This study compared the demographics and risk behaviors of two samples of men who have sex with men (MSM), using cross-sectional data that were collected via the Internet and through conventional bar-based outreach. The Internet sample was significantly older, more likely to identify as "bisexual," and less educated than the bar sample. After controlling for age and education, few differences were observed between the samples. However, three variables that markedly differentiated the samples were history of sexually transmitted disease infection, HIV serostatus, and sources utilized to obtain health information. No difference in Internet use was found. Based on the possible decreased social desirability promoted by the use of electronic data collection methodologies, these findings provide preliminary evidence that Internet and bar respondents are similar and that the Internet may serve as an expedient as well as reliable methodology to increase understanding of risk among MSM.