A Study on Warning Message Effect on Survey Experiments

Abstract

This paper examines the effects of warning messages issued before the survey experiment. This paper examines two kinds of effects: first, the effect of informing the respondents that the results of the survey will be widely publicized, and second, the effect of warning to the respondents who did not read the survey well. Previous studies on this theme have been conducted outside of Japan, and it was not clear how effective it was in Japan. This study suggests that the warning messages had no effect on the experimental effects of any kind of warning message, and that the identification of bad respondents by the IMCs did not lead directly to the fact that they skipped the experiment in question, and that it is most important to design the survey experiment to avoid cognitive load.