Partnering Across the Life Course: Sex, Relationships, and Mate Selection

Abstract

Marital delay, relationship dissolution and churning, and high divorce rates have extended the amount of time individuals in search of romantic relationships spend outside of marital unions. The scope of research on intimate partnering now includes studies of “hooking up,” Internet dating, visiting relationships, cohabitation, marriage following childbirth, and serial partnering, as well as more traditional research on transitions into marriage. Collectively, we know much more about relationship formation and development, but research often remains balkanized among scholars employing different theoretical approaches, methodologies, or disciplinary perspectives. The study of relationship behavior is also segmented into particular life stages, with little attention given to linkages between stages over the life course. Recommendations for future research are offered.

The nature and process of forming intimate relationships has changed in important ways over the past few decades. Previous “Decade in Review” articles focused on various aspects of relationship formation, ranging from adolescent pregnancy, premarital relationships, and mate selection to sexuality in relationships and families formed outside of marriage. These reviews dichotomized relationship behavior into romantic attachments preceding marriage and partnering that produced children. But dramatic changes in the timing and sequencing of relationship stages have made the study of intimate partnering more complex today than in the past. The scope of research has expanded to include studies of hookups and Internet dating, visiting relationships, cohabitation, marriage following childbirth, and serial partnering as well as more traditional research on transitions into marriage.

A unique challenge of reviewing research on partnering arises from changes in the marital behavior of Americans. Marital delay, relationship dissolution and churning, and high divorce rates have extended the amount of time substantial proportions of adults spend outside of formal marriage. Individuals select from a veritable smorgasbord of romantic options, including entering into casual, short-term sexual relationships; dating as an end toward finding a long-term partner; entering into shared living with a romantic partner (cohabitation) as an alternative to living alone; forming a cohabiting union as a precursor to marriage; or living with a partner as a substitute for formal marriage. Even though marriage remains among the most venerated of options (Cherlin, 2004; Edin, Kefalas, & Reed, 2004; Lichter, Batson, & Brown, 2004; Smock, 2004), it increasingly serves as a relationship capstone that takes place well after sexual involvement, shared living, and even childbearing and parenting (Carlson, McLanahan, & England, 2004) and may not even be a desired goal (Byrne & Carr, 2005; DePaulo & Morris, 2005).

A common thread unifying all relationships is a desire for intimacy—whether emotional or sexual. Involvement in romantic relationships, as a spouse, a cohabiting partner, or in a steady dating partnership, is beneficial to mental and physical health and sense of well-being (Kamp Dush & Amato, 2005; Williams & Umberson, 2004), though the benefits vary by race, gender, social class, parental status, and union type. Partnering behaviors change over the life course, both for structural (e.g., economic barriers or limited marriage market opportunities) and behavioral reasons (e.g., changing marital aspirations). Research published in social science journals over the last decade suggests that the behaviors and goals of emerging and young adults are widely divergent from older single adults. The topics studied differ dramatically, as does the frequency of coverage and how partnering behavior is framed as problematic or beneficial. Scholars from many different disciplines study partnering and parenting, but seldom is the research truly interdisciplinary, synergistic, or even complementary. This disciplinary balkanization is reflected in the theoretical approaches utilized, the data sources employed, and ultimately the knowledge produced.

This review identifies, synthesizes, and critiques the theoretical, methodological, and substantive research on heterosexual partnering. It examines the research on the formation and development of voluntary romantic relationships marked by expressions of affection, including physical intimacy and the expectation or experience of sexual relations (though see Donnelly & Burgess, 2008, for a study of celibacy in committed relationships). Following the major tropes introduced by the research, this review covers various life course stages. Like previous “Reviews,” it highlights recent research on relationship formation among adolescents and emerging (or young) adulthood, but it also examines changing patterns among midlife and older adults. Given the scope of the topic, it is not possible to review all forms of partnering behavior. This review therefore is limited to heterosexual partnering in the United States. Country-specific policies regarding union formation and parenting often differentiate relationship processes and outcomes, and other reviews in this issue cover same-sex partnerships and postmarital relationships.

Partnering: Examining Couple Formation From the Perspective of the Individual

Though seldom addressed from a dyadic perspective, research on the process of how two individuals become a couple is a central focus of much research on premarital relationships, sexuality, and mate selection. The bulk of research on Americans’ relationship formation behavior relies on data and analysis of individuals, though sometimes partners are examined in tandem, as when marital communication is observed or both partners are surveyed. Studies of partnering behavior may take the form of examining the traits preferred in partners (Stewart, Stinnett, & Rosenfeld, 2000), the behaviors engaged in during the preliminary stages of getting to know someone (O’Sullivan, Cheng, Harris, & Brooks-Gunn, 2007), what causes respondents to be more or less happy or satisfied in relationships (Arriaga, 2001; Sprecher, 2001), or the factors leading up to the decision to become sexually intimate, live together, or marry (Guzzo, 2006; Manning & Smock, 2002; Meier, 2007; Sassler, 2004; Uecker, 2008).

People imbue relationships with different meanings and approach them with varying goals. Individuals may desire particular attributes in a partner and actively seek them, without success, whereas not all physical intimacy between two individuals results in the establishment of a romantic relationship. Defining what constitutes partnering behavior is challenging; the unit of analysis can shift depending upon the research question. Studies of relationship quality, for example, often combine cohabitors and marrieds, even as research on transitions to marriage frequently groups cohabitors with singles who may or may not be in dating relationships (Surra, Gray, Cottle, & Boettcher, 2004). For the purpose of this review, I focus on partnering among unmarried adults, defining partnering as the formation and development of intimate relationships, which may be short in duration or lead to a stable marriage. This perspective encompasses the behaviors engaged in pursuit of that goal, the processes that enhance or impede the development of intimate relationships, and factors differentiating the union types entered.

Methodological Advances

Several advances characterize the research on partnering in the first decade of the 21st century. Large data collections, including longitudinal panel data, have reshaped conventional theoretical approaches to partnering behaviors. Many scholars have also gathered their own data and conducted smaller scale experiments. Even as the release of new nationally representative data has greatly expanded what is known about relationship formation and development, the study of relationship behavior has become increasingly balkanized into particular life stages, with certain behaviors studied for one population but not another.

Research on adolescents and emerging adults (spanning the early teens through the mid-20s) has proliferated in the past decade, abetted by the supplementation of several longitudinal data collections. The third wave of the National Longitudinal Study of Adolescent Health (Add Health), released in 2002, in conjunction with the earlier waves (1995 and 1996), allows researchers to examine the partnering behaviors of youth from middle-school (beginning with Grade 7) through their mid-20s. The 10th round of the National Longitudinal Survey of Youth 1997 (NLSY97), which was gathered in 2006 – 2007, provides detailed information on a similarly aged cohort (those born between 1980 and 1984). Wave 3 of the National Survey of Families and Households (NSFH), gathered in 2002, permits the study of the focal children of householders from the initial wave of the NSFH. Focusing on a particular age cohort, Cycle 6 (2002) of the National Survey of Family Growth (NSFG) provides detailed information on the sexual partnering and fertility experiences of respondents age 15 to 45, for the first time including data on men as well as women.

Several of these data sets, such as the NSFG and NSFH, are also useful for examining the partnering processes for adults in their 20s through midlife. Another widely used data source, the NLSY79, which follows men and women who were age 14 to 22 in 1979, has also been supplemented with new waves of data; as of Round 22 (2006), respondents were age 40 or older. Though many of these studies also contain information on important relationship dates, they are more often limited to cohabiting and marital unions and parenting.

What of those who have aged out of the reproductive years? Although their population share is projected to increase dramatically over the next few decades, information on partnering behavior is most limited for adults 45 years and older. The NSFH includes data on older respondents, as does the General Social Survey. But these sources include far less detail on the formation of sexual relationships than do data collections targeted at younger populations. Though it has not yet been extensively mined, the National Social Life, Health, and Aging Project (NSHAP), which explores the health and well-being of American men and women age 57 to 84, allows for the study of intimacy and sexuality among older adults. These data should be used to expand research on the partnering behavior of mature adults.

Family scholars have also turned to well-designed longitudinal data collections—including quantitative and qualitative components— focused on particular populations. Two large-scale multisite surveys have been the source of numerous studies of the partnering behavior of low-income and single parents. The Fragile Families and Child Well-Being Study followed a cohort of nearly 5,000 children born in large U.S. cities between 1998 and 2000 through the 5th year of life and included information on the relationship processes of the mothers and fathers of these children. It also included a qualitative component (Time, Love, and Cash in Couples with Children) consisting of four waves of individual and couple interviews with a parent who experienced a birth in 2000 (cf. England & Edin, 2007). The Three-City Study focused on the well-being of low-income children and their families in Boston, Chicago, and San Antonio, including surveys conducted in 1999, 2001, and 2005, and an ethnographic study of 256 children and families living in the same neighborhoods as the survey sample. Regional data collections have also increased. Scholars have been prolific in their use of the four-wave Toledo Adolescent Relationship Study (TARS), collected between 2001 and 2007 and including both survey and qualitative components. Several large qualitative data collections, such as that funded by the MacArthur Network on Transitions to Adulthood (cf. Settersten, Furstenberg, & Rumbaut, 2005) and the collaboration between scholars at the University of Michigan and Bowling Green State University on the meaning of cohabitation (cf. Manning & Smock, 2005), have also provided new sources for understanding the interpersonal processes involved in young adults’ partnering behaviors.

Advances in the analytical approaches utilized to assess the partnering behavior of adults across the life course have unfortunately not kept pace with the increased availability of rich data sources. The expansion of multiwave longitudinal data collections has extended the use of repeat pooled time-series analyses. Utilization of newer analytic advances designed to account for time-invariant sources of heterogeneity (fixed effects analysis), selection (propensity score matching, difference-in-difference models), or changes in trajectories of repeated measures (latent class analysis or growth curve models) are, however, underutilized in the extant research. This is surprising because these new data sources include sufficient detail on sexual and coresidential history to at least account for within-person change that might reduce omitted variable bias, to consider the potential bias introduced by differential selection into (or out of) particular behaviors or union statuses (Meier, 2007), or to assess varied pathways into partnerships and parenthood (Amato et al., 2008). Even though increased utilization of technical “fixes” risks further reifying disciplinary boundaries, their broader dissemination could also lower such barriers by simulating experimental approaches, enabling cross-disciplinary discussion, and better approximating causal processes.

Theoretical Frameworks

A number of scholarly theories are dominant among those who explore the processes underlying heterosexual attraction, partnering, and mate selection. Close relationship or interpersonal process models of heterosexual partnering provide theoretical guidance to many. They are popular because they acknowledge the diverse contexts in which relationships develop (see Cate, Levin, & Richmond, 2002; Conger, Cui, Bryant, & Elder, 2000). For example, family scholars have utilized attachment theory to examine various behavioral and affective phenomena in relationship formation, with an emphasis on dating and the selection of marital partners (e.g., Eastwick & Finkel, 2008a; Steinberg, Davila, & Fincham, 2006). Several comprehensive reviews have suggested that these studies are often intellectually segregated, however; scholars seldom reference research from outside their own disciplinary specialties (Surra et al., 2004).

Other established approaches such as exchange theory are most often the purview of sociologists and demographers. A social exchange perspective is based on the premise that relationship development and advancement is based on the satisfactory trade of rewards between partners, costs associated with involvement, and alternative possibilities. Much of the research on relationships that cross racial boundaries, for example, relies on an exchange perspective (e.g., Qian & Lichter, 2007). But variants such as equity theory are increasingly utilized by other disciplines to examine relationship progression, satisfaction, commitment, and stability (e.g., Rhoades, Stanley, & Markman, 2006; Sprecher, 2001).

More recently, theoretical approaches to assessing relationship formation have incorporated life course and feminist perspectives. The life course framework examines how individuals’ transitions and trajectories are linked across the age span and has been applied to topics such as how the race of initial romantic partners affects subsequent mate choice (King & Bratter, 2007) and the impact of prior marital and parenting experiences on entrance into postmarital unions (Lichter & Qian, 2008; Waller & Peters, 2008). Feminist theory, which studies how gender is reproduced through individual socialization and interpersonal actions, has also emerged as an approach to assessing the establishment and progression of intimate relationships (e.g., Bentley, Galliher, &Ferguson, 2007; Smiler, 2008). And sociobiological theory, which emphasizes the ways that evolutionary factors govern sexual and romantic preferences in mate selection, continues to emerge in studies of partner preferences (Buunk, Dijkstra, Fetchenhauer, & Kenrick, 2002; Stewart et al., 2000), though psychologists using experimental designs are challenging the very premises central to the approach (e.g., Eastwick & Finkel, 2008b). Collectively, these contributions have deepened and expanded the research literature on partnering. Other theoretical perspectives are utilized to study partnering behavior, of course, but are often specific to particular disciplines and as a result are not reviewed here.

Partnering Across the Life Course

The decision to enter into a romantic relationship, preferences for partner attributes, and goals for relationships vary widely across the life course. Adolescents and emerging adults pursue partnerships with different goals than do older single adults or previously married middle-age individuals; time horizons and desired ends also shape relationship behaviors among individuals of similar ages. For example, both women and men are less selective when asked about desired attributes for short-term versus long-term relationship partners; more minimal levels of relationship involvement yield stated preferences for lower levels of education, physical attractiveness, and (among men) relative intelligence (Buunk et al., 2002; Stewart et al., 2000). Emerging adults who desire marriage in their early 20s engage in different relationship patterns than do those whose marital horizons are later; not only do they express more conservative sexual attitudes and engage in fewer risky behaviors (binge drinking, cigarette smoking, and use of illegal drugs; Carroll et al., 2007), but they are less likely to engage in premarital sexual activity (Gaughan, 2002; Uecker, 2008). Finally, because the marriage market changes with age, preferences for desired partner attributes and methods of finding romantic partners shift. Adults who are parents or previously married are more tolerant of prospective mates who are divorced or have children (Goldscheider, Kaufman, & Sassler, 2009). Individuals are less likely to find romantic partners at school, and the workplace is often gender segregated. New approaches to finding romantic partners include Internet dating and speed-dating events (Eastwick & Finkel, 2008b; Feliciano, Robnett, & Komaie, 2009).

Along with disciplinary differences in how and what partnering behavior is studied, the availability of information from varying data sources has conditioned the production of the last decade’s research among those utilizing large nationally representative surveys. The cumulative results of research on the patterns and progression of adolescents’ partnering behaviors provide a far more comprehensive portrait of sexual progression than is available for older adults, including data on intimate fondling (touching partners under or without clothes or touching genitals), talking about birth control or sexually transmitted infections (STIs), and experiences with oral, anal, as well as vaginal intercourse (Brewster & Tillman, 2008; O’Sullivan et al., 2007). This influences what is studied at different life stages, with an (over)emphasis on sexual partnering among younger adults (those in their teens through mid-20s) in comparison to the transitions into shared living, whether cohabitation or marriage, and relationship quality among slightly older Americans. Current research also has given short shrift to (re)partnering at older ages. There is, therefore, much room to even out what is studied across the life course.

Partnering in Adolescence and Emerging Adulthood

The increase in the median age at first marriage in the United States means that most young adults will form romantic relationships—perhaps many relationships—well before they wed. Scholarly research on partnering in the last decade found that forming romantic relationships and selecting mates for the long term were central preoccupations for adolescents and emerging adults (Crissey, 2005; Nieder & Seiffge-Krenke, 2001). By age 15, nearly half of adolescents reported having engaged in a romantic relationship within the past 18 months, a figure that increased to nearly 70% by age 18 (Carver, Joyner, & Udry, 2003). Although learning how to be in a relationship is a normative developmental task of adolescents (those younger than age 18), the research suggests that this learning period extends to those who are emerging adults (those 18 to 25 years; Raley, Crissey, & Muller, 2007). Psychologists have documented how the emphasis of relationships shifts from companionship and affiliation among adolescents to trust and support in young adulthood (Collins, 2003; Furman, 2002; Shulman & Kipnis, 2001).

There has been a resurgence of interest in the trajectory of involvement within adolescent relationships, as scholars reject the premise that such relationships are developmentally insignificant or “trivial and transitory” (Collins, 2003, p. 4; Furman, 2002). More is known about the extent to which adolescents and emerging adults date, how dating behavior evolves over time, and relationship formation and progression (Carver et al., 2003; Mongeau, Jacobsen, & Donnerstein, 2007). As they age, adolescents’ growing involvement with mixed-gender friendship groups facilitates increased dating activity (Connolly, Furman, & Konarski, 2000). Social and romantic activities are important components of the relationship development sequence for the majority of adolescents. Hanging out with their partner and friends, meeting a partner’s parents, holding hands, and telling others they were in a relationship generally preceded sexual involvement (O’Sullivan et al., 2007), though the content of adolescent relationships varied by race and ethnicity. White adolescents were significantly more likely than Black, Asian, and Hispanic youth to report being introduced to their partner’s parent, holding hands, and informing friends that they were part of a couple, whereas Asian and Hispanic adolescents did not engage in precursor sexual events, such as intimate fondling, to the same extent as their White and Black counterparts (O’Sullivan et al.). Black youth also reported less interaction and disclosure with romantic partners and were less likely to be exclusive than their White counterparts (Giordano, Manning, & Longmore, 2005). Further justifying this growing emphasis on earlier stages of the life course are several studies whose findings document continuity between adolescent and young adult relationship experiences. Participation in serious romantic relationships in adolescence increase White youths’ marital expectations (Crissey, 2005); they also affect subsequent partnering behavior because those involved in romantic and sexual relationships during high school have an increased likelihood of forming cohabiting and marital unions by their early 20s (Gassanov, Nicholson, & Koch-Turner, 2008; Raley et al., 2007; Uecker & Stockes, 2008).

The past decade has also experienced a surge of interest in the attributes of partners selected and the impact this has on relationship acceptance, stability, and quality. Adolescents generally select romantic partners who are similar to themselves in terms of academic achievement, popularity, and attractiveness, which is important for subsequent developmental trajectories. Involvement with a partner who demonstrates delinquent behavior is significantly associated with self-reports of deviant behavior, and this finding is particularly salient for girls (Haynie, Giordano, Manning, & Longmore, 2005). On a more positive note, adolescents who date high-functioning partners tended to change more over the course of their relationship than those with low-functioning partners, such as gaining in popularity over time or exhibiting lower levels of depression or sadness (Simon, Aikins, & Prinstein, 2008). Giordano, Phelps, Manning, and Longmore (2008) also highlighted the reinforcing as well as motivating impact romantic partners can exert, particularly for boys; whereas some teens looked for a partying partner, others talked about the role their significant other played in encouraging them to do well in school.

Although romantic relationships among adolescents, like those of older adults, tend to be racially homogamous (Blackwell & Lichter, 2004), younger adults are the most likely to participate in relationships that cross racial lines (Joyner & Kao, 2005). Involvement in interracial relationships may have long-lasting effects. Young adults in interracial relationships received less social support from families and friends than did those in racially homogamous unions, and their relationships were more likely to dissolve (Vaquera & Kao, 2005; Wang, Kao, & Joyner, 2006). Interracial involvement also influences subsequent partner choice, as women whose first sexual experience was with a partner of a different race were significantly more likely to be in interracial marriages as adults (King & Bratter, 2007), though some groups of interracial couples also experience more marital instability (Bratter & King, 2008; Zhang & Van Hook, 2009).

Perhaps nowhere has the growth in research on partnering among adolescents and emerging adults been more evident than in studies of their sexual behaviors. This emphasis on adolescent sexuality, though generally concerned about adverse outcomes such as STIs and pregnancy, too often relies on a problem behavior perspective rather than viewing sexual engagement as a normative and appropriate developmental progression (Giordano et al., 2008). Although teen pregnancy and sexual coercion are critical social issues and the funding priorities of government agencies are problem oriented, it is important to ensure that research on adolescent behavior not neglect the more normative components of partnering. Though the sequencing of stages in intimate relationships does not always proceed in the expected order—from the formation of relationships that develop in intimacy and disclosure over time to sexual involvement—the normative pattern among teenagers is to date before engaging in sexual intimacy (Cooksey, Mott, & Neubauer, 2003; Longmore, Eng, Giordano, & Manning, 2009; O’Sullivan et al., 2007) and to share their first sexual experience with someone with whom they were “going steady” (Abma, Martinez, Mosher, & Dawson, 2004). Nonrelationship sexual partnering is practiced by a sizable minority of young adults, however, both for first sexual experiences and subsequently (Grello, Welsh, & Harper, 2006; Paul, McManus, & Hayes, 2000). Nonetheless, various studies utilizing different data sources reported that the most common pattern for teens who report sexual encounters outside of dating relationships is to choose friends or former significant others (Grello et al.; Manning, Giordano, & Longmore, 2006), with a subset harboring desires to kindle (or rekindle) a romance.

Notwithstanding the research evidence, the popular press frequently depicts contemporary young adults as engaging in partnering behavior that differs dramatically from previous generations—with more sexual activity and less desire for emotional connection (e.g., Stepp, 2007). The growing media coverage of hookups—casual sexual encounters that occur outside the context of a dating relationship and which can range from kissing to intercourse—is one manifestation of this belief. A closer look at hooking up behavior reveals its place on a broader continuum of sexual behaviors. Hookups are often thought to involve sexual intercourse, but several studies show otherwise. Paul et al. (2000) reported that though more than three fourths of their study participants had experienced at least one hookup during their college years, fewer than a third of their respondents had engaged in sexual intercourse with that partner—a result also found by Eshbaugh and Cute (2008), England and Thomas (2006), and others. Furthermore, as with first sexual experiences, casual sex occurred more often between friends than with strangers (Grello et al., 2006; Manning et al., 2006). Those whose hookup experience included sexual intercourse were more likely to be men, to report alcohol intoxication, and to adhere to a game-playing (i.e., ludic) love style (Grello et al.; Paul et al.). Casual sexual exploration was not without drawbacks; women who engaged in one-night stands expressed greater regret than did men (Campbell, 2008; Eshbaugh & Cute), and participants often engaged in behavior that exposed them to risks of STIs and pregnancy. Sexual encounters sometimes evolved into romantic attachments, though this is generally not the expected ordering of events (Manning et al., 2006), and the preferences of male partners more often ultimately determined whether sexual encounters led to serious romantic relationships (England & Thomas).

Of course, researchers continue to explore emerging adults’ transitions into marriage, though they increasingly include indicators of whether couples cohabited first (e.g., McGinnis, 2003). Most young Americans have positive attitudes about marriage, believe it will be in their futures, and see it as an important life achievement (Crissey, 2005; Gassanov et al., 2008; Manning, Longmore, & Giordano, 2007). In fact, only 5% of adolescents interviewed in 2000 for the Toledo Adolescent Relationship Study (TARS) did not expect to marry in the future, indicating that adolescents are not rejecting marriage as a future union formation experience (Manning et al., 2007). But recent studies have documented growing disparities in marital expectations by race, gender, and social class. Scholars utilizing data on unmarried young adults from the 1980s found few racial or ethnic differences in expectations for marriage once family background and social class variables were accounted for (e.g., McGinnis; Umana-Taylor & Fine, 2003). Research based on more recent data, however, found young Blacks reporting significantly lower expectations to wed than their White counterparts (Crissey; Gassanov et al.; Manning et al.). One study of African American adolescents, for example, found that they placed greater emphasis on their future careers than their romantic relationships, and felt they had more control over the former (McCabe & Barnett, 2000). The results for Hispanics are more mixed, though several studies find that they also articulate lower expectations of forming marital unions (Gassanov et al.; Manning et al.). Gender differentiates expectations for relationship behavior, with heterosexual women assigning greater value to lifelong commitment and faithfulness within marriage than do their male counterparts (Meier, Hull, & Ortyl, 2009). Personal experiences during childhood also shape marital expectations; individuals with divorced parents report more negative attitudes toward marriage (Riggio & Weiser, 2008), as do women who have experienced childhood sexual abuse (Larson & LaMont, 2005). Of note is that young adults with higher educational aspirations articulate the greatest expectations to marry (Manning et al.), suggesting that the growing educational disparities in marriage documented by demographers (e.g., Goldstein & Kenney, 2001) will continue to widen.

Research on union formation has changed in response to Americans’ delayed entrance into marriage. Studies of marriage among emerging adults often self-identifies as focusing on “early” marriage (Glick, Ruf, White, & Goldscheider, 2006; Uecker & Stokes, 2008). Fewer than one quarter of Americans now wed prior to the age of 25, in sharp contrast to previous generations. Those who choose to form early marital unions are more religious, are disproportionately drawn from disadvantaged families, have lower educational trajectories, and are more sexually conservative than those who defer marriage (Carroll et al., 2007; Gaughan, 2002; Uecker, 2008; Uecker & Stokes). Nonetheless, a sizable proportion of these unions have already dissolved by age 25 (Schoen, Landale, & Daniels, 2007). Cohabitation has become the more normative step among contemporary emerging adults, though these unions are often short-lived, with the majority not ending in marriage (Schoen et al.). Adolescents, in fact, often foresee cohabitation as part of their future life trajectory and view living together as a means to assess compatibility for marriage (Manning et al., 2007). Although most do not view living together as an alternative to marriage, cohabitation may become an intensive form of dating for young adults (Lichter, Turner, & Sassler, 2010).

Adult Transitions Into Coresidential Unions: Cohabitation and Marriage

If most research on adolescents and emerging adults focuses on dating and sexual exploration, the preponderance of studies on adults in their mid-20s through 40s concentrates on the formation of coresidential unions, how relationship commitment differs by the type of union formed, and relationship quality in coresidential unions. Even though sizable shares of adults in their 20s and beyond are not living with a partner, there is little scholarly attention to where this population meets dating partners or how relationships progress to coresidence. The growing prevalence of cohabitation is well documented. The majority of young adults have lived with a romantic partner by their mid-20s, and cohabitation is now the modal pathway into marriage (Kennedy & Bumpass, 2008). Whereas younger Americans express support for cohabitation as a means to assess compatibility for marriage, older adults appear increasingly likely to use cohabitation as an alternative to marriage, especially among less advantaged populations and those who have children or bear children outside of marriage (Gibson-Davis, Edin, & McLanahan, 2005; Musick, 2007; Reed, 2006). In recent years the proportion of cohabitors who marry their partners has decreased (Kennedy & Bumpass). Yet studies focused on the broader population assert that, notwithstanding increases in births to cohabiting women, cohabitation has not yet become a widespread alternative to marriage; it remains a relatively unstable living arrangement, and cohabitors continue to express preferences for parenting within marital unions (Raley, 2001; Sassler & Cunningham, 2008).

The presumption that living together serves as a precursor to marriage remains a dominant perspective in the literature. But a growing body of new, mainly qualitative, research has challenged this premise. This work documented that many cohabitors move in with partners very soon in the relationship, often because of changes in employment, housing exigencies, or convenience (Guzzo, 2006; Sassler, 2004) or in response to pregnancy (Reed, 2006; Sassler, Miller, & Favinger, 2009). Such rapid “slides” into shared living often preclude much discussion of the future (Manning & Smock, 2005; Sassler; Stanley, Rhoades, & Markham, 2006). Even though quantitative studies report that most cohabitors plan to wed their partners (Manning & Smock, 2002), qualitative research that explores the decision to move in together reports that marriage is often not considered a possibility until the couple has lived together for a while (Sassler) and partners have attained desired goals—school completion or obtaining a stable job or purchasing a house (Gibson-Davis et al., 2005; Reed; Smock, Manning, & Porter, 2005). Nonetheless, cohabiting adults express greater expectations of marrying their partner than do single adults who are not cohabiting with a romantic partner (Lichter et al., 2004; McGinnis, 2003).

Scholars have also begun to question whether entrance into shared living and marriage should be modeled as discrete choices or sequential decisions (Manning & Smock, 2005). Even though the majority of recently married couples lived together prior to the wedding, those who cohabit differ from individuals who marry directly; they are more economically disadvantaged, less religious, and less educationally and racially homogamous (Blackwell & Lichter, 2004; Sassler & Goldscheider, 2004). Attempts to understand the factors contributing to the decline in marriage among cohabitors has become a key focus of research, with researchers increasingly questioning whether standard economic explanations are adequate given marital delays across the social class spectrum. Men’s economic attributes play a less central role in the formation of cohabiting relationships than they do for marriage (Blackwell & Lichter; Oppenheimer, 2003; Sassler & Goldscheider). Qualitative studies based on low-income and working-class populations reported that a lack of money is frequently proffered as reason for not (yet) marrying, even among couples who live together and share parenting responsibilities (Edin et al., 2004; Gibson-Davis et al., 2005; Smock et al., 2005). Yet studies utilizing nationally representative data of transitions from cohabitation to marriage do not find a monotonic relationship between income or earnings and marriage; that is, cohabiting men with greater earnings were not more likely to wed than their less economically well-off counterparts (Oppenheimer; Sassler & McNally, 2003). Questions still to be answered include what level or combination of resources predict transitions to marriage as well as why fiscal barriers to childbearing are that much lower.

Other barriers delaying cohabitors’ transitions into marital unions include disagreement regarding the division of domestic labor, marriage plans, and how to resolve unplanned pregnancies. Using marital expectations reported by both partners, Sassler and McNally (2003) found that fewer than one third of cohabiting respondents concurred that they had definite plans to marry their partner; not surprisingly, couples who disagreed regarding their marriage plans were significantly less likely to wed. Dissonance in cohabiting partners’ views regarding how domestic work should be divided also increased the likelihood that cohabiting couples dissolved their relationship (Hohman-Marriott, 2006). Unintended pregnancies—higher among cohabitors than singles—both prolonged and destabilized unions (Reed, 2006; Sassler et al., 2009). Studies of disadvantaged populations also reported thatmental illness (Teitler&Reichman, 2008), fear of physical abuse (Cherlin, Burton, Hurt, & Purvin, 2004), and apprehension about divorce (Waller & Peters, 2008) reduced the odds that women married.

Concern with marital delay and the quality of current relationships is also reflected in an increasingly interdisciplinary body of research contrasting cohabiting and marital unions and assessing the impact of premarital cohabitation on marital quality. Cohabiting couples report higher levels of discord than do marrieds and lower levels of subjective well-being (Kamp Dush & Amato, 2005; Rhoades et al., 2006; Stafford, Kline,&Rankin, 2004; Williams et al., 2008). Scholars have sought to better understand to what extent such differences are the result of selection into cohabitation or what ensues after couples begin living together without marriage (or marriage plans; see Brown, 2004). Psychologists studying the impact of cohabitation on various aspects of relationship quality, including dedication, interaction, interpersonal commitment, relationship quality, and relationship confidence, found that cohabitors who were not engaged upon first moving in together were at significantly greater risk for poorer marital outcomes than were those who did not live together until after becoming engaged or getting married (Kline et al., 2004; Rhoades et al.). They attribute these findings to the inertia of cohabitation or the momentum that living together exerts on the likelihood of getting married, even in poor-quality relationships (Stanley et al., 2006). But more refined studies of the impact of cohabitation on relationship quality found that the difference in relationship quality between those cohabiting prior to marriage and marrying directly was largely driven by those experiencing births while cohabiting (Tach & Halpern-Meekin, 2009). As cohabitation prior to marriage becomes the normative experience among married couples, additional testing of this association is warranted.

Intimate Relationships in Later Life

As a result of divorce and rising proportions of those who have not married, in recent decades a large proportion of older adults are single. Among men, this increase results primarily from a doubling of the proportion who never married, whereas for older women the growth is largely because of divorce (Cooney & Dunne, 2001). Little is known about partnering—whether marriage, remarriage, cohabitation, or even dating—in later life (Cooney & Dunne). In part this is a vestige of data availability; nationally representative data sets that examine relationships tend to focus on younger adults at risk of childbearing. Yet many unmarried older adults are involved in intimate nonmarital relationships (Calasanti & Kiecolt, 2007; King & Scott, 2005; Mahay & Lewin, 2007) or are interested in repartnering (Carr, 2004; Mahay & Lewin).

The process of partnering differs for older adults, making theoretical assumptions applied to the search process of younger Americans less applicable (King & Scott, 2005; Mahay & Lewin, 2007). The partner market differs dramatically for men and women. Because women’s life expectancy is longer than men’s and men tend to partner with younger women, the sex ratio is particularly disadvantageous for older women looking to form relationships. Divorced and widowed men are more likely to remarry than their female counterparts (Ahrons, 2007), further diminishing the pool of mates available for unattached older women. Using data from the 2000 census, Calasanti and Kiecolt (2007) found that at age 65 and older, men were far more likely to be married than were women; 73% of the men were married, compared with only 43% of the women. Studies of the sexual activity of older adults also reported that unmarried women are less likely than men to have an intimate relationship (Lindau et al., 2007).

Marriage’s benefits also change significantly over the life course. The potential loss of social security, the challenges posed by merging households, and possible lack of support of adult children can affect decisions to marry at older ages (Mahay & Lewin, 2007). Older adults—particularly widows—may eschew the demands of marriage. Davidson (2001) found that older widows were more likely to associate widowhood with freedom and enjoy what they termed their “selfish” ability to do what they want. For these and other reasons, remarriage is uncommon in later life (Carr, 2004). But, as with younger Americans, older singles are forming alternative relationships, dating and establishing long-term supportive companionships that are not coresidential (termed “living apart together” in Europe; see Strohm, Seltzer, Cochran, & Mays, 2009, for a U.S. example) or, increasingly, cohabiting.

The dearth of data on the romantic aspirations and behaviors of older adults poses challenges to exploring their partnering behavior. Researchers have utilized disparate age ranges in their attempts to obtain adequate sample sizes of older individuals. Several studies of the likelihood of cohabiting among older adults focused on those age 51 and older (Brown, Bulanda, & Lee, 2005; Brown, Lee, & Bulanda, 2006; King & Scott, 2005), though Mahay and Lewin (2007) studied the partnering desires of those 55 to 69 with data from the General Social Survey. De Vos and Schwartzman (2008) defined the elderly as those age 60 or older, and other researchers utilize age 65 or older (Carr, 2004; Cooney & Dunne, 2001). Exploring whether older unmarried adults were even interested in forming new romantic attachments, Mahay and Lewin found that older single men and women were less desirous of marriage than their younger counterparts—though they are not rejecting relationships outright; as a sizable proportion were romantically involved. Older adults interested in forming new relationships may be selectively different from those uninterested in (re)partnering because they are better educated, physically healthier, and report fewer depressive symptoms (Carr; Mahay & Lewin). Gender differences emerged in the pace at which older bereaved adults are ready to reenter the partner market; within 6 months of bereavement, men were significantly more likely to express interest in either dating or marrying than were widowed women. Even though women’s desires to date rose with increased duration from their spouse’s death, gender gaps remained sizable. Within 18 months following their bereavement, nearly a quarter (23%) of widowers reported having gone on a date, compared with only 9% of widows.

As with younger Americans, the proportion of older adults who live with their romantic partner without being legally married has increased over the past few decades (Brown et al., 2005, 2006; Calasanti & Kiecolt, 2007; King & Scott, 2005). Cohabitation among the older population is most heavily concentrated among those age 51 to 59 and the previously married (Brown et al., 2006), with smaller proportions among those 60 and older. Calasanti and Kiecolt estimated that in 2005, cohabitors made up only 1% of men and 0.5% of women age 65 and older. Yet, as the population ages and more of those for whom cohabitation has become normative reenter the partner market following divorce or the dissolution of cohabiting or dating relationships, the proportion of older cohabiting adults should increase.

Cohabitation serves different functions for older respondents than for younger adults (Brown et al., 2006; King & Scott, 2005), being utilized more as an alternative to marriage than a precursor to it. Of note is that older adults living with their romantic partners assessed their relationships as more equitable, happier, and more stable and harmonious than did young cohabitors (King & Scott). Though older cohabitors fare better than their single counterparts, they nevertheless remain disadvantaged relative to remarried peers, for example, reporting significantly higher levels of depressive symptoms (Brown et al., 2005).

More research on partnering among older adults is needed, especially as the baby boom generation matures into retirement. To be sure, remarriage rates will remain low among this population, but various factors—increased life expectancy, good health, changing sexual attitudes, the growing acceptance of pharmaceutical sexual interventions (such as Viagra) targeted at older adults (predominantly men), the graying of baby boomer women used to expressing their sexual agency, and the rise in Internet dating and retirement communities—will undoubtedly change the romantic options available to older adults.

Repartnering Following the Dissolution of Cohabiting and Marital Unions

Because of high rates of union instability, many individuals reenter the partner market with prior cohabiting or marital experience. The challenges facing such individuals are varied; many already have children, both residential and living with an estranged partner (Carlson & Furstenberg, 2006; Goldscheider & Sassler, 2006; Graefe & Lichter, 2007); others are concerned that marriage will expose them to a partner’s poor financial history or to domestic abuse, challenge the way they parent, or reduce their receipt of government assistance (Reed, 2006; Waller & Peters, 2008). A separate article in this issue is addressing remarriage (Sweeney, 2010). Nonetheless, previous relationship experience has emerged as salient in the research on repartnering in several ways.

A small but expanding body of research has begun to assess the impact that living with multiple nonmarital partners exerts on subsequent union stability. One underlying premise of such work is that exposure to shared living experiences that end without marriage enforces the notion that unions are impermanent; a second is that individuals who live with multiple partners, termed serial cohabitation, may be selectively different from those who do not live with partners prior to marriage or only reside with the person who subsequently becomes their spouse. Support for the first premise has been found in several studies. Teachman (2003) reported that women with several cohabiting relationships who wed demonstrated an increased risk of divorce; premarital cohabitation limited to a woman’s husband, however, was not associated with an elevated risk of marital disruption. Extending this, Lichter and Qian (2008) found that serial cohabitors’ shared living situations were less likely to end in marriage than the coresidential unions of women who were living with their first partner; if serial cohabitors married, divorce rates were more than twice as high as for women who cohabited only with their eventual spouse. They also found support for the notion that serial cohabitation was selective, in that those who had lived with multiple partners were overrepresented among the economically disadvantaged, especially those with low income and education. But serial cohabitation is on the rise across the social class spectrum; between 1995 and 2002, women’s rates of serial cohabitation increased by nearly 40% (Lichter et al., 2010). Such change suggests the further uncoupling of cohabitation and marriage.

Another factor affecting those interested in forming relationships is the growing presence of parents among prospective partners, given high levels of divorce and nonmarital childbearing. Children have long been presumed to pose barriers to remarriage; men in particular are significantly less likely than women to express willingness to marry a partner who is a parent (Goldscheider et al., 2009). But when single fathers live with their children, they are substantially more likely to marry than are their female counterparts—even though mothers with coresidential children far outnumber their male counterparts. Single fathers and single mothers who marry often wed partners who are also parents (Goldscheider & Sassler, 2006).

Recent studies have also tried to better reflect the myriad forms parenting now takes and how that shapes repartnering. New research on the impact of children on union formation, for example, assessed whether they are residential or not or born within marriage (Goldscheider & Sassler, 2006; Guzzo & Furstenberg, 2007). Such studies found that nonresidential children do not increase the likelihood that parents remarry (Goldscheider & Sassler). Similarly, children born outside of marital unions retard the formation of parents’ new marriages (Carlson & Furstenberg, 2006; Carlson et al., 2004). Prior children, however, do not have as great a deterrent effect on the formation of new cohabiting unions, though such unions are often short lived (Goldscheider & Sassler; Guzzo & Furstenberg). The presence of children does reduce the quality of mates women repartner with, as well as whether they form a cohabiting or marital union (Graefe & Lichter, 2007; Lichter & Qian, 2008). The increase in multipartner fertility—when adults have children by more than one partner (Carlson & Furstenberg, 2006; Guzzo & Furstenberg)—therefore poses new challenges to the establishment of stable marital as well as cohabiting parenting unions.

Future Directions

Notwithstanding tremendous growth in studies of partnering behavior, the need for additional work is clear, especially research that is integrative and transcends disciplinary boundaries. This review has emphasized the scholarly balkanization of research on partnering; different family science disciplines have their own conceptual and theoretical lenses, distinct approaches to data collection, and favored topics. The past decade has brought progress, but more work needs to be done to unify what is known about partnering behaviors across the life course. Today’s adults are projected to spend a sizable proportion of their life outside of marital unions; additional study of the partnering behaviors of adults in their mid-20s and beyond is therefore necessary. To conclude this review, I propose several avenues for future research.

The Processes Behind Relationship Formation and Progression

Each relationship has its own unique trajectory, and relationships at one stage of the life course undoubtedly shape those at other stages in ways large and small. Yet relatively little is known about how early components of relationship progression shape subsequent union transitions. How do relationships progress from friendship to romance, and in what ways are relationships shaped by sexual involvement or coresidence? Measures of equity, sexual satisfaction, and commitment frequently utilized in psychological studies (e.g., Cate et al., 2002; Sprecher, 2001) are seldom incorporated into research conducted with nationally representative data on relationship progression prior to marriage. Qualitative studies of the tempo of cohabitors’ relationship progression from sexual involvement to shared living or parenting also must be replicated with nationally representative and longitudinal data.

Additional research is also needed to clarify how earlier relationships affect subsequent ones (e.g., Raley et al., 2007). Even though scholars have begun to explore prior relationship experience and marital stability in subsequent unions, these studies focused on the impact of cohabitation. The majority of Americans have engaged in sexual relations with someone other than their spouse, even if they have not lived with that partner. Greater attention to not just the number of prior sexual partners and coresidential unions but also the quality of those relationships could shed much light on the relationship patterns of today’s Americans and enable researchers to explore what individuals learn from prior (terminated) partnering experiences.

Variation in Partnering by Race, Ethnicity, Nativity, Social Class, and Gender

A reader unfamiliar with American society could easily conclude, on the basis of a cursory review of current literature, that this is a largely homogeneous country. Even though studies have explored the growing racial and ethnic diversity in childbearing, marriage, and family living arrangements (Lichter & Brown, 2009), in general, the research on close relationships assumes a largely White or middle-class template, even as our nation has become increasingly prismatic. Roughly 40% of America’s children today are racial or ethnic minorities or have immigrant parents (Hernandez, 2004). The youthful age new immigrant populations and racial minorities means that they will account for a growing share of young adults forming intimate relationships over the next few decades. Their presence, however, is not adequately represented in research on partnering.

Given the magnitude of this demographic change, we need to accelerate the shift from studies of “the family” to studies of families. Will America’s growing immigrant and minority youth emulate the relationship processes, family forms, and expectations of their origin families or will they embrace the patterns of the White population? What might it mean for foreign-born youth to become “Americanized” in terms of dating behavior, mate selection, and entrance into coresidential unions? How might the relationship behaviors of immigrant and minority youth influence the behaviors of the native-born population? Why are ethnic and racial disparities in dating behavior and marital expectations already evident among adolescents (Crissey, 2005; Vaquera & Kao, 2005) and are there variations within ethnic or racial groups? One promising area requiring additional attention is how generational status affects relationship processes. What research has been done suggests that American partnering patterns are learned behaviors. King and Harris (2007) found that foreign-born (first generation) youth were significantly less likely as adolescents to form romantic relationships than their third generation counterparts; second generation adolescents were still less likely than their third (or higher) generation counterparts to form romantic relationships, though such differences were not statistically significant. O’Sullivan and colleagues (2007) also found substantial ethnic variation in relationship progression patterns. Brown, Van Hook, and Glick (2008) reported that the likelihood of cohabiting increases with each generation in the United States, particularly among men; a unique contribution of this research is its presentation of results across various Hispanic groups (between Mexicans and Puerto Ricans, for example). Much more research in this vein is needed to better reflect the reality of Americans’ experiences.

Social class disparities in relationship behavior also deserve more attention, especially as inequality in the family formation behaviors of Americans has accelerated (McLanahan, 2004). Newly available data sources have increased the research focus on the partnering behaviors of low-income populations, but those from somewhat more advantaged families—what some have termed the “moderately educated” (Cherlin, 2009), who used to be described as the working-class or blue collar families—receive far less attention, even though they have also experienced dramatic changes in family building and formation processes. The heavy reliance of psychologists conducting experiments on college-based samples provides an incomplete and possibly misleading view of relationship quality, particularly for the large numbers of youth not enrolled in postsecondary schools. Nor do studies of the sexual behavior of college students—including the research on hooking up—shed light on the sexual experimentation of youth who do not attend college, or at least not 4-year residential schools. Whereas quantitative studies can shed light on some of the partnering behaviors of these populations, more experimental and qualitative study of young adults who do not pursue postsecondary schooling or who attend community colleges is needed. Finally, there has to date been an overemphasis on the romantic experiences of young women, which further reifies the belief that romance is less important or central to boys. Some studies have challenged that notion (Giordano, Longmore, & Manning, 2006; Smiler, 2008), but, given the two-sex nature of the issue, a better balance is needed.

The Importance of Parental Status and Type on Partnering

Just as extant research obscures the growing ethnic and racial diversity of American society, it also often fails to acknowledge that a sizable proportion of adults currently in the marriage market are parents from a previous relationship. Nearly one half of all recently formed unions include at least one adult who is a parent, whether from a prior marriage or sexual relationship (Stewart, Manning, & Smock, 2003). Although a sizable body of research has shown that parenthood is a deterrent in the marriage market, trends in divorce and childbearing outside of marriage have led to a partner market increasingly filled with parents.

Few studies have explored how being a parent influences relationship development and progression or how this varies by whether children are residential. Studies have, of course, included controls for the presence of children to determine their impact on remarriage and, increasingly, cohabitation; the evidence suggests that increases in the prevalence of being a single parent have reduced the negative effect of children on union formation. More than ever before, we need to know how children affect the earlier stages of relationships—such as decisions to enter into a dating relationship, the tempo of relationship progression to sexual involvement and coresidence, the form such unions take (marriage, cohabitation, or cohabitation that transitions to marriage), as well as the amount of time dating parents spend with new romantic partners and associations with quality, satisfaction, and commitment. Do parents engage in different relationship behaviors than do childless adults? In what ways does the partnering behavior of fathers with residential children differ from their more normative counterparts, men who do not live with their children, or women who have coresidential children? Answering these questions will require new data collection, as few large-scale studies include much information on nonresidential children or do not enquire about the child(ren) of a cohabiting partner if they are not coresident; data on the offspring of individuals who are dating are even thinner, particularly if the children do not reside with that parent. The growing body of research on multipartner fertility among fragile families has highlighted the salience of children to marriage and cohabitation. But there is room for more study of the impact children have on early partnering processes across the social class spectrum.

Grounding Research in a Historical Framework

Research over the past decade has been largely ahistorical. The popular press often portrays the romantic and sexual behavior among adolescents and emerging adults as different from that engaged in by previous generations, though there is little empirical foundation for such claims. In fact, teens in recent years have deferred sexual debut longer than did their counterparts in the previous decade (Abma et al., 2004). Such shifts highlight the need to deepen what is known about relationship processes during adolescence and emerging adulthood of earlier cohorts, such as the graying baby boomers and those who came of age in the more conservative Reagan years. Do hookups of the early 21st century differ from the “making out” of earlier generations? Are relationships progressing—to sexual involvement, coresidence, and marriage—more rapidly now than in the past? Has the function served by cohabitation changed as living together becomes a normative experience? Without comparing behaviors over time, it is difficult to determine whether these practices represent new developments or are simply extensions or modifications of prior behaviors.

New forms of dating—speed dating, Internet dating—also mandate more attention to studies that incorporate where individuals meet potential partners, how dating varies with age, and whether it then influences relationship progression, social support, marital stability, and relationship quality. The nascent body of research on how new technologies, including Internet and speed dating, shape relationship behavior suggests one fruitful avenue of study. Studies of Internet and speed dating have both reaffirmed why partnering processes remain gendered and racialized and challenged established wisdom regarding what men and women initially look for in mates. Internet daters, for example, winnow out prospective partners on the basis of preferences that are based on racialized images of masculinity and feminity; White men are more willing to date interracially than are White women (Feliciano et al., 2009). But speed daters’ stated preferences for partners do not always match up with choices exhibited upon meeting prospective partners (Eastwick & Finkel, 2008b), suggesting the variability that normative preferences might exert, especially during transitional periods. As new dating venues become more accepted, incorporating them into how family scholars examine partnering behavior is necessary.

Conclusion

A burgeoning social science literature reveals the myriad and changing forms of partnering in American society. The emphasis in the past on dating as a prelude to marriage has been replaced by a new focus on the fluidity of intimate relationships of all kinds. The formation and development of intimate relationships, nevertheless, have many commonalities over different stages of the life course. Indeed, this review has highlighted the central place of nonmarital sexual relations and cohabitation in the lives of most unmarried people at all phases of the life course. Marriage is only one of many contexts for sexual expression, emotional intimacy and commitment, coresidence, and childbearing and childrearing. Though the traditional functions of marriage clearly remain in place, they are now increasingly satisfied by other forms of intimate partnering, especially as the timing and trajectory of the marital life course have been reshaped by delayed marriage, divorce, and out-of-wedlock childbearing.

There are many reasons to be sanguine about this field of study. Scholars representing various disciplines have taken up the study of partnering behavior, and their work has been published in a wide array of journals. Innovative data collection efforts have provided an empirical basis for supplementing what we knew about partnering and broadening our conceptual and theoretical lenses from marital to nonmarital relationships. Considerable progress toward better understanding intimate relationships has been made in the past decade.

But this review of the research performed over the past decade has also uncovered the need to consider the connections between partnering behaviors over the life course, or how partnering at one stage shapes partnering at a subsequent juncture. Moreover, a White, middle-class template has become increasingly anachronistic with the growth of America’s racial minority and immigrant populations. Partnering processes undoubtedly vary among racial and ethnic minorities, immigrants and their descendants, and across the social class spectrum. Diverse populations and patterns of intimacy clearly call for new research that is truly interdisciplinary and that sets the stage for greater cross-fertilization across disciplinary boundaries. Americans will spend growing proportions of their adult lives outside of marital unions. As the options for emotional and sexual intimacy expand, so must our approaches to research on what is arguably one of the most dynamic areas of family social science today.