Kyrene Gibb and Dr. Christopher Karpowitz, Political Science
In the 2012 Presidential election cycle, independent expenditure committees, or Super PACs, had spent a more than $237 million advocating for one or the other presidential candidate (OpenSecrets.org, 2013). Given the large sums of money that independent expenditure committees are raising and spending to influence the outcome of presidential elections, it is worth investigating how the source of Super PAC ads affects the way they are received. As such, the purpose of this ORCA project was to investigate whether voters provided with specific identifying information for the source of an advertisement will evaluate a message differently than voters given less explicit or absent source cues.
In order to determine whether the sponsor of a Super PAC affects the way voters interpret Super PAC ads, I conducted a survey experiment through the Utah Voter Poll. This online survey of Utah voters allowed me to show a mock political mail advertisement to survey respondents. This advertisement was for Alan Marshall, a hypothetical congressional candidate running for the Wyoming At Large seat. I randomly assigned respondents into three conditions where the identifying information given for the source of the ad was varied. In the first condition, respondents were given the name of a generic Super PAC, “Americans for a Brighter Future PAC,” as the source of the ad. This served as a valuable control as I conducted my analyses. In the second and third conditions, respondents were told the name of the Super PAC presenting the ad as well as the industry that funds that Super PAC. I attributed funding for Americans for a Brighter Future PAC to ExxonMobil and the Sierra Club in the second and third conditions, respectively, because these two organizations represent distinctly different interests regarding the environment. Following these ads, I asked respondents whether the ad made them more or less likely to vote for the candidate. I also asked them to identify and evaluate the source of the ad to ensure that my experimental treatment was effective. I hypothesized that knowing the organization responsible for funding Americans for a Brighter Future PAC would change the way voters interpreted the political ad based on their prior opinion of these organizations.
My survey experiment was fielded in June of 2013 and I have since been able to analyze my data, finding some interesting results. Given only basic descriptive information about Alan Marshall, 54% of Utah voters said they were unsure of how likely they would be to vote for the hypothetical candidate, and 27% of voters said they would be somewhat likely to vote for him. After being presented with the Americans for a Brighter Future PAC ad without any additional sponsor information, 49% of voters said they would be more likely to vote for Marshall, even though 64% of voters were unsure of how they felt about the PAC. After viewing the ad in which they were told that Americans for a Brighter Future PAC was sponsored by ExxonMobil, 37% of voters were more likely to vote for Marshall and 30% were less likely to vote for him. With the additional information attributing the ad to ExxonMobil, only 31% of voters were still unsure about their opinion on the ad’s sponsors, which indicates that source cues did have an effect on the voters. The ad that told voters that Americans for a Brighter Future PAC was funded by the Sierra Club produced similar results. 41% of voters were more likely to vote for Marshall, and 29% were less likely to do so. Only 29% of voters were unsure how favorably or unfavorably they felt about the sponsors of the ad.
Using difference of means tests to compare groups, I found that seeing the control ad with only the Super PAC listed as a sponsor made voters only 6% more likely to vote for Alan Marshall than they were when given only a basic description of the candidate (significant at the p=0.01 level). However, when additional information was added, i.e. the funding for the Super PAC was attributed to either ExxonMobil or the Sierra Club, voters were less likely to vote for Alan Marshall (significant at p=0.01). Compared to the generic Super PAC ad, voters who saw the ExxonMobil treatment were nearly 12% less likely to vote for Alan Marshall, and voters who saw the Sierra Club ad were about 9% less likely to vote for Marshall.
Given this evidence that source cues did affect the way Utah voters evaluated this political message, I decided to analyze my results further. I found that among voters who had previously expressed a favorable opinion of oil companies, 38% were more likely to vote for Alan Marshall after seeing the ExxonMobil-sponsored ad. Similarly, if voters had previously expressed unfavorable opinions of oil companies, 35% were less likely to vote for Marshall after seeing the ExxonMobil-sponsored ad. I repeated my analysis for the Sierra Club treatment and found 65% of voters who had previously expressed a favorable opinion of the Sierra Club were more likely to vote for Alan Marshall after seeing the Sierra Club-sponsored ad. Among those who had expressed unfavorable opinions of the Sierra Club, 46% were less likely to vote for Marshall after seeing the ad attributed to the Sierra Club.
While there is still work to be done in the area of analyzing how Super PAC sponsors affect the way Super PAC ads are received and how much influence such ads have in actual elections, the research that ORCA funding has allowed me to conduct has shown that the source of Super PAC funding matters to voters. The more voters know about the source of the funding behind a Super PAC ad, the more capable they are of evaluating the ad and the candidate appropriately.