HIV-related behaviors, self-reported using Web-delivered or paper-pencil modes, were collected from two convenience samples of college students at a major university in the southeastern U.S. To enhance the equivalence of the comparisons, a subset pool of participants from each group, proportionally matched on key demographic variables including age, gender, and race, was randomly selected to be included in the analysis. Multiple-regression analyses showed similar self-reporting patterns on HIV testing-related behaviors and intention, as well as various sexual behaviors (oral, vaginal, and anal), initial age of sexual activity, number of partners, alcohol and condom use (Type I error adjusted), with the exception that prevalence of self-reported behaviors on alcohol use before and condom use during anal sex were higher from data collected via Web-delivered mode. Age and peer norms were consistently shown as significant positive predictors for the various sexual activities. The implication of using the Internet to collect HIV-related behavioral data are discussed. (Contains 3 tables.)