Institutional, Subcultural, and Individual Determinants of Same-Sex Sexual Contact Among College Women

Abstract

While the collegiate hookup literature identifies the prevalence and conditions under which women engage in same-sex sexual behaviors, less is understood about the macro- and mesolevel features of universities predictive of women’s engagement in these behaviors. How do predictors of same-sex contact differ for women who report engagement in nongenital versus genital sexual acts? Using the Online College Social Life Survey (OCSLS), this study examined 5,069 women from 21 U.S. colleges and universities who reported at least one same-sex encounter in their lifetime to empirically evaluate how same-sex sexual histories are predicted by university structure and membership in student subcultures. Results find subcultural participation to have a regulative effect on what types of sexual behaviors women engaged in with other women. While Greek-affiliated women were significantly more likely to have engaged in solely nongenital sexual behavior, women in team-based athletics were more likely to report a genital sexual history with women. This study’s focus on women’s sexual behaviors as opposed to sexual identity affirms the influence of institutional contexts, subcultural dynamics, and their variability across U.S. campuses in producing the conditions and opportunities for varied forms of sexual interactions among college women.

Acknowledgments

For their helpful comments and suggestions, the author would like to thank Sarah Thébaud, Paula England, Verta Taylor, Beth Schneider, Leila Rupp, Maria Charles, Alicia Cast, and anonymous reviewers.

Notes

¹ The data that support the findings of this study are publicly available at http://www.nyu.edu/projects/england/ocsls/.

² To determine the most party-heavy schools in the nation, the Princeton Review surveys students across institutions. Party schools boast a combination of high drug or alcohol usage, popularity of fraternities and sororities, and low numbers of personally reported study hours outside of class.

³ I also ran a series of models utilizing alternative macrolevel features of university spaces to include a school’s status as a public institution; school location in urban, rural, or suburban space; and an alternative categorization of a school’s undergraduate population utilizing reference categories (small institutions of 5,000 undergraduates or less, midsized institutions of 5,001 to 20,000 undergraduates, and large institutions comprised of 20,001 or more undergraduates). Inclusion of these alternative measures returned similar results as those presented, but weaker model fits.

⁴ The U.S Department of Homeland Security’s STEM degree program list draws upon the U.S. Department of Education’s National Center for Education Statistics to compile its list. The list may be accessed at https://www.ice.gov/sites/default/files/documents/Document/2014/stem-list.pdf.

⁵ Pooled VIFs for individual-level variables are 2.79 and below, and 1.56 for institutional-level variables. In addition, I ran a series of sensitivity checks to test for both individual and institutionally based outliers by removing freshman women and each of the 21 institutions represented in the data from the model, respectively. Each of these sensitivity checks returned results similar to those presented in this article.

⁶ Alternative models controlling for the percentage of students (male or female) per institution who identified as homosexual, bisexual, or unsure produced no significant result predicting a nongenital or genital sexual history.

⁷ These variables were included in the original OCSLS data and were the only sexual behaviors included on the sexual history portion of the survey that distinguished between public and private settings.