For my first column, I thought I would ask - and attempt to answer - a slightly controversial question: Does attending college increase young people's alcohol consumption, compared to if they never set foot on a university campus? Such a question can be justified on at least a couple of different grounds.
National college-student surveys conducted mainly in the 1990s by the Harvard School of Public Health found that roughly 44% of students reported recent heavy-drinking episodes. Thus, even before similar data from same-age non-college peers - which we now have - became available, it would have been natural to ask why such extensive heavy drinking was taking place in college.
Further, as critics such as Beer and Circus author Murray Sperber have pointed out, even though many university officials nationwide claimed to take strong stands against underage and excessive student drinking, such messages are contradicted by signs of a strong drinking culture on many campuses. For example, some schools' bookstores sell shot glasses featuring university logos and other drinking-related items.
The scientific method of choice for addressing whether college attendance affects student drinking would be a true experiment in which, for example, several hundred college-eligible high school graduates would be assigned at random to either attend college or stay away from the ivy-strewn quadrangles. The idea would be to create two groups that, in the aggregate, were equal on all characteristics - personality, socioeconomic status, etc. - except that one went to college and the other did not. That way, any differences observed between the two groups in drinking levels over the next few years would have to be attributed to attendance (or non-attendance) at college.
Such an experiment presumably never could be done, of course, for logistical and ethical reasons. Researchers have therefore tried to devise ways to compare the drinking of college- and non-college-attending youth, while keeping the two groups as demographically and behaviorally similar as possible. A team of Missouri researchers led by Wendy Slutske in 2004 studied female twin pairs, each consisting of one sibling who went to college and another who did not. Now that's going the extra mile to equate the college and non-college groups on all extraneous characteristics! These investigators found only limited alcohol-consumption differences between the two groups, with college-attending twins drinking more heavily.
Most other studies on this topic (including my own) stick to traditional survey methods, without seeking genetic comparability; these studies try to obtain large samples of young people who are representative of their community, state, or nation, to compare the drinking of those who do and do not attend college. Extraneous or incidental differences between the groups beyond their college attendance/non-attendance, such as unequal proportions of males and females, different racial-ethnic groups, and family histories of alcohol problems, are dealt with statistically. Studies that repeatedly contact the same participants over a period of a few years - from before to after college entry - provide a further benefit, namely evidence of whether arrival at college precedes increases in drinking.
If, once at college, students turn out to drink more heavily than their same-age non-college peers (especially if, as is sometimes found, the college-bound participants drank less as high school seniors than the non-college-bound), these results can be suggestive of a college effect. What prevents a full causal conclusion is that not all possible extraneous differences between college and non-college groups can be recognized in advance and measured by researchers, in order to control for them statistically.
Where does this research literature currently stand? A team of scholars from the University of South Florida published an article in the latest issue of the Journal of Studies on Alcohol and Drugs, which reviewed 18 prior comparative drinking studies of youth attending and not attending college. The task of compiling these findings was not as easy as might be imagined, due to differences in how researchers chose to categorize part-time students, those in two-year (as opposed to four-year) colleges, and "atypical" students (e.g., much younger or older than usual, married students, college dropouts).
The above limitations acknowledged, the authors concluded initially that, "Nearly all studies... found that college students consumed higher quantities of alcohol than noncollege peers or engaged in riskier consumption patterns." The authors qualified this conclusion somewhat, noting that, "several studies found this higher level of consumption particularly associated with college students living away from home, indicating the importance of living situation..." (quotes from p. 746).
A 2010 study by me and my colleagues (which came out too late to be included in the review article by the South Florida researchers) found a particularly sharp rise in college-bound students' drinking during their first year of college, but after that, they drank at levels similar to their non-college counterparts. This finding of college being a short-term impetus to heavier drinking seems to match with some of the studies discussed in the review article.
Ultimately, the South Florida reviewers suggest, "contextual influences" such as lack of direct parental supervision and the ready presence of drinking peers may play a more important role in youthful drinking than college status per se. Many of these contextual influences are indeed present among college students, but they would also likely exist for some non-students, as well.
All of these points are well-taken and highlight the complexity of judging whether college attendance, in and of itself, is a spur to heavier drinking than if the same individuals had never gone to college. I would say there's enough evidence to take seriously the idea of a college effect on drinking. But whether the case can be established beyond a reasonable doubt, that's another matter.
