The current study is a content analysis of 379 humorous political advertisements from 1952 to 1996 in every election level from civic to presidential. Percentages, Chi Square analyses and one-sample t-tests are employed to illustrate the different characteristics of the ads' content and the candidates who sponsor them.

Based upon the theory that humor may serve to mitigate potential voter backlash against the sponsoring candidate, this study predicted that the majority of humorous political advertisements would be negative in focus. Results of the present research support previous research findings that, indeed, the majority of humorous political advertisements attack an opponent. Previous findings that the majority of such ads are sponsored by white men and challenging candidates were also supported. Candidate image and campaign issues were equally represented in the ads. Similarly, ads were equally likely to use logical and emotional appeals, both of which were twenty percent more likely to occur than source credibility appeals. The ads were shown to utilize far fewer fear appeals, but to be three times more likely to employ unethical distortions of audio or video technology, than were general political advertisements studied in past content analyses (Kaid, 1987).

Results of this study support the theory that humor often serves to mitigate voter backlash against the sponsoring candidate which may result from negative advertisements. In addition, the results suggest that female and minority candidates are still hesitant to employ humorous strategies, which are viewed by practitioners as less traditional or "safe" than straightforward attacks. The implications and limitations of this research are discussed.