QUESTIONNAIRE DESIGN

Questionnaire Design



Questionnaire Design

Research Method

The 2001 CAS surveyed students at 120 of the colleges that participated in each of the previous three surveys. The participating schools were located in 38 states and the District of Columbia. In the original 1993 survey, we selected 140 schools from a list provided by the American Council on Education to provide a representative sample of accredited 4-year US colleges and universities. The attrition of 20 schools was primarily a result of institutions' inability to provide a sample of students and mailing addresses to meet the time constraints of the survey. Administrators at each participating school used the same procedure as that used in conducting previous CAS surveys to provide a list of 215 randomly selected students from all full-time undergraduate students enrolled during the 2000-2001 school year. Details of the previous sampling methods are described elsewhere.

In conducting the data analyses, we excluded 1 college with a response rate that was substantially lower than the others, leaving 119 schools. The inclusion criteria differed from previous survey years. We drew on results from 113 schools that we had surveyed earlier and reintroduced 6 schools that had been dropped from previous analyses. Thus, rates for earlier years cited in this report are slightly different from those reported in other articles that were based on CAS data. However, when we compared the bingedrinking rates of the 119 schools with the corresponding rates of all schools in each survey year, we found they were identical. Dropping the low-response schools did not change the results of the survey. The sample of 119 colleges represents a national crosssection of students enrolled at 4-year colleges.

Sixty-nine percent of the responders attended public colleges and 31% attended private colleges, which approximates the US national distribution of full-time 4-year college students, with 68% attending public and 32% attending private schools.23 Forty-seven percent of the responders attended large colleges (> 10,000 students), 23% were at mediumsized colleges (5,001-10,000 students), and 29% were enrolled in small colleges (< 5,001 students). The US national distribution is 37%, 24%, and 40%, respectively.23 Sixty-nine percent of the responders attended schools in large or medium-sized cities, compared with 71% of students nationwide.

Thirteen percent of the students attended schools with a religious affiliation, compared with 16% nationwide23; and 5% percent of the students attended all women's colleges. The 2001 sample consisted of 64% women, a rate higher than the national rate of undergraduate women, resulting in part from the inclusion of 5 women's colleges. About 3 in 4 students (74%) were White, and 1 in 2 (50.2%) was under the legal drinking age (ie, younger than 21 y). We noted significant changes in the demographic characteristics of the student samples over the 4 surveys. Therefore, to compare prevalence rates across survey years, we weighted the data to the population distribution of each school in 1993 in terms of sex/gender, age, and race/ethnicity. We used the weighted data in all analyses.

Questionnaire

The 2001 survey questionnaire asked respondents to answer a ...