Abstract

The rapid growth in cohabitation over the past quarter-century necessitates studies of changes in the stability and outcomes of cohabitation. We utilized data from the 1988 National Survey of Family Growth (NSFG) and the most recent NSFG data from 2011–2013 to examine the outcomes of two comparable cohorts of first premarital cohabiting women (1983–1988 and 2006–2013). Our results showed that cohabitations formed between 2006 and 2013 lasted longer—18 months, on average—than those formed in the mid-1980s, which lasted for an average of 12 months. We found that the lengthening of cohabitation over time cuts across sociodemographic characteristics—race/ethnicity, education, and motherhood status—and resulted mostly from the declining rate of transitioning to marriage. We found some support for the diverging destinies perspective in that disparities in the outcomes of cohabitation by education and by cohabiting birth have widened over time. Our analyses showed that changes in the outcomes of first premarital cohabiting unions over the past three decades were not due to compositional shifts in cohabitors. These results demonstrate the evolving dynamics of cohabitation over a 30-year window.

Introduction

In the past quarter-century, one of the most striking family changes in the United States has been the rapid growth in cohabitation. The share of American women aged 30–34 who had cohabited at least once nearly doubled, from 40 % in 1987 to 75 % in 2013 (Manning and Stykes 2015). Since the early 1990s, cohabitation has become the dominant pathway to forming a first coresidential union (Manning 2013). A defining feature of cohabitation in the United States is its relatively short duration (Heuveline and Timberlake 2004): cohabiting unions last less than two years, on average (Copen et al. 2013). Given this growth in cohabitation, a key question is whether there have been changes in the stability and pathways out of cohabiting unions (marriage or dissolution).

Despite the growth in cohabitation, the change has been uneven across sociodemographic groups, resulting in a shift in the composition of cohabitors: increasing childbearing in cohabiting unions, larger increases in cohabitation among whites and Hispanics, and greater increases in cohabitation among the modestly educated (Bumpass and Lu 2000; Kennedy and Bumpass 2008; Kuo and Raley 2016: Manning et al. 2015). We drew on two key perspectives used to assess family change, diverging destinies, and diffusion (Liefbroer and Dourleijn 2006; McLanahan 2004) to examine changes in cohabitation. We examined how the changing composition of cohabitors is associated with shifts in the stability of first premarital cohabiting unions and whether the sociodemographic divide in cohabiting outcomes is converging or diverging over time.

Unlike previous research on the stability of cohabitation, the present study compared two cohorts of first premarital cohabitors spanning a 30-year period (1983–1988 and 2006–2013). For the early cohort, cohabitation was relatively uncommon (with only 35 % of women having ever cohabited), and the late Baby Boom birth cohort was in their 20s. The recent cohort represents a period when cohabitation was common (with 65 % of women having ever cohabited) and Millennials were in their 20s. To construct these cohabitation cohorts, we relied on the 1988 and 2011–2013 National Survey of Family Growth (NSFG). We utilized life table techniques and event history models as well as regression decomposition to analyze the trends and relative contributions of compositional factors to changes in the stability of first premarital cohabitation over the past three decades. Understanding the trends in stability and transitions from premarital cohabitation to either marriage or separation provides a lens into the evolving relationship between cohabitation and marriage. Further, given that cohabitation is increasingly a context of childbearing and childrearing, it is important that researchers consider how the stability of cohabiting unions differs for those with and those without children.

Background

The growth in cohabitation over the past quarter-century, popularly referred to as the cohabitation revolution (Smock and Manning 2010), has been well documented. Nearly two-thirds (65 %) of women aged 19–44 had experienced cohabitation in 2013, compared with only one-third (33 %) in 1987. Similarly, 69 % of recent marriages among women aged 19–44 were preceded by cohabitation, a significant increase from 41 % in 1987 (Kennedy and Bumpass 2011; Manning and Stykes 2015). Scholarly interest in cohabitation is high partly because of the pace of change in cohabitation, which has shifted from a minority to majority experience over a short time span. However, a challenge in studying cohabitation is that although it has become widespread, it does not last long on average. The median duration of first premarital cohabitation among women aged 15–44 in the 2006–2010 NSFG is less than two years (Copen et al. 2013). Thus, at any given point in time, there may not be many individuals cohabiting, but a high proportion have spent some time in cohabiting union(s) (i.e., they ever cohabited). Although cohabitation tends to be short-lived, it nevertheless can have a substantial influence on the family lives of adults and children.

Previous studies have documented shifts in union transitions that point to changing stability levels of cohabitation. In the 1980s, cohabiting unions more often ended in marriage than dissolution (Bumpass and Sweet 1989). Researchers have reported a reversal of this trend by the late 1990s as cohabiting unions less often transitioned to marriage (Guzzo 2014; Lichter et al. 2006). Since the late 1990s, the lengthening of cohabitation (Kennedy and Bumpass 2008, 2011) has coincided with a continued decline in the rate of transition to marriage from premarital cohabiting unions (Guzzo 2014; Kuo and Raley 2016). This study reexamines changes in cohabitation outcomes over the past 30 years, exploring the notion of increasing institutionalization of cohabiting unions as an acceptable family arrangement, as cohabitation has become ubiquitous. According to the institutionalization perspective (Cherlin 2004; Heuveline and Timberlake 2004; Kiernan 2004), lengthening of cohabiting relationships suggests that the meaning of cohabitation is changing with the increasing prevalence of cohabiting unions.

A contribution of our study is to empirically evaluate the role of compositional changes in union transitions among women ages 15–39 in two key cohabitation cohorts (1983–1988 and 2006–2013). According to the diffusion perspective (Liefbroer and Dourleijn 2006), the risk of union instability among cohabitors depends on the extent to which cohabitation has spread within a society; the stability of cohabiting unions increases with increasing popularity of cohabitation. Possible explanations include reduced selectivity of cohabitors relative to noncohabitors and changing norms regarding unmarried cohabitation. Based on the diffusion perspective, cohabitors and noncohabitors become increasingly similar with the growing prevalence of cohabitation. Early on, cohabitors were more select in terms of their less-traditional attitudes and values, weaker commitment to marriage or higher expectations of romantic relationships, and precarious socioeconomic conditions (Axinn and Thornton 1992; Bumpass et al. 1991; DeMaris and Rao 1992; Manning and Lichter 1996). With cohabitation becoming more common and now a normative part of the family life course, individuals who cohabit are less select than they were a quarter-century ago. Also, societal norms regarding the practice of unmarried cohabitation may shift such that unmarried cohabitors are less stigmatized and less pressured to transition to marriage or dissolve their unions. The increasing popularity of cohabitation may also make it normative to have children within cohabiting relationships. The spread of cohabitation across the U.S. population could have resulted from a number of factors, including increased acceptance of cohabitation, generational changes, and/or postponement of marriage with increasing emphasis on economic buoyancy as a requisite for marriage. The demographic characteristics of cohabitors could have also shifted in response to compositional changes in the larger population.

An assessment of change in the stability of cohabitation requires attention to shifts in the composition of cohabitors over time. Although cohabitation increased across racial/ethnic groups between the 1987 National Survey of Families and Households and the 2011–2013 NSFG surveys, the greatest increase in cohabitation was among Hispanics (Manning and Stykes 2015). Cohabitation is increasingly common among all racial minorities (Bumpass and Sweet 1989; Copen et al. 2013; Manning et al. 2014) but is a steppingstone to marriage more so for whites than for blacks (Brown 2000; Guzzo 2009; Lichter et al. 2006; Manning and Smock 1995). Cohabitation is a more typical pathway to parenthood and a more common family context for raising children among Hispanics and blacks compared with whites (Manning 2001; Manning et al. 2015; Musick 2002; Wildsmith and Raley 2006). As such, cohabitation tends to last longer among Hispanic and black women (Copen et al. 2013; Kennedy and Bumpass 2008).

The experiences of cohabitation in the United States vary by social class. Cohabitation has been described as a more economical route to forming a coresidential union (Furstenberg 1996). Also, economic resources, particularly the male partner’s economic stability, promote the transition to marriage among cohabiting couples (Smock and Manning 1997; Smock et al. 2005). Middle-class cohabitors are more likely to be engaged to their cohabiting partners than their working-class counterparts (Sassler and Miller 2011). The present study focuses on educational attainment as an indicator of social class. Data limitations preclude the inclusion of other indicators. Adults with lower levels of education are more likely to cohabit than those with a college degree, and the educational gap in cohabitation experiences of women in the United States has widened over the past few decades. In 1987, among women aged 19–44, 43 % of those with less than a high school diploma and 31 % of those with a college degree had ever cohabited. In a recent period (2011–2013), more than three-quarters (76 %) of women with less than a high school diploma had ever cohabited compared with less than one-half (42 %) of college-educated women (Manning and Stykes 2015). Relative to no high school diploma, having a high school diploma or higher is positively associated with marriage among cohabitors (Carlson et al. 2004; Guzzo 2014; Kennedy and Bumpass 2008). A college education is associated with the highest odds of marriage among cohabiting women (Copen et al. 2013). College-educated women rarely have a child while cohabiting, in contrast to nearly one in three women with a high school diploma (Manning et al. 2015).

An important shift in cohabiting unions is the presence of children. Nearly one-half of children are expected to spend some time in a cohabiting family (Brown et al. 2016). The share of births to unmarried women in the United States has doubled since the 1980s, and nearly all the increase in nonmarital childbearing over the past few decades was due to increasing births among cohabiting women (Lichter et al. 2014; Manning et al. 2015). In the early 1980s, only 6 % of children were born to cohabiting parents, but recently as many as one-quarter of American children (25 %) were born to cohabiting parents (Manning et al. 2015). As a form of relationship-specific capital, children may act as a deterrent against separation and help to cement the relationship (Becker 1990; Manning 2004; Wu 1995). Alternatively, children can be a source of strain and stress, resulting in potentially greater levels of union instability (Evenson and Simon 2005; Nomaguchi and Milkie 2003). Children who were born prior to cohabitation may be destabilizing because they are not the biological offspring of both parents, resulting in potential tensions caused by ambiguous roles, financial obligations, and expectations among stepfathers and stepchildren (Brown and Manning 2009; Mahoney 2006; Marsiglio 2004; Sweeney 2010; Teachman 2008). Although several studies have considered the outcomes of cohabiting unions involving children (Lichter et al. 2016; Musick and Michelmore 2015), the influence of parenthood status on the outcome of cohabitation is less researched. We examine the role of three compositional shifts (race/ethnicity, education, and parenthood status) in changes in outcomes of first premarital cohabitation.

This study further examines patterns of change in the outcomes of cohabitation across sociodemographic characteristics. Prior work points to variation in cohabitation by race/ethnicity, education, and parenthood status (Brown 2000; Carlson et al. 2004; Guzzo 2009; Kennedy and Bumpass 2008; Kuo and Raley 2016; Lichter et al. 2006; Manning and Smock 2002), and we examine whether union outcomes across these dimensions are converging or diverging. The changing composition of cohabitors in terms of education, race/ethnicity, and parenthood status over the past few decades suggests potential sources of shifts in the outcomes of cohabiting unions. Thus, based on the diffusion perspective, we expect that patterns of transitions out of cohabitation may have converged across these demographic characteristics (education, race/ethnicity, and parenthood status) with declining selectivity. Early adopters of cohabitation in the United States were economically disadvantaged (Bumpass and Sweet 1989), less likely to be of Hispanic origin (Manning and Stykes 2015), and were mostly nonparents (Manning et al. 2015). The evolution of cohabitation from a minority to a majority experience suggests increasing similarities in the outcomes of cohabitation across groups.

Conversely, given the roles of race/ethnicity, education, and parenthood status in access to socioeconomic resources in the United States, a diverging destinies approach (McLanahan 2004) predicts growing disparity in cohabiting union transitions across racial/ethnic groups, levels of education, and parenthood statuses. The diverging destinies perspective postulates social class variations in changes in cohabitation and, therefore, outcomes of cohabiting unions. Scholars have argued that economic changes over the past few decades (e.g., disappearance of manufacturing jobs) disproportionately affected American men with no college degree, thereby precipitating what Andrew Cherlin described as an hourglass economy. Marriage has also become a status symbol, with the prerequisites for marriage (e.g., assets) increasingly beyond the reach of many working-class Americans (Cherlin 2014). In view of these changes, the diverging destinies perspective argues that changes in family patterns have been greater among the more economically disadvantaged than among those with more economic resources (McLanahan 2004). According to this view, the forces driving the trends of the second demographic transition (e.g., increasing cohabitation and unmarried childbearing) are producing different family experiences across social class (McLanahan 2004). Thus, the growth in cohabitation in the United States has been uneven across sociodemographic groups (Gibson-Davis and Rackin 2014; Goldstein and Kenney 2001; Manning et al. 2014; Raley 2000).

Current Investigation

The aim of our study is to establish the trends in stability and transitions out of cohabiting unions over the past 30 years. We focused on whether there has been convergence (diffusion perspective) or divergence (diverging destinies perspective) in four compositional features of cohabiting women: race/ethnicity, education, parenthood, and age at cohabitation. Relying on life table, event history, and regression decomposition and standardization techniques, we determined the probability of transitioning out of first premarital cohabitation into either marriage or separation relative to continuing to cohabit for five years. Our analyses accounted for key correlates associated with cohabitors’ union outcomes: family structure while growing up, nativity status, and age at first sex. Previous research has documented differences in outcomes of cohabitation by family background and by nativity status (Guzzo 2014). Also, sexual experiences in adolescence are significantly related to cohabitation experiences in early adulthood (Meier and Allen 2009; Raley et al. 2007). We documented patterns and differences in the stability of cohabiting unions during a period of rapid socioeconomic and attitudinal changes in the United States: 1983–2013.

Although researchers have established patterns of cohabitors’ union transitions and stability at specific time points (Kuo and Raley 2016), few have empirically examined how the compositional changes in cohabitation influence cohabitation outcomes over a critical time span (30 years). Assessments of diffusion and diverging destinies perspectives require analyses across cohorts. We conducted tests for significant intercohort variation in the effects of the correlates of stability of first premarital cohabitation. Because women in each cohort were sampled relatively close to the date of their cohabitations, we hoped to adequately capture the stability of cohabiting unions in each period. Further, the stability of cohabiting unions with children has received empirical attention (Lichter et al. 2016; Musick and Michelmore 2015). We extended the literature by focusing on the experiences of cohabiting unions with and without children. This distinction by motherhood status is critical because although cohabitation is increasingly a family context for children (Manning et al. 2015), the majority of cohabiting couples do not have children.

Data and Methods

We used data from Cycle 4 (1988) of the National Survey of Family Growth (NSFG) and the most recent 2011–2013 NSFG data. The NSFG is a repeated cross-sectional household survey with a nationally representative sample of reproductive-aged women (aged 15–44) in the United States. The NSFG provides information about union formation, union dissolution, fertility patterns, and other aspects of family life in the United States. Details about the design of the NSFG and its data collection procedures have been documented elsewhere (see U.S. Department of Health and Human Services 1994, 2014).

The NSFG interviewed 8,450 noninstitutionalized women between January and August 1988 for Cycle 4. There were 3,032 women who had cohabited with at least one partner. Of those ever-cohabited women, 682 experienced their first nonmarital cohabiting union after having been married and divorced and thus were excluded from our analysis. We limited our sample to a single cohort of first premarital cohabitors: 742 women who formed their unions within five years of the survey, between January 1983 and the date of interview (January to August 1988). As in previous studies (e.g., Lichter et al. 2006), we relied on a five-year period to minimize problems of age truncation and underreporting of cohabitation, which increases over time (Hayford and Morgan 2008).

Our goal was to analyze change in the stability of cohabitation between the 1980s when cohabitation was still a minority experience (fewer than one-half of U.S. adults had ever cohabited) and the recent years when majority (roughly two-thirds) of women had experienced cohabitation. In these two periods, late Baby Boomers and Millennials, respectively, were roughly in their 20s. To capture the changes in the stability of first premarital cohabitation over a period of 30 years, we used the 2011–2013 NSFG to estimate the duration of premarital cohabiting unions formed by women between 2006 and 2013.

Between September 2011 and September 2013, the NSFG interviewed 5,601 women, of whom 2,934 cohabited prior to first marriage. We sampled 794 women (aged 15–44) who formed their first premarital cohabiting unions within five years of their interviews using January as the benchmark as in the earlier cohort (i.e., January 2006 to month of interview in 2011 for those interviewed in 2011, and January 2008 to interview month in 2013 for those interviewed in 2013). Our approach was to analyze data collected near the referent period to minimize problems of recall of cohabitation start and end dates as well as to ensure cohabitation cohorts include the full age range.1 We excluded from our sample 7 respondents with missing age at first sex, 14 respondents who began cohabiting before reaching age 15, and 12 respondents aged 40 or older at the time of interview. The upper age limit of the sample meant that women over age 40 were not captured in the starting points of our cohabitation cohorts. Thus, our final analytic sample comprises 1,503 women aged 15–39 (729 in the 1983–1988 cohort, and 774 in the 2006–2013 cohort).

In both NSFG surveys, women provided detailed histories of their union formations and dissolutions, including the start and end dates of each union. The duration of first premarital cohabitation equals the difference (in months) between the start and end dates of the cohabiting union corresponding to a woman’s first premarital cohabitation. Outcomes of first premarital cohabiting union are intact cohabitation, marriage, or dissolved cohabitation (i.e., separation).

Respondents’ racial/ethnic categories include Hispanic, non-Hispanic white, non-Hispanic black, and “non-Hispanic other or multiple” race/ethnicity. Although we included the “other” racial/ethnic category in our analyses, we report the stability of first premarital cohabiting unions for Hispanic, non-Hispanic white (reference), and non-Hispanic black women. There were only 19 (2.6 %) women of “other” race/ethnicity in the 1983–1988 cohort and 62 (8.1 %) in the 2006–2013 cohort.

We analyzed four educational categories: less than high school diploma or GED, high school diploma or GED, some college (including two-year degree), and college degree or higher (reference). Education was established at the time of interview and may not reflect respondents’ education at start of cohabitation. Our indicator is a proxy given that the NSFG does not include full education histories. The recent NSFG includes measures of timing of high school as well as college graduation, and we found that the completed education level corresponds to education at time of interview.2

Birth timing is measured in three categories: (1) no birth before and during cohabitation (reference), (2) any birth during cohabitation, and (3) any birth prior to (but not during) cohabitation. The first category is cohabiting women with no birth (that is, zero parity), hereafter referred to as childless cohabitors. In the second category are women who had a birth during first cohabitation if the date of the outcome of any reported pregnancy that resulted in a live birth was between the start and end dates of the first premarital cohabitation. Also included in this second category are women who in addition to giving birth during cohabitation also had another childbirth before cohabitation. Women whose first biological child was born prior to the start date of their first premarital cohabitation and who did not report any other birth occurring while cohabiting were put in the third category of birth before cohabitation.

The multivariate models included key control variables. To capture family history, we included an indicator of living arrangement at age 14, coded as 1 if a woman reported any living arrangement other than with both biological/adoptive parents at age 14, and 0 otherwise. Nativity status is a binary variable coded as 1 if a woman was born outside the United States, and 0 otherwise. We categorized our sample into three groups for age at first premarital cohabitation: less than 20, 20–24 (reference), and 25–39. Age at first sex is a dummy variable coded as 1 if the respondent had sex before reaching age 16 (early sex), and 0 otherwise (later sex).

We compared first premarital cohabiting unions formed between 1983 and 1988 with those formed about three decades later between 2006 and 2013. First, we described the characteristics of women in the two cohabitation cohorts. Then we applied the techniques of multiple decrement life tables to estimate women’s probability of transitioning from first premarital cohabiting union to marriage or separation. We tracked each cohabiting union for a period of five years to see whether it remained intact (was censored), dissolved, or transitioned to marriage. An individual was censored if she remained in her first premarital cohabitation until the end of the fifth year or until the date of the interview.

In the second part of our analyses, we used a series of discrete-time multinomial logistic regression models to estimate women’s risks of marrying or separating from their first premarital cohabiting partner. Model 1 includes the cohort indicator, and Model 2 includes both the cohort measure and other predictors of the stability of cohabiting unions. We tested for variation in the associations between the sociodemographic indicators by cohabitation cohort.

The last set of analyses assessed the association between compositional factors and shifts in the outcomes of cohabiting unions. We estimated monthly conditional probabilities of transitioning from first premarital cohabitation to marriage or separation while varying the sociodemographic characteristics of cohabiting women (i.e., while holding the covariates for different cohorts at their weighted mean values). We then used the monthly conditional probabilities to estimate the cumulative predicted probabilities of transitioning to marriage and separation within five years of cohabiting at the different levels of sociodemographic characteristics.

All models accounted for duration of the cohabiting union. Each woman with a history of first premarital cohabitation contributed person-months, measured in discrete intervals of months from the start date of her cohabitation until the date she married, dissolved the union, or was censored. The 1,503 women in our sample contributed 25,251 person-months of data. All analyses were weighted to account for unequal probability of selection into the sample and to adjust for differential coverage and response rates.

Results

Table 1 presents the descriptive statistics of our focal variables by cohabitation cohort and showcases the significant compositional shifts in the cohabiting women across cohorts. The share of cohabiting women who were Hispanic nearly doubled across the period, from 11 % to 21 %; the percentage of blacks remained about the same, at 13 % and 14 %; and there were fewer whites in the recent cohort (73 % vs. 55 %). The educational composition of cohabiting women shifted. More women in the 2006–2013 cohort (28 %) than in the 1983–1988 cohort (20 %) were college graduates. The increased educational attainment among cohabiting women in our sample largely mirrors the general increase in education among women (results not shown). Our study reaffirms the growth in childbearing and child-rearing within cohabiting unions. The share of first premarital cohabiting women with children increased across the two cohorts, from 19 % in 1983–1988 to 30 % in 2006–2013. The rise in child-rearing among first premarital cohabitors over the past three decades is almost entirely due to increasing births within first cohabitation. Only a minority of single mothers (10 % to 11 %) transitioned to cohabiting relationships in both cohorts. Perhaps reflecting increased union instability in their parental generation, fewer cohabiting women in the recent cohabiting cohort (56 %) than in the 1980s cohort (66 %) lived with two biological or adoptive parents until age 14. The share of foreign-born cohabitors in the sample increased from 5 % in the 1980s to 9 % in 2006–2013. Age at first cohabitation changed little across cohorts. The share of premarital cohabitors who experienced early sex increased from 25 % in the 1983–1988 cohort to 35 % in the 2006–2013 cohort.

Table 2 presents our life table estimates of transitions out of first premarital cohabiting unions formed between 1983–1988 and 2006–2013. Although still relatively short-lived, cohabiting unions are lasting longer nowadays in the United States. Less than one-half of the earlier cohort of cohabitors celebrated their one-year anniversary, but two-thirds of the more recent cohabitation cohort did so. Similarly, the proportion of cohabiting unions surviving until the end of the fifth year nearly doubled from 23 % in the 1980s to 43 % in 2006–2013 (Table 2, panel A). The lengthening of first premarital cohabitation over the past three decades cuts across sociodemographic groups. The rate of transitioning to marriage among cohabitors declined over time. More than two out of every five women (42 %) married their first cohabiting partners within 5 years in the 1980s, but only one in five women (22 %) did so approximately 30 years later. Between 1983–1988 and 2006–2013, the probability of dissolving a first premarital cohabitation in the first year fell by 26 %, whereas by the end of the fifth year of cohabiting, similar shares of cohabitors had separated in both cohorts. This suggests that cohabitors were taking relatively longer to separate in the more recent years than in the 1980s. Nevertheless, whereas marriage served as the modal exit from premarital cohabitation in the 1980s, more cohabitors separated than married between 2006 and 2013.

As shown in panel B of Table 2, the increased duration of first premarital cohabitation between 1983–1988 and 2006–2013 was more pronounced for Hispanics than for whites and blacks. In both cohorts, a smaller share of black premarital cohabitors than whites and Hispanics transitioned to marriage. Over time, the modal pathway from first premarital cohabitation changed from dissolution to marriage for Hispanics and from marriage to dissolution for whites. In both periods, more black cohabitors dissolved their unions than married their partners. Considering the relatively small size of Hispanic population in our 1983–1988 cohort (11 %), the racial/ethnic differences should be interpreted with some caution.

The education gap in the share of women transitioning from a first premarital cohabitation to marriage grew over the period of study (panel C of Table 2). The proportion of cohabitors marrying their partners was 164 % higher for college graduates than for high school dropouts in 2006–2013, compared with a much smaller gap in the 1980s (63 % higher level for college graduates). In both periods, more cohabiting unions formed by women with less than a high school diploma ended in separation than transitioned to marriage. Conversely, for college-educated women, greater shares of premarital cohabiting relationships transitioned to marriage than dissolved in both 1983–1988 and 2006–2013. This suggests that among highly educated women in the United States, premarital cohabitation is still typically a prelude to marriage.

Over the study period, a first premarital cohabiting union persisted longer if the woman had one or more children while cohabiting than among childless cohabitors and single mothers who transitioned to cohabiting relationships (panel D of Table 2). Also, in both cohabitation cohorts, fewer women with a birth during cohabitation married their cohabiting partners compared with both childless cohabitors and those who had a biological child at the start date of their first premarital cohabitation. Births within cohabitation appear more likely to delay marriage than do pre-cohabitation births. We found some changes in the role of motherhood on the outcome of first premarital cohabiting unions over time. The gap in the proportions transitioning to marriage between women with cohabiting births and their counterparts with no birth and those with only pre-cohabitation births has widened over time, from less than 20 % in the 1980s to more than 50 % in 2006–2013. Further, in the 1980s, greater shares of childless cohabitors and those with cohabiting births transitioned to marriage than separated. Slightly more women with births before first premarital cohabitation dissolved their unions than married in the 1980s. In the recent cohort (2006–2013), more cohabitors separated than married regardless of their parenthood status.

In the 1980s, more cohabiting unions initiated at younger ages (15–24) than at older ages (25–39) persisted for five years; in the more recent cohort, a larger share of older cohabitors remained with their partners until the end of the fifth year (panel E of Table 2). In both cohorts, teen cohabitors transitioned to marriage at lower rates than women who delayed their first premarital unions until their mid-20s or later. Across the study period, the rate of dissolution increased among teenage cohabitors, remained the same among women who formed their unions in their early to mid-20s, and declined among older cohabitors (aged 25 or older).

Next, we used a series of discrete-time multinomial logistic regression models to examine the relative risks of first premarital cohabiting unions transitioning to marriage or to dissolution. Table 3 presents the results of models predicting the odds of transitioning to marriage or separation from a first premarital cohabiting union versus continuing cohabitation for five years. The results in Model 1 suggest a significant change in the distribution of the outcomes of first premarital cohabitation in the United States between 1983–1988 and 2006–2013. When we controlled for union duration, members of the 2006–2013 cohort of cohabitors were significantly more likely than the 1983–1988 cohort to continue cohabiting with their partners than to either marry or separate. The cohort difference in the risk of transitioning to marriage or dissolution persisted after we accounted for changes in the other predictors of cohabitation outcomes, as shown in Model 2.

The significant correlates of the outcomes of first premarital cohabitation in this study are education, birth timing, foreign-born status, and age at first cohabitation. Compared with those who graduated from college, women with a high school diploma or less education were more likely to dissolve their unions than to continue cohabiting. In a bivariate model (results not shown), the risk of transitioning to marriage was about twice as high for college graduates as their counterparts with no college degree, but this was mostly due to their lower likelihood of having cohabiting births and their delayed union formation.3 Only a minority of college graduates in both cohabiting cohorts (<10 %) had children before or during their first premarital cohabiting relationships. Also, compared with more than two-thirds of those with less than a high school diploma, only 5 % of women with a college degree in both cohabitation cohorts were teenagers (results not shown).

A cohabiting birth tended to prolong a first premarital cohabitation. We found significantly reduced odds of marriage and separation among women who gave birth while cohabiting relative to childless cohabitors, but birth before first premarital cohabitation was not significantly associated with the risks of marriage and dissolution. Foreign-born cohabitors had higher chances of marriage than cohabiting women born in the United States. Teenage cohabitors were more likely to separate from their partners than women who formed their first premarital cohabiting relationships in their early to mid-20s.

We further examined intercohort differences in the predictors of outcomes of premarital cohabitation. Our analyses showed that only the effects of education and birth timing significantly changed across the two cohabitation cohorts. The results of the interactions between respondents’ educational attainment and cohort presented in Model 3 (Table 3) suggest increasing educational divergence in the outcomes of cohabitation over time. Whereas college-educated cohabitors were not significantly different from their counterparts with lower levels of education in their risks of transitioning to marriage and dissolution in the 1980s, having a college degree (relative to less than college education) was associated with significantly higher risks of marriage in 2006–2013. Also, college-educated women had significantly lower risks of dissolution than high school graduates in the 2006–2013 cohort but not in the 1983–1988 cohort. The nonsignificant main effects of cohort in Model 3 indicate no significant change in the outcomes of cohabitation for college-educated women over the study period. Further tests of significant intercohort differences across educational groups showed that the risks of marriage declined significantly over time for all but college-educated women, whereas the risks of dissolution were significantly reduced only among women with less than a high school diploma or some college education (results not shown).

Model 4 (Table 3) shows the results of the interactions between the indicators of birth timing and cohort. The findings suggest an increase in the marriage-inhibiting or marriage-delaying effect of a cohabiting birth over time. Having one or more children while cohabiting was associated with a significantly lower likelihood of marriage in 2006–2013 than in the 1980s. The majority of women who cohabited did not have or raise children in a cohabiting union, and the pathways out of cohabitation shifted such that the risks of continued cohabitation, relative to transitions to marriage and dissolution among women with no children, increased significantly across cohorts.

We estimated the predicted probabilities of transitioning from first premarital cohabiting union to marriage or separation within five years of cohabiting at varying levels of the sociodemographic characteristics. The predicted probabilities presented in Table 4 are based on Model 2 in Table 3 with the covariates held at weighted mean values for each predictor in the model. Had no change in the sociodemographic characteristics of cohabiting women occurred across the two cohorts (all covariates held at their 1983–1988 mean values), the decline in the probability of transitioning from first premarital cohabitation to marriage within five years would have been slightly greater. The predicted probability of marriage would have declined by 42 % (from 53 % in 1983–1988 to 31 % in 2006–2013) as opposed to the observed 38 % decline (from 53 % to 33 %). Similarly, changes in the composition of cohabiting women over the past three decades minimized the increased risks of dissolution of first premarital cohabitation. The 20 % increase in the predicted probability of separation, from 44 % in 1983–1988 to 53 % in 2006–2013, would have been higher (23 %) if there had been no change in the sociodemographic composition of first premarital cohabitors (i.e., all covariates held at their 1983–1988 mean values).

Changes in racial composition across the two cohorts produced minimal change in the outcomes of cohabiting unions, with slightly lower levels of marriage and separation (with race/ethnicity means held at the 1983–1988 levels and other covariates held at the 2006–2013 levels). With regard to women’s educational attainment, the decline in the rate of transitioning to marriage from first premarital cohabitation would have been more pronounced, and there would have been no difference in the probability of separation if there had been no change in the educational attainment of cohabiting women across cohorts. Had no changes in birth timing occurred (with birth timing held at 1983–1988 mean values and other covariates held at the 2006–2013 means), the percentage change in the predicted probability of marriage would have been slightly higher, and the predicted probability of separation would have shifted minimally. Taken together, these standardization results indicate that changes in outcomes of first premarital cohabitation over the past three decades were mostly due to factors other than changing composition of cohabiting women.

Discussion

Decades after the onset of the cohabitation revolution (Smock and Manning 2010), the duration and outcomes of cohabiting unions have changed. Our results showed that cohabitations formed between 2006 and 2013 lasted longer, on average, than those formed in the mid-1980s. We also found that the lengthening of cohabitation over time resulted mostly from the declining rate of transitioning to marriage; close to one-half (42 %) of first premarital cohabitors married their partners in the 1980s, but only 22 % of recent cohabitors did so. It is important to note that increased age at first marriage (Manning et al. 2014) might account for part of the reduced rate of transitioning to marriage from first premarital cohabitation found in this study. The increase in average duration of first premarital cohabiting union over the study period cuts across sociodemographic groups—race/ethnicity, education, and motherhood status.

Despite racial/ethnic differences in the outcomes of cohabiting unions at the bivariate level, with longer average durations for whites and Hispanics than blacks, these racial/ethnic patterns have not changed over time. But, consistent with the diverging destinies perspective, we documented an increasing educational divergence in the outcomes of cohabitation over time. The risks of transitioning to marriage declined significantly over time for all but college-educated cohabitors. Thus, with regard to social class, the divide in the American family appears to be growing. Although an increasing proportion of cohabitors are college-educated, these findings suggest that college-educated women may be more often treating cohabitation as a pathway to marriage, and those with more modest educations are not. Future analyses of variations in the experiences of cohabitation should further explore the growing socioeconomic inequality among different groups of cohabitors.

We introduced the possibility that changes in the outcomes of cohabiting unions may be due to the fact that cohabitation is more widespread and less selective (diffusion perspective). The empirical support for this approach has been documented in Europe (Liefbroer and Dourleijn 2006) and has been applied to some U.S. analyses of marital dissolution (Manning and Cohen 2012; see also Killewald 2016). Our analyses indicated that only a small fraction of the change in outcomes of cohabitation was due to the changing composition of cohabiting couples.

Cohabiting unions with children (particularly cohabiting unions in which a birth occurred) last longer than those without children; women who had children while cohabiting experienced lower rates of transitioning to marriage or separation. Further, whereas the effect of a birth before cohabitation on cohabitation outcomes changed little over time, cohabiting births were linked to significantly lower risks of transitioning to marriage, relative to continued cohabitation, in 2006–2013 than in 1983–1988. Given the concentration of cohabiting births among less economically advantaged women (Kennedy and Bumpass 2008), these findings reinforce the growing divide in outcomes of premarital cohabitation across social class in the United States, aligning with the diverging destinies perspective. The composition of first premarital cohabiting women has shifted to include more Hispanics, college-educated women, and mothers. However, contrary to the diffusion perspective, the changing sociodemographic characteristics of cohabiting women did not account for most of the changes in the outcomes of first premarital cohabitation. The delinking of cohabitation from marriage and the declining rate of dissolution of first premarital cohabiting unions result from general changes in the U.S. population rather than behavioral changes specific to a group of cohabitors.

Understanding how the duration of cohabiting unions is changing is important for several reasons. First, it provides us with a broader perspective on the institutionalization of cohabitation and its changing role in the U.S. family life course. The lengthening of premarital cohabitation, coupled with the increased rate of childbearing and childrearing among premarital cohabitors over the past 30 years (from 19 % to 30 %), suggests that cohabitation is now more institutionalized as a unique family form in the United States. Cohabitation is increasingly serving a role similar to that of traditional marriage in offering a viable context for childbearing and child-rearing, particularly among women without a college degree. Compared with only 9 % of women with a college degree, 59 % of women with less than a high school diploma, 44 % of high school graduates, and 24 % of women with some college education had children before or during their first premarital cohabiting relationships in the 2006–2013 cohabitation cohort. The education gradient in childbearing during cohabitation has implications for assessments of children’s experiences in cohabiting-parent families and showcases potential differences in the meaning of cohabitation for education groups. Considering that educational attainment was assessed at the time of interview, our estimates of cohabiting births among women with low levels of education are conservative.

Second, deciphering the trends in transitions from premarital cohabitation to marriage provides a lens into the evolving relationship between cohabitation and marriage. Our finding of a declining rate of transitioning to marriage from first premarital cohabitation among women with no college degree diminishes the traditional view of cohabitation as a prelude to marriage.

Finally, the implications of cohabitation for the well-being of adults and children may shift given that cohabitation has become less of a transitory experience for women with cohabiting births. Distinguishing the meaning and implications of short-term versus long-term cohabiting unions, particularly for those with children, is an avenue for future research.

Although our study provides new insights into the changing nature of cohabitation, it has some limitations. First, our analyses are based on retrospective reports of the timing of first premarital cohabitations. Given that the start and end dates of cohabitation are often fluid, retrospective recollection of cohabitation dates may not be totally accurate (Hayford and Morgan 2008; Manning and Smock 2001). We restricted our focus to cohabiting unions that occurred within five years of interview to minimize recall bias. Second, the age limit in our sample (15–39) means that our findings may not be generalizable to older first-time premarital cohabitors, although few first-time cohabitations are expected to occur at these older ages. Third, we limited the analyses to first premarital cohabitation. Further attention to serial cohabitation is warranted. Also, to avoid conflating divorce (a second transition after marriage following cohabitation) with premarital union dissolution, we focused on first transitions out of first premarital cohabitation rather than the overall stability of relationships that began as cohabitations. It is important for future research to observe romantic partners from the onset of their union until they separate rather than transitions within the same relationship. Additionally, comparisons of changes in premarital and postmarital union dissolutions are an important avenue for future research on relationship stability in the United States.

Our analyses span about 30 years (1983–2013), representing two key birth cohorts as well as a period of rapid growth in cohabitation. These periods roughly align with the experiences of the late Baby Boomers (the youngest of whom were born in 1964 and were aged 19–24 between 1983 and 1988) and the Millennial birth cohort (the oldest of whom were aged 28–33 between 2008 and 2013). Given that the ages at first cohabitation during both periods remained relatively unchanged (at age 22; Manning et al. 2014), these periods capture the experiences of cohabitation for two key birth cohorts. Some women in our 2006–2013 cohabitation cohort formed their unions around the Great Recession. Hence, we examined the effect of the recession on our findings. We compared cohabiting unions in the 2006–2013 cohort that were formed prior to the recession with those formed after the recession. In both our bivariate and multivariate analyses, we found no significant differences in the outcomes of first premarital cohabiting relationships consummated during (December 2007–June 2009) or up to six months after the recession and those formed prior to the recession (January 2006–November 2007).

Fourth, because of the absence of full educational histories in the NSFG, respondents’ educational attainment was assessed at the time of interview rather than at the start of cohabitation. Because women may complete their education after they begin cohabitation, we may have incorrectly categorized their education. Data limitations prevented us from directly addressing the issue of temporal ordering created by the assessment of respondents’ educational attainment at the time of interview rather than at the beginning of cohabiting unions. Even though our analyses identified an important educational gradient in the experiences of first premarital cohabitation, the results do not imply a causal relationship between education and cohabitation outcome. More importantly, understanding of social class variations in the experiences of cohabitation requires broader measures than just educational attainment. Our analyses highlight the need for collection of more detailed information about labor market outcomes and educational attainment in fertility and family surveys. The NSFG data do not have information on any other measure of socioeconomic status that also contains dates that could be tracked prior to the start of the union. Our study offers additional evidence of the increasing divergence in family processes and family behaviors between college-educated Americans and their less-educated counterparts reported in previous studies (e.g., Gibson-Davis and Rackin 2014; Goldstein and Kenney 2001; McLanahan 2004; Raley 2000).

Fifth, data limitations precluded us from accounting for partners’ characteristics in the multivariate models. Future analyses of changes in cohabitation among unmarried mothers would also benefit from additional information about the children born to cohabiting women. Although a large share of mothers in this study, particularly in the recent cohort, had one or more births within their first premarital cohabitation, we could not establish biological ties between the children and their mothers’ cohabiting partners. Also, our supplemental analyses showed that 20 % of women who had children prior to the start of (but not during) a first premarital cohabitation transitioned to a coresidential union within six months of a birth, suggesting that some of the women with a birth before cohabitation later cohabited with the biological father of the child. Nonetheless, the NSFG is the optimal data source to track changes in the stability of cohabitation. Last, our analyses were based on relatively small samples of cohabiting women—729 in 1983–1988 and 774 in 2006–2013. Further attention to racial and ethnic variation in cohabiting couple outcomes is particularly warranted. In some cases, our analyses were limited to relatively small numbers of some subgroups (Hispanics in the 1983–1988 cohort).

Over the past 30 years, first premarital cohabiting unions have changed. Today’s unions are less likely to eventuate in marriage and more likely to persist, suggesting that the meaning and purpose of cohabitation in the United States is evolving, particularly among women without a college degree. Millennials and Baby Boomers have had distinct cohabitation experiences. Further, the significant changes in the outcomes of cohabiting unions, according to the presence of children as well as women’s education, provide insights into potential future family trajectories. Although first premarital cohabitations increasingly serve as a family context, tending to endure over time and often including children, this remains a minority experience. Clearly, social class increasingly shapes the experiences of cohabitation. For most college-educated women, cohabitation serves as a transitory union, one that rarely includes children. As premarital cohabitation has diffused across the U.S. population, its contours have altered, reshaping the meaning of cohabitation among less-educated Americans from a prelude to marriage to an increasingly important family form in its own right.

Acknowledgments

This research was supported in part by the Center for Family and Demographic Research (CFDR) at Bowling Green State University. The CFDR has core funding from The Eunice Kennedy Shriver National Institute of Child Health & Human Development (P2CHD050959).

Publisher’s Note

Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Notes

1

Because of the upper age limit of the NSFG, the retrospective construction of cohorts is problematic, resulting in analyses that represent experiences of only older respondents. For example, the 2011–2013 NSFG cannot be used to analyze outcomes of cohabitations formed in 1985 because it would reflect the experiences of only respondents aged 42–44 (15–17 in 2012). Similarly, analyses of cohabitations formed in 1990 using the same data would reflect the experiences of respondents who were 15–22 years old in 1990 (37–44 in 2012).

2

Our comparisons of the timing of college graduation (available in the 2011–2013 but not in the 1988 NSFG) and the timing of first premarital cohabitation showed that 89 % of the college-educated women in the 2006–2013 cohort obtained their degrees before or during their first premarital cohabitations; all of them were already in college when they started cohabiting. Similarly, based on the timing of high school graduation in the 2011–2013 NSFG, the majority (80 %) of women with a high school diploma in our sample had the same levels of education at the time of cohabitation. Educational attainment at the time of interview appears to largely reflect education at the time of cohabitation.

3

To address the issue of temporal ordering in the measurement of respondent’s education, we considered replacing respondent’s education with mother’s education. We found that mother’s education does not fully capture respondent’s educational attainment. As shown in Table A1 of the online appendix, less than one-third of respondents with a college degree had a college-educated mother. In the multivariate analyses (Table A2 of the online appendix), relying on solely mother’s education and not accounting for respondent’s education results in a slightly different conclusion about social class. In our exploratory models, we found that controlling for maternal education as an indicator of social class did not change our results or alter our conclusions. We excluded maternal education from our analyses to minimize collinearity problems.

References

Axinn, W. G., & Thornton, A. (
1992
).
The relationship between cohabitation and divorce: Selectivity or causal influence?
Demography
,
29
,
357
374
. 10.2307/2061823.
Becker, G. (
1990
).
A treatise on the family
.
Cambridge, MA
:
Harvard University Press
.
Brown, S. L. (
2000
).
Union transitions among cohabitors: The significance of relationship assessments and expectations
.
Journal of Marriage and the Family
,
62
,
833
846
. 10.1111/j.1741-3737.2000.00833.x.
Brown, S. L., & Manning, W. D. (
2009
).
Family boundary ambiguity and the measurement of family structure: The significance of cohabitation
.
Demography
,
46
,
85
101
. 10.1353/dem.0.0043.
Brown, S. L., Stykes, B. J., & Manning, W. D. (
2016
).
Trends in children’s family instability, 1995–2010
.
Journal of Marriage and Family
,
78
,
1173
1183
. 10.1111/jomf.12311.
Bumpass, L., & Lu, H.-H. (
2000
).
Trends in cohabitation and implications for children s family contexts in the United States
.
Population Studies
,
54
,
29
41
.
Bumpass, L. L., & Sweet, J. A. (
1989
).
National estimates of cohabitation
.
Demography
,
26
,
615
626
. 10.2307/2061261.
Bumpass, L. L., Sweet, J. A., & Cherlin, A. (
1991
).
The role of cohabitation in declining rates of marriage
.
Journal of Marriage and the Family
,
53
,
913
927
. 10.2307/352997.
Carlson, M., McLanahan, S., & England, P. (
2004
).
Union formation in fragile families
.
Demography
,
41
,
237
261
. 10.1353/dem.2004.0012.
Cherlin, A. J. (
2004
).
The deinstitutionalization of American marriage
.
Journal of Marriage and Family
,
66
,
848
861
. 10.1111/j.0022-2445.2004.00058.x.
Cherlin, A. J. (
2014
).
Labor’s love lost: The rise and fall of the working-class family in America
.
New York, NY
:
Russell Sage Foundation
.
Copen, C. E., Daniels, K., & Mosher, W. D. (
2013
).
First premarital cohabitation in the United States: 2006–2010 National Survey of Family Growth
(National Health Statistics Reports No. 64).
Hyattsville, MD
:
National Center for Health Statistics
.
DeMaris, A., & Rao, K. V. (
1992
).
Premarital cohabitation and subsequent marital stability in the United States: A reassessment
.
Journal of Marriage and the Family
,
54
,
178
190
. 10.2307/353285.
Evenson, R. J., & Simon, R. W. (
2005
).
Clarifying the relationship between parenthood and depression
.
Journal of Health and Social Behavior
,
46
,
341
358
. 10.1177/002214650504600403.
Furstenberg, F. F. (
1996
).
The future of marriage
.
American Demographics
,
18
(
6
),
34
40
.
Gibson-Davis, C., & Rackin, H. (
2014
).
Marriage or carriage? Trends in union context and birth type by education
.
Journal of Marriage and Family
,
76
,
506
519
. 10.1111/jomf.12109.
Goldstein, J. R., & Kenney, C. T. (
2001
).
Marriage delayed or marriage forgone? New cohort forecasts of first marriage for U.S. women
.
American Sociological Review
,
66
,
506
519
. 10.2307/3088920.
Guzzo, K. B. (
2009
).
Marital intentions and the stability of first cohabitations
.
Journal of Family Issues
,
30
,
179
205
. 10.1177/0192513X08323694.
Guzzo, K. B. (
2014
).
Trends in cohabitation outcomes: Compositional changes and engagement among never-married young adults
.
Journal of Marriage and Family
,
76
,
826
842
. 10.1111/jomf.12123.
Hayford, S. R., & Morgan, S. P. (
2008
).
The quality of retrospective data on cohabitation
.
Demography
,
45
,
129
141
. 10.1353/dem.2008.0005.
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
. 10.1111/j.0022-2445.2004.00088.x.
Kennedy, S., & Bumpass, L. (
2011
,
April
).
Cohabitation and trends in the structure and stability of children’s family lives
.
Paper presented at the 2011 annual meeting of the Population Association of America
,
Washington, DC
.
Kennedy, S., & Bumpass, L. L. (
2008
).
Cohabitation and children’s living arrangements: New estimates from the United States
.
Demographic Research
,
19
,
1663
1692
. https://doi.org/10.4054/DemRes.2008.19.47.
Kiernan, K. (
2004
).
Redrawing the boundaries of marriage
.
Journal of Marriage and Family
,
66
,
980
987
. 10.1111/j.0022-2445.2004.00068.x.
Killewald, A. (
2016
).
Money, work, and marital stability: Assessing change in the gendered determinants of divorce
.
American Sociological Review
,
81
,
696
719
. 10.1177/0003122416655340.
Kuo, J. C. L., & Raley, R. K. (
2016
).
Diverging patterns of union transition among cohabitors by race/ethnicity and education: Trends and marital intentions in the United States
.
Demography
,
53
,
921
935
. 10.1007/s13524-016-0483-9.
Lichter, D. T., Michelmore, K., Turner, R. N., & Sassler, S. (
2016
).
Pathways to a stable union? Pregnancy and childbearing among cohabiting and married couples
.
Population Research and Policy Review
,
35
,
377
399
. 10.1007/s11113-016-9392-2.
Lichter, D. L., Qian, Z. Z., & Mellott, L. M. (
2006
).
Marriage or dissolution? Union transitions among poor cohabiting women
.
Demography
,
43
,
223
240
. 10.1353/dem.2006.0016.
Lichter, D. T., Sassler, S., & Turner, R. N. (
2014
).
Cohabitation, post-conception unions, and the rise in nonmarital fertility
.
Social Science Research
,
47
,
134
147
. 10.1016/j.ssresearch.2014.04.002.
Liefbroer, A. C., & Dourleijn, E. (
2006
).
Unmarried cohabitation and union stability: Testing the role of diffusion using data from 16 European countries
.
Demography
,
43
,
203
221
. 10.1353/dem.2006.0018.
Mahoney, M. M. (
2006
).
Stepparents as third parties in relation to their stepchildren
.
Family Law Quarterly
,
40
,
81
108
.
Manning, W. D. (
2001
).
Childbearing in cohabiting unions: Racial and ethnic differences
.
Family Planning Perspectives
,
33
,
217
223
. 10.2307/2673785.
Manning, W. D. (
2004
).
Children and the stability of cohabiting couples
.
Journal of Marriage and Family
,
66
,
674
689
. 10.1111/j.0022-2445.2004.00046.x.
Manning, W. D. (
2013
).
Trends in cohabitation: Over twenty years of change, 1987–2010
(NCFMR Family Profiles, No. FP-13-12).
Bowling Green, OH
:
National Center for Family and Marriage Research
. Retrieved from http://www.bgsu.edu/content/dam/BGSU/college-of-arts-and-sciences/NCFMR/documents/FP/FP-13-12.pdf
Manning, W. D., Brown, S. L., & Payne, K. K. (
2014
).
Two decades of stability and change in age at first union formation
.
Journal of Marriage and Family
,
76
,
247
260
. 10.1111/jomf.12090.
Manning, W. D., Brown, S. L., & Stykes, B. (
2015
).
Trends in births to single and cohabiting mothers, 1980–2013
(NCFMR Family Profiles, No. FP-15-03).
Bowling Green, OH
:
National Center for Family and Marriage Research
. Retrieved from http://www.bgsu.edu/content/dam/BGSU/college-ofarts-and-sciences/NCFMR/documents/FP/FP-15-03-birthtrends-singlecohabitingmoms.pdf
Manning, W. D., & Cohen, J. A. (
2012
).
Premarital cohabitation and marital dissolution: An examination of recent marriages
.
Journal of Marriage and Family
,
74
,
377
387
. 10.1111/j.1741-3737.2012.00960.x.
Manning, W. D., & Lichter, D. T. (
1996
).
Parental cohabitation and children’s economic well-being
.
Journal of Marriage and the Family
,
58
,
998
1010
. 10.2307/353986.
Manning, W. D., & Smock, P. J. (
1995
).
Why marry? Race and the transition to marriage among cohabitors
.
Demography
,
32
,
509
520
. 10.2307/2061671.
Manning, W. D., & Smock, P. J. (
2002
).
First comes cohabitation and then comes marriage? A research note
.
Journal of Family Issues
,
23
,
1065
1087
. 10.1177/019251302237303.
Manning, W. D., & Smock, P. J. (
2005
).
Measuring and modeling cohabitation: New perspectives from qualitative data
.
Journal of Marriage and Family
,
67
,
989
1002
. 10.1111/j.1741-3737.2005.00189.x.
Manning, W. D., & Stykes, B. (
2015
).
Twenty-five years of change in cohabitation in the U.S., 1987–2013
(NCFMR Family Profiles, No. FP-15-01).
Bowling Green, OH
:
National Center for Family and Marriage Research
. Retrieved from https://www.bgsu.edu/content/dam/BGSU/college-of-arts-and-sciences/NCFMR/documents/FP/FP-15-01-twenty-five-yrs-cohab-us.pdf
Marsiglio, W. (
2004
).
When stepfathers claim stepchildren: A conceptual analysis
.
Journal of Marriage and Family
,
66
,
22
39
. 10.1111/j.1741-3737.2004.00002.x.
McLanahan, S. (
2004
).
Diverging destinies: How children are faring under the second demographic transition
.
Demography
,
41
,
607
627
. 10.1353/dem.2004.0033.
Meier, A., & Allen, G. (
2009
).
Romantic relationships from adolescence to young adulthood: Evidence from the National Longitudinal Study of Adolescent Health
.
Sociological Quarterly
,
50
,
308
335
. 10.1111/j.1533-8525.2009.01142.x.
Musick, K. (
2002
).
Planned and unplanned childbearing among unmarried women
.
Journal of Marriage and Family
,
64
,
915
929
. 10.1111/j.1741-3737.2002.00915.x.
Musick, K., & Michelmore, K. (
2015
).
Change in the stability of marital and cohabiting unions following the birth of a child
.
Demography
,
52
,
1463
1485
. 10.1007/s13524-015-0425-y.
Nomaguchi, K. M., & Milkie, M. A. (
2003
).
Costs and rewards of children: The effects of becoming a parent on adults’ lives
.
Journal of Marriage and Family
,
65
,
356
374
. 10.1111/j.1741-3737.2003.00356.x.
Raley, R. K. (
2000
).
Recent trends and differentials in marriage and cohabitation: The United States
. In L. J. Waite (Ed.),
The ties that bind: Perspectives on marriage and cohabitation
(pp.
19
39
).
New York, NY
:
Aldine de Gruyter
.
Raley, R. K., Crissey, S., & Muller, C. (
2007
).
Of sex and romance: Late adolescent relationships and young adult union formation
.
Journal of Marriage and Family
,
69
,
1210
1226
. 10.1111/j.1741-3737.2007.00442.x.
Sassler, S., & Miller, A. J. (
2011
).
Class differences in cohabitation processes
.
Family Relations
,
60
,
163
177
. 10.1111/j.1741-3729.2010.00640.x.
Smock, P. J., & Manning, W. D. (
1997
).
Cohabiting partners’ economic circumstances and marriage
.
Demography
,
34
,
331
341
. 10.2307/3038287.
Smock, P. J., & Manning, W. D. (
2010
).
New couples, new families: The cohabitation revolution in the United States
. In B. J. Risman & V. Rutter (Eds.),
Families as they really are
(pp.
131
139
).
New York, NY
:
Norton
.
Smock, P. J., Manning, W. D., & Porter, M. (
2005
).
“Everything’s there except money”: How money shapes decisions to marry among cohabitors
.
Journal of Marriage and Family
,
67
,
680
696
. 10.1111/j.1741-3737.2005.00162.x.
Sweeney, M. M. (
2010
).
Remarriage and stepfamilies: Strategic sites for family scholarship in the 21st century
.
Journal of Marriage and Family
,
72
,
667
684
. 10.1111/j.1741-3737.2010.00724.x.
Teachman, J. (
2008
).
Complex life course patterns and the risk of divorce in second marriages
.
Journal of Marriage and Family
,
70
,
294
305
. 10.1111/j.1741-3737.2008.00482.x.
U.S. Department of Health and Human Services
. (
1994
).
National Survey of Family Growth, Cycle IV, 1988
[Public use data tape documentation].
Hyattsville, MD
:
National Center for Health Statistics, Centers for Disease Control and Prevention
.
U.S. Department of Health and Human Services
. (
2014
).
2011–2013 National Survey of Family Growth
[Public use data tape documentation].
Hyattsville, MD
:
National Center for Health Statistics, Centers for Disease Control and Prevention
.
Wildsmith, E., & Raley, R. K. (
2006
).
Race-ethnic differences in nonmarital fertility: A focus on Mexican American women
.
Journal of Marriage and Family
,
68
,
491
508
. 10.1111/j.1741-3737.2006.00267.x.
Wu, Z. Z. (
1995
).
Premarital cohabitation and postmarital cohabiting union formation
.
Journal of Family Issues
,
16
,
212
232
. 10.1177/019251395016002005.

Supplementary data