Do Words Really Mean Things? Habitual Voters and the Framing Effects of Ballot Language

Abstract

The alleged dangers of direct democracy have been debated since the founding of the United States, and this discussion carries on into the current time whenever the subject of referendum elections is raised. While much scholarship in recent years has been devoted to the study of framing effects in news media and electioneering, relatively little attention has been given to the way ballot measures are written. Those studies that have been undertaken have shown either mixed or inconclusive results. One variable that appears to be underutilized is the sophistication of, or the habits of, the voter as it pertains to their ability to sift through loaded language and make an informed decision on election day. Using a survey experiment involving 500 respondents obtained through the Amazon Mechanical Turk service, I attempt to discover the extent to which political sophistication and habitual voting may alter the effects of framing in the language of ballot measures. The treatments are related to two state questions currently set for the November general election in Oklahoma. The findings should provide important insight into whether differences in ballot language might have a significant impact on the outcome of referendum elections.