The study uses a sub-set of the incoming class of 7,500 students starting their freshman year in 1997 and 1998 that entered the housing lottery. A subset of these students (those who chose not to live in an 'enrichment center' and who met the housing lottery deadline) were randomly assigned to rooms (and roommates). This left a sample of 1,357 students. These students put in preferences for (i) gender (ii) substance or smoking free (iii) number of roommates (iv) geographical area on campus. Random assignment was conditional on these preferences, and hence all regression specifications control for these preferences. An analysis done for a second group of students who requested their roommates was used to assess how large the selection bias problem would have been for a non-experimental study.
The primary outcome of interest was the cumulative GPA for all students at the end of summer of 1999. This would represent the freshman-year grades for one group in the sample, and sophomore- year grades for the rest. Thus for the latter group, the study also looks at continuing effect of the first-year roommate since in the second year only 17% choose to continue living with the same person.
Explanatory variables related to roommate drinking were obtained from the Entering Student Survey of the Cooperative Institutional Research Program (CIRP) which was administered to all admitted students during their orientation week in the summer prior to starting classes and had a response rate of 89%. This survey contains a section in which respondents are presented with a list of activities and asked whether they undertook the activities frequently, occasionally, or not at all during the last year. The list of activities includes "Drank beer" and "Drank wine or liquor." Respondents who answered "frequently" to at least one of the two drinking-related questions were classified as "frequent drinkers". Respondents who were not "frequent drinkers" but answered "occasionally" to at least one of the two drinking related questions were classified as "occasional drinkers". Students who reported not drinking beer, wine, or liquor in the last year were classified as "nondrinkers."