This study examines both unit and item nonresponse in a survey administered by mail and on the Web. A population of Oregon hunters was sampled to obtain opinions on current and proposed hunting regulations. A sample of 4,000 hunters was selected and split into two equal groups. Each group could receive up to four contacts: a pre-notification; initial mailing of the questionnaire; follow-up postcard; and a final mailing to nonrespondents. One group received all contacts by mail. The second group received a mixed- mode method of data collection. First, a pre-notification was sent by mail, followed by a letter directing the hunter to the Web site where the questionnaire was located. A follow-up postcard was sent by mail. The final contact for this group was a mailing that included a paper copy of the questionnaire. In the letter directing hunters to the Web site, a postcard was included asking the respondent to return it if he/she did not have access to the Internet. Those returning the postcard received a mail version of the questionnaire. Demographics for the two groups of 2,000 were evaluated to compare response rates across age and gender. In addition, similarities between groups in their responses to questions were analyzed to evaluate whether responses from the Web group differed from those in the mail group. The group receiving all contacts by mail had the highest response rate in this study. The results also suggest that the older age group were more slightly more likely to respond by mail rather using the Web.