Poll Illiteracy: How the Public Misunderstands Polls

Abstract

The century-long tradition of reporting poll results as vote shares does not provide direct answers to the very questions voters are most interested in (e.g., who will govern). The present study provides experimental evidence showing that voters are largely unable to derive such information from polls themselves. When estimating the likelihood of two electoral outcomes, participants failed to realize that small changes in poll numbers can have large electoral consequences. The results suggest that it may be time to revisit how to communicate poll results to the public.