This study was designed to examine response rates and bias among a sample of community college students who received a district-wide survey by standard mail or e-mail. Findings suggest that predictors of response and types of responses are not appreciably different across paper and online mail-out samples when these samples are "matched" in terms of key demographics. Rates of response, however, differ by mode of survey administration, gender, and race/ethnicity. (Contains 4 tables.)