Presumed influence, planned behaviors, and political advertising
Johnson, Elizabeth Adams
MetadataShow full item record
Low numbers of young voters heading to the polls each election cycle and decreased levels of political involvement among voters between the ages of 18 and 24 are disturbing trends demanding increased scholarly attention. Research has garnered a strong level of support for the perceptual component of the third‐person effect, which posits there will be a greater perceived influence of a message on the other than on the self. Less documented is the behavioral component of the third‐person effect, which predicts the self will act to control or mitigate the effects of the message on the other. Given its behavioral predictions, the third‐person effect seems to have a logical union with the Theory of Planned Behavior. This study tests both theories independently and jointly to reveal their power in predicting voting behaviors and responses to political advertising of the young electorate. A factorial experiment design was employed for this study (n=270) in which participants in randomly assigned conditions viewed Presidential campaign advertisements one week prior to the 2004 general election. During the week following the election, a follow‐up survey was distributed to measure self‐reported voting behaviors. The results of this study yield strong support for the perceptual component of the third‐person effect. This study also provides support for the behavioral component of the third‐person effect; intensity of third‐person effect significantly and positively correlated with intention to vote in the 2004 Presidential election. Perhaps most interestingly, significant interactions of message sponsorship and tone as well as of message tone and strategy predicted strength of perceptual bias. Results of structural equation modeling revealed that the perceived behavioral control component of the Theory of Planned Behavior was an extremely strong predictor of voting behaviors; however, the third‐person perceptual bias did not increase the variance explained in behavioral intentions to vote among the young electorate. Overall, in addition to their theoretical contributions, the results of this study have important applied value to inform message strategy in both voting enhancement campaigns and political advertising to ultimately increase the political involvement of the young electorate.