Mode effects for collecting alcohol and other drug use data: Web and US mail

Objective: The present study examined mode effects for collecting alcohol and other drug use data using a Web-based survey mode and a US. mail-based survey mode for comparison. Method: A survey regarding alcohol and other drugs was administered to a randomly selected sample of 7,000 undergraduate students attending a large midwestern research university in the spring of 2001. The sample was randomly assigned to either a Web-based survey mode (n = 3,500) or a US. mail-based mode (n = 3,500). Results: The Web survey mode of administration resulted in a final sample that more closely matched the target sample in gender mix than did the U.S. mail survey mode. The response rate for the Web survey mode was significantly higher than for the US. mail survey mode. Chi-square results indicated there were significant differences in response propensity by several sample characteristics including sex, race, class year and academic credit hours. Multivariate logistic regression results revealed significant racial and gender differences in the response propensity between and within modes. After controlling for design discrepancies, there were no significant differences between modes in data quality or substantive responses to substance-use variables. Conclusions: The findings of the present study provide strong evidence that Web surveys can be used as an effective mode for collecting alcohol and other drug use data among certain populations who have access to the Internet and high rates of use. Web surveys provide promise for enhancing survey research methodology among undergraduate college students.