Three quasi-experimental studies with nonstudent samples reveal that one's ability to identify with a character shown in an ad based on shared race depends on the construction of the ad and the context in which characters are depicted. Results show that race-based identification overshadows both gender-and role-based identification for a racially targeted ad for distinctive black subjects but occurs for both black and white subjects for a culturally ambiguous ad. Further, results show that race-based character identification is absent when black and white characters are depicted in a mainstream inclusive ad and that dominant cultural norms predominate. Theoretical and managerial implications regarding the contextuality of race-based identification, processing of source cues, and construction of ads in a pluralistic society are discussed. © 2009 Wiley Periodicals, Inc.Consider the results of an in-class discussion among a diverse mix of evening part-time MBA students in a Consumer Behavior elective. Throughout the lecture, the instructor showed a variety of ads that differed in content, format, and type, to illustrate different concepts. At one point, students were asked to consider Psychology & Marketing, Vol. 26 (11) an ad depicting a black man and boy, presumably father and son. Students were asked to raise their hands if the ad resonated with them and to share why. Black and white students in their 30s or older raised their hands. So, too, did younger black students in their early 20s, both African Americans and foreign nationals. Only younger white students failed to raise their hands.In the discussion that ensued, it became clear that the ad resonated with the older students (black and white) because they had the role of parent in common with the adult character depicted, not shared race (even if applicable). In contrast, the younger black students identified with both characters based on shared race (not role), a more relevant source of identification for these participants who did not yet have children. In this unscientific inquiry, younger white students perceived no basis for identification, including shared gender with the males shown, which seemed to have no effect on the remaining men in the class.In research exploring racial identification and similarity, researchers have presumed a priori that a match in race between a viewer and ad characters will lead to similarity and identification effects based on race, irrespective of other possible bases of similarity and identification that may be more relevant (Whittler & DiMeo, 1991;Williams & Qualls, 1989). The previous example shows that this may not be a valid presumption. Further, what happens when a character in an ad is similar in race, but different in gender or role? Will a black male viewer find an ad with a white male character or an ad with a black female character more appealing? Will a working mother be more likely to identify with a stay-at-home mom of the same race or a working mother of a different race? In an increasingly diverse m...