I recently discussed on the blog new evidence pointing towards a positive relationship between students' grades in college and their later-life health outcomes. A recent economics paper adds a new dimension in this area by producing findings on the effect of preceding health-risk behaviour on subsequent academic performance. The paper is by Scott E. Carrell, Mark Hoekstra, and James E. West; and is called "Does Drinking Impair College Performance? Evidence from a Regression Discontinuity Approach". Abstract: here. Paper: here. Discussion on the Freakonomics Blog: here.
The paper exploits a discontinuity in drinking at age 21 at a college in which the minimum legal drinking age is strictly enforced. A comparison is made between the grades of students who turned 21 before final exams to those who turned 21 just afterward. The authors find that drinking causes causes statistically and economically meaningful reductions in academic performance, particularly for the highest-performing students. The authors suggest that concern regarding the harmful effects of drinking in U.S. colleges is reflected by the Amethyst Initiative - in which 135 university presidents and chancellors argue that current policy has resulted in binge-drinking by students.