Inside College Parties: Surprising Findings About Drinking Behavior

"Most studies use survey methods that require people to recall their drinking behavior -- days, weeks or months prior -- and such recall is not always accurate," noted J.D. Clapp, director of the Center for Alcohol and Drug Studies and Services at San Diego State University and corresponding author for the study. "By going out into the field and doing observations and surveys, including breath tests for alcohol concentrations, we were able to mitigate many of the problems associated with recall of behavior and complex settings."

"In addition," said James A. Cranford, research assistant professor in the Department of Psychiatry at the University of Michigan, "this study is unique in its focus on both individual- and environmental-level predictors of alcohol involvement. Rather than relying on students' reports of the environment, researchers actually gained access to college-student parties and made detailed observations about the characteristics of these parties."

For three academic semesters, researchers conducted a multi-level examination of 1,304 young adults (751 males, 553females) who were attending 66 college parties in private residences located close to an urban public university in southern California. Measures included observations of party environments, self-administered questionnaires, and collection of blood-alcohol concentrations (BrACs).

"Both individual behavior and the environment matter when it comes to student-drinking behavior," said Clapp. "At the individual level, playing drinking games and having a history of binge drinking predicted higher BrACs. At the environmental level, having a lot of intoxicated people at a party and themed events predicted higher BrACs. One of the more interesting findings was that young women drank more heavily than males at themed events. It is rare to find any situation where women drink more than men, and these events tended to have sexualized themes and costumes."

"Conversely," added Cranford, "students who attended parties in order to socialize had lower levels of drinking. Interestingly, larger parties were associated with less drinking. Dr. Clapp and colleagues speculate that there may simply be less alcohol available at larger parties, and I suspect this may be the case."

Both Clapp and Cranford hope this study's design will help future research look at "the whole picture."

"From a methodological standpoint, our study illustrates that is possible and important to examine drinking behavior in real-world settings," noted Clapp. "It is more difficult than doing web surveys and the like, but provides a much richer data set. Secondly, environmental factors are important. Much of the current research on drinking behavior focuses on individual characteristics and ignores contextual factors. Yet both are important to our understanding of drinking behavior and problems."

On a more practical level, Clapp urged caution on the part of party hosts as well as guests. "Hosts should not allow drinking games and students should avoid playing them," he said. "Such games typically result in large amounts of alcohol being consumed very quickly - a dangerous combination." He and his colleagues are currently testing party-host interventions that may help, and also plan to further examine themed parties in greater detail, other alcohol-related problems occurring at all types of parties, and drinking in a bar environment.

Full results are published in the January issue of Alcoholism: Clinical & Experimental Research.

Co-authors of the ACER paper, "Person and Environment Predictors of Blood Alcohol Concentrations: A Multi-Level Study of College Parties," were: J.W. Min, A.M. Shillington, M.B. Reed and J.M. Ketchie, all of the Center for Alcohol and Other Drug Studies and Services at San Diego State University. The study was funded by the National Institute on Alcohol Abuse and Alcoholism.