Cumulative evidence is mixed regarding the effect of lottery incentives on survey participation; little is known about why this strategy sometimes works and other times fails. We examined two factors that can influence the effectiveness of lottery incentives as suggested by leverage-salience theory: emphasis of survey attributes in invitations and characteristics of target populations. We conducted a web survey experiment where one condition highlighted lottery incentives in the e-mail invitations (incentive-centered condition) and the other highlighted the value of the survey with a brief mention of the lottery (survey-centered condition). We found that the incentive-centered condition had a significantly higher response rate than the survey-centered condition, especially among individuals with a relatively low income. Although invitation emphasis affected respondent compositions regarding motives for participation, the differences in response quality between the two experimental conditions were small and mostly not significant.