Am I In? Influence of Viewers’ Race and Sex on Image Appeal for Higher Education Advertising

Diversification of recruitment advertising has become increasingly common across industries, including for institutions of higher education where promoting student body diversity is an on-going concern. In their recruitment activities, institutions go to extensive lengths to portray their campuses as racially and culturally diverse, including in promotional materials. Using custom images similar to those found in university promotional materials, this experimental study examines the influence of viewers’ race and sex on short exposure to race- and sex-specific images. Guided by Social Identity Theory’s in-group versus out-group premise, results suggest that participants’ race and sex have a significant influence on image appeal and that race represents a more salient group identifier than sex for certain groups. Findings are compared with previous race-related advertising research, and implications for higher education ad messaging and avenues for future research are discussed.

participation was voluntary and responses were treated as confidential, only aggregate data were reported and used in the analysis, and the study was approved by the university's Institutional Review Board.
One hundred and sixty-five participants completed the experiment. Upon completion of the experiment, participants answered an open-ended question to self-identify their race, and their responses were then divided in three categories: black, white and other. Out of the 165 participants, 141 met the study's requirements and were retained for the analysis. Thirtyseven participants self-identified as black (23 women and 14 men), and 104 participants selfidentified as white (60 women and 44 men). Responses from the 24 other participants were removed prior to the analysis because they racially self-identified as Hispanic or Middle-

Eastern.
A 2 (participant's race: black or white) X 2 (participant's sex: man or woman) experimental setting was used to test for influence of race and sex on perceived image appeal. Ten different images depicting two broadcast news anchors in professional attire were photographed exclusively for the study in order to control model race and sex, mirroring those used for journalism school recruitment purposes. The models' racial identity used in the experimental stimuli was solely based on their visible identity as black or white, men or women. The background of all ten images was the same, and all models were dressed alike to ensure that only the models' race and sex differed. All participants were exposed to the ten different images, divided into three categories: images of black students only; white students only; and black and white (integrated) students. The first two categories contained three images each, differentiated by model sex: men only; women only; and both a man and a woman. The category with both racial groups contained four images to account for all possible race-sex combinations.
Each image was presented on screen for six seconds, and the image presentation order was randomized using MediaLab software (Jarvis, 2006a). Instant appeal was measured with a continuous measurement method during image viewing using a hand-held linear potentiometer (slider) marked on a scale from 0 (very negative) to 9 (very positive) operating in Direct RT. Participants were told they would be exposed to a series of images and would be asked to rate their image appeal. Participants were then trained to use the slider and latter image, white participants displayed a stronger image appeal for the other two all-black images than they displayed for the WM-WM image (see Figure 1).
To explore the effect of participants' race and sex on image appeal, a series of two way 2 (participant's race) X 2 (participant's sex) analysis of variances (ANOVA) were conducted with image appeal as the dependent variable. Main effects for race, sex, and the interaction were then examined to test the hypotheses. H.1.a.: Black students will display a stronger appeal for the images depicting only black models than for the other images.
ANOVAs revealed a statistically significant main effect of race on participants' appeal to the three images depicting only black models (see Tables 1, 2      ANOVAs revealed a statistically significant main effect of race on participants' appeal to only one of the three images depicting exclusively white models (see Table 4). White participants (M = 3.60, SD = 0.82) displayed statistically significant higher appeal levels for the image featuring a white woman with a white man than did black participants (M = 3.03, Online Journal of Communication and Media Technologies Volume: 2 -Issue: 4 -October -2012 SD = 0.98). There were no statistically significant differences between white and black participants' appeal for the other two all-white images. Therefore, this hypothesis was only partially validated. Apart from the four images previously mentioned (three all-black and one all-white), no statistically significant differences were found between black and white participants' appeal to the other six images. In other words, participants' race did not affect their appeal levels for any of the four integrated images, thus invalidating this hypothesis.

R.Q.2.: How does viewer sex affect appeal to race-and sex-specific university recruitment images?
Next, participants were divided by sex to determine if sex accounted for differences in image appeal. Overall, women displayed higher appeal levels for the all-white images than did men.
Conversely, men displayed higher appeal levels for the all-black images than did women (see Figure 2).
Women favored the white woman/black man image the most, while men favored the black woman/black man image the most. However, participants' sex was determined to be statistically significant for only one image: the image depicting two black men (see Table 3).  Both women and men displayed the lowest appeal levels for the image depicting two white men. Women displayed higher appeal levels than did men for two of the three women-only images (WW-WW and WW-BW), while men displayed higher appeal levels than did women for two of the three men-only images (BM-BM and WM-BM) (see Figure 2).  H.2.b.: Black women will display a stronger appeal for the images featuring at least one black woman than for images with no black women.