Testing the race-of-institution effect on misreporting electoral behavior: A research note

This research note reports on an attempt to gain a greater understanding of voter misreporting. Is there a race-of-institution effect present in mail surveys that contributes to misreporting similar to the race-of-interviewer effect present in other survey techniques? To examine this research question a survey was developed. The research consisted of selecting two institutions (universities) within a southern U.S. city, one recognized as a ''traditionally Black'' institution and one as ''traditionally White.'' Equal systematic random samples of Black and White registered voters were selected from similar socioeconomic areas. The research provides evidence that voter misreporting may be because of response bias. Indeed, for both Black and White respondents, a race-of-institution effect exists. Differences in electoral contexts produce variations in the race-of-institution effect and, therefore, variations in misreporting.