Why Do People Underestimate Polling Effects? Examining the Gap Between Actual and Perceived Polling Effects

Abstract

ABSTRACT Two major challenges have limited the advance of our knowledge about people’s bias in perceived polling effects compared with actual polling effects. The first challenge comes from a methodological difficulty in measuring the gap between perceived and actual polling effects. Second, two contradicting polling effects (i.e. bandwagon and underdog effect) cancel out each other at the aggregate level, making reported findings shrink. Addressing these challenges, this study developed a method to systematically evaluate the gap between perceived and actual polling effects on the self. Two sets of two-wave panel surveys (total N = 1,001) were employed. Drawing on the literature on false uniqueness bias, motivated reasoning, and impression management theory, we expect people to underestimate the polling effects on themselves when their preexisting attitudes are weakened, but not when reinforced by the poll. Our two studies confirmed this pattern, observed in both cases that experienced bandwagon and underdog effects of the poll.