Search TESS:


Racial Appeals and Political Campaigns in a Diversifying Society

Download data and study materials


Principal Investigator(s):

Natalie Masuoka
Tufts University
Home page:

Sample size:590
Field period: 8/17/2009-11/25/2009



This proposal outlines an experimental design that seeks to expand our understanding of "racial attitudes" by considering how theories which explain white attitudes about blacks apply in more racially diverse settings. At the heart of this design is a replication of the Implicit-Explicit model proposed by Tali Mendelberg ( 2001) which claims that that while white Americans have increasingly developed more egalitarian values about race, many continue to feel resentment towards blacks but are unwilling to express these concerns openly for fear of being perceived as a racist. Thus, elites can effectively prime racial thinking through the use of implicit racial appeals, which communicate race indirectly through images or visuals rather than directly referrring to blacks by name. This study applies the Implicit-Explicit model on an alternative target group—Latinos—and a public policy issue that is currently assumed to be framed around non-black minorities--immigration. In this experiment, respondents will be asked to evaluate a potential mayoral candidate after being presented with one of three different treatments that vary in the level of implicitness in the racial appeal. The data demonstrate that anti-immigrant messages are interpreted as partisan which in turn influences whether or not an individual will vote for that candidate. Further, implicit racial appeals have the greatest effect on those who report more open attitudes about immigration.

Research Questions:

Are campaign appeals linking immigration to particular ethnic groups as effective at priming racial thinking among whites as those appeals that target blacks such as the infamous 1988 Willie Horton ad? In other words, can elites who want to prime racial thinking simply substitute one racial minority group for another in a racial appeal and get the same, predictable results?

There are many uses of implicit communications. What degree of "implicitness" is needed to activate racial thinking?

Experimental Manipulation:

This study uses a panel design taken in two waves. In the first wave, all respondents are asked an opinion survey of 8 questions. The second wave will be conducted at least one month later. In this second wave, respondents will be given an instruction screen informing them that they will be evaluating a potential mayoral candidate. Respondents will then be randomly assigned to one of four treatment groups. All respondents will be given a total of 3 different images (each image will be shown separately but sequentially). These images will be 3 pages of a hypothetical candidate's campaign website—images will include text and a photograph. All respondents will first be shown the same two images which present the candidate's position on education and transportation. The third image presented to respondents will represent the experimental treatment. There will be four different versions of the third image which presents the candidate's position on immigration. The treatment groups provide a range of "implicitness:" (Please see Appendix for copies of treatments)
1) Explicit treatment: text discusses "Latino immigrants," includes an image of 2 male Latino immigrants
2) Implicit-immigrants treatment: text discusses "immigrants," includes same image as in the explicit treatment
3) Implicit-language treatment: text discusses "immigrants," includes an image of a Spanish language sign
4) Control: text discusses housing affordability, includes image of suburban houses

Respondents will be given the opportunity to view images for as long as they desire and should be given the option to move back and forth between images. Respondents will then be told to proceed to the candidate evaluation survey when ready. Respondents will not have the option of returning to the images once they begin the survey. All respondents will then complete a survey of 9 questions.

Key Dependent Variables:

After respondents viewed the hypothetical candidate's platform, they were then asked to provide their evaluation of the candidate and provided ratings of the candidate's perceived ideology and party affiliation, tone of campaign and willingness to vote for the candidate. Respondents were also asked to provide their opinions on various issues related to immigration policy.

Summary of Findings:

Anti-immigrant messages are viewed as partisan and conservative. As a result, respondents use anti-immigrant messages in how they evaluate a candidate. However, the respondent's education level, ideology and level of perceived threat all moderate how the anti-immigrant messages influence candidate evaluation. Further, the results suggest that respondents do not see a difference between explicit and implicit appeals. In general, an anti-immigrant message is evaluated the same regardless of whether the candidate makes direct or indirect reference to Latinos.


Masuoka, Natalie. 2010. "Implicit Communications and Candidate Evaluation." Paper presented at the Western Political Science Association Annual Meeting. San Francisco, CA. March.

Masuoka, Natalie. 2009. "Implicit and Explicit Priming in the Context of Immigration Policy" Paper presented at the Immigration and Political Psychology Conference, University of Texas, Austin, April.

Copyright © 2014, TESS