Abstract
This experiment (N = 198), conducted just before the 2008 presidential election, set out to examine the effects of tone and sponsorship in current political advertising, the first such study since campaign law began requiring candidates to approve their ads explicitly. In another first, we also examined the role of reactance in responses to political advertising. With regard to tone, positive ads received higher ad evaluation and cognitive response valence scores and less reactance than negative ads, but negative ads led to a greater likelihood of turning out to vote. Moreover, those without a strong candidate preference were more likely to vote for a candidate supported by a negative ad. Sponsorship had little effect on its own, but there were some intriguing interactions with political knowledge such that high-knowledge respondents had less reactance and lower opponent ratings, whereas moderate-knowledge respondents had the opposite reaction. We also found that reactance appears to play a major role in the effects of political advertising. It was associated directly with more negative cognitive responses, ad, and candidate evaluations and indirectly with lower intention to vote for the candidate supported by the ad, but it had no relationship with intent to turn out to vote.
Notes
Note. Means and standard deviations with interest/attention, political predisposition and political knowledge held constant at their means. “Candidate” refers to the candidate supported in the ad. “Opponent” refers to the opposing candidate. Eta-squared reported is partial eta-squared.
a n = 100.
b n = 95.
*p < .05. ***p < .001.