The feasibility of a web-based surveillance system to collect health risk behavior data from college students.

This study examined the feasibility of collecting health risk behavior data from undergraduate students using a Web-based survey. Undergraduates were randomly selected and assigned randomly to a mail survey group and a Web survey group. There were no statistically significant differences between the two groups for demographics, response rates, item completion, and item completion errors. Yet differences were found for response time and sensitive item completion. This is the first study to demonstrate the feasibility of collecting health risk behavior data from undergraduates using the Web. Undergraduates are just as likely to respond to a Web survey compared with a mail survey and more likely to answer socially threatening items using this method. Also, the Web format and protocol required less time to administer. Researchers and practitioners conducting health survey research with college students or other homogeneous populations who have access to e-mail and the Web should consider using a Web-based survey design as an altemative to a mail, self-administered survey. In such a population, a Web-based survey should not discourage participation, particularly if participants are interested in the questionnaire content.