Effects of poll reports on voter preferences

Relative strengths of the bandwagon (or rally-around-the-winner) affect and its converse, the underdog effect, were tested. Study 1 was conducted with registered Republicans during 4 days immediately prior to the first major Republican primary of 1996. Bogus poll data showing Dole leading Forbes (or Dole trailing Forbes) were presented to voters who then Voted their prefer ence for Dole, Forbes, or neither. Findings showed a significantly greater tendency to vote for Dole over Forbes when the bogus poll showed Dole leadi ng Forbes than when it showed Dole trailing. Thus, results supported the ba ndwagon effect and, furthermore, showed it as explaining 6% of the variance in voter preferences. In Study 2, participants were given bogus poll data on 2 personally relevant issues and were then asked to vote their preferenc es on the issues. Participants had extremely strong consensus preferences o n one issue and were not influenced by bogus poll data. On the second issue that involved moderately strong consensus preferences, bogus polls signifi cantly affected votes, supporting the bandwagon effect. Bandwagon effects w ere stronger for women compared with men, and for 2 of 3 PAD (Pleasantness, Arousability, Dominance) basic temperament factors; that is, for individua ls with more arousable and less dominant temperaments. Implications for oth er personality variables were noted.