Incentives are commonly used to boost survey response rates, although the evidence is still limited as to whether they actually reduce nonresponse bias. Besides, we speculate that incentives might also increase measurement error if they attract respondents who participate only to obtain the incentive and so are unwilling to expend much effort. We explore the impacts of incentives on data quality by manipulating its salience in email invitation letters for a Web survey sent to university staff that asks about their use of information technology for work duties. Two designs of the invitations are compared. In one condition, we make incentives salient by starting the email subject line with the incentive and putting a statement about the incentive at the beginning of the invitation letters, in bold and larger font. In the other condition, importance of the survey is emphasized. Thus, both conditions inform the staff of the incentive and the importance of the survey, but with different emphases. To assess nonresponse error, we compare response rates between the users and non-users of a campus-wide, web-based information management system, where the usage status is known for the sampled staff and related to some of the survey variables. To evaluate measurement error, we ask two questions for which we have administrative records. We also examine the indirect indicators of response quality, such as completion times and length of open-ended responses. We find that making incentives salient increases the response rate, particularly among the non-users of the system, which suggests reductions in nonresponse bias for some survey estimates. The incentive salience also affects the make-up of the respondents on their motivations for taking the survey, which we find are related to their response quality. The findings suggest a potential link between nonresponse and measurement error driven by incentives.