Article: Are Drinking and Sexual Behaviors Different for People Who Go to College?

October 18, 2017:sv home

Despite the media coverage and research that have been devoted to risky behavior among young adults, many questions remain about which populations are at risk. In a recent article in Journal of Research on Adolescence, Methodology Center Investigators Sara Vasilenko, Ashley Linden-Carmichael, Stephanie Lanza, and Methodology Center Affiliate Megan Patrick examine differences between college attenders and their non-attending peers in terms of drinking behavior, sexual behavior, and the co-occurrence of heavy drinking and sex. This article also provides a thorough introduction to weighted time-varying-effect modeling (TVEM) and moderation in TVEM.

The authors analyzed data from 11,848 participants in The National Longitudinal Survey of Adolescent to Adult Health (Add Health). They examined whether—and how strongly—sexual behavior was predicted by frequent heavy episodic drinking (HED—occasions when a person has five or more drinks in a row) at different ages from 14-24. As noted above, they also examined whether there was a difference in behavior between college attenders and college non-attenders. They found that the association between HED and sexual behaviors is stronger for college attenders than it is for non-attenders both early in adolescence and during the early college years.

Lead author Sara Vasilenko said, “What I found interesting about this research is that, despite the popular belief that college is a hotbed of risky sexual behavior, college attenders were less likely to engage in risky sexual behavior than their non-attending peers. That said, the association between risky sexual behavior and HED is stronger for attenders than non-attenders from ages 18 to 20. So, while attending college does not appear to lead to greater levels of risky sexual behavior overall, college may be a unique environment for the co-occurrence of drinking and sex.

“In addition to the substantive contributions, this paper demonstrates two different ways to perform moderation in TVEM, and the article includes sample code. This will be useful to researchers who want to learn if time-varying effects differ for different types of people. In this example, we examined the differences between college attenders and non-attenders, but the same technique for dynamic moderation could be applied to any type of moderator.”

Open the article. (Journal access required.)


Vasilenko, S. A., Linden-Carmichael, A., Lanza, S. A. & Patrick. M. E. (in press). Sexual behavior and heavy episodic drinking across the transition to adulthood: Differences by college attendance. Journal of Research on Adolescence.