A quantitative review and extension of racial similarity effects in advertising

Date of Completion

January 2010


Business Administration, Marketing|Psychology, Social|Speech Communication|Sociology, Ethnic and Racial Studies




This dissertation provides a quantitative review and extension of racial similarity effects in advertising. It consists of two essays which explore the impact of the endorser's race on the advertising evaluations of African American and White consumers. The literature on ingroup favoritism, which describes the tendency to evaluate people perceived to belong to your own group more favorably than those belonging to other groups, provides the theoretical background to both essays. Essay 1 is a meta-analytic review of extant studies that examined the advertising responses of consumers exposed to ads featuring same (vs. different) race endorsers. A total of 84 statistically independent data sets are included, providing data from 9,496 participants (3,232 African Americans and 6,264 Whites), and summarizing forty years (1969-2009) of research work in this domain. Findings based primarily on explicit measures show that both African American and White participants prefer advertisements featuring same race endorsers, especially when they identify strongly with their racial ingroup. The overall weighted mean effect size for African American participants is d+ = -.53, and for White participants it is d + = .15. Additional moderators of advertising evaluations related to study and participant characteristics, as well to the methodology employed by researchers are discussed. ^ Essay 2 uses the Implicit Association Test (IAT) to assess African American and White participants' racial bias (i.e., preference for African American vs. White racial stimuli), and color bias (i.e., preference for the color white vs. the color black) on their implicit reactions to persuasive communications. Furthermore, this essay investigates the respective individual and joint effects of racial bias and color bias on participants' reactions to non-celebrity endorsers (study 1), celebrity political candidates (study 2), and identical products colored either black or white (study 3). The results from all three studies document that both African American and White consumers exhibit an implicit color association in favor of the color white (as compared to the color black), and an implicit racial association in favor of their racial ingroup. African Americans exhibited an overall pro-African American racial association across the three studies, whereas Whites exhibited a pro-White racial association. Furthermore, White participants' color associations predicted their racial associations, and both White and African American participants' racial associations predicted their product associations. Implicit measures (such as the IAT) are believed to measure culturally learned associations regarding African American and White endorsers, which are not likely to apply to evaluations of celebrities for whom individuals have developed opinions. Consistent with this rationale, findings show that White participants' racial associations predicted their advertising associations for non-celebrity endorsers; however White and African American participants' racial associations did not predict their associations for celebrity political candidates. ^