Predictors of Extradyadic Sexual Involvement in Unmarried Opposite-Sex Relationships

Article excerpt

Although there is much research on correlates of extradyadic sexual involvement (ESI) in marriage and on the negative impacts of this behavior (see Allen et al., 2005; Whisman, Gordon, & Chatav, 2007), there is virtually no information on factors associated with ESI in unmarried, adult dating relationships and few data that come from prospective samples. The research that does exist about ESI in unmarried relationships indicates that, for U.S. samples, it is more common and accepted in dating relationships than in marriage (Sheppard, Nelson, & Andreoli-Mathie, 1995; Wiederman, 1997; Wiederman & Hurd, 1999), but is not the norm, and is widely met with disapproval (Regnerus & Uecker, 2011). For example, English-speaking Americans, aged 18 to 59, who were either married or cohabiting were interviewed about their sexual attitudes and behaviors in the 1992 National Health and Social Life Survey, and on a 7-point scale ranging from 1 (completely acceptable) to 7 (completely unacceptable), individuals rated marital ESI scenarios an average of 6.68 and dating ESI scenarios an average of 5.96. Similarly, 94% of cohabiters expected sexual exclusivity, compared to 98% to 99% of married individuals (Treas & Giesen, 2000).

Examination of ESI in unmarried relationships is important due to the emotional (e.g., pain of betrayal) and physical (e.g., sexually transmitted diseases) risks. Further, researchers argue that ESI in U.S. college dating relationships may represent risks for ESI in later marital relationships (Drigotas, Safstrom, & Gentilia, 1999; Roscoe, Cavanaugh, & Kennedy, 1988). In addition, in a sample of predominantly White individuals, Allen et al. (2008) found that some premaritally measured factors (e.g., communication and sexual satisfaction) were related to the likelihood of later marital ESI for couples who went on to marry. Thus, assessing predictors of ESI in dating relationships may have implications for ESI in later married relationships, and could represent earlier opportunities to evaluate potential risks and to intervene.

Most prior studies of ESI "predictors" not only drew on married samples, but were cross-sectional, which makes it difficult to determine whether ESI is a cause or a consequence of poor relationship quality and other factors (Previti & Amato, 2004). In addition, several "predictive" studies have assessed perceived susceptibility to future ESI, rather than actual reports of later ESI (e.g., Buss & Shackelford, 1997; Lewandowski & Ackerman, 2006; Zak et al., 2002). Thus, in addition to addressing the dearth of research on unmarried ESI predictors, a primary aim of this study was to use a longitudinal model in which actual ESI was predicted from prior levels of individual and relationship functioning. Additionally, this study focused on predictors of ESI in unmarried, opposite-sex relationships in a nationally representative sample, rather than using a convenience sample as other research on unmarried ESI has done (e.g., Drigotas et al., 1999; Hall & Fincham, 2009). We assessed a wide range of possible predictors that prior research has examined related to ESI (typically using cross-sectional models and married samples), along with some lesser-studied variables, which may have important implications for ESI in dating relationships.

Prior Research on Predictors and Correlates of ESI

Allen et al. (2005) suggested an organizational framework to use for the conceptualization of variables related to ESI. Based on their labels and descriptions, we conceptualized our variables as representing predisposing involved-partner factors and relationship factors. We further organized our review of the literature and our results into three types of involved-partner factors (demographics, mental health, and sexual behavior) and four types of relationship factors (sexual relationship, relationship satisfaction, commitment, and negative interactions). …