Date Published: June 30, 2017
Publisher: Public Library of Science
Author(s): Gary A. Williams, AnaMarie C. Guichard, JungHa An, Philip Allen.
Priming with race-typed names and religious concepts have been shown to activate stereotypes and increase prejudice towards out-groups. We examined the effects of name and religious word priming on views of a specific and well-known person, President Barack Obama. We predicted that politically conservative participants primed with President Obama’s middle name (Hussein) would rate him more negatively and be more likely to view him as a Muslim than those not shown his middle name. We also examined whether conservatives primed with concrete religious words would rate President Obama more negatively and be more likely to view him as Muslim than those primed with other word types. Furthermore, we predicted that those who mis-identify President Obama as Muslim would rate him more negatively than would those who view him as Christian. The results provided mixed support for these hypotheses. Conservatives primed with President Obama’s middle name rated him significantly more negatively than did those in the control condition. This effect was not found for politically liberal or moderate participants. Name priming did not significantly affect views of President Obama’s religious affiliation. Although not statistically significant, conservatives primed with abstract religious words tended to rate President Obama more negatively than did those primed with other word types. Religious word priming significantly influenced views of President Obama’s religious affiliation; interestingly, participants primed with abstract religious words were more likely to think President Obama is Muslim than were those primed with religious agent or non-religious words. As predicted, participants who thought president Obama was Muslim rated him significantly more negatively than did those who thought he was Christian. Overall, our results provide some evidence that ethnic name and religious word priming can significantly influence opinions, even with a well-known and specific person.
Political discourse in the United States has been described as being marked by incivility  with efforts often focused on highlighting a candidate’s similarities with the electorate on one hand, and demonizing political opponents on the other. To that end, candidates and their supporters may use a variety of strategies, such as coded talk , to discretely remind voters that they share the same religious beliefs and values as their constituents. For example, conservative evangelical voters are likely to identify a political candidate who uses religious coded talk as an in-group member  and to express higher levels of support for that candidate. When Barack Obama entered the 2008 Presidential race he was the first African-American major party candidate and his out-group status, both in terms of his ethnicity and rumored religious (non-Christian) beliefs, was frequently discussed in the media. For example, as illustrated by the quote above, references to President Obama’s middle name were often used in an attempt to sway potential voters. We suspect that the use of religious coded talk and references to the President’s middle name could serve as primes, consequently influencing views of potential voters. The authors are interested in exploring the effects of both ethnic (e.g., name) and religious priming, similar to what may have occurred during the 2008 and 2012 Presidential elections, on people’s attitudes toward President Obama, a well-known figure in contemporary America.
First, we conducted a 2 (name prime) x 3 (political identity) ANOVA to test the prediction that conservative participants in the middle name present condition would rate President Obama more negatively than would those in the middle name absent condition. Results showed no significant main effect of the name prime manipulation on opinions of President Obama F(1,329) = 1.56, p = .213, partial eta-squared = .005. There was, however, a significant main effect of participant political identity on opinions of President Obama, F(2, 329) = 39.35, p < .001, partial eta-squared = .193. All political identity groups significantly differed, ps < .05. Conservative participants (n = 59, M = 3.12, SD = 1.17) rated the president more negatively than did liberal (n = 119, M = 4.70, SD = 1.00) and politically moderate participants (n = 157, M = 4.09, SD = 1.11). The predicted interaction between the name prime manipulation and participant political identity on opinions of the president was marginally significant, F(2, 329) = 2.80, p = .062, partial eta-squared = .017. Pairwise comparisons showed that conservative participants in the middle name present condition (n = 34, M = 2.87, SD = 1.04) rated President Obama significantly more negatively than did those in the middle name absent condition (n = 25, M = 3.47, SD = 1.26), p = .036, d = .527 (see Fig 1). Using conventions in psychology this is considered a medium effect size . There was no significant effect of name prime for liberal and moderate participants (ps > .28).
Not surprisingly, as compared to politically moderate and liberal participants, politically conservative participants had more negative views of a politically liberal president. Interestingly, conservatives were somewhat susceptible to the negative effects of race-type name priming whereas race-type name priming did not influence views of liberal and moderate participants. Among conservative participants, those who were primed with President Obama’s middle name had more negative views of him, as compared to conservatives not primed with his middle name.