Sorting out resistance to celebrity endorsements

Republicans and conservatives used to complain about celebrity involvement in politics, despite the fact that the Party brought us President Ronald Reagan and Congressmen Sonny Bono and Fred Grandy. After the Republican Party nominated a TV reality-show host with no public service experience for President in 2016, conservative criticism of entertainers continued, but generally did not include eventual President Donald Trump. 

Scholars have generated much research on the positive effects of celebrity political endorsements. By positive effects, I mean research that has shown that under the right circumstances, celebrity endorsements of political candidates and beliefs can influence members of the public to agree with the celebrity’s preferences. There has also been some evidence to suggest that taking political positions may even improve the standing of the celebrity in the public’s eyes. On the other hand, some research has found minimal or no effects of celebrity political endorsements. 

Recently I’ve become interested in those individuals on whom celebrity political endorsements appear not to work. Is there a group of potential voters who reject celebrity endorsements completely, and instead rely exclusively on other methods to make up their minds? To find out, I looked back into some survey data that my Bowling Green State University (BGSU) colleague Melissa Miller and I gathered just before the 2016 Ohio presidential primaries. The poll was conducted Oct. 16 and 17, 2015, by Zogby Analytics and included 804 likely 2016 general election voters, and has a margin of error of 3.5 percentage points. While the data is not new, it is still useful for helping us understand the important role of celebrities in politics. 

Respondents were asked if the hypothetical presidential endorsement of each of seven celebrities would increase their likelihood of voting for the candidate, reduce it, or have no effect. The celebrities whose potential endorsements were measured included George Clooney, Beyoncé, Trace Adkins, Oprah Winfrey, Ted Nugent, Eva Longoria, and Lena Dunham. These celebrities were chosen because of their racial, gender, partisan and ideological diversity. If a voter were to be influenced by a celebrity, certainly one of these should do it. In every case, the overwhelming majority of respondents indicated that the celebrity endorsement would have no effect, although there were circumstances where the celebrities could motivate votes.

To investigate this further, I decided to look at those who said every one of the seven celebrities’ potential endorsement would have no effect on their own vote. Think of these respondents as “celebrity resistant” or at least “celebrity indifferent.” Exactly 46.1 percent of respondents said each and every one of the celebrities’ endorsements would have no effect on their vote. Who are these people?

First, they are more likely to be women than men (48.1 percent to 41.8 percent) and they are more likely to be non-college educated than college educated (48.9 percent to 43.2 percent).  There were no differences with regard to age, but those unaffected by any celebrity endorsements came from the higher income brackets measured. Also, African Americans (37 percent) were significantly less likely to be unaffected by any celebrity endorsements than others (47.1 percent). In terms of religion, those who identify as born again were significantly less likely to be unaffected by endorsements than others (43.3 percent to 55.2 percent).

There were a number of significant differences in terms of political orientation too. Members of unions were much more likely to be unaffected by any celebrity endorsements than others (53.1 percent to 44.6 percent). Those who identify as independents were about ten percentage points more likely to be unaffected by celebrity endorsements than were those who identify as Democrats or Republicans. Those who identify as moderate (56.1 percent) were much more likely to be unaffected by celebrity endorsements than conservatives (38 percent) or liberals (41.9 percent). 

Some of these differences are more easily explained than others. This time, let’s start with the political variables. Independents and moderates likely know the partisan and ideological orientations of celebrities, most of whom lean left and Democratic, and consequently may reject their endorsements just as they may have rejected party affiliation and consistent ideological preferences overall. Independents and moderates may see celebrity endorsements mostly as proxies for Democratic and liberal endorsements. On the other hand, union members may be accustomed to taking political cues from union leadership, and therefore may be resistant to celebrity endorsements.   

In terms of the demographic variables, the difference between Black and other respondents is likely due to the presence of the hypothetical endorsement of Oprah Winfrey, one of the most consistently successful and powerful African American celebrities ever. The other differences are a little harder to explain. Are women inherently more skeptical of celebrities? Are the born again less so? Why would college educated respondents be less likely to be celebrity resistant? Is common sense more important in celebrity resistance than specialized knowledge? A possible partial explanation is that some of these demographic characteristics correlate with and interact with other political and social variables, creating a complex mix that will need sorting out in future research. 

Of course, there are a number of factors besides demographic and political characteristics that could influence one’s feelings about entertainment and politics, including media preferences, particularly social media use. Unfortunately, the survey data available does not include measures of this, but future research must. It’s also possible there are some people who just refuse to participate in celebrity culture for idiosyncratic reasons that may not show up in survey research.

The data reviewed here suggest that there is a core group of American voters who are unmoved by celebrity endorsements. They appear to be a minority of voters, however. While certain demographic characteristics and political beliefs help explain who is celebrity resistant or indifferent, we cannot say those characteristics cause the individual to be unmoved by any celebrity endorsements. As celebrity involvement in politics continues to grow, will more voters become celebrity resistant, or at least more critical? Will there be more celebrity candidates and endorsements from both the left and right? Since the next election is always just around the corner, we don’t have to wait long to find out. 

David J. Jackson is professor of political science at Bowling Green State University in Ohio. His major research area is the relationship between entertainment and politics, in particular the role of celebrity endorsements in politics. He has also written about the Polish diaspora in North America and U.S. organized labor’s electoral strategies. He is the author of the book “Entertainment and Politics.”