Methodological variables in Web-based research that may affect results: sample type, monetary incentives, and personal information.

Abstract

There are many methodological differences between Web-based studies, differences that could substantially affect the results. The present study investigated whether sample type, offering payment through a lottery, and requiring participants to enter personal information would affect dropout rates and/or the substantive results in a study of jury decision making in capital cases. Asking participants to enter their e-mail addresses increased dropout rates, and offering payment through a lottery tended to do so as well. Participants offered payment tended to be less likely to give death sentences, and sample type moderated the influence of attitudes toward the death penalty on verdicts.