Segments then were divided into clusters of approximately four to eight housing units, and all occupied housing units were included in the survey. Within each household, one person age 18 or older was randomly selected to participate in the survey. Oversampling of young adults (i.e., people ages 18 to 29) was accomplished at that stage of sample selection to increase the representation of this heavy-drinking population subgroup. Thus, young adults were sampled at a ratio of 2.25 percent to 1.00 percent.
