divorce rates in the United States are uncommonly high (Martin & Kats, 2003); nearly one half of first marriages initiated in the 1990s are expected to end in divorce (Schoen & Standish, 2001). The implications for children are large. The latest available federal statistics on marriage indicate that 53% of all divorces involve minor children; more than 1 million children are affected each year by the divorce of their parents (see Clarke, 1995; London, 1989). In addition, most children of divorce experience the remarriages of their parents, which are often preceded by cohabitation. In 2000, 43% of cohabiting unions included coresidential minor children (Lichter & Qian, 2004). High rates of union dissolution and remarriage in the past have obvious implications for current marriage patterns. Demographic momentum is built into current family patterns through the intergenerational reproduction of divorce and family instability (Amato, 1996; Rogers & Amato, 1997; Teachman, 2002). The emerging consensus is that healthy, stable marriages improve family functioning, promote positive developmental outcomes among children, and provide day-to-day examples on which young adults can model their own intimate relationships (Brotherson & Duncan, 2004; Moore et  al., 2007). Children who grow up with both biological parents fare better on average than those whose parents were never married or who divorced and remarried (Moore et  al., 2007; Teachman, 2002). They score higher on measures of psychological adjustment, self-esteem, and academic success (Amato, 2005; Brown, 2006; McLanahan & Sandefur, 1994; Ross & Mirowsky, 1999), and they are less likely to be involved in precocious adolescent sexual activity or to experience nonmarital teen childbearing (Wu & Martinson, 1993). They also experience fewer behavioral and academic difficulties than do children from cohabiting or married stepparent families (Brown, 2004, 2006; Manning & Lamb, 2003; Morrison & Ritualo, 2000). These short-term effects potentially have important long-term ramifications for marriage and family life. One implication seems clear: Young adults whose biological parents divorced, then cohabited, and/or remarried may experience more unstable relationships themselves and be less satisfied with their own intimate relationships. Indeed, research clearly documents a strong statistical association between having experienced a parental marital disruption during childhood and later union instability during adulthood (Amato, 1996; Cherlin, Kiernan, & ChaseLansdale, 1995; Kiernan, 1992; McLanahan & Sandefur, 1994; Wolfinger,

Downloaded from http://jfi.sagepub.com at CORNELL UNIV on May 12, 2009

760   Journal of Family Issues

2000). Unambiguous interpretations of this statistical association are difficult to make, in part because previous studies rarely measure and evaluate the effects of family processes and transitions in the aftermath of divorce. Most divorced parents (of reproductive age) quickly enter into new postdivorce relationships that lead them back to the altar and into new lives as stepfamilies (Goldscheider & Sassler, 2006). The majority of remarriages today are preceded by cohabitation (Bumpass & Lu, 2000). In fact, cohabitation rates among the divorced population exceed rates among the never married; cohabitation apparently has become an alternative to traditional marriage, especially among those with nonresidential children (Goldscheider & Sassler, 2006; Stewart, 2001; Stewart, Manning, & Smock, 2003). At the same time, few empirical studies have systematically evaluated whether the kinds of unions parents enter into following marital disruption have divergent effects on children’s own relationships and patterns of union formation. Clearly, not all intimate relationships are equivalent. Recent research suggests that some forms of parental relationships result in more advantageous outcomes for offspring (Aquilino, 1996; Cooksey, 1997; Dunifon & Kowaleski-Jones, 2002; Morrison & Ritualo, 2000). For example, parental remarriage is associated with more favorable long-term economic and behavioral outcomes for children than is cohabitation (Brown, 2006; Dunifon & Kowaleski-Jones, 2002; Hofferth & Anderson, 2003; Morrison & Ritualo, 2000), but this may simply reflect the greater economic selectivity of marriage over singlehood or cohabitation (Hofferth & Anderson, 2003). Even after accounting for economic resources, however, residing in cohabiting families undermines adolescents’ school engagement (Brown, 2006), with potentially important ramifications for later life achievement. Whether the experience of parental cohabitation during childhood can explain patterns of union formation among young adults today, however, requires additional study. One argument is that cohabitation among adult children is less stigmatized or negatively sanctioned by parents who have cohabited (Kapinus, 2005; Thornton, 1991). Studies also show that experiencing family disruption often accelerates children’s departure from the parental home (Goldscheider & Goldscheider, 1998). This may in part account for the bifurcated marriage patterns of children of divorced parents (Wolfinger, 2000); the adult children of divorced parents are slightly less likely to marry than those from intact families, but they also are more likely to form unions at either an earlier or later age (Teachman, 2003; Wolfinger, 2003). Young adults who grew up with divorced parents are also more likely to cohabit or remain single relative to marrying (Clarkberg, 1999; Sassler & Goldscheider, 2004; Thornton, 1991)

Downloaded from http://jfi.sagepub.com at CORNELL UNIV on May 12, 2009

Sassler et  al. / Patterns of Union Formation   761

and are more likely to cohabit prior to marriage (Teachman, 2004). The effects of growing up in a stepfamily, however, often differ by gender. Men who grew up in a stepfamily situation are more likely to replicate their childhood experiences by entering into cohabiting unions with single mothers (Goldscheider & Sassler, 2006), while women who grew up with remarried parents marry earlier. Women also acquire a greater number of risk factors for subsequent union dissolution (such as experiencing a teen birth) than women who grew up in stably married families or those who experienced a parental death (Teachman, 2002). As we have described, previous research on longer term family outcomes has examined mostly intergenerational effects on marriage, cohabitation, or divorce. A small but growing body of research has focused on the intergenerational transmission of marital quality (Frisch & Hviid, 2006; Teachman, 2002). This research examines how the exposure to poor-quality parental relationships during childhood can shape young adults’ own romantic attachments and relationship quality (Teachman, 2003). For example, young adults who experienced parental conflict and divorce are themselves more likely to report marital problems and express lower satisfaction in their own marriages (Amato, 1996; McLeod, 1991; Ross & Mirowsky, 1999). The adult children of divorce are more likely than other adults to be in unhappy relationships (Ross & Mirowsky, 1999). Unfortunately, we know little about the specific pathways or mechanisms that link parental and filial patterns of union formation and quality. Until now, data limitations have prevented researchers from assessing how these childhood experiences have shaped the quality of their own adult relationships.

Current Study To sum up, our study addresses a straightforward question: How have the family forms experienced during childhood and the increasing array of romantic options shaped young adults’ own intimate relationships? We hypothesize that exposure to different family forms has the putative effect of increasing young adults’ acceptance of lifestyle alternatives and may subsequently shape their own relationship decision making and the quality of their relationships. We also expect that the impact of relationship instability on young adults’ own experiences will be uneven, with young adults who grew up in divorced families being less likely to marry than their counterparts from intact families. But more significantly, we anticipate that young adults whose parents remarried will be more likely to be married themselves,

Downloaded from http://jfi.sagepub.com at CORNELL UNIV on May 12, 2009

762   Journal of Family Issues

though parental remarriage may also be associated with elevated odds of cohabitation (before marrying). Parental cohabitation, on the other hand, is expected to reduce the likelihood of marriage among their adult offspring, while increasing the likelihood of cohabitation or other types of relationships (e.g., romantically involved but not living together). For those young adults whose parents remained romantically unattached after divorce, we expect that they will also be less likely to form romantic unions—dating, cohabitation, or marriage. We also assess the intergenerational effects of parental relationships following divorce on the relationship quality of their young adult children. Most studies of transitions to marriage have focused on the effects of contemporaneous social conditions and processes (e.g., Lichter, Qian, & Mellott, 2006; Sassler & Goldscheider, 2004). Yet current patterns of marriage and cohabitation are also rooted in the past, reflecting learned experiences from childhood. Our guiding hypothesis is that children who grew up with divorced parents who cohabitated or remarried are likely to report lower relationship satisfaction, higher disagreement, and a heightened perception that their own relationships will end, relative to youth who grew up in intact families. Their relationship quality should also reflect the type of unions their parents entered postdivorce. We expect young adults whose parents cohabited to report lower relationship quality than those whose parents remarried or remained single. If our hypotheses are born out, the implication is that family instability has built-in momentum for future instability in America’s married population. The preceding discussion has focused on intergenerational family patterns and relationship quality, that is, linkages between parental and filial generations. Of course, many other risk factors affect the type, timing, and quality of unions formed in early adulthood. Our multivariate models (described in the next section) therefore control for a number of important family background and early adult characteristics: family social class background, young adults’ own indicators of adolescent adjustment and maturation, gender, stage in the life course, race, and educational pursuits and attainment. Experiences with divorce have been found to extend to the third generation, which suggests the need to consider parents’ own family structures during childhood (Amato & Cheadle, 2005). Parents with substantial resources, such as education and earnings, may discourage early marriage among their children but facilitate marriage at later ages (Sassler & Goldscheider, 2004; Waite & Spitze, 1981). Youth from more advantaged backgrounds are no more likely to marry than they are to form cohabiting unions, once their current pursuits and life cycle stage are accounted for (Sassler & Goldscheider, 2004). The

Downloaded from http://jfi.sagepub.com at CORNELL UNIV on May 12, 2009

Sassler et  al. / Patterns of Union Formation   763

intergenerational link in family structure may also be reinforced by negative or positive adolescent adjustment. For example, spending time in a singleparent or stepparent family is associated with precocious sexual activity (Cooksey, Mott, & Neubauer, 2002; Upchurch, Levy-Storms, Sucoff, & Aneshensel, 1998). And early sexual debut may be associated with more sexual partners, a greater likelihood of cohabitation, and relationship instability (Sassler & Kamp Dush, 2007). Many other demographic attributes are associated with the timing and stability of nonmarital relationships and marital unions. Men generally marry later than women (Oppenheimer, 1988), and the proportion of unmarried adults declines with age. It is well established that African Americans are less likely to be married than their White and Hispanic counterparts (Cherlin, 2004; Lichter & Qian, 2004), although racial difference in cohabitation or in sexually involved but noncoresidential unions are not well understood (Raley, 1996). Highly educated men and women are more likely to be married, particularly relative to cohabiting (Lichter & Qian, 2004; Sassler & Goldscheider, 2004), whereas enrollment in school apparently reduces the likelihood of marriage. The link between education and cohabitation is mixed. Some studies find that school enrollment reduces the likelihood of cohabiting (Thornton, Axinn, & Teachman, 1995), whereas others report that young adults in school are significantly more likely to cohabit than they are to marry (Sassler & Goldscheider, 2004). Clearly, any conclusions about the intergenerational reproduction of family patterns must account for many confounding factors that link parental and filial generations (e.g., race).

Data and Method Intergenerational panel data on union formation and dissolution are required to achieve our objectives. Several previous studies of intergenerational family patterns have been drawn from the Detroit Area Study (Axinn & Thornton, 1992, 1996; M. Cunningham & Thornton, 2006), but a White sample in a single large metropolitan city is not nationally representative. Fortunately, nationally representative data on intergenerational family patterns are now available from responses of the adult children included in the newly released third wave of the NSFH. These data, collected in 2002, provide unusually rich information on early adult relationships and can be linked to previously collected information on parents and children from two earlier waves of the NSFH (collected in 1987-1988 and 1992-1994). The original NSFH panel consisted of a national probability sample of 13,008

Downloaded from http://jfi.sagepub.com at CORNELL UNIV on May 12, 2009

764   Journal of Family Issues

individuals aged 19 and older, plus an oversample of minorities, singleparent families, recently married couples, and cohabiting couples (Sweet, Bumpass, & Call, 1988). These earlier data waves recorded parents’ relationship and marital histories and the quality of their relationships, tracked the changing family living conditions of children, and measured many different aspects of their childhood well-being and adolescent development. Our analyses are based on reports from children from Wave 1, who at the time of their interviews at Wave 3 were between the ages of 18 and 34.1 The final sample includes 1,571 children.2 We focus on two main outcomes for this analysis: the union status of these young adults at Wave 3 and their assessment of their relationship quality (conditional on being in a relationship). We rely on detailed life histories to identify young adults who are currently married, cohabiting, in a steady romantic (but noncoresidential) relationship, or are not currently in a romantic union. To assess these young adults’ relationship quality, we provide a measure of the perceived likelihood of divorce or relationship dissolution and then construct two multi-item scales of relationship quality (i.e., Satisfaction and Conflict). We then explore how parental union transitions shape the relationships of children who are romantically involved at Wave 3. Excluding those not currently in a romantic relationship reduces the sample size for this part of the analysis to 1,087. Our first measure of union quality centers on perceptions that the relationship will end (chance of separating). Respondents are asked, “It is always difficult to predict what will happen in a relationship, but realistically, what do you think the chances are that you and your partner/ spouse will eventually separate/divorce?” Responses range from very low (1) to very high (5). Higher mean values represent a higher perceived chance of the relationship breaking up. Relationship satisfaction is measured by a summated rating scale from responses to questions about (a) the understanding received from the partner, (b) the love and affection received from the partner, (c) the amount of time spent with the partner, (d) the demands the partner places on the respondent, (e) the way the partner spends money, (f) the sexual relationship, and (g) the work the partner does around the house. Possible responses varied from 1 (very unhappy) to 7 (very happy). When summed, satisfaction scores range from 7 to 49, with higher scores indicating greater relationship satisfaction (Cronbach’s α = .77 for married persons, .81 for cohabiters). For those in steady relationships, the questions are limited to satisfaction with understanding, love and affection, time spent with the partner, and the sexual relationship. Scores range from 4 to 28; the internal consistency and reliability of this measure are somewhat lower (α = .68).

Downloaded from http://jfi.sagepub.com at CORNELL UNIV on May 12, 2009

Sassler et  al. / Patterns of Union Formation   765

The third measure focuses on the assessment of conflict and is limited to those in either cohabiting relationships or marriages. Respondents were asked, “Please tell me how often, if at all, you and your partner have had an OPEN disagreement about each of the following in the past year?” We calculate a summated rating scale from responses (1 = never to 6 = almost every day) to the following potential sources of discord: (a) household tasks, (b) money, (c) spending time together, (d) sex, and (e) in-laws. Scores range from 1 to 28; higher scores indicate more pervasive disagreement (α = .66 for married persons, .76 for cohabiters).

Measuring Childhood Family Structure and Resources Our primary independent variable of interest—family structure of the young adult while growing up—relies on information about parental relationship histories from Waves 1 and 2. Initially, we examine whether the child’s reporting parent remained married between survey waves. For parents who divorced, we determine from their relationship histories (at either Wave 1 if they were divorced at the initial interview or Wave 2 if they divorced between survey waves) whether they subsequently entered another coresidential union. We identify four postdivorce transitions: (a) remarriage, (b) cohabitation followed by remarriage, (c) cohabitation, and (d) remaining single (no coresidential unions). The number of mothers who were never married at the child’s birth was too small (n = 66) to adequately assess the statistical relationship between subsequent union transitions and their children’s union formation patterns. Previous studies have examined the age at which parental union disruption occurred, whereas others focus on the number of transitions that parents experience in their marital histories (Amato, 1996; Teachman, 2003, 2004). We focus here on the type of union entered following parental divorce, but data limitations prevent us from measuring the age at which the child experienced the dissolution or subsequent union transitions of the parents.3 For a small number of parents, information at Wave 2 was affected by sample attrition; 95 of those who were married at Wave 1 and 36 of those who were divorced or widowed were lost to follow-up. Statistical tests, however, revealed no significant differences between the outcomes for children whose parents did not respond between waves and those with data at both time periods. The married parents who experienced sample attrition are therefore assigned to the stably married category, while parents who were divorced at Wave 1 and were subsequently lost to follow-up are grouped with divorced parents who experienced no union transitions between waves.

Downloaded from http://jfi.sagepub.com at CORNELL UNIV on May 12, 2009

766   Journal of Family Issues

Measures of familial resources during childhood include mother’s educational attainment, whether the mother’s parents had divorced, and maternal employment at Wave 1.4 Whereas some studies utilize averages of mothers’ and fathers’ educational levels (e.g., Goldscheider & Waite, 1986; Sassler & Goldscheider, 2004), maternal attributes are more robust predictors of children’s outcomes, particularly for daughters (Axinn & Thornton, 1992, 1993). For each child, we also include measures of (logged) family income at Wave 1; the number of siblings, both full and half;5 and nonmetropolitan residence. To evaluate whether adolescent emotional and psychological well-being mediates the link between parents’ and children’s union formation, we include three measures from Waves 2 and 3. To measure self-esteem, children were asked in Wave 2 whether they agreed with the following four questions: (a) “I can do just about anything I really set my mind to do”; (b) “I feel that I’m a person of worth, at least on an equal basis with others”; (c) “I am able to do things as well as most other people”; and (d) “On the whole, I am satisfied with myself.” Answers ranging from 1 (strongly agree) to 4 (strongly disagree) were summed, with higher scores indicating greater levels of self-esteem (α = .67).6 A second intervening variable— adolescent adjustment—is the response to the following question asked at Wave 2: “Taking all things together, on a scale from 0 to 10, where 0 means really bad and 10 means absolutely perfect, how would you say things are for you these days?”7 Missing data were assigned the mean scores of children who had experienced similar parental union transitions.8 A third intervening variable, age at first sex, was ascertained from Wave 3 data.9 Salient characteristics of the child as an adult are drawn from Wave 3 and include age, race/ethnicity, educational attainment, and school enrollment. We also examine two measures known to influence relationship satisfaction and union dissolution: presence of coresidential children (yes or no) and depression. Depression is measured with 10 items from the Center for Epidemiological Studies Short Depression Scale (CES-D 10; Anresen, Malmgren, Carter, & Patrick, 1994). Responses to questions range from 0 to 7 days per week; when summed, the resulting scale ranges from 0 to 70 (α = .80). Variable descriptions, weighted means, and standard deviations for all variables are presented in Table 1.10

Analytic Approach Our analysis proceeds in two stages. First, we estimate multinomial logistic regression models to examine union types of the adult children at Wave 3

Downloaded from http://jfi.sagepub.com at CORNELL UNIV on May 12, 2009

767

Downloaded from http://jfi.sagepub.com at CORNELL UNIV on May 12, 2009

SD

Family attributes Parental union transitions Stable marriage Still married to same partner at Wave 2 .752 .495 .729 Divorced Focal child’s parents no longer married by Wave 2 .248 .497 .287 Remarriage (no cohabitation)    Entered new marriage by Wave 2 .046 .268 .052 Cohabit to remarry    Entered cohabiting union and then marriage .053 .290 .063 Cohabit (no remarriage)    Entered cohabiting union .038 .235 .038 Remain single    Never cohabited or remarried by Wave 2 .112 .400 .119 Mother’s educational attainment Education of mother, by years of school and degree Less than high school    Less than 12 years of school .087 .315 .093 High school graduate    12 years of school, high school diploma, or GED .430 .496 .448 Some college    Associates degree or some years of college .246 .433 .236 BA or more    Bachelor’s degree or more .237 .404 .224 Family characteristics Maternal experience with divorce Mother experienced own parents’ divorce .207 .408 .200 Maternal employment Mother worked at Wave 1 .668 .464 .677 Family income (in US$) Combined family income at Wave 1 $42,881 $38,789 $42,220 Logged family income Combined family income at Wave 1, logged $10.159 $2.294 $10.102 Rural residence Lived in nonmetropolitan area (1980) at Wave 1 .328 .472 .333 Number of focal children’s siblings Siblings (both full and half) 2.288 2.029 2.380

(continued)

.405 .460 $40,740 $2.397 .478 2.005

.321 .497 .433 .395

.498 .499 .278 .314 .234 .401

SD

M

M

Variable

Description

Among Those in a Relationship

Total

Table 1 Means and Standard Deviations of Variables

768

Downloaded from http://jfi.sagepub.com at CORNELL UNIV on May 12, 2009

Description M

SD

M

Note: CES-D 10 = 10-item Center for Epidemiological Studies Short Depression Scale. Source: National Survey of Families and Households.

.257 .458 .488 .427 .396 .499 6.655

4.378 .495 .355 .299 .214

1.698

1.352 1.303

SD

Among Those in a Relationship

Focal child attributes Self-esteem Measured at Wave 2, ranges from 4 to 16 13.167 1.356 13.166 Assessment of how things are going Measured at Wave 2, ranges from 0 (really bad) 7.488 1.384 7.504    to 10 (absolutely perfect) Age at first sex Age first reported sexual intercourse, measured at Wave 3 16.893 2.000 16.915 Demographic attributes Age Age in years at Wave 3 25.850 4.448 26.827 Female Female = 1, male = 0 .531 .499 .559 Race (non-Hispanic White) White = 1, Other = 0 .887 .379 .904 Black Black = 1, Other = 0 .068 .325 .056 Hispanic Hispanic = 1, Other = 0 .042 .224 .040 Educational attainment Education of focal child, by Wave 3 Less than high school    Less than 12 years of school .051 .238 .064 High school graduate    12 years of school, high school diploma, or GED .276 .460 .268 Some college    Associates degree or some years of college .426 .493 .387 BA or more    Bachelor’s degree or more .248 .416 .281 Currently enrolled in school    Attending part- or full-time at Wave 3 .244 .420 .193 Has child Was a parent as of Wave 3 .345 .484 .451 Depression CES-D 10, range of 0 to 70, from Wave 3 13.484 7.626 12.797 N   1,571 1,087

Variable

Total

Table 1  (continued)

Sassler et  al. / Patterns of Union Formation   769

(Maddala, 1983). Possible outcomes include marriage, cohabiting union, steady dating relationship, or no romantic involvement.11 Exponentiated coefficients show the odds of being in a specific kind of union relative to the reference category (i.e., married). The second part of the analysis, based on ordinary least squares regression, focuses on the relationship quality among adult children currently involved in romantic unions. We present results and discuss explained variance on the relationship types about which questions are asked. About one quarter of these children had experienced their parents’ divorce by the second interview.12 Of the parents whose marriages ended, the largest share had not (yet) entered another coresidential relationship at the second interview. About 45% were single and not currently cohabiting. Nearly 40% of the parents had remarried, and the majority of these remarriages were preceded by cohabitation. Furthermore, about 15% of adult children’s parents who had experienced a divorce were cohabiting at the time of the second interview. Clearly, the children in the NSFH experienced a diverse set of family trajectories as they transitioned into adulthood.

Results The adult children in our sample were distributed across a wide array of relationship statuses at Wave 3. Women were somewhat more likely to be married than men (37% vs. 30%) and were also more likely to be in cohabit­ ing relationships (17% vs. 13%). More than one fifth of the remaining young adults reported being in a steady dating relationship. Men are more likely than women (36% vs. 25%) to not be romantically involved (data not shown).

Intergenerational Reproduction of Family Patterns Table 2 displays exponentiated logistic regression coefficients, that is, the odds ratios of being in a cohabiting relationship, having a steady romantic partner, or not being romantically involved, relative to being married. Results are presented in two sequential models. We first control for parental attributes and family characteristics, then we incorporate intervening measures of adolescent adjustment and individual characteristics that mediate the relationship between family background characteristics and Wave 3 marital outcomes. We present outcomes in two panels; Panel A presents the coefficients for experiencing a parental divorce, net of other controls, while Panel B distinguishes among the type of union experiences witnessed by children whose parents divorced.

Downloaded from http://jfi.sagepub.com at CORNELL UNIV on May 12, 2009

770

Downloaded from http://jfi.sagepub.com at CORNELL UNIV on May 12, 2009

Versus Married

None

Cohabit

0.962 0.963 1.430 0.830 1.289 1.012 1.035 0.610**

0.733 0.490* 1.185 1.209

0.870

None

(continued)

Versus Married

Steady

Model 2 

Panel A: Divorced 1.718** 0.642** 0.720* 1.703** 0.817 Panel B: Parental union transitions (vs. stable marriage) Divorce to: Remarriage (no cohabitation) 0.929 0.422** 0.572* 0.911 0.541† Cohabit to remarriage 1.942** 0.655 0.467** 1.622* 0.698 Cohabit (no remarriage) 2.078* 0.731 1.025 1.808† 0.911 Remain single 1.617* 0.738 0.921 1.795* 1.044 Other family characteristics Mother’s educational attainment (high school graduate) Less than high school 0.936 1.053 0.973 0.900 1.198 Some college 1.364 1.235 1.246 1.265 0.932 1.545† BA or more 1.383 1.750** 1.557* 1.543† Maternal employment as child 1.009 0.688* 0.729* 1.127 0.811 Mother’s parents’ divorced 1.493* 1.163 1.378* 1.544* 1.052 Family income (logged) 0.940† 0.956 0.998 0.965 0.971 Number of other children (siblings) 1.048 0.996 0.992 1.072† 1.045 Rural residence 1.155 0.729* 0.708* 1.047 0.624**

Independent Variables

Steady



Cohabit

Model 1 



Table 2 Odds Ratios for Focal Children’s Union Outcomes

771

Downloaded from http://jfi.sagepub.com at CORNELL UNIV on May 12, 2009

Versus Married

None

Cohabit Versus Married

Steady

Model 2 

Source: National Survey of Families and Households. † p < .10. *p < .05. **p < .01. ***p < .001.

Focal child attributes Self-esteem 0.993 1.037 Assessment of how things are going 0.985 1.030 Age at first sex 0.919 1.093** Female 0.946 0.508*** Age 0.852*** 0.746*** Race (non-Hispanic White) Black 1.255 4.591*** Hispanic 1.090 0.770 Educational attainment (high school graduate) Less than high school 1.760* 0.563 Some college 0.991 0.897 BA or more 0.943 0.987 Currently enrolled in school 1.957** 2.708*** Intercept (unexponentiated) –0.851* 0.321 0.173 4.645*** 5.490*** Log likelihood –2,056.626 –1,770.295 N 1,571  1,571 

Independent Variables

Steady



Cohabit

Model 1 



Table 2  (continued)

0.506* 0.971 0.727 2.197*** 5.221***

3.910*** 1.266

1.050 0.860** 1.152*** 0.399*** 0.766***

None

772   Journal of Family Issues

Experiencing a parental divorce is associated with significantly higher rates of cohabitation relative to marriage (odds = 1.7) among young adults, even after controlling for familial background, adolescent adjustment and behaviors, and personal attributes (Panel A). Parental divorce also increases the odds of being married relative to being either in a steady dating relationship or not romantically involved, although there is no longer any significant difference after adding controls for children’s own attributes. These findings on the intergenerational consequences of parental divorce largely replicate results from previous studies (cf. Teachman, 2003; Wolfinger, 2000). We also recognize that unmeasured factors (e.g., IQ or temperament) that are selective of parental divorce may also be associated with their children’s patterns of union formation. Our main substantive contribution, however, is to evaluate the intergenerational linkages between the postdivorce relationships of parents and adult children’s own intimate relationships. Indeed, the results in Panel B of Table 2 highlight the need to better understand parents’ romantic behaviors after they divorce and the long-term implications for their growing children. Although most young adults who experienced a parental divorce demonstrate elevated odds of cohabiting relative to being married, for example, those whose parents remarried without first living with their partner are no more (or less) likely to be cohabiting than young adults whose parents remained in intact marriages (Model 1). However, they are significantly less likely to be in steady dating relationships or romantically involved than to be married, relative to those who grew up with stably married parents. Some additional analyses that test for significant differences between various parental marital status transitions also reveal that young adults whose parents remarried directly (i.e., without first cohabiting) differ significantly from those whose parents first lived with their new spouse. Indeed, young adults whose parents cohabited following their divorce, whether prior to or in lieu of marriage, are more than twice as likely to be cohabiting as are children whose parents remarried without first living with their new spouse. They are also substantially more likely to form cohabiting unions (relative to marital ones) than adult children who grew up with stably married parents (odds = 1.94 for those who cohabit prior to remarriage, 2.08 for those who cohabit without remarrying). Clearly, cohabitation has an intergenerational dimension.13 Our results also indicate that an economically advantaged childhood is associated with slower entry into marriage and cohabitation. Young adults with highly educated mothers, for example, are more than 1.5 times as likely to be in a dating relationship or not romantically involved as they are

Downloaded from http://jfi.sagepub.com at CORNELL UNIV on May 12, 2009

Sassler et  al. / Patterns of Union Formation   773

to be married. We also find evidence of intergenerational patterns of union formation. When parents divorce, adult children have elevated rates of cohabitation relative to marrying. Moreover, they are 1.5 times more likely to cohabit than to marry when their mothers also experienced parents’ divorce as a child and are significantly less likely to be in a romantic relationship than to be married. Including measures of adolescent development and the adult child’s current attributes in the models attenuates the longer term consequences of parental union transitions (Model 2, Table 2). The substantive implication is clear: Adolescent well-being links, at least in part, the intergenerational reproduction of family structure. The impact of experiencing a parental cohabitation, whether prior to a remarriage or not, remains significant, albeit only weakly for adult children whose parents cohabited but did not remarry. Experiences with parental cohabitation, as well as having a parent who does not form a new coresidential union, elevate children’s own likelihood of cohabiting relative to being married. The inclusion of statistical controls has the effect of reducing but not eliminating the impact of parental remarriage on the likelihood of entering into noncoresidential unions. The unions formed by divorced parents apparently are linked to their children’s romantic attachments long after the divorce and its aftermath. The long-term effects of family background are also attenuated after including the full range of controls. The intergenerational impact of maternal divorce on young adults’ likelihood of cohabiting remains sizable and statistically significant, as does the impact of growing up in rural areas. We find little relationship between the child’s adolescent adjustment measures or current pursuits and union outcomes. Adolescents who were more positive regarding how their life was going are significantly more likely as young adults to be married relative to having no relationship, whereas those who experienced a sexual debut at a later age are slightly more likely to be dating or not in a romantic relationship relative to being married. Women also are more likely than men to be married than dating and about half as likely as men to be uninvolved romantically, relative to being married. The individual factors operate largely as hypothesized. Young adults are significantly more likely as they age to be married than to be cohabiting, dating, or romantically uninvolved. Blacks are about four times more likely to be dating or not in romantic relationships, relative to being married, than are non-Hispanic White respondents (cf. Raley, 1996). The impact of educational attainment on young adults’ union formation pattern is weak; high

Downloaded from http://jfi.sagepub.com at CORNELL UNIV on May 12, 2009

774   Journal of Family Issues

school dropouts are more likely than high school graduates to be married than not in a relationship; they are also 1.7 times more likely to be cohabiting than married, though these effects are only weakly significant. We find no statistical relationship between having attended some college or attained a college degree and getting married, which may indicate the weakening impact of economic markers on marriage (Sassler & Goldscheider, 2004). But young adults who are currently attending school are more than twice as likely to be in a steady dating relationship or romantically unattached than to be married, relative to those who are not enrolled; they are also almost twice as likely to be cohabiting.

Family Background and Relationship Quality We now turn to the question of how young adults’ current relationship quality is shaped by parental union transitions during childhood. Table 3 provides means and standard deviations on three dimensions of children’s relationships: perceptions of the chance the relationship will break up, satisfaction, and levels of disagreement. Not surprisingly, married respondents are least likely to believe their relationships will end. Cohabiters occupy an intermediate position, with steady daters most likely to believe their relationships will dissolve. Married men and women also report the highest levels of relationship satisfaction, although differences from cohabiters are statistically insignificant. The scaling was different for those in steady relationships, who were asked only four questions. When it comes to conflict—the amount of disagreement about the household division of labor, finances, time spent together, sex, or relationships with the partner’s parents—cohabiters report about the same levels as do married respondents. We regress the measures of relationship quality on the independent variables defined earlier (see Table 1). Results are presented in two panels in Table 4. Panel A presents the coefficient for having experienced a parental divorce, net of other controls. Panel B distinguishes between the type of union experiences of parents who divorced and incorporates the other control variables. In addition to the variables included in Table 2, we also incorporate measures of whether the adult child has children or reports depressive symptoms. Data in Panel A indicate that married children and steady daters whose parents divorced are more likely than others to believe that their relationships also will break up. This relationship was not found among cohabiting couples. Parental divorce is also unrelated to satisfaction among young adults who are married or cohabiting. It is, however, significantly associated with less satisfying dating relationships. Last, married

Downloaded from http://jfi.sagepub.com at CORNELL UNIV on May 12, 2009

Sassler et  al. / Patterns of Union Formation   775

Table 3 Self-Reported Assessment of Focal Child’s Relationship Qualitya Variable

Total Description

M

Chances of divorce/separation Married personsb,c 1 = very low, 5 = very high 1.411 Cohabitersd 1 = very low, 5 = very high 1.774 Steady daters 1 = very low, 5 = very high 2.084 Satisfaction Married persons Ranges from 7 to 49 39.752 Cohabiters Ranges from 7 to 49 39.017 Steady daters Ranges from 4 to 28 23.781 Disagreement Married persons Ranges from 1 to 28 11.138 Cohabiters Ranges from 1 to 28 11.185 Steady daters NA n   1,087

SD 0.761 0.952 1.020 6.269 7.295 3.815 3.314 4.419

Note: Excludes focal children who are not in a romantic relationship (n = 474). a. Results are not weighted. b. Denotes significant difference between married persons and cohabiters. c. Denotes significant difference between married persons and steady daters. d. Denotes significant difference between cohabiters and steady daters.

and cohabiting young adults who experienced the disruption of their parents’ marriage have no higher levels of disagreement than those whose parents remained in intact marriages. The results presented in Panel B of Table 4 indicate that the types of unions parents form after divorcing are associated with the relationship quality of their children. For example, married and dating young adults whose parent entered a cohabiting union following divorce are significantly more likely to think their own relationships will end. In contrast, married respondents whose parents remarried without first living together, as well as those whose parents remained single following marital disruption, are no more likely to think their current relationship will end than those whose parents remained in intact marriages. Parental union transitions never attain statistical significance for cohabiting young adults; however, results (not shown) indicate that cohabiting young adults are significantly more likely to believe that their relationship will end than their married counterparts (b = .30, p < .001), even after accounting for variation in family class background, individual

Downloaded from http://jfi.sagepub.com at CORNELL UNIV on May 12, 2009

776

Downloaded from http://jfi.sagepub.com at CORNELL UNIV on May 12, 2009

–0.104 0.196 –0.087 0.147 0.069 0.001 –0.003 –0.002 0.129** 0.001 –0.146** –0.077† –0.004 0.018*

0.046 0.105 0.167† 0.042 0.026 0.013 0.003 0.121† –0.014 –0.036 0.001

0.142 0.138 0.010

–0.539 0.522 0.307

0.120 –0.375 2.359 –0.052 –0.762 –1.192 0.402** –0.484 –0.568 –0.034 –0.825 –1.089 –0.103 0.726 0.746 0.037 0.143 0.071 –0.046 0.113 0.057 –0.316* –1.280* 0.667

0.273 0.550 –0.040 0.240 –0.423 –0.006 –0.070 0.380

–0.131 0.160 –0.748 0.017

–0.266 –0.128 0.517** –0.034 0.037 0.018

0.512 0.266 0.546 0.203 0.509 0.002 0.105 –0.922*

–1.634 –2.009* –2.113* –0.159

(continued)

–0.016 –0.351 0.050

–0.678 –0.426 –0.150 –0.266 –0.814 –0.020 0.105 –0.397

0.086 0.705 –2.071† 0.133

–0.092

2.968 0.588 4.079* 1.442

–0.185

–1.617† –1.611† 0.567 –0.695

0.403 0.531* 0.698* 0.300†

–0.146 –0.090 –0.339 –0.043

0.144 0.256* 0.257† 0.150

–1.135*

1.211

–0.079

0.158*

0.426** –0.831

Disagreement

Panel A: Divorced Panel B: Parental union transition (vs. stable marriage) Remarriage (no cohabitation) Cohabit to remarriage Cohabit (no remarriage) Remain single Maternal education (high school graduate) Less than high school Some college BA or more Maternal employment as child Mother’s parents’ divorced Family income (logged) Number of other children (siblings) Rural residence Focal child attributes Self-esteem Assessment of how things are going Age at first sex

Satisfaction

Steady Steady Dating Married Cohabiting Dating Married Cohabiting

Chance of Separating

Married Cohabiting



Table 4 Ordinary Least Squares Regression of Focal Child’s Relationship Quality, by Parental Union Transitions

777

Downloaded from http://jfi.sagepub.com at CORNELL UNIV on May 12, 2009

p < .10. *p < .05. **p < .01. ***p < .001.



–0.086 –0.369

–0.004 –1.307

–2.214 –2.009

0.048 0.544 –0.386 –2.277

–0.058 –0.026

Steady Dating

0.998 0.705

–0.056 –0.236

Married

2.981* 1.600

–0.064 –0.620

Cohabiting

Disagreement

0.207*** 0.273 –0.597* –0.309 –0.717 –0.159 0.489 0.005 –0.177* 0.329* –0.006 0.074 –2.429 –0.375 –0.277 0.902 –0.181† –0.028 –0.202 0.398 –0.824 –1.304† –0.208 0.420 0.163 –0.067 –0.063 –0.918 1.881 –0.739 0.571 0.369 0.052 0.201 –0.198 –2.313*** –2.113† –0.341 1.057*** 0.977 0.015*** 0.020*** 0.017*** –0.143*** –0.162*** –0.087*** 0.096*** 0.128*** 1.176* 0.563 1.967* 45.801*** 38.169*** 26.160*** 12.521*** 13.098** .139 .175 .147 .148 .152 .165 .161 .193 525 237 335 525 237 335 525 237

–0.041 0.412

0.185 0.644

–0.203* –0.107

0.000 –0.014

0.007 –0.166

0.006 –0.025

Female Age Race (non-Hispanic White) Black Hispanic Educational attainment (high school   graduate) Less than high school Some college BA or more Currently enrolled in school Has child Depression Intercept R2 n

Satisfaction

Steady Dating Married Cohabiting

Chance of Separating

Married Cohabiting



Table 4  (continued)

778   Journal of Family Issues

adjustment measures, and their own demographic attributes. Young adults who are dating are also far more likely than married persons to believe their relationship will end (b = .69, p < .001). Subsequent analyses suggest how experiences with parental union transitions may shape the dimensions of intimate relationships in ways that undermine their stability. Parental union transitions appear to exert quite opposite effects on the relationship satisfaction of married and cohabiting respondents. Cohabiting young adults who experienced a parental cohabitation without remarriage report relationship satisfaction that is 4 points greater than cohabiters whose parents remained in an intact marriage; this may be the result of greater parental support for children’s union choices. On the other hand, married young adults whose parents remarried, whether preceded by a spell of cohabitation or not, report considerably lower levels of relationship satisfaction than those whose parents remained in an intact marriage (b = 1.6 points), though both are significant only at the .10 level. In previous research, parental remarriage apparently expedites young adults’ leaving the home and entering marriage (Goldscheider & Waite, 1986; Wolfinger, 2003). The rapid entrance into romantic relationships results in lower quality unions, suggesting that these couples may be poorly matched. Alternatively, married young adults’ lower levels of satisfaction may be the result of poor socialization in how to find a good match. In some additional analysis (not shown), we found that cohabiters express significantly lower levels of satisfaction with their current relationship than do their married counterparts, even after accounting for family background characteristics and their own demographic attributes. Satisfaction for steady daters was measured in terms of mutual understanding, love and affection, time spent with the partner, and quality of the sexual relationship. Experiencing a parental cohabitation (both preceding marriage and with no remarriage) lowers daters’ levels of relationship satisfaction (on the order of 2 points lower) relative to their counterparts from intact families. One interpretation is that they may have a heightened awareness—based on their childhood experiences—that relationships often end unhappily. They also may be less well equipped to form satisfying relationships.14 Questions about disagreement or conflict were asked only of married and cohabiting adult children. Whereas the experience of parental divorce does not significantly shape disagreement levels for either married or cohabiting young adults, a closer examination of parental union transitions reveals that one particular union type—entrance into a cohabiting union not

Downloaded from http://jfi.sagepub.com at CORNELL UNIV on May 12, 2009

Sassler et  al. / Patterns of Union Formation   779

followed by a remarriage—is negatively associated with disagreement. However, this statistical association is only weakly significant and evident only among cohabiters. These young adults, who lived through their parents’ divorce and subsequent cohabiting unions, may view relationships as highly fragile and disagreement or conflict in the relationship as a major threat to stability.15

Discussion and Conclusion A primary objective of this article has been to examine intergenerational patterns of union formation and relationship quality. Specifically, how are the relationships and marital patterns of young adults today shaped by their childhood experiences of parental divorce, cohabitation, and remarriage? The recently released data on adult children in the third wave of the NSFH, when combined with earlier waves of the NSFH, provide an opportunity to evaluate the intergenerational transmission of family patterns. Unlike previous studies, we incorporated an extensive array of familial background characteristics, intervening psychosocial adjustment measures of the children in adolescence, and current achieved and ascribed characteristics of the child in early adulthood. Our working hypothesis was that instability in marriage and intimate relationships is passed from generation to generation. In fact, our results suggest that experiencing a parental divorce during childhood elevates the risk of entering a cohabiting union (relative to marriage). But not all parental union transitions exert the same influence on children’s own adult relationship trajectories. Significantly, if parents remarried following a divorce without first cohabiting, their children as young adults exhibited union patterns that were little different from adults who grew up with both parents. They are no more likely to cohabit or be married. By their behaviors, these young adults replicate their parents’ optimism regarding marriage. On the other hand, children whose parents entered cohabiting unions are more likely to cohabit themselves, as are youth whose parents remained single following their divorce. This cohabitation effect persists even when parents subsequently married. Furthermore, experiencing parental cohabitation also elevates perceptions that one’s own relationship will break up and reduces relationship satisfaction, at least among married young adults or those in steady dating relationships. These findings are consistent with the conventional view that unstable family backgrounds make today’s young adults more cautious about marriage, elevate their

Downloaded from http://jfi.sagepub.com at CORNELL UNIV on May 12, 2009

780   Journal of Family Issues

uncertainty about their ability to weather relationship hardships, and reduce satisfaction with their own unions. Our working hypothesis was that parental postdivorce relationships may have long-term implications for their children that persist into adulthood. Children who were exposed to the emotional roller-coaster associated with their parents’ dating, cohabiting, and remarriage undoubtedly faced unfamiliar and sometimes troubling situations that their counterparts in intact marriages did not experience. One view is that these children learn to be optimistic about relationships, notwithstanding the fact that they do not always work out. A less sanguine view is that the uncertainty and role ambiguity associated with reconfigured homes may push young adults out of the nest, while undermining their ability to form healthy and satisfying relationships that end in a stable marriage. We find support for both perspectives, as young adults whose parents divorce are more likely to form relationships, but they also express greater concerns about relationship stability and are less satisfied than those from intact families. Some researchers have suggested that the number of transitions is what is important, perhaps even more so than the type. Our analyses suggest the need to rethink this claim. Children whose parents remained single following divorce (one transition) are no less likely to be cohabiting rather than married than are young adults who experienced parents’ divorce and subsequent cohabitation (two transitions), or cohabitation to remarriage (three). Future studies should focus on the kinds of relationships parents form rather than solely on the number of union transitions. Nonetheless, our results indicated that youth whose parents experienced multiple transitions were less satisfied with their own relationships than those whose parents remained in stable marriages or remained single following a divorce. The increasingly diverse family experiences of today’s young adults pose new theoretical and methodological challenges to studying their longterm consequences. We recognize that our study of intergenerational patterns is not without some limitations. For example, the diversity of childhood experiences means that there are too few cases of some types to draw strong conclusions (and statistical significance). Delays in marriage also mean that a substantial proportion of our adult children have not yet married, although most will marry at some point. For them, the verdict is still out regarding the long-term consequences of family instability and the intergenerational transmission of family patterns. We also considered only a few measures of adjustment to parental divorce during childhood or adolescence; identical questions were often not asked of children at different ages. Also missing is information on whether the young adult’s partners experienced parental divorce and union transitions. Finally, our analyses

Downloaded from http://jfi.sagepub.com at CORNELL UNIV on May 12, 2009

Sassler et  al. / Patterns of Union Formation   781

suggest the need to distinguish between adult children who cohabited prior to marriage and those who did not. Previous studies suggest large differences between couples that cohabited only with their marital partners and those who cohabit serially (Lichter & Qian, 2008; Teachman, 2002). The concern is that simple binary measurement of cohabitation—whether respondents are cohabiting or not—can mask heterogeneity among cohabiters themselves, grouping those who reside only with a subsequent marital partner with those who bounce from one relationship to the next. Our study also cannot make strong causal attributions on the basis of the empirical analyses presented here. The findings reinforce the conventional view that current patterns of marital dissatisfaction and unhappiness can take on a momentum of their own if they are passed along to children. But we have not controlled for all sources of unobserved variation or heterogeneity among young adults (i.e., the children of the original NSFH respondents). Nor is it likely that we have identified all of the potential pathways that connect the early family experiences with later young adult marital outcomes. This study thus provides a point of departure for new studies that shifts the focus from intergenerational patterns of divorce to other increasingly prevalent familial processes, including parental cohabitation and remarriage. Our study has also centered on the adult children born to a specific cohort of parents. The intergenerational lessons drawn from the adult children of the NSFH’s original sample of parents may or may not apply to current or future generations. As cohabitation and remarriage become institutionalized in American society, more commonplace, and less highly selective of at-risk populations, intergenerational correlations between parental and filial generations may weaken. Indeed, our empirical results suggest that this may already be happening. To sum up, whereas much research has focused on the potentially deleterious short- and long-term effects of divorce on children, we have placed emphasis on the subsequent union transitions of divorced parents and their implications for their adult children. Previous studies have documented that parental remarriage is often beneficial to children. Our results provide additional confirmation to this finding, at least with respect to children’s own intimate relationships and the confidence they have in the stability of their own unions. Nonetheless, not all remarriages are equivalent. Parental cohabi­ tation prior to remarriage, which is an increasingly normative steppingstone to marriage, conveys a message to children that is quite different from direct transitions from divorce to remarriage. Children whose parents cohabited prior to remarrying are more likely themselves to cohabit (relative

Downloaded from http://jfi.sagepub.com at CORNELL UNIV on May 12, 2009

782   Journal of Family Issues

to being married), and if they are married, they also express less confidence in the strength of their own unions and lower levels of satisfaction. To be sure, the specter of divorce looms large over all young adults. But relationship uncertainty is more likely to be expressed by those who have experienced parental divorce and the formation of a new relationship that is less committed, such as cohabitation. This raises new questions about the conventional wisdom regarding the inevitability of the intergenerational transmission of family structure. The data presented here suggest a far more complex picture.

Notes   1. At the initial (Wave 1) interview, one “focal” child was randomly selected from among eligible children in each household in which the main respondent had any biological, adopted, step-, or foster children who were under the age of 18.   2. An additional 174 children were omitted. A total of 130 were stepchildren of the respondent or children of the respondent’s live-in partner. Information on the characteristics of the biological parent was therefore missing. Also excluded were small numbers of children whose mother was never married at the time of the child’s birth.   3. We also are unable to examine the number of union transitions. National Survey of Families and Households (NSFH) parents experienced a multitude of union transitions. To limit the number of less populated categories and to ensure relatively stable parameter estimates in our models, we restrict the analysis to coresidential relationships (distinguishing between unions that subsequently were legalized via marriage or remained cohabitations). Although a substantial share of these subsequent unions dissolved, small sample sizes prevent us from additional analyses of the impact of multiple relationship transitions. Twelve of the 136 divorced parents who remarried a partner without first living with him or her had divorced by the second interview, as did 21 of the 162 divorced parents who cohabited and then remarried.   4. We also included a measure of maternal age at first marriage. The main effect of including this variable was to increase the likelihood that young adults were not in a romantic relationship, while decreasing the impact of mother’s having a college education so that it was only weakly significant. Later maternal age at marriage also increases the likelihood of cohabiting over being married, at the .10 level. No other variables change substantively. We therefore elect not to include maternal age at first marriage in the analyses, as we do not estimate marriage number, and it is associated with maternal education.   5. We used responses to this question from Wave 3. Although the question was asked at Wave 2, the considerable number of children who did not respond makes reliance on the third wave of data collection preferable. Although siblings may have been born when the young adult was older, or out of the house, such births could also condition relationships with parents (and may be more likely among parents experiencing union disruption).   6. An additional measure of self-efficacy was also constructed but had a low Cronbach’s α (.41).   7. Questions were asked separately of younger and older children. The questions we rely on to construct these measures are identical for both groups—unlike many of the other questions that could be used to ascertain emotional well-being.

Downloaded from http://jfi.sagepub.com at CORNELL UNIV on May 12, 2009

Sassler et  al. / Patterns of Union Formation   783

  8. Preliminary analyses (not shown) indicate that children missing responses to these questions were substantially less likely to be from families with married parents at Wave 1, and more likely to have experienced divorce and have cohabiting parents, than children whose responses were recorded. A total of 398 respondents were missing data on self-esteem (179 women and 219 men), while 356 did not respond to the question on global well-being (199 women and 156 men). Setting respondents to the mean according to parental union transitions did not substantially alter the coefficients from the model that set missing responses to the neutral category of 2.5 and included a dummy variable in the model to indicate item nonresponse.   9. About 11% of the sample (n = 188) report never having had sex, while 115 refused to answer the question. To keep the virgins in the sample, we set their mean age at sex equal to their current age plus 1. Because a substantial share of those refusing to answer the question on age at first sex were either cohabiting or married, we set them equal to the mean for their union status. 10. Means by young adults’ relationship status are available by request. 11. These outcomes are treated as separate risks. Unless noted, the married group serves as the reference category. 12. Although this figure seems low, many of the children are still quite young and may yet experience the dissolution of their parents’ marriage. At the time of the second interview, children in this sample were as young as 8. 13. Further comparisons (not shown) also reveal no significant differences in the likelihood of being in cohabiting or dating relationships when parents remained single or cohabited. 14. When the satisfaction scale for married and cohabiting young adults is limited to the four dimensions available for the steady daters, parental union transitions no longer exert a significant effect on either married persons’ or cohabiters’ relationship satisfaction. Additional analyses pooling married, cohabiting, and dating young adults’ responses to these four dimensions of satisfaction reveal no significant differences across groups. 15. Additional analyses (not shown) also revealed no significant differences in the levels of disagreement reported by married and cohabiting respondents. Cohabiters typically have many fewer common domains about which to argue (e.g., child rearing, spending and saving, and in-laws), yet they disagree as much as married couples. In this case, the similarity between cohabiting and married couples is telling.

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., & Cheadle, J. (2005). The long reach of divorce: Divorce and child well-being across three generations. Journal of Marriage and Family, 67, 191-206. Andresen, E. M., Malmgren, J. A., Carter, W. B., & Patrick, D. L. (1994). Screening for depression in well older adults: Evaluation of a short form of the CES-D (Center for Epidemiologic Studies Depression Scale). American Journal of Preventive Medicine, 10, 77-84. Aquilino, W. S. (1996). The life course of children born to unmarried mothers: Childhood living arrangements and young adult outcomes. Journal of Marriage and the Family, 58, 293-310.

Downloaded from http://jfi.sagepub.com at CORNELL UNIV on May 12, 2009

784   Journal of Family Issues

Axinn, W. G., & Thornton, A. (1992). The relationship between cohabitation and divorce: Selectivity or causal influence? Demography, 29, 357-374. Axinn, W. G., & Thornton, A. (1993). Mothers, children, and cohabitation: The intergenerational effects of attitudes and behavior. American Sociological Review, 58, 233-246. Axinn, W. G., & Thornton, A. (1996). The influence of parents’ marital dissolutions on children’s attitudes toward family formation. Demography, 33, 66-81. Brotherson, S. E., & Duncan, W. C. (2004). Rebinding the ties that bind: Government efforts to preserve and promote marriage. Family Relations, 53, 459-468. Brown, S. L. (2004). Family structure and child well-being: The significance of parental cohabitation. Journal of Marriage and Family, 66, 351-368. Brown, S. L. (2006). Family structure transitions and adolescent well-being. Demography, 43, 447-461. Bumpass, L. L. (1984). Children and marital disruption: A replication and update. Demography, 71, 71-82. Bumpass, L. L., & Lu, H. H. (2000). Trends in cohabitation and implications for children’s family contexts. Population Studies, 54, 29-41. Cherlin, A. J. (2004). The deinstitutionalization of American marriage. Journal of Marriage and Family, 66, 848-861. Cherlin, A. J., Kiernan, K. E., & Chase-Lansdale, P. L. (1995). Parental divorce in childhood and demographic outcomes in young adulthood. Demography, 32, 299-318. Clarkberg, M. (1999). The price of partnering: The role of economic well-being in young adults’ union experiences. Social Forces, 77, 945-968. Clarke, S. C. (1995). Advance report of final divorce statistics, 1989 and 1990. Monthly Vital Statistics Report, 43(9 Suppl.). Cooksey, E. C. (1997). Consequences of young mothers’ marital histories for children’s cognitive development. Journal of Marriage and the Family, 59, 245-261. Cooksey, E. C., Mott, F. L., & Neubauer, S. A. (2002). Friendships and early relationships: Links to sexual initiation among American adolescents born to young mothers. Perspectives on Sexual and Reproductive Health, 34, 118-126. Cunningham, M., & Thornton, A. (2006). The influence of parents’ marital quality on adult children’s attitudes toward marriage and its alternative: Main and moderating effects. Demography, 43, 659-672. Dunifon, R., & Kowaleski-Jones, L. (2002). Who’s in the house? Race differences in cohabitation, single parenthood, and child development. Child Development, 73, 1249-1265. Fomby, P., & Cherlin, A. J. (2007). Family instability and child well-being. American Sociological Review, 72, 181-204. Frisch, M., & Hviid, A. (2006). Childhood family correlates of heterosexual and homosexual marriages: A national cohort study of two million Danes. Archives of Sexual Behavior, 35, 533-547. Goldscheider, F. K., & Goldscheider, C. (1998). The effects of childhood family structure on leaving and returning home. Journal of Marriage and the Family, 60, 745-756. Goldscheider, F. K., & Sassler, S. (2006). Creating step-families: Integrating children into the study of union formation. Journal of Marriage and Family, 68, 1-17. Goldscheider, F. K., & Waite, L. J. (1986). Sex differences in the entry into marriage. American Journal of Sociology, 92, 91-109. Graefe, D. R., & Lichter, D. T. (1999). Life course transitions of American children: Parental cohabitation, marriage, and single motherhood. Demography, 36, 205-217.

Downloaded from http://jfi.sagepub.com at CORNELL UNIV on May 12, 2009

Sassler et  al. / Patterns of Union Formation   785

Heuveline, P., & Timberlake, J. M. (2004). The role of cohabitation in family formation: The United States in comparative perspective. Journal of Marriage and Family, 66, 1214-1230. Hofferth, S. L., & Anderson, K. G. (2003). Are all dads equal? Biology versus marriage as a basis for paternal investment. Journal of Marriage and Family, 65, 213-232. Kapinus, C. A. (2005). The effect of parental marital quality on young adults’ attitudes toward divorce. Sociological Perspectives, 48, 319-335. Kiernan, K. E. (1992). The impact of family disruption in childhood on transitions made in young adult life. Population Studies, 46, 213-224. Lichter, D. T., & Qian, Z. (2004). Marriage and family in a multiracial society. New York: Russell Sage Foundation; Washington, DC: Population Reference Bureau. Lichter, D. T., Qian, Z., & Mellott, L. (2006). Marriage or dissolution? Union transitions among poor cohabiting women. Demography, 43, 223-240. Lichter, D. T. & Qian, Z. (2008). Serial cohabitation and the marital life course. Journal of Marriage and Family, 70, 861-878. London, K. A. (1989). Children of divorce (National Vital Statistics Series 21, No. 46; DHHS Pub. No. (PHS) 89–1294). Washington, DC: Government Printing Office. Maddala, G. S. (1983). Limited-dependent and qualitative variables in economics. Cambridge, UK: Cambridge University Press. Manning, W. D., & Lamb, K. A. (2003). Adolescent well-being in cohabiting, married, and single-parent families. Journal of Marriage and Family, 68, 876-894. Manning, W. D., & Lichter, D. T. (1996). Parental cohabitation and children’s economic wellbeing. Journal of Marriage and the Family, 58, 998-1010. Manning, W. D., Longmore, M. A., & Giordano, P. C. (2007). Changing institution of marriage: Adolescents’ expectations to cohabit or to marry. Journal of Marriage and Family, 69, 559-576. Martin, G., & Kats, V. (2003, September). Families and work in transition in 12 countries, 1980-2001. Monthly Labor Review, 1-29. McLanahan, S., & Sandefur, G. (1994). Growing up with a single parent: What hurts, what helps. Cambridge, MA: Harvard University Press. McLeod, J. D. (1991). Childhood parental loss and adult depression. Journal of Health and Social Behavior, 32, 205-220. Moore, K. A., Bronte-Tinkew, J., Jekielek, S., Guzman, L., Ryan, S., Redd, Z., et  al. (2007). Developing measures of healthy marriages and relationships. In S. L. Hofferth & L. M. Casper (Eds.), Handbook of measurement issues in family research (pp. 101-121). Mahwah, NJ: Lawrence Erlbaum. Morrison, D. R., & Ritualo, A. (2000). Routes to children’s economic recovery after divorce: Are cohabitation and remarriage equivalent? American Sociological Review, 65, 560-580. Oppenheimer, V. K. (1988). A theory of marriage timing. American Journal of Sociology, 94, 563-591. Raley, R. K. (1996). A shortage of marriageable men? A note on the role of cohabitation in Black-White differences in marriage rates. American Sociological Review, 61, 973-983. Raley, R. K., & Wildsmith, E. (2004). Cohabitation and children’s family instability. Journal of Marriage and Family, 66, 210-219. Rogers, S. J., & Amato, P. R. (1997). Is marital quality declining? The evidence from two generations. Social Forces, 75, 1089-1100. Ross, C. E., & Mirowsky, J. (1999). Parental divorce, life course disruption, and adult depression. Journal of Marriage and the Family, 61, 1034-1045.

Downloaded from http://jfi.sagepub.com at CORNELL UNIV on May 12, 2009

786   Journal of Family Issues

Sassler, S., & 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. Sassler, S., & Kamp Dush, C. (2007, August). The pace of relationship progression: Does timing to sexual involvement matter? Paper presented at the annual meeting of the American Sociological Association, New York. Schoen, R., & Standish, N. (2001). The retrenchment of marriage: Results from marital status life tables for the United States, 1995. Population & Development Review, 27, 553-563. Stewart, S., Manning, W. D., & Smock, P. J. (2003). Union formation among men in the U.S.: Does having prior children matter? Journal of Marriage and Family, 65, 90-104. Stewart, S. D. (2001). Contemporary American stepparenthood: Integrating cohabiting and nonresident stepparents. Population Research and Policy Review, 20, 345-364. Sweet, J. A., Bumpass, L. L, & Call, V. (1988). The design and content of the National Survey of Families and Households (NSFH Working Paper No. 1). Madison: University of Wisconsin. Teachman, J. D. (2002). Childhood living arrangements and the intergenerational transmission of divorce. Journal of Marriage and Family, 64, 717-729. Teachman, J. D. (2003). Childhood living arrangements and the formation of coresidential unions. Journal of Marriage and Family, 65, 507-524. Teachman, J. D. (2004). The childhood living arrangements of children and the characteristics of their marriages. Journal of Family Issues, 25, 86-111. Thornton, A. (1991). Influence of the marital history of parents on the marital and cohabitational experiences of children. American Journal of Sociology, 96, 868-894. Thornton, A., Axinn, W., & Teachman, J. D. (1995). The influence of school enrollment and accumulation on cohabitation and marriage in early adulthood. American Sociological Review, 60, 762-774. Thornton, A., & Young-DeMarco, L. (2001). Four decades of trends in attitudes toward family issues in the United States: The 1960s through the 1990s. Journal of Marriage and Family, 63, 1009-1037. Upchurch, D. M., Levy-Storms, L., Sucoff, C. A., & Aneshensel, C. S. (1998). Gender and ethnic differences in the timing of first sexual intercourse. Family Planning Perspectives, 30(3), 121-127. Waite, L., & Spitze, G. D. (1981). Young women’s transition to marriage. Demography, 18, 681-694. Webster, P. S., Orbuch, T. L., & House, J. S. (1995). Effects of childhood family background and adult marital quality and perceived stability. American Journal of Sociology, 101, 404-432. Wolfinger, N. H. (2000). Beyond the intergenerational transmission of divorce: Do people replicate the patterns of marital instability they grew up with? Journal of Family Issues, 21, 1061-1086. Wolfinger, N. H. (2003). Parental divorce and offspring marriage: Early or late? Social Forces, 82, 337-353. Wu, L. L., & Martinson, B. C. (1993). Family structure and the risk of a pre-marital birth. American Sociological Review, 58, 210-232.

Downloaded from http://jfi.sagepub.com at CORNELL UNIV on May 12, 2009