A National Portrait of Stepfamilies in Later Life

A National Portrait of Stepfamilies in Later Life Abstract Objectives Scholars have documented increases in the prevalence and complexity of stepfamilies earlier in the life course, but no one has systematically investigated U.S. stepfamily structure in later life. Guided by a family systems approach, we described the prevalence and composition of later-life stepfamilies. Method The analysis was based on 6,250 married and cohabiting couples participating in the 2012 Health and Retirement Study. We identified the prevalence of later-life stepfamilies, decomposed stepfamily structures, and compared the sociodemographic characteristics and relationship quality of the couples in stepfamilies with those in married families (with only joint children and no stepchildren), paying attention to differences between married and cohabiting stepfamilies. Results Roughly 40% of middle-aged and older couples with children were in stepfamilies. Of all stepfamilies, 86% were married couples and 14% were cohabiting couples. Cohabiting stepfamilies more often included children from both partners’ previous relationships, but couples in married stepfamilies more often had joint children. Cohabiting stepfamilies appeared to be the most socially and economically disadvantaged, followed by married stepfamilies, and lastly married families. Despite these compositional differences, partner relationship quality was largely similar across married families, married stepfamilies, and cohabiting stepfamilies. Discussion This study underscores the high prevalence and complexity of later-life stepfamilies and foregrounds the urgency of additional research on this topic. Blended family, Cohabitation, Remarriage, Stepchildren Today’s older adults came of age during 1970s and 1980s, an era marked by high divorce and remarriage rates. Many have experienced multiple union transitions over the adult life course. In 2015, close to 30% of individuals aged 50 and older had experienced two or more marriages (Authors’ calculation based on the American Community Survey). Cohabitation is also on the rise among older adults, with the number cohabiting doubling in the past decade (Stepler, 2017). Nearly all cohabitors aged 50 and older experienced the dissolution of a prior marital union (Brown, Lee, & Bulanda, 2006). Whether remarried or cohabiting, the vast majority of these repartnered middle-aged and older adults (hereafter older adults) have children from previous relationships, meaning that they are part of a stepfamily. Scholars have carefully documented increases in the prevalence and complexity of stepfamilies earlier in the life course (Sweeney, 2010), but no one has systematically investigated U.S. stepfamily configuration later in life. Stepfamily structure is complex not only because one or both partners have children from previous relationships but also because the couple may subsequently have children together (hereafter joint children) and form a blended family. When only one partner has children from a previous relationship, the stepfamily consists of one biological parent and one stepparent. When both partners have children from previous relationships, each partner is a biological parent of their own children and a stepparent of their partner’s children. When the couple subsequently has joint children, each partner is a biological parent of their own children from previous relationships, a stepparent of their partner’s children from previous relationships, and a biological parent of the children with their current partner. Prior studies on later-life stepfamilies often focus on individual parent–child dyads, overlooking these varied constellations of the stepfamily environment (Blieszner & Voorpostel, 2016). Understanding stepfamily configuration is important because the more complex the stepfamily structure, the more ambiguous the roles and expectations are, creating more opportunities for conflict and disagreement between partners (Hobart, 1991). Prior research has compared relationship quality among older adults in first marriages versus remarriages (Bulanda, 2011; Cooney, Proulx, & Snyder-Rivas, 2016) and in remarriages versus cohabiting unions (Brown & Kawamura, 2010), but these studies do not speak directly to how partner relationship quality varies across stepfamily types. Stepfamilies can be formed through first marriages, and remarriages do not always result in stepfamilies. The objective of this study is to use data from the 2012 Health and Retirement Study to construct a national portrait of later-life stepfamilies, documenting the prevalence and complexity of various stepfamily forms. Guided by a family systems approach (Cox & Paley, 1997) that takes into account both parents’ relationships with each child in the family, we examine how the marital biographies, sociodemographic characteristics, economic resources, and health statuses of the couples in stepfamilies differ from those in married families with only joint children and no stepchildren, hereafter referred to as married families (we do not use the term “intact families” to avoid the appearance of any value judgment). We pay close attention to differences between married and cohabiting stepfamilies. We also investigate the extent to which relationship quality varies by stepfamily configuration. This study provides the first systematic examination of U.S. later-life stepfamilies by elucidating the prevalence and composition of different stepfamily forms. Background The Growth in Older Adults Who Are Remarried or Cohabiting Today’s older adults are more likely to be remarried or cohabiting than any previous generation in U.S. history. In 1980, 7 million married individuals aged 50 and older were in a higher-order marriage (Authors’ calculation based on the Census). The number has risen to 20 million in 2015 (Authors’ calculation based on the American Community Survey). This growth is not merely an artifact of the aging of the population. In 1980, 19% of older adults were in higher-order marriages and by 2015 the share had climbed to 30%. At the same time, cohabitation among older adults has accelerated, quadrupling from less than 1 million to 4 million persons from 2000 to 2016 (Brown et al., 2006; Stepler, 2017). Regardless of whether they are remarried or cohabiting, most of these individuals have children from previous relationships, suggesting that a significant proportion of older adults are in stepfamilies. Yet, we lack national estimates of the prevalence and complexity of later-life stepfamilies. Nor do we have any understanding of partner relationship quality in these families. This oversight is likely attributable to two reasons. First, most of the attention surrounding stepfamilies has focused on minor children, on the premise that family instability resulting from parental divorce and repartnering is detrimental to child development (Amato, 2010). Later-life stepfamilies tend to be ignored because most children in these families have launched into independent adulthood. Second, the rise in cohabitation among older adults is a relatively new phenomenon that was not well recognized and documented by social scientists until the beginning of this century (King & Scott, 2005; Brown et al., 2006). Little attention has been given to later-life cohabiting stepfamilies. Several sociodemographic trends, however, suggest that later-life stepfamilies can no longer be ignored. The divorce rate after age 50 has doubled over the past two decades. Even if the rate remains constant over the next 20 years, the number of older adults who would experience divorce in 2030 would rise by one third (Brown & Lin, 2012). Divorced older adults are more likely to repartner than their widowed counterparts (Brown, Lin, Hammersmith, & Wright, 2016). Not only has cohabitation increased in later life, but older adult attitudes toward cohabitation have also become more favorable in recent decades (Brown & Wright, 2016). Together, the number of later-life stepfamilies is likely to surge in the coming years, but scholars are confronted with a startling paucity of knowledge about these families. Stepfamily Configuration Stepfamily structure is complex not only because one or both partners have children from previous relationships but also because the couple may subsequently have joint children. Stepfamilies in which only one partner has children from a previous relationship are simple stepfamilies, in contrast to complex stepfamilies in which both partners have children from previous relationships (Steinbach & Hank, 2016; van der Pas & van Tilburg, 2010). Whether simple or complex, the couple may subsequently have joint children and form a blended family. The composition of stepfamily structure is likely to vary depending on whether the couple is married or cohabiting. The presence of children from previous relationships reduces the odds that parents marry their partner (Manning, 2004), particularly when both partners have children from previous relationships. Thus, cohabiting stepfamilies are more often complex families compared with married stepfamilies. On the other hand, having joint children provides parents the motivation to marry (Stewart, 2007). Therefore, married stepfamilies more often have joint children relative to cohabiting stepfamilies. Surprisingly, given the growth in older adults who are remarried or cohabiting, no study has documented the prevalence and complexity of later-life stepfamilies. A handful of studies have examined stepparent–stepchild relationships in later life, but these studies tend to use individual parent–child dyads as the unit of the analysis (e.g., Pezzin, Pollak, & Schone, 2013; Seltzer, Yahirun, & Bianchi, 2013), disregarding the various constellations of the stepfamily environment. Some biological parent–biological child dyads may be part of a stepfamily in which the parent is married to or living with a partner who is a stepparent of the child. Depending on which parent–child dyad is selected for the study, the prevalence of stepfamilies may be underestimated. Without treating the family as a system (Cox & Paley, 1997) that takes into account how both parents are related to each child in the family, it is impossible to fully gauge the diverse range of stepfamily arrangements. Partner Relationship Quality in Stepfamilies Unpacking stepfamily configuration provides insights into how the family functions. Stepfamilies have been viewed as incomplete institutions (Cherlin, 1978), in which family norms and obligations are not clearly defined or broadly shared. This lack of institutionalized guidelines for stepfamilies can lead to disagreement, undermining partner relationship quality. Further, the role of parenting is often perceived as more ambiguous by cohabiting than married parents (Stewart, 2007), suggesting that cohabiting stepfamilies are the least institutionalized and most unstable type of stepfamily (Ganong & Coleman, 2017). Prior research has not directly compared couples’ relationship quality in married families, married stepfamilies, and cohabiting stepfamilies, but has shown that couples in a higher-order marriage are more likely to break up than couples in a first marriage (Booth & Edwards, 1992). Cohabitors tend to report poorer relationship quality than do married individuals (Brown & Booth, 1996; Nock, 1995; Skinner, Bahr, Crane, & Call, 2002). Nonetheless, recent studies on older adults suggest no difference in relationship quality between first marriages and remarriages (Bulanda, 2011; Cooney et al., 2016), and older cohabitors share similar relationship quality as remarried older adults (Brown & Kawamura, 2010). The union type differences in partner relationship quality are negligible probably because older couples have a longer union duration and older cohabitors tend to cohabit as an alternative to marriage (King & Scott, 2005). In addition, children are a major source of relationship strains in stepfamilies (Stanley, Markman, & Whitton, 2002), but most children have grown up and left home in older families. Not only does marital status matter, but stepfamily structure is also likely to affect partner relationship quality. When partners have children from previous relationships, oftentimes confusion arises regarding who is part of the family and what roles individuals within the family should play (Boss, Bryant, & Mancini, 2016). Having joint children further blurs family boundaries. Boundary ambiguity is stressful for family members and creates opportunities for conflict. Thus, the more complex the stepfamily structure, the less clear the guideline and norms are for role performance, leading to more strained relationships. Prior studies have revealed poorer partner relationship quality in complex stepfamilies than in simple stepfamilies (Schultz, Schultz, & Olson, 1991), but these studies are limited to small, convenience samples. Having a joint child may or may not improve relationship quality. Some scholars propose the value of children argument (Friedman, Hechter, & Kanazawa, 1994), suggesting that having a joint child reduces uncertainty in the relationship and increases commitment to each other. However, increasing the complexity of family subsystems with a combination of full-, step-, and half-siblings may result in more ambiguity (Ward, Spitze, & Deane, 2009). Empirical evidence thus far has centered on younger stepfamilies and is mixed, showing positive associations (Downs, 2003), negative associations (Kurdek, 1999), or no association (Ganong & Coleman, 1988). The Present Study This study has three objectives. First, we establish a national portrait of later-life stepfamilies by identifying the share of couples (in which at least one spouse or partner is aged 51 or older) in stepfamilies, distinguishing between married stepfamilies and cohabiting stepfamilies. Within stepfamilies, we further investigate stepfamily structure by considering whether it is a simple or complex stepfamily and whether the couple has a joint child. Next, we examine how the marital biographies, sociodemographic characteristics, economic resources, and health statuses of the couples in stepfamilies differ from those in married families. Prior studies have shown that stepparents are distinct from parents who have biological children (hereafter biological parents) in several regards (Stewart, 2001; Thomson, 1994). Stepparents are more likely to be in a higher-order marriage and have a shorter union duration than biological parents. Because cohabitation is more common at younger ages and men tend to repartner with younger women, stepparents are often younger than biological parents. The gap in age between partners is greater among repartnered couples and a larger share of stepparents are nonwhite. Relative to biological parents, stepparents have less education and lower earnings, despite higher levels of full-time employment. Moreover, cohabiting stepparents appear to be more socially and economically disadvantaged compared with married stepparents. Because these studies are based on younger samples, the extent to which these patterns persist among later-life stepfamilies is unknown. A key advantage of our study is that we have information on the characteristics of both partners and thus can examine couple-level indicators which aligns with a family systems perspective. Last, we examine how both positive and negative dimensions of partner relationship quality differ among those in married families, married stepfamilies, and cohabiting stepfamilies. Three relationship quality indicators gauge whether the relationship with the partner is very close, positive support received from the partner, and negative support received from the partner. A lack of closeness, lower levels of positive support, and higher levels of negative support signal poorer partner relationship quality. Prior research has revealed the highest relationship quality is typically in first marriages, followed by remarriages, and lastly cohabiting unions (Brown & Booth, 1996; Nock, 1995; Skinner et al., 2002). However, these differentials may be smaller or even negligible among older adults for whom there are few relationship quality differentials between married and cohabiting individuals (Brown & Kawamura, 2010) as well as first married and remarried individuals (Bulanda, 2011; Cooney et al., 2016), although studies to date have not differentiated by stepfamily status. We also anticipate that the more complex the stepfamily structure, the poorer the partner relationship quality. In other words, couples in complex stepfamilies are less likely to have a very close relationship and experience less positive support and more negative support than couples in simple stepfamilies, and having a joint child is expected to be associated with poorer relationship quality. Method We used data from the 2012 round of the Health and Retirement Study (HRS), a longitudinal study of a nationally representative, continuous cohort of individuals aged 51 or older and their partners in the United States. The HRS began interviewing in 1992 with a cohort of individuals born in 1931–1941 and re-interviews have been conducted every other year. In 1998, three additional cohorts were added to the study to make the sample representative of the target population. Refresher samples aged 51–56 were added in 2004 and 2010 to maintain a nationally representative sample of individuals aged 51 or older. The response rates for the baseline interviews of various cohorts hover around 70%–82% and roughly 90% or higher for follow-up interviews. Blacks, Hispanics, and Florida residents were oversampled (Health and Retirement Study, 2011). The HRS includes rich information on respondents’ and their partners’ marital history, demographic characteristics, employment history, wealth, and health status, making the data ideal for the purposes of this study. We focused on the 2012 HRS because this is the most recent wave in the family data file (2012, Version 1), cleaned and processed by RAND, at the time when this analysis was conducted. The file includes information delineating how each child mentioned by HRS respondents is related to both the HRS respondents and their partners at each wave. We took advantage of all available longitudinal data to measure family type and stepfamily configuration. We captured parent–child relationships using the most frequent relationship code across waves. In cases where there were ties or that respondents were first interviewed in 2012, we used the relationship code in 2012. We also checked our estimates against the newly released 2014 Survey of Income and Program Participation (SIPP) which contains questions asking respondents about their number of childbearing unions and whether their current union is a childbearing one. Our estimates are very similar to those based on the 2014 SIPP (results not shown but available upon request). In total, 20,554 respondents were interviewed in 2012, of which 13,080 lived in coupled households, constituting 6,842 couples. Same-sex (n = 34) and childless (n = 253) couples were excluded. We also removed 187 couples for whom one or both partners’ relationships to their children could not be ascertained in the RAND HRS family data file. In addition, we omitted 48 couples who provided inconsistent information about their marital status and 46 couples with a sample weight of zero. Finally, 24 cohabiting couples who had joint children only were excluded, because it is too small of a group to permit a separate analysis. After eliminating these couples, we had 6,250 couples with children in the analysis, of which 3,486 were married families, 2,346 were married stepfamilies, and 418 were cohabiting stepfamilies. Measures Stepfamily configuration was gauged using couples’ marital status, whether they had children from previous relationships, and whether they subsequently had a joint child. We created three family types: married families (in which married couples have only joint children and no stepchildren), married stepfamilies (reference category), and cohabiting stepfamilies. We also constructed two measures to capture stepfamily structure: whether only one or both partners had children from previous relationships (0 = only one partner, 1 = both partners) and whether the couple had a joint child (0 = No, 1 = Yes). Partner relationship quality was assessed using respondents’ reports of whether they were very close to their partner, as well as positive and negative support received from their partner, with an absence of being very close, less positive support, and more negative support indicating poorer relationship quality. In the HRS psychosocial and lifestyle survey, respondents were asked how close their relationship is with their partner (very close, quite close, not very close, or not at all close). We created a dichotomous measure such that those who were very close were coded 1 and all others were coded 0. Moreover, respondents were asked of three questions regarding positive support (understanding the way you feel about things, relying on them if you have a serious problem, and opening up to them if you need to talk about your worries) and four questions about negative support (making too many demands on you, criticizing you, letting you down when you are counting on them, and getting on your nerves) received from their partners. The response categories were a lot (= 1), some (= 2), a little (= 3), and not at all (= 4). We reverse coded all items and averaged the scores separately for positive and negative support. Following the recommendation of the HRS (Smith, Ryan, Sonnega, & Weir, 2017), we set the average scores to missing when the respondent did not report more than one item for positive support or when the respondent did not report more than two items for negative support. To reduce respondent burden, relationship quality questions were asked of a randomly selected half of the HRS sample. Thus, the closeness, positive support, and negative support questions were available for only 2,313, 2,309, and 2,310 couples, respectively. For couples in which both partners were selected for interviews in 2012 and provided their evaluations of relationship quality, we randomly selected one partner’s report. Because women and men tend to perceive their relationship quality differently, we included a dichotomous variable in the multivariate analysis to indicate whether a woman’s or a man’s report was used. Marital biographies included marital history and union duration. Marital history was appraised by whether both partners were previously married (reference category), only one was previously married, and neither was previously married. The duration of the current union was measured in years. Sociodemographic characteristics consisted of the man’s age and age heterogamy, racial heterogamy, and the presence of children in the household. Man’s age was measured in years. Age heterogamy was coded into three categories that are consistent with prior research (Brown, Manning, & Payne, 2017; Shafer, 2013): the man and the woman shared similar ages (i.e., the woman was less than 2 years older than the man or the man was less than 5 years older than the woman, reference category), the man was older (i.e., the man was 5 or more years older than the woman), and the woman was older (i.e., the woman was 2 or more years older than the man). Racial heterogamy was defined as both partners were non-Hispanic white (reference category), both partners were black or Hispanic, and the partners were of different race and ethnicity. The presence of children in the household was assessed using three categories: no resident child (reference category), the youngest resident child was less than 18 years old, and the youngest resident child was aged 18 or older. Economic resources included the man’s education and educational heterogamy, employment heterogamy, home ownership, and assets. Man’s educational attainment was composed of four categories: less than high school (reference category), high school graduate, some college, and college or more. Educational heterogamy was captured by whether both partners had achieved the same level of education (reference category), the man had achieved more education than the woman, and the woman had achieved more education than the man. Employment heterogamy was assessed by whether both partners were working, only the man was working, only the woman was working, and neither was working (reference category). Home ownership was a dichotomous measure (0 = No, 1 = Yes). The couple’s assets were measured in dollars and logged in the multivariate analyses to correct for skewness. Health status was measured by whether both partners reported their health was good, very good, or excellent (good health, reference category), both partners reported their health was fair or poor (poor health), the man reported poor health but the woman reported good health, and the woman reported poor health but the man reported good health. Analytic Strategy We conducted three analyses. First, we described the prevalence and complexity of later-life stepfamily structure using percentages. Second, we provided a national portrait of later-life stepfamilies by examining compositional differences among married families, married stepfamilies, and cohabiting stepfamilies using means and percentages (as appropriate). Third, we performed logistic regressions (for closeness) and ordinary least squares regressions (for positive and negative support) to examine whether and how partner relationship quality varied across different family type and stepfamily structure, adjusting for compositional differences. A multiple imputation procedure was used to handle missing cases except for relationship quality (ranging from less than 1% for any child residing in the household to 11% for home ownership), such that the missing value for a single covariate was imputed as a function of other covariates in the analysis (Acock, 2005). To preserve the randomness of imputed variables, the study results were based on 10 random, multiple-imputed replicates. All estimates were weighted using the 2012 household weights to adjust for the unequal probability of selection, nonresponse, and sample attrition (Ofstedal, Weir, Chen, & Wagner, 2011). Results Stepfamily Configuration In 2012, two of every five couples in which at least one partner was aged 51 or older were in stepfamilies (41%), as presented in Table 1. Of all stepfamilies, 86% were married couples and 14% were cohabiting couples. Among married stepfamilies, 54% were simple stepfamilies and 46% were complex stepfamilies. The corresponding numbers for cohabiting stepfamilies were 43% and 57%, respectively. The proportion of married stepfamilies with a joint child was nearly four times the proportion in cohabiting stepfamilies (38% vs 10%). Consistent with our expectations, cohabiting stepfamilies were more often complex than married stepfamilies, but married stepfamilies more often had a joint child than cohabiting stepfamilies. Table 1. Weighted Percentages for Stepfamily Configuration in 2012 Family type  Married family  Married stepfamily  Cohabiting stepfamily  Total  59.1  35.4  5.5  Stepfamily  —  86.4  13.6  Stepfamily configuration   Simple stepfamily  —  54.3  42.8   Complex stepfamilya  —  45.7  57.2   Does not have a joint child  —  62.3  89.7   Has a joint childa  —  37.7  10.3  Unweighted N  3,486  2,346  418  Family type  Married family  Married stepfamily  Cohabiting stepfamily  Total  59.1  35.4  5.5  Stepfamily  —  86.4  13.6  Stepfamily configuration   Simple stepfamily  —  54.3  42.8   Complex stepfamilya  —  45.7  57.2   Does not have a joint child  —  62.3  89.7   Has a joint childa  —  37.7  10.3  Unweighted N  3,486  2,346  418  aMarried stepfamilies significantly differ from cohabiting stepfamilies at p < .01. View Large Table 1. Weighted Percentages for Stepfamily Configuration in 2012 Family type  Married family  Married stepfamily  Cohabiting stepfamily  Total  59.1  35.4  5.5  Stepfamily  —  86.4  13.6  Stepfamily configuration   Simple stepfamily  —  54.3  42.8   Complex stepfamilya  —  45.7  57.2   Does not have a joint child  —  62.3  89.7   Has a joint childa  —  37.7  10.3  Unweighted N  3,486  2,346  418  Family type  Married family  Married stepfamily  Cohabiting stepfamily  Total  59.1  35.4  5.5  Stepfamily  —  86.4  13.6  Stepfamily configuration   Simple stepfamily  —  54.3  42.8   Complex stepfamilya  —  45.7  57.2   Does not have a joint child  —  62.3  89.7   Has a joint childa  —  37.7  10.3  Unweighted N  3,486  2,346  418  aMarried stepfamilies significantly differ from cohabiting stepfamilies at p < .01. View Large Variation Among Married Families, Married Stepfamilies, and Cohabiting Stepfamilies The characteristics of married families, married stepfamilies, and cohabiting stepfamilies are displayed in Table 2. As for marital biographies, couples in married families were the least likely to be in a higher-order marriage and were in unions of the longest average duration, followed by couples in married stepfamilies, and finally couples in cohabiting stepfamilies. Specifically, only 13% of the married families had at least one partner who was previously married, whereas nearly all of those in married stepfamilies (94%) or cohabiting stepfamilies (96%) were previously married. Indeed, couples in which both partners were previously married was the modal form of later-life stepfamilies, comprising approximately 62% of married stepfamilies and 80% of cohabiting stepfamilies. The average union duration was 41 years for married families, in sharp contrast to 20 years for married stepfamilies and 7 years for cohabiting stepfamilies. Table 2. Weighted Means or Percentages for the Characteristics of Coupled Family Types in 2012   Married family  Married stepfamily  Cohabiting stepfamily  Variable  (1)  (2)  (3)  Marital biographies   Marital history12,13,23    Both previously married  2.7  61.9  80.0    Only one previously married  10.0  32.5  16.2    Neither previously married  87.3  5.6  3.9   Union duration12,13,23  40.7  20.1  7.3  Sociodemographic characteristics   Man’s age12,13,23  65.7  63.3  61.4   Age heterogamy12,13    Both similar age  70.3  35.4  29.3    Man older  21.8  44.4  45.9    Woman older  7.9  20.2  24.9   Racial heterogamy12,13,23    Both non-Hispanic white  81.7  72.6  58.7    Both black or Hispanic  14.1  18.5  28.8    Different race/ethnicity  4.2  8.8  12.5   Resident child status12    No resident child  65.7  69.7  70.4    Youngest resident child younger than 18  10.3  10.9  10.4    Youngest resident child 18 or older  24.0  19.4  19.2  Economic resources   Man’s education12,13,23    Less than high school  10.8  13.4  18.6    High school  28.9  30.9  31.0    Some college  22.2  27.7  31.1    College or more  38.1  28.0  19.4   Educational heterogamy12    Both have same education  51.7  41.8  44.6    Man has more education  27.1  27.9  32.0    Woman has more education  21.1  30.2  23.4   Employment heterogamy12,13    Both working  38.8  42.1  45.2    Only man working  17.5  16.0  16.5    Only woman working  9.7  14.5  13.4    Neither working  34.0  27.3  24.9   Owns home12,13,23  92.1  83.8  64.6   Assets (in 10,000 median)12,13,23  31.2  17.4  8.7  Health heterogamy12,13,23   Both good health  68.2  61.5  50.6   Both poor health  6.9  8.2  13.6   Man poor health, woman good health  13.3  15.9  15.9   Woman poor health, man good health  11.6  14.4  19.9  Relationship quality   Very close12,13,23  66.8  56.7  43.4   Positive support12  3.5  3.4  3.4   Negative support  1.9  2.0  1.9  Unweighted N  3,486  2,346  418    Married family  Married stepfamily  Cohabiting stepfamily  Variable  (1)  (2)  (3)  Marital biographies   Marital history12,13,23    Both previously married  2.7  61.9  80.0    Only one previously married  10.0  32.5  16.2    Neither previously married  87.3  5.6  3.9   Union duration12,13,23  40.7  20.1  7.3  Sociodemographic characteristics   Man’s age12,13,23  65.7  63.3  61.4   Age heterogamy12,13    Both similar age  70.3  35.4  29.3    Man older  21.8  44.4  45.9    Woman older  7.9  20.2  24.9   Racial heterogamy12,13,23    Both non-Hispanic white  81.7  72.6  58.7    Both black or Hispanic  14.1  18.5  28.8    Different race/ethnicity  4.2  8.8  12.5   Resident child status12    No resident child  65.7  69.7  70.4    Youngest resident child younger than 18  10.3  10.9  10.4    Youngest resident child 18 or older  24.0  19.4  19.2  Economic resources   Man’s education12,13,23    Less than high school  10.8  13.4  18.6    High school  28.9  30.9  31.0    Some college  22.2  27.7  31.1    College or more  38.1  28.0  19.4   Educational heterogamy12    Both have same education  51.7  41.8  44.6    Man has more education  27.1  27.9  32.0    Woman has more education  21.1  30.2  23.4   Employment heterogamy12,13    Both working  38.8  42.1  45.2    Only man working  17.5  16.0  16.5    Only woman working  9.7  14.5  13.4    Neither working  34.0  27.3  24.9   Owns home12,13,23  92.1  83.8  64.6   Assets (in 10,000 median)12,13,23  31.2  17.4  8.7  Health heterogamy12,13,23   Both good health  68.2  61.5  50.6   Both poor health  6.9  8.2  13.6   Man poor health, woman good health  13.3  15.9  15.9   Woman poor health, man good health  11.6  14.4  19.9  Relationship quality   Very close12,13,23  66.8  56.7  43.4   Positive support12  3.5  3.4  3.4   Negative support  1.9  2.0  1.9  Unweighted N  3,486  2,346  418  Note. Column total may not equal 100% due to rounding error. Superscripts indicate that the two groups statistically differ at p < .05. View Large Table 2. Weighted Means or Percentages for the Characteristics of Coupled Family Types in 2012   Married family  Married stepfamily  Cohabiting stepfamily  Variable  (1)  (2)  (3)  Marital biographies   Marital history12,13,23    Both previously married  2.7  61.9  80.0    Only one previously married  10.0  32.5  16.2    Neither previously married  87.3  5.6  3.9   Union duration12,13,23  40.7  20.1  7.3  Sociodemographic characteristics   Man’s age12,13,23  65.7  63.3  61.4   Age heterogamy12,13    Both similar age  70.3  35.4  29.3    Man older  21.8  44.4  45.9    Woman older  7.9  20.2  24.9   Racial heterogamy12,13,23    Both non-Hispanic white  81.7  72.6  58.7    Both black or Hispanic  14.1  18.5  28.8    Different race/ethnicity  4.2  8.8  12.5   Resident child status12    No resident child  65.7  69.7  70.4    Youngest resident child younger than 18  10.3  10.9  10.4    Youngest resident child 18 or older  24.0  19.4  19.2  Economic resources   Man’s education12,13,23    Less than high school  10.8  13.4  18.6    High school  28.9  30.9  31.0    Some college  22.2  27.7  31.1    College or more  38.1  28.0  19.4   Educational heterogamy12    Both have same education  51.7  41.8  44.6    Man has more education  27.1  27.9  32.0    Woman has more education  21.1  30.2  23.4   Employment heterogamy12,13    Both working  38.8  42.1  45.2    Only man working  17.5  16.0  16.5    Only woman working  9.7  14.5  13.4    Neither working  34.0  27.3  24.9   Owns home12,13,23  92.1  83.8  64.6   Assets (in 10,000 median)12,13,23  31.2  17.4  8.7  Health heterogamy12,13,23   Both good health  68.2  61.5  50.6   Both poor health  6.9  8.2  13.6   Man poor health, woman good health  13.3  15.9  15.9   Woman poor health, man good health  11.6  14.4  19.9  Relationship quality   Very close12,13,23  66.8  56.7  43.4   Positive support12  3.5  3.4  3.4   Negative support  1.9  2.0  1.9  Unweighted N  3,486  2,346  418    Married family  Married stepfamily  Cohabiting stepfamily  Variable  (1)  (2)  (3)  Marital biographies   Marital history12,13,23    Both previously married  2.7  61.9  80.0    Only one previously married  10.0  32.5  16.2    Neither previously married  87.3  5.6  3.9   Union duration12,13,23  40.7  20.1  7.3  Sociodemographic characteristics   Man’s age12,13,23  65.7  63.3  61.4   Age heterogamy12,13    Both similar age  70.3  35.4  29.3    Man older  21.8  44.4  45.9    Woman older  7.9  20.2  24.9   Racial heterogamy12,13,23    Both non-Hispanic white  81.7  72.6  58.7    Both black or Hispanic  14.1  18.5  28.8    Different race/ethnicity  4.2  8.8  12.5   Resident child status12    No resident child  65.7  69.7  70.4    Youngest resident child younger than 18  10.3  10.9  10.4    Youngest resident child 18 or older  24.0  19.4  19.2  Economic resources   Man’s education12,13,23    Less than high school  10.8  13.4  18.6    High school  28.9  30.9  31.0    Some college  22.2  27.7  31.1    College or more  38.1  28.0  19.4   Educational heterogamy12    Both have same education  51.7  41.8  44.6    Man has more education  27.1  27.9  32.0    Woman has more education  21.1  30.2  23.4   Employment heterogamy12,13    Both working  38.8  42.1  45.2    Only man working  17.5  16.0  16.5    Only woman working  9.7  14.5  13.4    Neither working  34.0  27.3  24.9   Owns home12,13,23  92.1  83.8  64.6   Assets (in 10,000 median)12,13,23  31.2  17.4  8.7  Health heterogamy12,13,23   Both good health  68.2  61.5  50.6   Both poor health  6.9  8.2  13.6   Man poor health, woman good health  13.3  15.9  15.9   Woman poor health, man good health  11.6  14.4  19.9  Relationship quality   Very close12,13,23  66.8  56.7  43.4   Positive support12  3.5  3.4  3.4   Negative support  1.9  2.0  1.9  Unweighted N  3,486  2,346  418  Note. Column total may not equal 100% due to rounding error. Superscripts indicate that the two groups statistically differ at p < .05. View Large These three family types also differ in their sociodemographic characteristics. Compared with men in married families (66 years old), men in stepfamilies tended to be younger (63 and 61 years old for men in married stepfamilies and cohabiting stepfamilies, respectively). Age heterogamy was highest among cohabiting stepfamilies (70%), followed by married stepfamilies (64%), and then married families (29%). Relative to couples in married families, couples in stepfamilies were more likely to be minorities or of different racial and ethnic backgrounds, and this was more pronounced for cohabiting stepfamilies than married stepfamilies. Married stepfamilies were less likely than married families to have a child residing in the household, particularly an adult child (19% vs 24%). Concerning economic resources, men in married families had achieved the most education (38% with college or more), followed by men in married stepfamilies (28%) and then men in cohabiting stepfamilies (19%). Education heterogamy was more prevalent in married stepfamilies (58%) than in married families (48%), but cohabiting stepfamilies were no different from married families and married stepfamilies. Compared with married families, stepfamilies more often had both partners working or only women working, but no difference was found between married and cohabiting stepfamilies. Cohabiting stepfamilies were the least likely to own a home and accrued the lowest assets, followed by married stepfamilies and married families. Finally, couples in married families were the most likely to report that both partners had good health, while couples in cohabiting stepfamilies were most likely to report that both partners had poor health. In sum, couples in married families fare the best socially and economically, followed by couples in married stepfamilies and lastly couples in cohabiting stepfamilies. In the next section, we examine whether and how partner relationship quality varies by family type and stepfamily structure. Relationship Quality by Family Type and Stepfamily Structure As shown at the bottom of Table 2, couples in married families most often reported their relationship was very close (67%), followed by couples in married stepfamilies (57%) and then couples in cohabiting stepfamilies (43%). Moreover, couples in married families reported receiving slightly more positive support from their spouses than did couples in married stepfamilies. Nevertheless, there was no difference in positive support between married stepfamilies and cohabiting stepfamilies. Nor were there any differences in negative support across family types. The few significant bivariate associations largely disappeared after taking into account the differentials in marital biographies, sociodemographic characteristics, economic resources, and health status, as shown in Table 3. Although couples in cohabiting stepfamilies reported a lower likelihood of being very close compared with either couples in married families (−0.53 vs 0.37, p < .05) or remarried stepfamilies (−0.53 p < .05), married families and married stepfamilies were similarly likely to report their relationship was very close. Levels of both positive and negative support from the partner were comparable across married families, married stepfamilies, and cohabiting stepfamilies. Table 3. Coefficients From Logistic or OLS Regressions of Relationship Quality by Family Type and Selected Characteristics Variable  Very close  Positive support  Negative support  Married family  0.37  0.11  −0.06  Married stepfamily (ref)        Cohabiting stepfamily  −0.53*  0.01  −0.02  Marital biographies   Marital history    Both previously married (ref)    Only one previously married  −0.27  −0.09  0.16*    Neither previously married  −0.12  −0.17*  0.12   Union duration/10  0.01  0.03  0.01  Sociodemographic characteristics   Man’s age/10  −0.05  −0.04  −0.03   Age heterogamy    Both similar age (ref)    Man older  0.01  0.02  0.05    Woman older  0.00  0.01  −0.03   Racial heterogamy    Both non-Hispanic white (ref)    Both black or Hispanic  −0.11  −0.08  0.10*    Different race/ethnicity  −0.40  −0.17  0.19*   Resident child status    No resident child (ref)          Resident child younger than 18  −0.22  −0.12  0.08    Resident child 18 or older  0.19  −0.01  0.04  Economic resources   Man’s education    Less than high school (ref)    High school  0.27  0.14*  −0.02    Some college  0.21  0.17*  −0.07    College or more  0.05  0.17*  −0.03   Educational heterogamy    Both have same education (ref)    Man has more education  −0.33*  −0.10**  0.05    Woman has more education  −0.34*  −0.02  0.09   Employment heterogamy    Both working  −0.44*  −0.04  0.01    Only man working  −0.11  0.00  −0.03    Only woman working  −0.47*  −0.04  −0.01    Neither working (ref)   Owns home  −0.22  −0.03  0.03   Assets  −0.02  0.04  −0.06*  Health heterogamy   Both good health (ref)   Both poor health  −0.64**  −0.24**  0.28**   Man poor health, woman good health  −0.34*  −0.14**  0.14**   Woman poor health, man good health  −0.36  −0.24**  0.10  Man’s report  0.27*  0.17***  −0.08*  Intercept  1.45  3.07***  2.90***  Unweighted N  2,313  2,309  2,310  Variable  Very close  Positive support  Negative support  Married family  0.37  0.11  −0.06  Married stepfamily (ref)        Cohabiting stepfamily  −0.53*  0.01  −0.02  Marital biographies   Marital history    Both previously married (ref)    Only one previously married  −0.27  −0.09  0.16*    Neither previously married  −0.12  −0.17*  0.12   Union duration/10  0.01  0.03  0.01  Sociodemographic characteristics   Man’s age/10  −0.05  −0.04  −0.03   Age heterogamy    Both similar age (ref)    Man older  0.01  0.02  0.05    Woman older  0.00  0.01  −0.03   Racial heterogamy    Both non-Hispanic white (ref)    Both black or Hispanic  −0.11  −0.08  0.10*    Different race/ethnicity  −0.40  −0.17  0.19*   Resident child status    No resident child (ref)          Resident child younger than 18  −0.22  −0.12  0.08    Resident child 18 or older  0.19  −0.01  0.04  Economic resources   Man’s education    Less than high school (ref)    High school  0.27  0.14*  −0.02    Some college  0.21  0.17*  −0.07    College or more  0.05  0.17*  −0.03   Educational heterogamy    Both have same education (ref)    Man has more education  −0.33*  −0.10**  0.05    Woman has more education  −0.34*  −0.02  0.09   Employment heterogamy    Both working  −0.44*  −0.04  0.01    Only man working  −0.11  0.00  −0.03    Only woman working  −0.47*  −0.04  −0.01    Neither working (ref)   Owns home  −0.22  −0.03  0.03   Assets  −0.02  0.04  −0.06*  Health heterogamy   Both good health (ref)   Both poor health  −0.64**  −0.24**  0.28**   Man poor health, woman good health  −0.34*  −0.14**  0.14**   Woman poor health, man good health  −0.36  −0.24**  0.10  Man’s report  0.27*  0.17***  −0.08*  Intercept  1.45  3.07***  2.90***  Unweighted N  2,313  2,309  2,310  *p < .05. **p < .01. ***p < .001. View Large Table 3. Coefficients From Logistic or OLS Regressions of Relationship Quality by Family Type and Selected Characteristics Variable  Very close  Positive support  Negative support  Married family  0.37  0.11  −0.06  Married stepfamily (ref)        Cohabiting stepfamily  −0.53*  0.01  −0.02  Marital biographies   Marital history    Both previously married (ref)    Only one previously married  −0.27  −0.09  0.16*    Neither previously married  −0.12  −0.17*  0.12   Union duration/10  0.01  0.03  0.01  Sociodemographic characteristics   Man’s age/10  −0.05  −0.04  −0.03   Age heterogamy    Both similar age (ref)    Man older  0.01  0.02  0.05    Woman older  0.00  0.01  −0.03   Racial heterogamy    Both non-Hispanic white (ref)    Both black or Hispanic  −0.11  −0.08  0.10*    Different race/ethnicity  −0.40  −0.17  0.19*   Resident child status    No resident child (ref)          Resident child younger than 18  −0.22  −0.12  0.08    Resident child 18 or older  0.19  −0.01  0.04  Economic resources   Man’s education    Less than high school (ref)    High school  0.27  0.14*  −0.02    Some college  0.21  0.17*  −0.07    College or more  0.05  0.17*  −0.03   Educational heterogamy    Both have same education (ref)    Man has more education  −0.33*  −0.10**  0.05    Woman has more education  −0.34*  −0.02  0.09   Employment heterogamy    Both working  −0.44*  −0.04  0.01    Only man working  −0.11  0.00  −0.03    Only woman working  −0.47*  −0.04  −0.01    Neither working (ref)   Owns home  −0.22  −0.03  0.03   Assets  −0.02  0.04  −0.06*  Health heterogamy   Both good health (ref)   Both poor health  −0.64**  −0.24**  0.28**   Man poor health, woman good health  −0.34*  −0.14**  0.14**   Woman poor health, man good health  −0.36  −0.24**  0.10  Man’s report  0.27*  0.17***  −0.08*  Intercept  1.45  3.07***  2.90***  Unweighted N  2,313  2,309  2,310  Variable  Very close  Positive support  Negative support  Married family  0.37  0.11  −0.06  Married stepfamily (ref)        Cohabiting stepfamily  −0.53*  0.01  −0.02  Marital biographies   Marital history    Both previously married (ref)    Only one previously married  −0.27  −0.09  0.16*    Neither previously married  −0.12  −0.17*  0.12   Union duration/10  0.01  0.03  0.01  Sociodemographic characteristics   Man’s age/10  −0.05  −0.04  −0.03   Age heterogamy    Both similar age (ref)    Man older  0.01  0.02  0.05    Woman older  0.00  0.01  −0.03   Racial heterogamy    Both non-Hispanic white (ref)    Both black or Hispanic  −0.11  −0.08  0.10*    Different race/ethnicity  −0.40  −0.17  0.19*   Resident child status    No resident child (ref)          Resident child younger than 18  −0.22  −0.12  0.08    Resident child 18 or older  0.19  −0.01  0.04  Economic resources   Man’s education    Less than high school (ref)    High school  0.27  0.14*  −0.02    Some college  0.21  0.17*  −0.07    College or more  0.05  0.17*  −0.03   Educational heterogamy    Both have same education (ref)    Man has more education  −0.33*  −0.10**  0.05    Woman has more education  −0.34*  −0.02  0.09   Employment heterogamy    Both working  −0.44*  −0.04  0.01    Only man working  −0.11  0.00  −0.03    Only woman working  −0.47*  −0.04  −0.01    Neither working (ref)   Owns home  −0.22  −0.03  0.03   Assets  −0.02  0.04  −0.06*  Health heterogamy   Both good health (ref)   Both poor health  −0.64**  −0.24**  0.28**   Man poor health, woman good health  −0.34*  −0.14**  0.14**   Woman poor health, man good health  −0.36  −0.24**  0.10  Man’s report  0.27*  0.17***  −0.08*  Intercept  1.45  3.07***  2.90***  Unweighted N  2,313  2,309  2,310  *p < .05. **p < .01. ***p < .001. View Large Few control variables were associated with the three indicators of relationship quality. For example, couples in which both partners were in a higher-order marriage received more positive support and less negative support from their partners relative to couples in which both partner were in their first marriage or only one partner was remarried, respectively. Racial-ethnic minority couples and couples of different racial-ethnic backgrounds tended to report higher levels of negative support than white couples. Education was related to positive support from the partner with those without a high school diploma reporting the least positive support, on average. Educational heterogamy was negatively associated with both of the positive indicators of relationship quality, whereas employment heterogamy was only negatively related to the likelihood of reporting a very close relationship. Assets were largely unrelated to relationship quality, but couples with fewer assets reported higher levels of negative support from their partners. Health was consistently associated with the three dimensions of relationship quality such that couples in which both partners reported good health enjoyed the highest levels of positive quality and the lowest levels of negative quality. Finally, aligning with prior research on marital quality, men were more likely to characterize their unions as very close and tended to report higher levels of positive support and lower levels of negative support compared with women. Because married stepfamilies and cohabiting stepfamilies differ in the prevalence of simple versus complex stepfamilies and the presence of a joint child, we further examined whether and how relationship quality differed by stepfamily structure. As shown in Table 4, stepfamily configuration was not significantly associated with the three indicators of relationship quality after taking into account compositional differences. Couples in simple versus complex stepfamilies were similarly likely to report being very close and they reported comparable levels of positive and negative support from partners. Additionally, having a joint child was unrelated to perceptions of closeness as well as positive and negative support from the partner. Notably, the presence and age of the youngest child were not associated with relationship quality. For the most part, the relationships between the control variables and the three indicators of relationship quality remained largely unchanged from Table 3, although fewer achieved statistical significance (perhaps reflecting the smaller sample size or the inclusion of the measures of stepfamily configuration). There were no significant interactions between being in a simple versus complex stepfamily and having a joint child or between family type and stepfamily structure (results not shown). Table 4. Coefficients from Logistic or OLS Regressions of Relationship Quality by Stepfamily Structure and Selected Characteristics Variable  Very close  Positive support  Negative support  Cohabiting stepfamily (vs married stepfamily)  −0.59*  −0.03  0.03  Both have a stepchild (vs only one has a stepchild)  0.21  −0.07  0.03  Has a joint child (vs no joint child)  −0.06  −0.02  0.05  Marital biographies   Marital history    Both previously married (ref)          Only one previously married  −0.29  −0.06  0.13    Neither previously married  −0.30  −0.13  0.14   Union duration/10  −0.01  −0.02  0.04  Sociodemographic characteristics   Man’s age/10  −0.04  −0.00  −0.04   Age heterogamy    Both similar age (ref)          Man older  0.28  0.05  0.02    Woman older  0.39  0.08  −0.07   Racial heterogamy    Both non-Hispanic white (ref)          Both black or Hispanic  −0.13  −0.06  0.04    Different race/ethnicity  −0.23  −0.10  0.11   Resident child status    No resident child (ref)          Resident child younger than 18  0.30  −0.01  0.02    Resident child 18 or older  0.40  0.06  0.01  Economic resources   Man’s education    Less than high school (ref)          High school  −0.25  0.14  0.02    Some college  −0.26  0.19  −0.05    College or more  −0.27  0.22  −0.04   Educational heterogamy    Both have same education (ref)          Man has more education  −0.62*  −0.21**  0.09    Woman has more education  −0.78**  −0.13  0.15   Employment heterogamy    Both working  −0.51*  −0.09  0.10    Only man working  −0.36  −0.01  0.01    Only woman working  −0.53  −0.08  −0.01    Neither working (ref)         Owns home  −0.41  −0.06  0.07   Assets  −0.09  0.00  −0.03  Health heterogamy   Both good health (ref)         Both poor health  −0.85*  −0.33**  0.39**   Man poor health, woman good health  −0.35  −0.17*  0.18*   Woman poor health, man good health  −0.92**  −0.36**  0.11  Man’s report  0.40*  0.21**  −0.11  Intercept  2.93  3.52***  2.40***  Unweighted N  1,006  1,005  1,006  Variable  Very close  Positive support  Negative support  Cohabiting stepfamily (vs married stepfamily)  −0.59*  −0.03  0.03  Both have a stepchild (vs only one has a stepchild)  0.21  −0.07  0.03  Has a joint child (vs no joint child)  −0.06  −0.02  0.05  Marital biographies   Marital history    Both previously married (ref)          Only one previously married  −0.29  −0.06  0.13    Neither previously married  −0.30  −0.13  0.14   Union duration/10  −0.01  −0.02  0.04  Sociodemographic characteristics   Man’s age/10  −0.04  −0.00  −0.04   Age heterogamy    Both similar age (ref)          Man older  0.28  0.05  0.02    Woman older  0.39  0.08  −0.07   Racial heterogamy    Both non-Hispanic white (ref)          Both black or Hispanic  −0.13  −0.06  0.04    Different race/ethnicity  −0.23  −0.10  0.11   Resident child status    No resident child (ref)          Resident child younger than 18  0.30  −0.01  0.02    Resident child 18 or older  0.40  0.06  0.01  Economic resources   Man’s education    Less than high school (ref)          High school  −0.25  0.14  0.02    Some college  −0.26  0.19  −0.05    College or more  −0.27  0.22  −0.04   Educational heterogamy    Both have same education (ref)          Man has more education  −0.62*  −0.21**  0.09    Woman has more education  −0.78**  −0.13  0.15   Employment heterogamy    Both working  −0.51*  −0.09  0.10    Only man working  −0.36  −0.01  0.01    Only woman working  −0.53  −0.08  −0.01    Neither working (ref)         Owns home  −0.41  −0.06  0.07   Assets  −0.09  0.00  −0.03  Health heterogamy   Both good health (ref)         Both poor health  −0.85*  −0.33**  0.39**   Man poor health, woman good health  −0.35  −0.17*  0.18*   Woman poor health, man good health  −0.92**  −0.36**  0.11  Man’s report  0.40*  0.21**  −0.11  Intercept  2.93  3.52***  2.40***  Unweighted N  1,006  1,005  1,006  *p < .05. **p < .01. ***p < .001. View Large Table 4. Coefficients from Logistic or OLS Regressions of Relationship Quality by Stepfamily Structure and Selected Characteristics Variable  Very close  Positive support  Negative support  Cohabiting stepfamily (vs married stepfamily)  −0.59*  −0.03  0.03  Both have a stepchild (vs only one has a stepchild)  0.21  −0.07  0.03  Has a joint child (vs no joint child)  −0.06  −0.02  0.05  Marital biographies   Marital history    Both previously married (ref)          Only one previously married  −0.29  −0.06  0.13    Neither previously married  −0.30  −0.13  0.14   Union duration/10  −0.01  −0.02  0.04  Sociodemographic characteristics   Man’s age/10  −0.04  −0.00  −0.04   Age heterogamy    Both similar age (ref)          Man older  0.28  0.05  0.02    Woman older  0.39  0.08  −0.07   Racial heterogamy    Both non-Hispanic white (ref)          Both black or Hispanic  −0.13  −0.06  0.04    Different race/ethnicity  −0.23  −0.10  0.11   Resident child status    No resident child (ref)          Resident child younger than 18  0.30  −0.01  0.02    Resident child 18 or older  0.40  0.06  0.01  Economic resources   Man’s education    Less than high school (ref)          High school  −0.25  0.14  0.02    Some college  −0.26  0.19  −0.05    College or more  −0.27  0.22  −0.04   Educational heterogamy    Both have same education (ref)          Man has more education  −0.62*  −0.21**  0.09    Woman has more education  −0.78**  −0.13  0.15   Employment heterogamy    Both working  −0.51*  −0.09  0.10    Only man working  −0.36  −0.01  0.01    Only woman working  −0.53  −0.08  −0.01    Neither working (ref)         Owns home  −0.41  −0.06  0.07   Assets  −0.09  0.00  −0.03  Health heterogamy   Both good health (ref)         Both poor health  −0.85*  −0.33**  0.39**   Man poor health, woman good health  −0.35  −0.17*  0.18*   Woman poor health, man good health  −0.92**  −0.36**  0.11  Man’s report  0.40*  0.21**  −0.11  Intercept  2.93  3.52***  2.40***  Unweighted N  1,006  1,005  1,006  Variable  Very close  Positive support  Negative support  Cohabiting stepfamily (vs married stepfamily)  −0.59*  −0.03  0.03  Both have a stepchild (vs only one has a stepchild)  0.21  −0.07  0.03  Has a joint child (vs no joint child)  −0.06  −0.02  0.05  Marital biographies   Marital history    Both previously married (ref)          Only one previously married  −0.29  −0.06  0.13    Neither previously married  −0.30  −0.13  0.14   Union duration/10  −0.01  −0.02  0.04  Sociodemographic characteristics   Man’s age/10  −0.04  −0.00  −0.04   Age heterogamy    Both similar age (ref)          Man older  0.28  0.05  0.02    Woman older  0.39  0.08  −0.07   Racial heterogamy    Both non-Hispanic white (ref)          Both black or Hispanic  −0.13  −0.06  0.04    Different race/ethnicity  −0.23  −0.10  0.11   Resident child status    No resident child (ref)          Resident child younger than 18  0.30  −0.01  0.02    Resident child 18 or older  0.40  0.06  0.01  Economic resources   Man’s education    Less than high school (ref)          High school  −0.25  0.14  0.02    Some college  −0.26  0.19  −0.05    College or more  −0.27  0.22  −0.04   Educational heterogamy    Both have same education (ref)          Man has more education  −0.62*  −0.21**  0.09    Woman has more education  −0.78**  −0.13  0.15   Employment heterogamy    Both working  −0.51*  −0.09  0.10    Only man working  −0.36  −0.01  0.01    Only woman working  −0.53  −0.08  −0.01    Neither working (ref)         Owns home  −0.41  −0.06  0.07   Assets  −0.09  0.00  −0.03  Health heterogamy   Both good health (ref)         Both poor health  −0.85*  −0.33**  0.39**   Man poor health, woman good health  −0.35  −0.17*  0.18*   Woman poor health, man good health  −0.92**  −0.36**  0.11  Man’s report  0.40*  0.21**  −0.11  Intercept  2.93  3.52***  2.40***  Unweighted N  1,006  1,005  1,006  *p < .05. **p < .01. ***p < .001. View Large Discussion Prior research has extensively examined stepfamilies earlier in the life course, but social scientists lack a basic understanding of the prevalence and complexity of stepfamilies in later life (van der Pas, van Tilburg, & Silverstein, 2013). In addition, past studies examining stepfamilies usually focus exclusively on married stepfamilies, ignoring cohabiting stepfamilies (Ganong & Coleman, 2017). The extent to which cohabiting stepfamilies are similar to or different from married stepfamilies is under-studied. This omission is likely to be increasingly consequential as the share of older adults in cohabiting unions accelerates and the proportion married declines (Lin & Brown, 2012; Stepler, 2017). To fill these notable gaps in the literature, we used a nationally representative sample from the 2012 HRS to describe the prevalence and composition of later-life stepfamilies, paying particular attention to married and cohabiting stepfamilies. We showed that roughly 40% of middle-aged and older couples with children were in stepfamilies in 2012. Slightly more than 10% of the couples in later-life stepfamilies were cohabiting. Cohabiting stepfamilies were more likely than married stepfamilies to have children from both partners’ previous relationships, but married stepfamilies were more likely than cohabiting stepfamilies to have a joint child. Our findings align with prior research showing that among younger people, stepfamilies are often more socially and economically disadvantaged than married families (Stewart, 2001; Thomson, 1994). Moreover, cohabiting stepfamilies had fewer social and economic resources than married stepfamilies. Despite these compositional differences, partner relationship quality was largely similar across married families, married stepfamilies, and cohabiting stepfamilies, which echoed earlier work showing that older individuals reported comparable marital quality in first versus remarriages (Bulanda, 2011,Cooney et al., 2016). Additionally, prior work indicated that across most dimensions of relationship quality, including partner support, older remarried and cohabiting adults reported similar levels of quality (Brown & Kawamura, 2010). The only exception was that cohabitors were less likely to be very happy than remarried individuals (Brown & Kawamura, 2010), which closely parallels our study findings. The sole family type differential we found was that couples in cohabiting stepfamilies were less likely than couples in either married families or married stepfamilies to report that their relationship with their partner was very close. In short, our overall picture of comparable relationship quality by family type aligns with prior research on older adult relationship quality. Our study offers new insights on relationship quality in later life by accounting not just for marriage order and family type but also stepfamily configuration. Contrary to our expectations, stepfamily configuration is immaterial to couple relationship quality. Couples in simple and complex stepfamilies enjoyed the same relationship quality. Likewise, having a joint child was unrelated to stepfamily relationship quality after the marital biographies, sociodemographic characteristics, economic resources, and health statuses of couples were taken into account. These patterns signal that stepfamily composition may be less salient later in the life course, at least in terms of relationship quality dynamics. This study has some limitations. First, we were not able to identify family types for 187 couples, and thus these couples were excluded from the analyses. The absence of clear information on the relationships of children to cohabiting parents likely reflects the boundary ambiguity of stepfamilies (Stewart, 2005). Because the majority of cohabiting couples are stepfamilies, our results presented here may slightly underestimate cohabiting stepfamilies. Second, the questions gauging relationship quality were only asked of roughly one-half (randomly selected) of the 2012 HRS sample. Consequently, we were not able to construct couple-level measures of relationship quality which arguably would have provided a richer, more nuanced portrait of the relationship dynamics characterizing married families, married stepfamilies, and cohabiting stepfamilies. Nonetheless, our study improved on prior research on later-life relationship quality by differentiating both by family type and stepfamily configuration. Indeed, we established that a substantial share of older couples with children are in stepfamilies. Our picture of how older married and cohabiting stepfamilies are faring compared with married families is mixed. On the one hand, married stepfamilies and cohabiting stepfamilies typically fared worse than married families across every social and economic indicator we examined. This pattern of disadvantage could have implications for individual well-being. On the other hand, the relationship quality of older couples in stepfamilies does not appreciably differ from that of older married couples, which may help to minimize family type variation in other domains of well-being. Regardless, we can expect the number of later-life stepfamilies to grow rapidly given the recent increases in older adult divorce and cohabitation, underscoring the need for additional research on the ramifications of stepfamilies for later-life well-being. Moreover, the rise in older adult stepfamilies raises important questions about caregiving and support. Cohabiting partners, for example, are much less likely to provide care to one another than are married spouses (Noël-Miller, 2011). Prior studies have shown that not only are adult stepchildren less likely than biological children to help their parents, but the presence of a stepsibling also discourages adult children from helping their biological parents in times of need (Pezzin, Pollak, & Schone, 2008, 2013). Current old-age policies and programs likely will need to be re-designed in response to the changing demographics of older adults who are increasingly living in a complex later-life family structure which is likely to increase demand for institutional care. Stepfamilies constitute two out of every five co-resident partnered relationships with children in later life, yet remain overlooked by gerontologists and family scholars alike. Our study offers some notable insights to later-life stepfamilies by establishing a demographic profile of married and cohabiting stepfamilies in comparison to married families. Stepfamilies tend to be socially and economically disadvantaged compared with couples in married families. Yet, the three family types enjoy largely similar relationship quality. Our work serves as a launching point for urgently needed future research on later-life stepfamilies, which are likely to comprise a growing share of older adults in the coming years. Funding This research was supported by a grant to the first two authors from the National Institute on Aging (R15AG047588). Additional support was provided by the Center for Family and Demographic Research, Bowling Green State University, which has core funding from the Eunice Kennedy Shriver National Institute of Child Health and Human Development (P2CHD0509059). Author Contributions I.-F. Lin and S. L. Brown conceived the study, planned the analyses, and wrote the article. I.-F. Lin and C. J. Cupka prepared the data file and performed the analyses. Conflict of Interest Lin serves on Journal Editorial Board. Acknowledgments An earlier version of this article was presented at the annual meeting of the Population Association of America, Chicago, Illinois, April 2017. Any opinions expressed here are solely those of the authors and not of the funding agency or center. References Acock, A. C . ( 2005). Working with missing values. Journal of Marriage and Family , 67, 1012– 1028. doi: 10.1111/j.1741-3737.2005.00191.x Google Scholar CrossRef Search ADS   Amato, P. R . ( 2010). Research on divorce: Continuing trends and new developments. Journal of Marriage and Family , 72, 650– 666. doi: 10.1111/j.1741-3737.2010.00723.x Google Scholar CrossRef Search ADS   Blieszner, R. , & Voorpostel, M . ( 2016). Families and aging: Toward an interdisciplinary family-level approach. In V. L. Bengtson & R. A. Settersten, Jr (Eds.), Handbook of theories of aging  ( 3rd ed., pp. 327– 348). New York: Springer. Booth, A. , & Edwards, J. N . ( 1992). Starting over: Why remarriages are more unstable. Journal of Family Issues , 13, 179– 194. doi: 10.1177/019251392013002004 Google Scholar CrossRef Search ADS PubMed  Boss, P. , Bryant, C. M. , & Mancini, J. A . ( 2016). Family stress management: A contextual approach  ( 3rd ed.). Thousand Oaks, CA: Sage. Brown, S. L. , & Booth, A . ( 1996). Cohabitation versus marriage: A comparison of relationship quality. Journal of Marriage and the Family , 58, 668– 678. doi: 10.2307/353727 Google Scholar CrossRef Search ADS   Brown, S. L., & Kawamura, S . ( 2010). Relationship quality among cohabitors and marrieds in older adulthood. Social Science Research , 39, 777– 786. doi: 10.1016/j.ssresearch.2010.04.010 Google Scholar CrossRef Search ADS PubMed  Brown, S. L. , Lee, G. R. , & Bulanda, J. R . ( 2006). Cohabitation among older adults: A national portrait. The Journals of Gerontology, Series B: Psychological Sciences and Social Sciences , 61, S71– S79. doi: 10.1093/geronb/61.2.S71 Google Scholar CrossRef Search ADS   Brown, S. L., & Lin, I.-F . ( 2012). The gray divorce revolution: Rising divorce among middle-aged and older adults, 1990–2010. The Journals of Gerontology, Series B: Psychological Sciences and Social Sciences , 67, 731– 741. doi: 10.1093/geronb/gbs089 Google Scholar CrossRef Search ADS   Brown, S. L. , Lin, I.-F. , Hammersmith, A. M. , & Wright, M. R. (2016). Later life marital dissolution and repartnership status: A national portrait. The Journals of Gerontology, Series B: Psychological Sciences and Social Sciences . doi: 10.1093/geronb/gbw051 Brown, S. L. , Manning, W. D. , & Payne, K. K . ( 2017). Relationship quality among cohabiting versus married couples. Journal of Family Issues , 38, 1730– 1753. doi: 10.1177/0192513x15622236 Google Scholar CrossRef Search ADS   Brown, S. L. , & Wright, M. R . ( 2016). Older adults’ attitudes toward cohabitation: Two decades of change. The Journals of Gerontology, Series B: Psychological Sciences and Social Sciences , 71, 755– 764. doi: 10.1093/geronb/gbv053 Google Scholar CrossRef Search ADS   Bulanda, J. R . ( 2011). Gender, marital power, and marital quality in later life. Journal of Women & Aging , 23, 3– 22. doi: 10.1080/08952841.2011.540481 Google Scholar CrossRef Search ADS PubMed  Cherlin, A . ( 1978). Remarriage as an incomplete institution. American Journal of Sociology , 84, 634– 650. doi: 10.1086/226830 Google Scholar CrossRef Search ADS   Cooney, T. M. , Proulx, C. M. , & Snyder-Rivas, L. A . ( 2016). A profile of later life marriages: Comparisons by gender and marriage order. In G. Gianesini & S. L. Blair (Eds.), Divorce, separation, and remarriage: The transformation of family  (Vol. 10, pp. 3– 37). Bingley, UK: Emerald Group. doi: 10.1108/S1530-353520160000010002 Google Scholar CrossRef Search ADS   Cox, M. J. , & Paley, B . ( 1997). Families as systems. Annual Review of Psychology , 48, 243– 267. doi: 10.1146/annurev.psych.48.1.243 Google Scholar CrossRef Search ADS PubMed  Downs, K. J. M . ( 2003). Family commitment, role perceptions, social support, and mutual children in remarriage: A test of uncertainty reduction theory. Journal of Divorce & Remarriage , 40, 35– 53. doi: 10.1300/J087v40n01_03 Google Scholar CrossRef Search ADS   Friedman, D. , Hechter, M. , & Kanazawa, S . ( 1994). A theory of the value of children. Demography , 31, 375– 401. doi: 10.2307/2061749 Google Scholar CrossRef Search ADS PubMed  Ganong, L. H. , & Coleman, M . ( 1988). Do mutual children cement bonds in stepfamilies? Journal of Marriage and the Family , 50, 687– 698. doi: 10.2307/352638 Google Scholar CrossRef Search ADS   Ganong, L. , & Coleman, M . ( 2017). Stepfamily relationships: Development, dynamics, and interventions . New York: Springer. doi: 10.1007/978-1-4899-7702-1 Google Scholar CrossRef Search ADS   Health and Retirement Study . ( 2011). Sample sizes and response rates . University of Michigan. Retrieved from http://hrsonline.isr.umich.edu/sitedocs/sampleresponse.pdf Hobart, C . ( 1991). Conflict in remarriage. Journal of Divorce & Remarriage , 15, 69– 86. doi: 10.1300/J087v15n03_04 Google Scholar CrossRef Search ADS   King, V. , & Scott, M. E . ( 2005). A comparison of cohabiting relationships among older and younger adults. Journal of Marriage and Family , 67, 271– 285. doi: 10.1111/j.0022-2445.2005.00115.x Google Scholar CrossRef Search ADS   Kurdek, L. A . ( 1999). The nature and predictors of the trajectory of change of marital quality for husbands and wives over the first 10 years of marriage. Developmental Psychology , 35, 1283– 1296. doi: 10.1037/0012-1649.35.5.1283 Google Scholar CrossRef Search ADS PubMed  Lin, I.-F. , & Brown, S. L . ( 2012). Unmarried Boomers confront old age: a national portrait. The Gerontologist , 52, 153– 165. doi: 10.1093/geront/gnr141 Google Scholar CrossRef Search ADS PubMed  Manning, W. D . ( 2004). Children and the stability of cohabiting couples. Journal of Marriage and Family , 66, 674– 689. doi: 10.1111/j.0022-2445.2004.00046.x Google Scholar CrossRef Search ADS   Noël-Miller, C. M . ( 2011). Partner caregiving in older cohabiting couples. The Journals of Gerontology, Series B: Psychological Sciences and Social Sciences , 66, 341– 353. doi: 10.1093/geronb/gbr027 Google Scholar CrossRef Search ADS   Nock, S. L . ( 1995). A comparison of marriages and cohabiting relationships. Journal of Family Issues , 16, 53– 76. doi: 10.1177/019251395016001004 Google Scholar CrossRef Search ADS   Ofstedal, M. B. , Weir, D. R. , Chen, K. T. , & Wagner, J . ( 2011). Updates to HRS sample weights  (HRS Report No. DR-013). University of Michigan. Retrieved from http://hrsonline.isr.umich.edu/sitedocs/userg/dr-013.pdf Pezzin, L. E. , Pollack, R. A. , & Schone, B. S . ( 2008). Parental marital disruption, family type, and transfers to disabled elderly parents. The Journals of Gerontology, Series B: Psychological Sciences and Social Sciences , 63, S349– S358. doi: 10.1093/geronb/63.6.S349 Google Scholar CrossRef Search ADS   Pezzin, L. E. , Pollak, R. A. , & Schone, B. S . ( 2013). Complex families and late-life outcomes among elderly persons: Disability, institutionalization, and longevity. Journal of Marriage and Family , 75, 1084– 1097. doi: 10.1111/jomf.12062 Google Scholar CrossRef Search ADS PubMed  Schultz, N. C. , Schultz, C. L. , & Olson, D. H . ( 1991). Couple strengths and stressors in complex and simple stepfamilies in Australia. Journal of Marriage and the Family , 53, 555– 564. doi: 10.2307/352732 Google Scholar CrossRef Search ADS   Seltzer, J. A., Yahirun, J. J., & Bianchi, S. M . ( 2013). Coresidence and geographic proximity of mothers and adult children in stepfamilies. Journal of Marriage and the Family , 75, 1164– 1180. doi: 10.1111/jomf.12058 Google Scholar CrossRef Search ADS PubMed  Shafer, K . ( 2013). Disentangling the relationship between age and marital history in age-assortative mating. Marriage & Family Review , 49, 83– 114. doi: 10.1080/01494929.2012.728557 Google Scholar CrossRef Search ADS   Skinner, K. B. , Bahr, S. J. , Crane, D. R. , & Call, V. R. A . ( 2002). Cohabitation, marriage, and remarriage: A comparison of relationship quality over time. Journal of Family Issues , 23, 74– 90. doi: 10.1177/0192513X02023001004 Google Scholar CrossRef Search ADS   Smith, J. , Ryan, L. , Sonnega, A. , & Weir, D . ( 2017). Psychosocial and lifestyle questionnaire 2006–2016: Documentation report core section LB . University of Michigan. Retrieved from https://hrs.isr.umich.edu/sites/default/files/biblio/HRS%202006–2016%20SAQ%20Documentation_07.06.17.pdf Stanley, S. M. , Markman, H. J. , & Whitton, S. W . ( 2002). Communication, conflict, and commitment: Insights on the foundations of relationship success from a national survey. Family Process , 41, 659– 675. doi: 10.1111/j.1545-5300.2002.00659.x Google Scholar CrossRef Search ADS PubMed  Steinbach, A. , & Hank, K . ( 2016). Intergenerational relations in older stepfamilies: A comparison of France, Germany, and Russia. The Journals of Gerontology, Series B: Psychological Sciences and Social Sciences , 71, 880– 888. doi: 10.1093/geronb/gbw046 Google Scholar CrossRef Search ADS   Stepler, R . ( 2017). Number of U.S. adults cohabiting with a partner continues to rise, especially among those 50 and older . Fact Tank: Pew Research Center. Retrieved from http://www.pewresearch.org/fact-tank/2017/04/06/number-of-u-s-adults-cohabiting-with-a-partner-continues-to-rise-especially-among-those-50-and-older/ Stewart, S. D . ( 2001). Contemporary American stepparenthood: Integrating cohabiting and nonresident stepparents. Population Research and Policy Review , 20, 345– 364. doi: 10.1023/A:1011895216970 Google Scholar CrossRef Search ADS   Stewart, S. D . ( 2005). Boundary ambiguity in stepfamilies. Journal of Family Issues , 26, 1002– 1029. doi: 10.1177/0192513X04273591 Google Scholar CrossRef Search ADS   Stewart, S. D . ( 2007). Brave new stepfamilies: Diverse paths toward stepfamily living . Thousand Oaks, CA: Sage. Sweeney, M. M . ( 2010). Remarriage and stepfamilies: Strategic sites for family scholarship in the 21st century. Journal of Marriage and Family , 72, 667– 684. doi: 10.1111/j.1741-3737.2010.00724.x Google Scholar CrossRef Search ADS   Thomson, E . ( 1994). “Settings” and “development” from a demographic point of view. In A. Booth & J. Dunn (Eds.), Stepfamilies: Who benefits? Who does not?  (pp. 89– 96). Hillsdale, NJ: Lawrence Erlbaum. van der Pas, S. , & van Tilburg, T. G . ( 2010). The influence of family structure on the contact between older parents and their adult biological children and stepchildren in the Netherlands. The Journals of Gerontology, Series B: Psychological Sciences and Social Sciences , 65, 236– 245. doi: 10.1093/geronb/gbp108 Google Scholar CrossRef Search ADS   van der Pas, S. , van Tilburg, T. G. , & Silverstein, M . ( 2013). Stepfamilies in later life. Journal of Marriage and Family , 75, 1065– 1069. doi: 10.1111/jomf.12054 Google Scholar CrossRef Search ADS   Ward, R. A. , Spitze, G. , & Deane, G . ( 2009). The more the merrier? Multiple parent–adult child relations. Journal of Marriage and Family , 71, 161– 173. doi: 10.1111/j.1741-3737.2008.00587.x Google Scholar CrossRef Search ADS   © The Author(s) 2017. Published by Oxford University Press on behalf of The Gerontological Society of America. All rights reserved. For permissions, please e-mail: journals.permissions@oup.com. http://www.deepdyve.com/assets/images/DeepDyve-Logo-lg.png The Journals of Gerontology Series B: Psychological Sciences and Social Sciences Oxford University Press

Loading next page...
 
/lp/ou_press/a-national-portrait-of-stepfamilies-in-later-life-y0wfzAQpS2
Publisher
Oxford University Press
Copyright
© The Author(s) 2017. Published by Oxford University Press on behalf of The Gerontological Society of America. All rights reserved. For permissions, please e-mail: journals.permissions@oup.com.
ISSN
1079-5014
eISSN
1758-5368
D.O.I.
10.1093/geronb/gbx150
Publisher site
See Article on Publisher Site

Abstract

Abstract Objectives Scholars have documented increases in the prevalence and complexity of stepfamilies earlier in the life course, but no one has systematically investigated U.S. stepfamily structure in later life. Guided by a family systems approach, we described the prevalence and composition of later-life stepfamilies. Method The analysis was based on 6,250 married and cohabiting couples participating in the 2012 Health and Retirement Study. We identified the prevalence of later-life stepfamilies, decomposed stepfamily structures, and compared the sociodemographic characteristics and relationship quality of the couples in stepfamilies with those in married families (with only joint children and no stepchildren), paying attention to differences between married and cohabiting stepfamilies. Results Roughly 40% of middle-aged and older couples with children were in stepfamilies. Of all stepfamilies, 86% were married couples and 14% were cohabiting couples. Cohabiting stepfamilies more often included children from both partners’ previous relationships, but couples in married stepfamilies more often had joint children. Cohabiting stepfamilies appeared to be the most socially and economically disadvantaged, followed by married stepfamilies, and lastly married families. Despite these compositional differences, partner relationship quality was largely similar across married families, married stepfamilies, and cohabiting stepfamilies. Discussion This study underscores the high prevalence and complexity of later-life stepfamilies and foregrounds the urgency of additional research on this topic. Blended family, Cohabitation, Remarriage, Stepchildren Today’s older adults came of age during 1970s and 1980s, an era marked by high divorce and remarriage rates. Many have experienced multiple union transitions over the adult life course. In 2015, close to 30% of individuals aged 50 and older had experienced two or more marriages (Authors’ calculation based on the American Community Survey). Cohabitation is also on the rise among older adults, with the number cohabiting doubling in the past decade (Stepler, 2017). Nearly all cohabitors aged 50 and older experienced the dissolution of a prior marital union (Brown, Lee, & Bulanda, 2006). Whether remarried or cohabiting, the vast majority of these repartnered middle-aged and older adults (hereafter older adults) have children from previous relationships, meaning that they are part of a stepfamily. Scholars have carefully documented increases in the prevalence and complexity of stepfamilies earlier in the life course (Sweeney, 2010), but no one has systematically investigated U.S. stepfamily configuration later in life. Stepfamily structure is complex not only because one or both partners have children from previous relationships but also because the couple may subsequently have children together (hereafter joint children) and form a blended family. When only one partner has children from a previous relationship, the stepfamily consists of one biological parent and one stepparent. When both partners have children from previous relationships, each partner is a biological parent of their own children and a stepparent of their partner’s children. When the couple subsequently has joint children, each partner is a biological parent of their own children from previous relationships, a stepparent of their partner’s children from previous relationships, and a biological parent of the children with their current partner. Prior studies on later-life stepfamilies often focus on individual parent–child dyads, overlooking these varied constellations of the stepfamily environment (Blieszner & Voorpostel, 2016). Understanding stepfamily configuration is important because the more complex the stepfamily structure, the more ambiguous the roles and expectations are, creating more opportunities for conflict and disagreement between partners (Hobart, 1991). Prior research has compared relationship quality among older adults in first marriages versus remarriages (Bulanda, 2011; Cooney, Proulx, & Snyder-Rivas, 2016) and in remarriages versus cohabiting unions (Brown & Kawamura, 2010), but these studies do not speak directly to how partner relationship quality varies across stepfamily types. Stepfamilies can be formed through first marriages, and remarriages do not always result in stepfamilies. The objective of this study is to use data from the 2012 Health and Retirement Study to construct a national portrait of later-life stepfamilies, documenting the prevalence and complexity of various stepfamily forms. Guided by a family systems approach (Cox & Paley, 1997) that takes into account both parents’ relationships with each child in the family, we examine how the marital biographies, sociodemographic characteristics, economic resources, and health statuses of the couples in stepfamilies differ from those in married families with only joint children and no stepchildren, hereafter referred to as married families (we do not use the term “intact families” to avoid the appearance of any value judgment). We pay close attention to differences between married and cohabiting stepfamilies. We also investigate the extent to which relationship quality varies by stepfamily configuration. This study provides the first systematic examination of U.S. later-life stepfamilies by elucidating the prevalence and composition of different stepfamily forms. Background The Growth in Older Adults Who Are Remarried or Cohabiting Today’s older adults are more likely to be remarried or cohabiting than any previous generation in U.S. history. In 1980, 7 million married individuals aged 50 and older were in a higher-order marriage (Authors’ calculation based on the Census). The number has risen to 20 million in 2015 (Authors’ calculation based on the American Community Survey). This growth is not merely an artifact of the aging of the population. In 1980, 19% of older adults were in higher-order marriages and by 2015 the share had climbed to 30%. At the same time, cohabitation among older adults has accelerated, quadrupling from less than 1 million to 4 million persons from 2000 to 2016 (Brown et al., 2006; Stepler, 2017). Regardless of whether they are remarried or cohabiting, most of these individuals have children from previous relationships, suggesting that a significant proportion of older adults are in stepfamilies. Yet, we lack national estimates of the prevalence and complexity of later-life stepfamilies. Nor do we have any understanding of partner relationship quality in these families. This oversight is likely attributable to two reasons. First, most of the attention surrounding stepfamilies has focused on minor children, on the premise that family instability resulting from parental divorce and repartnering is detrimental to child development (Amato, 2010). Later-life stepfamilies tend to be ignored because most children in these families have launched into independent adulthood. Second, the rise in cohabitation among older adults is a relatively new phenomenon that was not well recognized and documented by social scientists until the beginning of this century (King & Scott, 2005; Brown et al., 2006). Little attention has been given to later-life cohabiting stepfamilies. Several sociodemographic trends, however, suggest that later-life stepfamilies can no longer be ignored. The divorce rate after age 50 has doubled over the past two decades. Even if the rate remains constant over the next 20 years, the number of older adults who would experience divorce in 2030 would rise by one third (Brown & Lin, 2012). Divorced older adults are more likely to repartner than their widowed counterparts (Brown, Lin, Hammersmith, & Wright, 2016). Not only has cohabitation increased in later life, but older adult attitudes toward cohabitation have also become more favorable in recent decades (Brown & Wright, 2016). Together, the number of later-life stepfamilies is likely to surge in the coming years, but scholars are confronted with a startling paucity of knowledge about these families. Stepfamily Configuration Stepfamily structure is complex not only because one or both partners have children from previous relationships but also because the couple may subsequently have joint children. Stepfamilies in which only one partner has children from a previous relationship are simple stepfamilies, in contrast to complex stepfamilies in which both partners have children from previous relationships (Steinbach & Hank, 2016; van der Pas & van Tilburg, 2010). Whether simple or complex, the couple may subsequently have joint children and form a blended family. The composition of stepfamily structure is likely to vary depending on whether the couple is married or cohabiting. The presence of children from previous relationships reduces the odds that parents marry their partner (Manning, 2004), particularly when both partners have children from previous relationships. Thus, cohabiting stepfamilies are more often complex families compared with married stepfamilies. On the other hand, having joint children provides parents the motivation to marry (Stewart, 2007). Therefore, married stepfamilies more often have joint children relative to cohabiting stepfamilies. Surprisingly, given the growth in older adults who are remarried or cohabiting, no study has documented the prevalence and complexity of later-life stepfamilies. A handful of studies have examined stepparent–stepchild relationships in later life, but these studies tend to use individual parent–child dyads as the unit of the analysis (e.g., Pezzin, Pollak, & Schone, 2013; Seltzer, Yahirun, & Bianchi, 2013), disregarding the various constellations of the stepfamily environment. Some biological parent–biological child dyads may be part of a stepfamily in which the parent is married to or living with a partner who is a stepparent of the child. Depending on which parent–child dyad is selected for the study, the prevalence of stepfamilies may be underestimated. Without treating the family as a system (Cox & Paley, 1997) that takes into account how both parents are related to each child in the family, it is impossible to fully gauge the diverse range of stepfamily arrangements. Partner Relationship Quality in Stepfamilies Unpacking stepfamily configuration provides insights into how the family functions. Stepfamilies have been viewed as incomplete institutions (Cherlin, 1978), in which family norms and obligations are not clearly defined or broadly shared. This lack of institutionalized guidelines for stepfamilies can lead to disagreement, undermining partner relationship quality. Further, the role of parenting is often perceived as more ambiguous by cohabiting than married parents (Stewart, 2007), suggesting that cohabiting stepfamilies are the least institutionalized and most unstable type of stepfamily (Ganong & Coleman, 2017). Prior research has not directly compared couples’ relationship quality in married families, married stepfamilies, and cohabiting stepfamilies, but has shown that couples in a higher-order marriage are more likely to break up than couples in a first marriage (Booth & Edwards, 1992). Cohabitors tend to report poorer relationship quality than do married individuals (Brown & Booth, 1996; Nock, 1995; Skinner, Bahr, Crane, & Call, 2002). Nonetheless, recent studies on older adults suggest no difference in relationship quality between first marriages and remarriages (Bulanda, 2011; Cooney et al., 2016), and older cohabitors share similar relationship quality as remarried older adults (Brown & Kawamura, 2010). The union type differences in partner relationship quality are negligible probably because older couples have a longer union duration and older cohabitors tend to cohabit as an alternative to marriage (King & Scott, 2005). In addition, children are a major source of relationship strains in stepfamilies (Stanley, Markman, & Whitton, 2002), but most children have grown up and left home in older families. Not only does marital status matter, but stepfamily structure is also likely to affect partner relationship quality. When partners have children from previous relationships, oftentimes confusion arises regarding who is part of the family and what roles individuals within the family should play (Boss, Bryant, & Mancini, 2016). Having joint children further blurs family boundaries. Boundary ambiguity is stressful for family members and creates opportunities for conflict. Thus, the more complex the stepfamily structure, the less clear the guideline and norms are for role performance, leading to more strained relationships. Prior studies have revealed poorer partner relationship quality in complex stepfamilies than in simple stepfamilies (Schultz, Schultz, & Olson, 1991), but these studies are limited to small, convenience samples. Having a joint child may or may not improve relationship quality. Some scholars propose the value of children argument (Friedman, Hechter, & Kanazawa, 1994), suggesting that having a joint child reduces uncertainty in the relationship and increases commitment to each other. However, increasing the complexity of family subsystems with a combination of full-, step-, and half-siblings may result in more ambiguity (Ward, Spitze, & Deane, 2009). Empirical evidence thus far has centered on younger stepfamilies and is mixed, showing positive associations (Downs, 2003), negative associations (Kurdek, 1999), or no association (Ganong & Coleman, 1988). The Present Study This study has three objectives. First, we establish a national portrait of later-life stepfamilies by identifying the share of couples (in which at least one spouse or partner is aged 51 or older) in stepfamilies, distinguishing between married stepfamilies and cohabiting stepfamilies. Within stepfamilies, we further investigate stepfamily structure by considering whether it is a simple or complex stepfamily and whether the couple has a joint child. Next, we examine how the marital biographies, sociodemographic characteristics, economic resources, and health statuses of the couples in stepfamilies differ from those in married families. Prior studies have shown that stepparents are distinct from parents who have biological children (hereafter biological parents) in several regards (Stewart, 2001; Thomson, 1994). Stepparents are more likely to be in a higher-order marriage and have a shorter union duration than biological parents. Because cohabitation is more common at younger ages and men tend to repartner with younger women, stepparents are often younger than biological parents. The gap in age between partners is greater among repartnered couples and a larger share of stepparents are nonwhite. Relative to biological parents, stepparents have less education and lower earnings, despite higher levels of full-time employment. Moreover, cohabiting stepparents appear to be more socially and economically disadvantaged compared with married stepparents. Because these studies are based on younger samples, the extent to which these patterns persist among later-life stepfamilies is unknown. A key advantage of our study is that we have information on the characteristics of both partners and thus can examine couple-level indicators which aligns with a family systems perspective. Last, we examine how both positive and negative dimensions of partner relationship quality differ among those in married families, married stepfamilies, and cohabiting stepfamilies. Three relationship quality indicators gauge whether the relationship with the partner is very close, positive support received from the partner, and negative support received from the partner. A lack of closeness, lower levels of positive support, and higher levels of negative support signal poorer partner relationship quality. Prior research has revealed the highest relationship quality is typically in first marriages, followed by remarriages, and lastly cohabiting unions (Brown & Booth, 1996; Nock, 1995; Skinner et al., 2002). However, these differentials may be smaller or even negligible among older adults for whom there are few relationship quality differentials between married and cohabiting individuals (Brown & Kawamura, 2010) as well as first married and remarried individuals (Bulanda, 2011; Cooney et al., 2016), although studies to date have not differentiated by stepfamily status. We also anticipate that the more complex the stepfamily structure, the poorer the partner relationship quality. In other words, couples in complex stepfamilies are less likely to have a very close relationship and experience less positive support and more negative support than couples in simple stepfamilies, and having a joint child is expected to be associated with poorer relationship quality. Method We used data from the 2012 round of the Health and Retirement Study (HRS), a longitudinal study of a nationally representative, continuous cohort of individuals aged 51 or older and their partners in the United States. The HRS began interviewing in 1992 with a cohort of individuals born in 1931–1941 and re-interviews have been conducted every other year. In 1998, three additional cohorts were added to the study to make the sample representative of the target population. Refresher samples aged 51–56 were added in 2004 and 2010 to maintain a nationally representative sample of individuals aged 51 or older. The response rates for the baseline interviews of various cohorts hover around 70%–82% and roughly 90% or higher for follow-up interviews. Blacks, Hispanics, and Florida residents were oversampled (Health and Retirement Study, 2011). The HRS includes rich information on respondents’ and their partners’ marital history, demographic characteristics, employment history, wealth, and health status, making the data ideal for the purposes of this study. We focused on the 2012 HRS because this is the most recent wave in the family data file (2012, Version 1), cleaned and processed by RAND, at the time when this analysis was conducted. The file includes information delineating how each child mentioned by HRS respondents is related to both the HRS respondents and their partners at each wave. We took advantage of all available longitudinal data to measure family type and stepfamily configuration. We captured parent–child relationships using the most frequent relationship code across waves. In cases where there were ties or that respondents were first interviewed in 2012, we used the relationship code in 2012. We also checked our estimates against the newly released 2014 Survey of Income and Program Participation (SIPP) which contains questions asking respondents about their number of childbearing unions and whether their current union is a childbearing one. Our estimates are very similar to those based on the 2014 SIPP (results not shown but available upon request). In total, 20,554 respondents were interviewed in 2012, of which 13,080 lived in coupled households, constituting 6,842 couples. Same-sex (n = 34) and childless (n = 253) couples were excluded. We also removed 187 couples for whom one or both partners’ relationships to their children could not be ascertained in the RAND HRS family data file. In addition, we omitted 48 couples who provided inconsistent information about their marital status and 46 couples with a sample weight of zero. Finally, 24 cohabiting couples who had joint children only were excluded, because it is too small of a group to permit a separate analysis. After eliminating these couples, we had 6,250 couples with children in the analysis, of which 3,486 were married families, 2,346 were married stepfamilies, and 418 were cohabiting stepfamilies. Measures Stepfamily configuration was gauged using couples’ marital status, whether they had children from previous relationships, and whether they subsequently had a joint child. We created three family types: married families (in which married couples have only joint children and no stepchildren), married stepfamilies (reference category), and cohabiting stepfamilies. We also constructed two measures to capture stepfamily structure: whether only one or both partners had children from previous relationships (0 = only one partner, 1 = both partners) and whether the couple had a joint child (0 = No, 1 = Yes). Partner relationship quality was assessed using respondents’ reports of whether they were very close to their partner, as well as positive and negative support received from their partner, with an absence of being very close, less positive support, and more negative support indicating poorer relationship quality. In the HRS psychosocial and lifestyle survey, respondents were asked how close their relationship is with their partner (very close, quite close, not very close, or not at all close). We created a dichotomous measure such that those who were very close were coded 1 and all others were coded 0. Moreover, respondents were asked of three questions regarding positive support (understanding the way you feel about things, relying on them if you have a serious problem, and opening up to them if you need to talk about your worries) and four questions about negative support (making too many demands on you, criticizing you, letting you down when you are counting on them, and getting on your nerves) received from their partners. The response categories were a lot (= 1), some (= 2), a little (= 3), and not at all (= 4). We reverse coded all items and averaged the scores separately for positive and negative support. Following the recommendation of the HRS (Smith, Ryan, Sonnega, & Weir, 2017), we set the average scores to missing when the respondent did not report more than one item for positive support or when the respondent did not report more than two items for negative support. To reduce respondent burden, relationship quality questions were asked of a randomly selected half of the HRS sample. Thus, the closeness, positive support, and negative support questions were available for only 2,313, 2,309, and 2,310 couples, respectively. For couples in which both partners were selected for interviews in 2012 and provided their evaluations of relationship quality, we randomly selected one partner’s report. Because women and men tend to perceive their relationship quality differently, we included a dichotomous variable in the multivariate analysis to indicate whether a woman’s or a man’s report was used. Marital biographies included marital history and union duration. Marital history was appraised by whether both partners were previously married (reference category), only one was previously married, and neither was previously married. The duration of the current union was measured in years. Sociodemographic characteristics consisted of the man’s age and age heterogamy, racial heterogamy, and the presence of children in the household. Man’s age was measured in years. Age heterogamy was coded into three categories that are consistent with prior research (Brown, Manning, & Payne, 2017; Shafer, 2013): the man and the woman shared similar ages (i.e., the woman was less than 2 years older than the man or the man was less than 5 years older than the woman, reference category), the man was older (i.e., the man was 5 or more years older than the woman), and the woman was older (i.e., the woman was 2 or more years older than the man). Racial heterogamy was defined as both partners were non-Hispanic white (reference category), both partners were black or Hispanic, and the partners were of different race and ethnicity. The presence of children in the household was assessed using three categories: no resident child (reference category), the youngest resident child was less than 18 years old, and the youngest resident child was aged 18 or older. Economic resources included the man’s education and educational heterogamy, employment heterogamy, home ownership, and assets. Man’s educational attainment was composed of four categories: less than high school (reference category), high school graduate, some college, and college or more. Educational heterogamy was captured by whether both partners had achieved the same level of education (reference category), the man had achieved more education than the woman, and the woman had achieved more education than the man. Employment heterogamy was assessed by whether both partners were working, only the man was working, only the woman was working, and neither was working (reference category). Home ownership was a dichotomous measure (0 = No, 1 = Yes). The couple’s assets were measured in dollars and logged in the multivariate analyses to correct for skewness. Health status was measured by whether both partners reported their health was good, very good, or excellent (good health, reference category), both partners reported their health was fair or poor (poor health), the man reported poor health but the woman reported good health, and the woman reported poor health but the man reported good health. Analytic Strategy We conducted three analyses. First, we described the prevalence and complexity of later-life stepfamily structure using percentages. Second, we provided a national portrait of later-life stepfamilies by examining compositional differences among married families, married stepfamilies, and cohabiting stepfamilies using means and percentages (as appropriate). Third, we performed logistic regressions (for closeness) and ordinary least squares regressions (for positive and negative support) to examine whether and how partner relationship quality varied across different family type and stepfamily structure, adjusting for compositional differences. A multiple imputation procedure was used to handle missing cases except for relationship quality (ranging from less than 1% for any child residing in the household to 11% for home ownership), such that the missing value for a single covariate was imputed as a function of other covariates in the analysis (Acock, 2005). To preserve the randomness of imputed variables, the study results were based on 10 random, multiple-imputed replicates. All estimates were weighted using the 2012 household weights to adjust for the unequal probability of selection, nonresponse, and sample attrition (Ofstedal, Weir, Chen, & Wagner, 2011). Results Stepfamily Configuration In 2012, two of every five couples in which at least one partner was aged 51 or older were in stepfamilies (41%), as presented in Table 1. Of all stepfamilies, 86% were married couples and 14% were cohabiting couples. Among married stepfamilies, 54% were simple stepfamilies and 46% were complex stepfamilies. The corresponding numbers for cohabiting stepfamilies were 43% and 57%, respectively. The proportion of married stepfamilies with a joint child was nearly four times the proportion in cohabiting stepfamilies (38% vs 10%). Consistent with our expectations, cohabiting stepfamilies were more often complex than married stepfamilies, but married stepfamilies more often had a joint child than cohabiting stepfamilies. Table 1. Weighted Percentages for Stepfamily Configuration in 2012 Family type  Married family  Married stepfamily  Cohabiting stepfamily  Total  59.1  35.4  5.5  Stepfamily  —  86.4  13.6  Stepfamily configuration   Simple stepfamily  —  54.3  42.8   Complex stepfamilya  —  45.7  57.2   Does not have a joint child  —  62.3  89.7   Has a joint childa  —  37.7  10.3  Unweighted N  3,486  2,346  418  Family type  Married family  Married stepfamily  Cohabiting stepfamily  Total  59.1  35.4  5.5  Stepfamily  —  86.4  13.6  Stepfamily configuration   Simple stepfamily  —  54.3  42.8   Complex stepfamilya  —  45.7  57.2   Does not have a joint child  —  62.3  89.7   Has a joint childa  —  37.7  10.3  Unweighted N  3,486  2,346  418  aMarried stepfamilies significantly differ from cohabiting stepfamilies at p < .01. View Large Table 1. Weighted Percentages for Stepfamily Configuration in 2012 Family type  Married family  Married stepfamily  Cohabiting stepfamily  Total  59.1  35.4  5.5  Stepfamily  —  86.4  13.6  Stepfamily configuration   Simple stepfamily  —  54.3  42.8   Complex stepfamilya  —  45.7  57.2   Does not have a joint child  —  62.3  89.7   Has a joint childa  —  37.7  10.3  Unweighted N  3,486  2,346  418  Family type  Married family  Married stepfamily  Cohabiting stepfamily  Total  59.1  35.4  5.5  Stepfamily  —  86.4  13.6  Stepfamily configuration   Simple stepfamily  —  54.3  42.8   Complex stepfamilya  —  45.7  57.2   Does not have a joint child  —  62.3  89.7   Has a joint childa  —  37.7  10.3  Unweighted N  3,486  2,346  418  aMarried stepfamilies significantly differ from cohabiting stepfamilies at p < .01. View Large Variation Among Married Families, Married Stepfamilies, and Cohabiting Stepfamilies The characteristics of married families, married stepfamilies, and cohabiting stepfamilies are displayed in Table 2. As for marital biographies, couples in married families were the least likely to be in a higher-order marriage and were in unions of the longest average duration, followed by couples in married stepfamilies, and finally couples in cohabiting stepfamilies. Specifically, only 13% of the married families had at least one partner who was previously married, whereas nearly all of those in married stepfamilies (94%) or cohabiting stepfamilies (96%) were previously married. Indeed, couples in which both partners were previously married was the modal form of later-life stepfamilies, comprising approximately 62% of married stepfamilies and 80% of cohabiting stepfamilies. The average union duration was 41 years for married families, in sharp contrast to 20 years for married stepfamilies and 7 years for cohabiting stepfamilies. Table 2. Weighted Means or Percentages for the Characteristics of Coupled Family Types in 2012   Married family  Married stepfamily  Cohabiting stepfamily  Variable  (1)  (2)  (3)  Marital biographies   Marital history12,13,23    Both previously married  2.7  61.9  80.0    Only one previously married  10.0  32.5  16.2    Neither previously married  87.3  5.6  3.9   Union duration12,13,23  40.7  20.1  7.3  Sociodemographic characteristics   Man’s age12,13,23  65.7  63.3  61.4   Age heterogamy12,13    Both similar age  70.3  35.4  29.3    Man older  21.8  44.4  45.9    Woman older  7.9  20.2  24.9   Racial heterogamy12,13,23    Both non-Hispanic white  81.7  72.6  58.7    Both black or Hispanic  14.1  18.5  28.8    Different race/ethnicity  4.2  8.8  12.5   Resident child status12    No resident child  65.7  69.7  70.4    Youngest resident child younger than 18  10.3  10.9  10.4    Youngest resident child 18 or older  24.0  19.4  19.2  Economic resources   Man’s education12,13,23    Less than high school  10.8  13.4  18.6    High school  28.9  30.9  31.0    Some college  22.2  27.7  31.1    College or more  38.1  28.0  19.4   Educational heterogamy12    Both have same education  51.7  41.8  44.6    Man has more education  27.1  27.9  32.0    Woman has more education  21.1  30.2  23.4   Employment heterogamy12,13    Both working  38.8  42.1  45.2    Only man working  17.5  16.0  16.5    Only woman working  9.7  14.5  13.4    Neither working  34.0  27.3  24.9   Owns home12,13,23  92.1  83.8  64.6   Assets (in 10,000 median)12,13,23  31.2  17.4  8.7  Health heterogamy12,13,23   Both good health  68.2  61.5  50.6   Both poor health  6.9  8.2  13.6   Man poor health, woman good health  13.3  15.9  15.9   Woman poor health, man good health  11.6  14.4  19.9  Relationship quality   Very close12,13,23  66.8  56.7  43.4   Positive support12  3.5  3.4  3.4   Negative support  1.9  2.0  1.9  Unweighted N  3,486  2,346  418    Married family  Married stepfamily  Cohabiting stepfamily  Variable  (1)  (2)  (3)  Marital biographies   Marital history12,13,23    Both previously married  2.7  61.9  80.0    Only one previously married  10.0  32.5  16.2    Neither previously married  87.3  5.6  3.9   Union duration12,13,23  40.7  20.1  7.3  Sociodemographic characteristics   Man’s age12,13,23  65.7  63.3  61.4   Age heterogamy12,13    Both similar age  70.3  35.4  29.3    Man older  21.8  44.4  45.9    Woman older  7.9  20.2  24.9   Racial heterogamy12,13,23    Both non-Hispanic white  81.7  72.6  58.7    Both black or Hispanic  14.1  18.5  28.8    Different race/ethnicity  4.2  8.8  12.5   Resident child status12    No resident child  65.7  69.7  70.4    Youngest resident child younger than 18  10.3  10.9  10.4    Youngest resident child 18 or older  24.0  19.4  19.2  Economic resources   Man’s education12,13,23    Less than high school  10.8  13.4  18.6    High school  28.9  30.9  31.0    Some college  22.2  27.7  31.1    College or more  38.1  28.0  19.4   Educational heterogamy12    Both have same education  51.7  41.8  44.6    Man has more education  27.1  27.9  32.0    Woman has more education  21.1  30.2  23.4   Employment heterogamy12,13    Both working  38.8  42.1  45.2    Only man working  17.5  16.0  16.5    Only woman working  9.7  14.5  13.4    Neither working  34.0  27.3  24.9   Owns home12,13,23  92.1  83.8  64.6   Assets (in 10,000 median)12,13,23  31.2  17.4  8.7  Health heterogamy12,13,23   Both good health  68.2  61.5  50.6   Both poor health  6.9  8.2  13.6   Man poor health, woman good health  13.3  15.9  15.9   Woman poor health, man good health  11.6  14.4  19.9  Relationship quality   Very close12,13,23  66.8  56.7  43.4   Positive support12  3.5  3.4  3.4   Negative support  1.9  2.0  1.9  Unweighted N  3,486  2,346  418  Note. Column total may not equal 100% due to rounding error. Superscripts indicate that the two groups statistically differ at p < .05. View Large Table 2. Weighted Means or Percentages for the Characteristics of Coupled Family Types in 2012   Married family  Married stepfamily  Cohabiting stepfamily  Variable  (1)  (2)  (3)  Marital biographies   Marital history12,13,23    Both previously married  2.7  61.9  80.0    Only one previously married  10.0  32.5  16.2    Neither previously married  87.3  5.6  3.9   Union duration12,13,23  40.7  20.1  7.3  Sociodemographic characteristics   Man’s age12,13,23  65.7  63.3  61.4   Age heterogamy12,13    Both similar age  70.3  35.4  29.3    Man older  21.8  44.4  45.9    Woman older  7.9  20.2  24.9   Racial heterogamy12,13,23    Both non-Hispanic white  81.7  72.6  58.7    Both black or Hispanic  14.1  18.5  28.8    Different race/ethnicity  4.2  8.8  12.5   Resident child status12    No resident child  65.7  69.7  70.4    Youngest resident child younger than 18  10.3  10.9  10.4    Youngest resident child 18 or older  24.0  19.4  19.2  Economic resources   Man’s education12,13,23    Less than high school  10.8  13.4  18.6    High school  28.9  30.9  31.0    Some college  22.2  27.7  31.1    College or more  38.1  28.0  19.4   Educational heterogamy12    Both have same education  51.7  41.8  44.6    Man has more education  27.1  27.9  32.0    Woman has more education  21.1  30.2  23.4   Employment heterogamy12,13    Both working  38.8  42.1  45.2    Only man working  17.5  16.0  16.5    Only woman working  9.7  14.5  13.4    Neither working  34.0  27.3  24.9   Owns home12,13,23  92.1  83.8  64.6   Assets (in 10,000 median)12,13,23  31.2  17.4  8.7  Health heterogamy12,13,23   Both good health  68.2  61.5  50.6   Both poor health  6.9  8.2  13.6   Man poor health, woman good health  13.3  15.9  15.9   Woman poor health, man good health  11.6  14.4  19.9  Relationship quality   Very close12,13,23  66.8  56.7  43.4   Positive support12  3.5  3.4  3.4   Negative support  1.9  2.0  1.9  Unweighted N  3,486  2,346  418    Married family  Married stepfamily  Cohabiting stepfamily  Variable  (1)  (2)  (3)  Marital biographies   Marital history12,13,23    Both previously married  2.7  61.9  80.0    Only one previously married  10.0  32.5  16.2    Neither previously married  87.3  5.6  3.9   Union duration12,13,23  40.7  20.1  7.3  Sociodemographic characteristics   Man’s age12,13,23  65.7  63.3  61.4   Age heterogamy12,13    Both similar age  70.3  35.4  29.3    Man older  21.8  44.4  45.9    Woman older  7.9  20.2  24.9   Racial heterogamy12,13,23    Both non-Hispanic white  81.7  72.6  58.7    Both black or Hispanic  14.1  18.5  28.8    Different race/ethnicity  4.2  8.8  12.5   Resident child status12    No resident child  65.7  69.7  70.4    Youngest resident child younger than 18  10.3  10.9  10.4    Youngest resident child 18 or older  24.0  19.4  19.2  Economic resources   Man’s education12,13,23    Less than high school  10.8  13.4  18.6    High school  28.9  30.9  31.0    Some college  22.2  27.7  31.1    College or more  38.1  28.0  19.4   Educational heterogamy12    Both have same education  51.7  41.8  44.6    Man has more education  27.1  27.9  32.0    Woman has more education  21.1  30.2  23.4   Employment heterogamy12,13    Both working  38.8  42.1  45.2    Only man working  17.5  16.0  16.5    Only woman working  9.7  14.5  13.4    Neither working  34.0  27.3  24.9   Owns home12,13,23  92.1  83.8  64.6   Assets (in 10,000 median)12,13,23  31.2  17.4  8.7  Health heterogamy12,13,23   Both good health  68.2  61.5  50.6   Both poor health  6.9  8.2  13.6   Man poor health, woman good health  13.3  15.9  15.9   Woman poor health, man good health  11.6  14.4  19.9  Relationship quality   Very close12,13,23  66.8  56.7  43.4   Positive support12  3.5  3.4  3.4   Negative support  1.9  2.0  1.9  Unweighted N  3,486  2,346  418  Note. Column total may not equal 100% due to rounding error. Superscripts indicate that the two groups statistically differ at p < .05. View Large These three family types also differ in their sociodemographic characteristics. Compared with men in married families (66 years old), men in stepfamilies tended to be younger (63 and 61 years old for men in married stepfamilies and cohabiting stepfamilies, respectively). Age heterogamy was highest among cohabiting stepfamilies (70%), followed by married stepfamilies (64%), and then married families (29%). Relative to couples in married families, couples in stepfamilies were more likely to be minorities or of different racial and ethnic backgrounds, and this was more pronounced for cohabiting stepfamilies than married stepfamilies. Married stepfamilies were less likely than married families to have a child residing in the household, particularly an adult child (19% vs 24%). Concerning economic resources, men in married families had achieved the most education (38% with college or more), followed by men in married stepfamilies (28%) and then men in cohabiting stepfamilies (19%). Education heterogamy was more prevalent in married stepfamilies (58%) than in married families (48%), but cohabiting stepfamilies were no different from married families and married stepfamilies. Compared with married families, stepfamilies more often had both partners working or only women working, but no difference was found between married and cohabiting stepfamilies. Cohabiting stepfamilies were the least likely to own a home and accrued the lowest assets, followed by married stepfamilies and married families. Finally, couples in married families were the most likely to report that both partners had good health, while couples in cohabiting stepfamilies were most likely to report that both partners had poor health. In sum, couples in married families fare the best socially and economically, followed by couples in married stepfamilies and lastly couples in cohabiting stepfamilies. In the next section, we examine whether and how partner relationship quality varies by family type and stepfamily structure. Relationship Quality by Family Type and Stepfamily Structure As shown at the bottom of Table 2, couples in married families most often reported their relationship was very close (67%), followed by couples in married stepfamilies (57%) and then couples in cohabiting stepfamilies (43%). Moreover, couples in married families reported receiving slightly more positive support from their spouses than did couples in married stepfamilies. Nevertheless, there was no difference in positive support between married stepfamilies and cohabiting stepfamilies. Nor were there any differences in negative support across family types. The few significant bivariate associations largely disappeared after taking into account the differentials in marital biographies, sociodemographic characteristics, economic resources, and health status, as shown in Table 3. Although couples in cohabiting stepfamilies reported a lower likelihood of being very close compared with either couples in married families (−0.53 vs 0.37, p < .05) or remarried stepfamilies (−0.53 p < .05), married families and married stepfamilies were similarly likely to report their relationship was very close. Levels of both positive and negative support from the partner were comparable across married families, married stepfamilies, and cohabiting stepfamilies. Table 3. Coefficients From Logistic or OLS Regressions of Relationship Quality by Family Type and Selected Characteristics Variable  Very close  Positive support  Negative support  Married family  0.37  0.11  −0.06  Married stepfamily (ref)        Cohabiting stepfamily  −0.53*  0.01  −0.02  Marital biographies   Marital history    Both previously married (ref)    Only one previously married  −0.27  −0.09  0.16*    Neither previously married  −0.12  −0.17*  0.12   Union duration/10  0.01  0.03  0.01  Sociodemographic characteristics   Man’s age/10  −0.05  −0.04  −0.03   Age heterogamy    Both similar age (ref)    Man older  0.01  0.02  0.05    Woman older  0.00  0.01  −0.03   Racial heterogamy    Both non-Hispanic white (ref)    Both black or Hispanic  −0.11  −0.08  0.10*    Different race/ethnicity  −0.40  −0.17  0.19*   Resident child status    No resident child (ref)          Resident child younger than 18  −0.22  −0.12  0.08    Resident child 18 or older  0.19  −0.01  0.04  Economic resources   Man’s education    Less than high school (ref)    High school  0.27  0.14*  −0.02    Some college  0.21  0.17*  −0.07    College or more  0.05  0.17*  −0.03   Educational heterogamy    Both have same education (ref)    Man has more education  −0.33*  −0.10**  0.05    Woman has more education  −0.34*  −0.02  0.09   Employment heterogamy    Both working  −0.44*  −0.04  0.01    Only man working  −0.11  0.00  −0.03    Only woman working  −0.47*  −0.04  −0.01    Neither working (ref)   Owns home  −0.22  −0.03  0.03   Assets  −0.02  0.04  −0.06*  Health heterogamy   Both good health (ref)   Both poor health  −0.64**  −0.24**  0.28**   Man poor health, woman good health  −0.34*  −0.14**  0.14**   Woman poor health, man good health  −0.36  −0.24**  0.10  Man’s report  0.27*  0.17***  −0.08*  Intercept  1.45  3.07***  2.90***  Unweighted N  2,313  2,309  2,310  Variable  Very close  Positive support  Negative support  Married family  0.37  0.11  −0.06  Married stepfamily (ref)        Cohabiting stepfamily  −0.53*  0.01  −0.02  Marital biographies   Marital history    Both previously married (ref)    Only one previously married  −0.27  −0.09  0.16*    Neither previously married  −0.12  −0.17*  0.12   Union duration/10  0.01  0.03  0.01  Sociodemographic characteristics   Man’s age/10  −0.05  −0.04  −0.03   Age heterogamy    Both similar age (ref)    Man older  0.01  0.02  0.05    Woman older  0.00  0.01  −0.03   Racial heterogamy    Both non-Hispanic white (ref)    Both black or Hispanic  −0.11  −0.08  0.10*    Different race/ethnicity  −0.40  −0.17  0.19*   Resident child status    No resident child (ref)          Resident child younger than 18  −0.22  −0.12  0.08    Resident child 18 or older  0.19  −0.01  0.04  Economic resources   Man’s education    Less than high school (ref)    High school  0.27  0.14*  −0.02    Some college  0.21  0.17*  −0.07    College or more  0.05  0.17*  −0.03   Educational heterogamy    Both have same education (ref)    Man has more education  −0.33*  −0.10**  0.05    Woman has more education  −0.34*  −0.02  0.09   Employment heterogamy    Both working  −0.44*  −0.04  0.01    Only man working  −0.11  0.00  −0.03    Only woman working  −0.47*  −0.04  −0.01    Neither working (ref)   Owns home  −0.22  −0.03  0.03   Assets  −0.02  0.04  −0.06*  Health heterogamy   Both good health (ref)   Both poor health  −0.64**  −0.24**  0.28**   Man poor health, woman good health  −0.34*  −0.14**  0.14**   Woman poor health, man good health  −0.36  −0.24**  0.10  Man’s report  0.27*  0.17***  −0.08*  Intercept  1.45  3.07***  2.90***  Unweighted N  2,313  2,309  2,310  *p < .05. **p < .01. ***p < .001. View Large Table 3. Coefficients From Logistic or OLS Regressions of Relationship Quality by Family Type and Selected Characteristics Variable  Very close  Positive support  Negative support  Married family  0.37  0.11  −0.06  Married stepfamily (ref)        Cohabiting stepfamily  −0.53*  0.01  −0.02  Marital biographies   Marital history    Both previously married (ref)    Only one previously married  −0.27  −0.09  0.16*    Neither previously married  −0.12  −0.17*  0.12   Union duration/10  0.01  0.03  0.01  Sociodemographic characteristics   Man’s age/10  −0.05  −0.04  −0.03   Age heterogamy    Both similar age (ref)    Man older  0.01  0.02  0.05    Woman older  0.00  0.01  −0.03   Racial heterogamy    Both non-Hispanic white (ref)    Both black or Hispanic  −0.11  −0.08  0.10*    Different race/ethnicity  −0.40  −0.17  0.19*   Resident child status    No resident child (ref)          Resident child younger than 18  −0.22  −0.12  0.08    Resident child 18 or older  0.19  −0.01  0.04  Economic resources   Man’s education    Less than high school (ref)    High school  0.27  0.14*  −0.02    Some college  0.21  0.17*  −0.07    College or more  0.05  0.17*  −0.03   Educational heterogamy    Both have same education (ref)    Man has more education  −0.33*  −0.10**  0.05    Woman has more education  −0.34*  −0.02  0.09   Employment heterogamy    Both working  −0.44*  −0.04  0.01    Only man working  −0.11  0.00  −0.03    Only woman working  −0.47*  −0.04  −0.01    Neither working (ref)   Owns home  −0.22  −0.03  0.03   Assets  −0.02  0.04  −0.06*  Health heterogamy   Both good health (ref)   Both poor health  −0.64**  −0.24**  0.28**   Man poor health, woman good health  −0.34*  −0.14**  0.14**   Woman poor health, man good health  −0.36  −0.24**  0.10  Man’s report  0.27*  0.17***  −0.08*  Intercept  1.45  3.07***  2.90***  Unweighted N  2,313  2,309  2,310  Variable  Very close  Positive support  Negative support  Married family  0.37  0.11  −0.06  Married stepfamily (ref)        Cohabiting stepfamily  −0.53*  0.01  −0.02  Marital biographies   Marital history    Both previously married (ref)    Only one previously married  −0.27  −0.09  0.16*    Neither previously married  −0.12  −0.17*  0.12   Union duration/10  0.01  0.03  0.01  Sociodemographic characteristics   Man’s age/10  −0.05  −0.04  −0.03   Age heterogamy    Both similar age (ref)    Man older  0.01  0.02  0.05    Woman older  0.00  0.01  −0.03   Racial heterogamy    Both non-Hispanic white (ref)    Both black or Hispanic  −0.11  −0.08  0.10*    Different race/ethnicity  −0.40  −0.17  0.19*   Resident child status    No resident child (ref)          Resident child younger than 18  −0.22  −0.12  0.08    Resident child 18 or older  0.19  −0.01  0.04  Economic resources   Man’s education    Less than high school (ref)    High school  0.27  0.14*  −0.02    Some college  0.21  0.17*  −0.07    College or more  0.05  0.17*  −0.03   Educational heterogamy    Both have same education (ref)    Man has more education  −0.33*  −0.10**  0.05    Woman has more education  −0.34*  −0.02  0.09   Employment heterogamy    Both working  −0.44*  −0.04  0.01    Only man working  −0.11  0.00  −0.03    Only woman working  −0.47*  −0.04  −0.01    Neither working (ref)   Owns home  −0.22  −0.03  0.03   Assets  −0.02  0.04  −0.06*  Health heterogamy   Both good health (ref)   Both poor health  −0.64**  −0.24**  0.28**   Man poor health, woman good health  −0.34*  −0.14**  0.14**   Woman poor health, man good health  −0.36  −0.24**  0.10  Man’s report  0.27*  0.17***  −0.08*  Intercept  1.45  3.07***  2.90***  Unweighted N  2,313  2,309  2,310  *p < .05. **p < .01. ***p < .001. View Large Few control variables were associated with the three indicators of relationship quality. For example, couples in which both partners were in a higher-order marriage received more positive support and less negative support from their partners relative to couples in which both partner were in their first marriage or only one partner was remarried, respectively. Racial-ethnic minority couples and couples of different racial-ethnic backgrounds tended to report higher levels of negative support than white couples. Education was related to positive support from the partner with those without a high school diploma reporting the least positive support, on average. Educational heterogamy was negatively associated with both of the positive indicators of relationship quality, whereas employment heterogamy was only negatively related to the likelihood of reporting a very close relationship. Assets were largely unrelated to relationship quality, but couples with fewer assets reported higher levels of negative support from their partners. Health was consistently associated with the three dimensions of relationship quality such that couples in which both partners reported good health enjoyed the highest levels of positive quality and the lowest levels of negative quality. Finally, aligning with prior research on marital quality, men were more likely to characterize their unions as very close and tended to report higher levels of positive support and lower levels of negative support compared with women. Because married stepfamilies and cohabiting stepfamilies differ in the prevalence of simple versus complex stepfamilies and the presence of a joint child, we further examined whether and how relationship quality differed by stepfamily structure. As shown in Table 4, stepfamily configuration was not significantly associated with the three indicators of relationship quality after taking into account compositional differences. Couples in simple versus complex stepfamilies were similarly likely to report being very close and they reported comparable levels of positive and negative support from partners. Additionally, having a joint child was unrelated to perceptions of closeness as well as positive and negative support from the partner. Notably, the presence and age of the youngest child were not associated with relationship quality. For the most part, the relationships between the control variables and the three indicators of relationship quality remained largely unchanged from Table 3, although fewer achieved statistical significance (perhaps reflecting the smaller sample size or the inclusion of the measures of stepfamily configuration). There were no significant interactions between being in a simple versus complex stepfamily and having a joint child or between family type and stepfamily structure (results not shown). Table 4. Coefficients from Logistic or OLS Regressions of Relationship Quality by Stepfamily Structure and Selected Characteristics Variable  Very close  Positive support  Negative support  Cohabiting stepfamily (vs married stepfamily)  −0.59*  −0.03  0.03  Both have a stepchild (vs only one has a stepchild)  0.21  −0.07  0.03  Has a joint child (vs no joint child)  −0.06  −0.02  0.05  Marital biographies   Marital history    Both previously married (ref)          Only one previously married  −0.29  −0.06  0.13    Neither previously married  −0.30  −0.13  0.14   Union duration/10  −0.01  −0.02  0.04  Sociodemographic characteristics   Man’s age/10  −0.04  −0.00  −0.04   Age heterogamy    Both similar age (ref)          Man older  0.28  0.05  0.02    Woman older  0.39  0.08  −0.07   Racial heterogamy    Both non-Hispanic white (ref)          Both black or Hispanic  −0.13  −0.06  0.04    Different race/ethnicity  −0.23  −0.10  0.11   Resident child status    No resident child (ref)          Resident child younger than 18  0.30  −0.01  0.02    Resident child 18 or older  0.40  0.06  0.01  Economic resources   Man’s education    Less than high school (ref)          High school  −0.25  0.14  0.02    Some college  −0.26  0.19  −0.05    College or more  −0.27  0.22  −0.04   Educational heterogamy    Both have same education (ref)          Man has more education  −0.62*  −0.21**  0.09    Woman has more education  −0.78**  −0.13  0.15   Employment heterogamy    Both working  −0.51*  −0.09  0.10    Only man working  −0.36  −0.01  0.01    Only woman working  −0.53  −0.08  −0.01    Neither working (ref)         Owns home  −0.41  −0.06  0.07   Assets  −0.09  0.00  −0.03  Health heterogamy   Both good health (ref)         Both poor health  −0.85*  −0.33**  0.39**   Man poor health, woman good health  −0.35  −0.17*  0.18*   Woman poor health, man good health  −0.92**  −0.36**  0.11  Man’s report  0.40*  0.21**  −0.11  Intercept  2.93  3.52***  2.40***  Unweighted N  1,006  1,005  1,006  Variable  Very close  Positive support  Negative support  Cohabiting stepfamily (vs married stepfamily)  −0.59*  −0.03  0.03  Both have a stepchild (vs only one has a stepchild)  0.21  −0.07  0.03  Has a joint child (vs no joint child)  −0.06  −0.02  0.05  Marital biographies   Marital history    Both previously married (ref)          Only one previously married  −0.29  −0.06  0.13    Neither previously married  −0.30  −0.13  0.14   Union duration/10  −0.01  −0.02  0.04  Sociodemographic characteristics   Man’s age/10  −0.04  −0.00  −0.04   Age heterogamy    Both similar age (ref)          Man older  0.28  0.05  0.02    Woman older  0.39  0.08  −0.07   Racial heterogamy    Both non-Hispanic white (ref)          Both black or Hispanic  −0.13  −0.06  0.04    Different race/ethnicity  −0.23  −0.10  0.11   Resident child status    No resident child (ref)          Resident child younger than 18  0.30  −0.01  0.02    Resident child 18 or older  0.40  0.06  0.01  Economic resources   Man’s education    Less than high school (ref)          High school  −0.25  0.14  0.02    Some college  −0.26  0.19  −0.05    College or more  −0.27  0.22  −0.04   Educational heterogamy    Both have same education (ref)          Man has more education  −0.62*  −0.21**  0.09    Woman has more education  −0.78**  −0.13  0.15   Employment heterogamy    Both working  −0.51*  −0.09  0.10    Only man working  −0.36  −0.01  0.01    Only woman working  −0.53  −0.08  −0.01    Neither working (ref)         Owns home  −0.41  −0.06  0.07   Assets  −0.09  0.00  −0.03  Health heterogamy   Both good health (ref)         Both poor health  −0.85*  −0.33**  0.39**   Man poor health, woman good health  −0.35  −0.17*  0.18*   Woman poor health, man good health  −0.92**  −0.36**  0.11  Man’s report  0.40*  0.21**  −0.11  Intercept  2.93  3.52***  2.40***  Unweighted N  1,006  1,005  1,006  *p < .05. **p < .01. ***p < .001. View Large Table 4. Coefficients from Logistic or OLS Regressions of Relationship Quality by Stepfamily Structure and Selected Characteristics Variable  Very close  Positive support  Negative support  Cohabiting stepfamily (vs married stepfamily)  −0.59*  −0.03  0.03  Both have a stepchild (vs only one has a stepchild)  0.21  −0.07  0.03  Has a joint child (vs no joint child)  −0.06  −0.02  0.05  Marital biographies   Marital history    Both previously married (ref)          Only one previously married  −0.29  −0.06  0.13    Neither previously married  −0.30  −0.13  0.14   Union duration/10  −0.01  −0.02  0.04  Sociodemographic characteristics   Man’s age/10  −0.04  −0.00  −0.04   Age heterogamy    Both similar age (ref)          Man older  0.28  0.05  0.02    Woman older  0.39  0.08  −0.07   Racial heterogamy    Both non-Hispanic white (ref)          Both black or Hispanic  −0.13  −0.06  0.04    Different race/ethnicity  −0.23  −0.10  0.11   Resident child status    No resident child (ref)          Resident child younger than 18  0.30  −0.01  0.02    Resident child 18 or older  0.40  0.06  0.01  Economic resources   Man’s education    Less than high school (ref)          High school  −0.25  0.14  0.02    Some college  −0.26  0.19  −0.05    College or more  −0.27  0.22  −0.04   Educational heterogamy    Both have same education (ref)          Man has more education  −0.62*  −0.21**  0.09    Woman has more education  −0.78**  −0.13  0.15   Employment heterogamy    Both working  −0.51*  −0.09  0.10    Only man working  −0.36  −0.01  0.01    Only woman working  −0.53  −0.08  −0.01    Neither working (ref)         Owns home  −0.41  −0.06  0.07   Assets  −0.09  0.00  −0.03  Health heterogamy   Both good health (ref)         Both poor health  −0.85*  −0.33**  0.39**   Man poor health, woman good health  −0.35  −0.17*  0.18*   Woman poor health, man good health  −0.92**  −0.36**  0.11  Man’s report  0.40*  0.21**  −0.11  Intercept  2.93  3.52***  2.40***  Unweighted N  1,006  1,005  1,006  Variable  Very close  Positive support  Negative support  Cohabiting stepfamily (vs married stepfamily)  −0.59*  −0.03  0.03  Both have a stepchild (vs only one has a stepchild)  0.21  −0.07  0.03  Has a joint child (vs no joint child)  −0.06  −0.02  0.05  Marital biographies   Marital history    Both previously married (ref)          Only one previously married  −0.29  −0.06  0.13    Neither previously married  −0.30  −0.13  0.14   Union duration/10  −0.01  −0.02  0.04  Sociodemographic characteristics   Man’s age/10  −0.04  −0.00  −0.04   Age heterogamy    Both similar age (ref)          Man older  0.28  0.05  0.02    Woman older  0.39  0.08  −0.07   Racial heterogamy    Both non-Hispanic white (ref)          Both black or Hispanic  −0.13  −0.06  0.04    Different race/ethnicity  −0.23  −0.10  0.11   Resident child status    No resident child (ref)          Resident child younger than 18  0.30  −0.01  0.02    Resident child 18 or older  0.40  0.06  0.01  Economic resources   Man’s education    Less than high school (ref)          High school  −0.25  0.14  0.02    Some college  −0.26  0.19  −0.05    College or more  −0.27  0.22  −0.04   Educational heterogamy    Both have same education (ref)          Man has more education  −0.62*  −0.21**  0.09    Woman has more education  −0.78**  −0.13  0.15   Employment heterogamy    Both working  −0.51*  −0.09  0.10    Only man working  −0.36  −0.01  0.01    Only woman working  −0.53  −0.08  −0.01    Neither working (ref)         Owns home  −0.41  −0.06  0.07   Assets  −0.09  0.00  −0.03  Health heterogamy   Both good health (ref)         Both poor health  −0.85*  −0.33**  0.39**   Man poor health, woman good health  −0.35  −0.17*  0.18*   Woman poor health, man good health  −0.92**  −0.36**  0.11  Man’s report  0.40*  0.21**  −0.11  Intercept  2.93  3.52***  2.40***  Unweighted N  1,006  1,005  1,006  *p < .05. **p < .01. ***p < .001. View Large Discussion Prior research has extensively examined stepfamilies earlier in the life course, but social scientists lack a basic understanding of the prevalence and complexity of stepfamilies in later life (van der Pas, van Tilburg, & Silverstein, 2013). In addition, past studies examining stepfamilies usually focus exclusively on married stepfamilies, ignoring cohabiting stepfamilies (Ganong & Coleman, 2017). The extent to which cohabiting stepfamilies are similar to or different from married stepfamilies is under-studied. This omission is likely to be increasingly consequential as the share of older adults in cohabiting unions accelerates and the proportion married declines (Lin & Brown, 2012; Stepler, 2017). To fill these notable gaps in the literature, we used a nationally representative sample from the 2012 HRS to describe the prevalence and composition of later-life stepfamilies, paying particular attention to married and cohabiting stepfamilies. We showed that roughly 40% of middle-aged and older couples with children were in stepfamilies in 2012. Slightly more than 10% of the couples in later-life stepfamilies were cohabiting. Cohabiting stepfamilies were more likely than married stepfamilies to have children from both partners’ previous relationships, but married stepfamilies were more likely than cohabiting stepfamilies to have a joint child. Our findings align with prior research showing that among younger people, stepfamilies are often more socially and economically disadvantaged than married families (Stewart, 2001; Thomson, 1994). Moreover, cohabiting stepfamilies had fewer social and economic resources than married stepfamilies. Despite these compositional differences, partner relationship quality was largely similar across married families, married stepfamilies, and cohabiting stepfamilies, which echoed earlier work showing that older individuals reported comparable marital quality in first versus remarriages (Bulanda, 2011,Cooney et al., 2016). Additionally, prior work indicated that across most dimensions of relationship quality, including partner support, older remarried and cohabiting adults reported similar levels of quality (Brown & Kawamura, 2010). The only exception was that cohabitors were less likely to be very happy than remarried individuals (Brown & Kawamura, 2010), which closely parallels our study findings. The sole family type differential we found was that couples in cohabiting stepfamilies were less likely than couples in either married families or married stepfamilies to report that their relationship with their partner was very close. In short, our overall picture of comparable relationship quality by family type aligns with prior research on older adult relationship quality. Our study offers new insights on relationship quality in later life by accounting not just for marriage order and family type but also stepfamily configuration. Contrary to our expectations, stepfamily configuration is immaterial to couple relationship quality. Couples in simple and complex stepfamilies enjoyed the same relationship quality. Likewise, having a joint child was unrelated to stepfamily relationship quality after the marital biographies, sociodemographic characteristics, economic resources, and health statuses of couples were taken into account. These patterns signal that stepfamily composition may be less salient later in the life course, at least in terms of relationship quality dynamics. This study has some limitations. First, we were not able to identify family types for 187 couples, and thus these couples were excluded from the analyses. The absence of clear information on the relationships of children to cohabiting parents likely reflects the boundary ambiguity of stepfamilies (Stewart, 2005). Because the majority of cohabiting couples are stepfamilies, our results presented here may slightly underestimate cohabiting stepfamilies. Second, the questions gauging relationship quality were only asked of roughly one-half (randomly selected) of the 2012 HRS sample. Consequently, we were not able to construct couple-level measures of relationship quality which arguably would have provided a richer, more nuanced portrait of the relationship dynamics characterizing married families, married stepfamilies, and cohabiting stepfamilies. Nonetheless, our study improved on prior research on later-life relationship quality by differentiating both by family type and stepfamily configuration. Indeed, we established that a substantial share of older couples with children are in stepfamilies. Our picture of how older married and cohabiting stepfamilies are faring compared with married families is mixed. On the one hand, married stepfamilies and cohabiting stepfamilies typically fared worse than married families across every social and economic indicator we examined. This pattern of disadvantage could have implications for individual well-being. On the other hand, the relationship quality of older couples in stepfamilies does not appreciably differ from that of older married couples, which may help to minimize family type variation in other domains of well-being. Regardless, we can expect the number of later-life stepfamilies to grow rapidly given the recent increases in older adult divorce and cohabitation, underscoring the need for additional research on the ramifications of stepfamilies for later-life well-being. Moreover, the rise in older adult stepfamilies raises important questions about caregiving and support. Cohabiting partners, for example, are much less likely to provide care to one another than are married spouses (Noël-Miller, 2011). Prior studies have shown that not only are adult stepchildren less likely than biological children to help their parents, but the presence of a stepsibling also discourages adult children from helping their biological parents in times of need (Pezzin, Pollak, & Schone, 2008, 2013). Current old-age policies and programs likely will need to be re-designed in response to the changing demographics of older adults who are increasingly living in a complex later-life family structure which is likely to increase demand for institutional care. Stepfamilies constitute two out of every five co-resident partnered relationships with children in later life, yet remain overlooked by gerontologists and family scholars alike. Our study offers some notable insights to later-life stepfamilies by establishing a demographic profile of married and cohabiting stepfamilies in comparison to married families. Stepfamilies tend to be socially and economically disadvantaged compared with couples in married families. Yet, the three family types enjoy largely similar relationship quality. Our work serves as a launching point for urgently needed future research on later-life stepfamilies, which are likely to comprise a growing share of older adults in the coming years. Funding This research was supported by a grant to the first two authors from the National Institute on Aging (R15AG047588). Additional support was provided by the Center for Family and Demographic Research, Bowling Green State University, which has core funding from the Eunice Kennedy Shriver National Institute of Child Health and Human Development (P2CHD0509059). Author Contributions I.-F. Lin and S. L. Brown conceived the study, planned the analyses, and wrote the article. I.-F. Lin and C. J. Cupka prepared the data file and performed the analyses. Conflict of Interest Lin serves on Journal Editorial Board. Acknowledgments An earlier version of this article was presented at the annual meeting of the Population Association of America, Chicago, Illinois, April 2017. Any opinions expressed here are solely those of the authors and not of the funding agency or center. References Acock, A. C . ( 2005). Working with missing values. Journal of Marriage and Family , 67, 1012– 1028. doi: 10.1111/j.1741-3737.2005.00191.x Google Scholar CrossRef Search ADS   Amato, P. R . ( 2010). Research on divorce: Continuing trends and new developments. Journal of Marriage and Family , 72, 650– 666. doi: 10.1111/j.1741-3737.2010.00723.x Google Scholar CrossRef Search ADS   Blieszner, R. , & Voorpostel, M . ( 2016). Families and aging: Toward an interdisciplinary family-level approach. In V. L. Bengtson & R. A. Settersten, Jr (Eds.), Handbook of theories of aging  ( 3rd ed., pp. 327– 348). New York: Springer. Booth, A. , & Edwards, J. N . ( 1992). Starting over: Why remarriages are more unstable. Journal of Family Issues , 13, 179– 194. doi: 10.1177/019251392013002004 Google Scholar CrossRef Search ADS PubMed  Boss, P. , Bryant, C. M. , & Mancini, J. A . ( 2016). Family stress management: A contextual approach  ( 3rd ed.). Thousand Oaks, CA: Sage. Brown, S. L. , & Booth, A . ( 1996). Cohabitation versus marriage: A comparison of relationship quality. Journal of Marriage and the Family , 58, 668– 678. doi: 10.2307/353727 Google Scholar CrossRef Search ADS   Brown, S. L., & Kawamura, S . ( 2010). Relationship quality among cohabitors and marrieds in older adulthood. Social Science Research , 39, 777– 786. doi: 10.1016/j.ssresearch.2010.04.010 Google Scholar CrossRef Search ADS PubMed  Brown, S. L. , Lee, G. R. , & Bulanda, J. R . ( 2006). Cohabitation among older adults: A national portrait. The Journals of Gerontology, Series B: Psychological Sciences and Social Sciences , 61, S71– S79. doi: 10.1093/geronb/61.2.S71 Google Scholar CrossRef Search ADS   Brown, S. L., & Lin, I.-F . ( 2012). The gray divorce revolution: Rising divorce among middle-aged and older adults, 1990–2010. The Journals of Gerontology, Series B: Psychological Sciences and Social Sciences , 67, 731– 741. doi: 10.1093/geronb/gbs089 Google Scholar CrossRef Search ADS   Brown, S. L. , Lin, I.-F. , Hammersmith, A. M. , & Wright, M. R. (2016). Later life marital dissolution and repartnership status: A national portrait. The Journals of Gerontology, Series B: Psychological Sciences and Social Sciences . doi: 10.1093/geronb/gbw051 Brown, S. L. , Manning, W. D. , & Payne, K. K . ( 2017). Relationship quality among cohabiting versus married couples. Journal of Family Issues , 38, 1730– 1753. doi: 10.1177/0192513x15622236 Google Scholar CrossRef Search ADS   Brown, S. L. , & Wright, M. R . ( 2016). Older adults’ attitudes toward cohabitation: Two decades of change. The Journals of Gerontology, Series B: Psychological Sciences and Social Sciences , 71, 755– 764. doi: 10.1093/geronb/gbv053 Google Scholar CrossRef Search ADS   Bulanda, J. R . ( 2011). Gender, marital power, and marital quality in later life. Journal of Women & Aging , 23, 3– 22. doi: 10.1080/08952841.2011.540481 Google Scholar CrossRef Search ADS PubMed  Cherlin, A . ( 1978). Remarriage as an incomplete institution. American Journal of Sociology , 84, 634– 650. doi: 10.1086/226830 Google Scholar CrossRef Search ADS   Cooney, T. M. , Proulx, C. M. , & Snyder-Rivas, L. A . ( 2016). A profile of later life marriages: Comparisons by gender and marriage order. In G. Gianesini & S. L. Blair (Eds.), Divorce, separation, and remarriage: The transformation of family  (Vol. 10, pp. 3– 37). Bingley, UK: Emerald Group. doi: 10.1108/S1530-353520160000010002 Google Scholar CrossRef Search ADS   Cox, M. J. , & Paley, B . ( 1997). Families as systems. Annual Review of Psychology , 48, 243– 267. doi: 10.1146/annurev.psych.48.1.243 Google Scholar CrossRef Search ADS PubMed  Downs, K. J. M . ( 2003). Family commitment, role perceptions, social support, and mutual children in remarriage: A test of uncertainty reduction theory. Journal of Divorce & Remarriage , 40, 35– 53. doi: 10.1300/J087v40n01_03 Google Scholar CrossRef Search ADS   Friedman, D. , Hechter, M. , & Kanazawa, S . ( 1994). A theory of the value of children. Demography , 31, 375– 401. doi: 10.2307/2061749 Google Scholar CrossRef Search ADS PubMed  Ganong, L. H. , & Coleman, M . ( 1988). Do mutual children cement bonds in stepfamilies? Journal of Marriage and the Family , 50, 687– 698. doi: 10.2307/352638 Google Scholar CrossRef Search ADS   Ganong, L. , & Coleman, M . ( 2017). Stepfamily relationships: Development, dynamics, and interventions . New York: Springer. doi: 10.1007/978-1-4899-7702-1 Google Scholar CrossRef Search ADS   Health and Retirement Study . ( 2011). Sample sizes and response rates . University of Michigan. Retrieved from http://hrsonline.isr.umich.edu/sitedocs/sampleresponse.pdf Hobart, C . ( 1991). Conflict in remarriage. Journal of Divorce & Remarriage , 15, 69– 86. doi: 10.1300/J087v15n03_04 Google Scholar CrossRef Search ADS   King, V. , & Scott, M. E . ( 2005). A comparison of cohabiting relationships among older and younger adults. Journal of Marriage and Family , 67, 271– 285. doi: 10.1111/j.0022-2445.2005.00115.x Google Scholar CrossRef Search ADS   Kurdek, L. A . ( 1999). The nature and predictors of the trajectory of change of marital quality for husbands and wives over the first 10 years of marriage. Developmental Psychology , 35, 1283– 1296. doi: 10.1037/0012-1649.35.5.1283 Google Scholar CrossRef Search ADS PubMed  Lin, I.-F. , & Brown, S. L . ( 2012). Unmarried Boomers confront old age: a national portrait. The Gerontologist , 52, 153– 165. doi: 10.1093/geront/gnr141 Google Scholar CrossRef Search ADS PubMed  Manning, W. D . ( 2004). Children and the stability of cohabiting couples. Journal of Marriage and Family , 66, 674– 689. doi: 10.1111/j.0022-2445.2004.00046.x Google Scholar CrossRef Search ADS   Noël-Miller, C. M . ( 2011). Partner caregiving in older cohabiting couples. The Journals of Gerontology, Series B: Psychological Sciences and Social Sciences , 66, 341– 353. doi: 10.1093/geronb/gbr027 Google Scholar CrossRef Search ADS   Nock, S. L . ( 1995). A comparison of marriages and cohabiting relationships. Journal of Family Issues , 16, 53– 76. doi: 10.1177/019251395016001004 Google Scholar CrossRef Search ADS   Ofstedal, M. B. , Weir, D. R. , Chen, K. T. , & Wagner, J . ( 2011). Updates to HRS sample weights  (HRS Report No. DR-013). University of Michigan. Retrieved from http://hrsonline.isr.umich.edu/sitedocs/userg/dr-013.pdf Pezzin, L. E. , Pollack, R. A. , & Schone, B. S . ( 2008). Parental marital disruption, family type, and transfers to disabled elderly parents. The Journals of Gerontology, Series B: Psychological Sciences and Social Sciences , 63, S349– S358. doi: 10.1093/geronb/63.6.S349 Google Scholar CrossRef Search ADS   Pezzin, L. E. , Pollak, R. A. , & Schone, B. S . ( 2013). Complex families and late-life outcomes among elderly persons: Disability, institutionalization, and longevity. Journal of Marriage and Family , 75, 1084– 1097. doi: 10.1111/jomf.12062 Google Scholar CrossRef Search ADS PubMed  Schultz, N. C. , Schultz, C. L. , & Olson, D. H . ( 1991). Couple strengths and stressors in complex and simple stepfamilies in Australia. Journal of Marriage and the Family , 53, 555– 564. doi: 10.2307/352732 Google Scholar CrossRef Search ADS   Seltzer, J. A., Yahirun, J. J., & Bianchi, S. M . ( 2013). Coresidence and geographic proximity of mothers and adult children in stepfamilies. Journal of Marriage and the Family , 75, 1164– 1180. doi: 10.1111/jomf.12058 Google Scholar CrossRef Search ADS PubMed  Shafer, K . ( 2013). Disentangling the relationship between age and marital history in age-assortative mating. Marriage & Family Review , 49, 83– 114. doi: 10.1080/01494929.2012.728557 Google Scholar CrossRef Search ADS   Skinner, K. B. , Bahr, S. J. , Crane, D. R. , & Call, V. R. A . ( 2002). Cohabitation, marriage, and remarriage: A comparison of relationship quality over time. Journal of Family Issues , 23, 74– 90. doi: 10.1177/0192513X02023001004 Google Scholar CrossRef Search ADS   Smith, J. , Ryan, L. , Sonnega, A. , & Weir, D . ( 2017). Psychosocial and lifestyle questionnaire 2006–2016: Documentation report core section LB . University of Michigan. Retrieved from https://hrs.isr.umich.edu/sites/default/files/biblio/HRS%202006–2016%20SAQ%20Documentation_07.06.17.pdf Stanley, S. M. , Markman, H. J. , & Whitton, S. W . ( 2002). Communication, conflict, and commitment: Insights on the foundations of relationship success from a national survey. Family Process , 41, 659– 675. doi: 10.1111/j.1545-5300.2002.00659.x Google Scholar CrossRef Search ADS PubMed  Steinbach, A. , & Hank, K . ( 2016). Intergenerational relations in older stepfamilies: A comparison of France, Germany, and Russia. The Journals of Gerontology, Series B: Psychological Sciences and Social Sciences , 71, 880– 888. doi: 10.1093/geronb/gbw046 Google Scholar CrossRef Search ADS   Stepler, R . ( 2017). Number of U.S. adults cohabiting with a partner continues to rise, especially among those 50 and older . Fact Tank: Pew Research Center. Retrieved from http://www.pewresearch.org/fact-tank/2017/04/06/number-of-u-s-adults-cohabiting-with-a-partner-continues-to-rise-especially-among-those-50-and-older/ Stewart, S. D . ( 2001). Contemporary American stepparenthood: Integrating cohabiting and nonresident stepparents. Population Research and Policy Review , 20, 345– 364. doi: 10.1023/A:1011895216970 Google Scholar CrossRef Search ADS   Stewart, S. D . ( 2005). Boundary ambiguity in stepfamilies. Journal of Family Issues , 26, 1002– 1029. doi: 10.1177/0192513X04273591 Google Scholar CrossRef Search ADS   Stewart, S. D . ( 2007). Brave new stepfamilies: Diverse paths toward stepfamily living . Thousand Oaks, CA: Sage. Sweeney, M. M . ( 2010). Remarriage and stepfamilies: Strategic sites for family scholarship in the 21st century. Journal of Marriage and Family , 72, 667– 684. doi: 10.1111/j.1741-3737.2010.00724.x Google Scholar CrossRef Search ADS   Thomson, E . ( 1994). “Settings” and “development” from a demographic point of view. In A. Booth & J. Dunn (Eds.), Stepfamilies: Who benefits? Who does not?  (pp. 89– 96). Hillsdale, NJ: Lawrence Erlbaum. van der Pas, S. , & van Tilburg, T. G . ( 2010). The influence of family structure on the contact between older parents and their adult biological children and stepchildren in the Netherlands. The Journals of Gerontology, Series B: Psychological Sciences and Social Sciences , 65, 236– 245. doi: 10.1093/geronb/gbp108 Google Scholar CrossRef Search ADS   van der Pas, S. , van Tilburg, T. G. , & Silverstein, M . ( 2013). Stepfamilies in later life. Journal of Marriage and Family , 75, 1065– 1069. doi: 10.1111/jomf.12054 Google Scholar CrossRef Search ADS   Ward, R. A. , Spitze, G. , & Deane, G . ( 2009). The more the merrier? Multiple parent–adult child relations. Journal of Marriage and Family , 71, 161– 173. doi: 10.1111/j.1741-3737.2008.00587.x Google Scholar CrossRef Search ADS   © The Author(s) 2017. Published by Oxford University Press on behalf of The Gerontological Society of America. All rights reserved. For permissions, please e-mail: journals.permissions@oup.com.

Journal

The Journals of Gerontology Series B: Psychological Sciences and Social SciencesOxford University Press

Published: Nov 28, 2017

There are no references for this article.

You’re reading a free preview. Subscribe to read the entire article.


DeepDyve is your
personal research library

It’s your single place to instantly
discover and read the research
that matters to you.

Enjoy affordable access to
over 18 million articles from more than
15,000 peer-reviewed journals.

All for just $49/month

Explore the DeepDyve Library

Search

Query the DeepDyve database, plus search all of PubMed and Google Scholar seamlessly

Organize

Save any article or search result from DeepDyve, PubMed, and Google Scholar... all in one place.

Access

Get unlimited, online access to over 18 million full-text articles from more than 15,000 scientific journals.

Your journals are on DeepDyve

Read from thousands of the leading scholarly journals from SpringerNature, Elsevier, Wiley-Blackwell, Oxford University Press and more.

All the latest content is available, no embargo periods.

See the journals in your area

DeepDyve

Freelancer

DeepDyve

Pro

Price

FREE

$49/month
$360/year

Save searches from
Google Scholar,
PubMed

Create lists to
organize your research

Export lists, citations

Read DeepDyve articles

Abstract access only

Unlimited access to over
18 million full-text articles

Print

20 pages / month

PDF Discount

20% off