New Evidence About the Existence of a Bandwagon Effect in the Opinion Formation Process

This study undertakes an empirical test of the "bandwagon effect"—individuals rallying to the majority opinion. The study is done outside the electoral context on two issues: abortion and the constitutional future of Quebec. A panel is used, as well as an experimental design in which respondents are told the state and direction of public opinion. Three methodological criteria are used as minimal requirements for a satisfac tory test of the bandwagon thesis. "Underdogging," as well as opinion movement due to factors outside the experiment, are both accounted for. Though the reasons for a bandwagon remain unclear, the authors demon strate that a bandwagon effect of 5-7 percent existed on both issues.