p. 15

This analysis revealed a significant main effect of model [F(1,110) = 38.66, p < 0.001, h2 = 0.19], indicating, as in Study 1, that attitudes toward the product were more positive when there was a model in the ad (M = 4.73, SD = 1.09) than when the model was absent (M = 3.75, SD = 0.96). There was also a main effect of product relevance [F(1,110) = 13.10, p < 0.001,h2 = 0.06], indicating that participants had more favorable attitudes toward the shampoo (M = 4.54, SD = 1.21) than toward the computer (M = 3.96, SD = 0.98). There was also an interaction between model and product relevance [F(1,110) = 16.95, p < 0.001, h2 = 0.08]. This interaction indicated that attitudes toward the relevant product (shampoo) were more positive when it was depicted along with an attractive model (M = 5.34, SD = 0.83) compared to no model [M = 3.71, SD = 0.97; F(1,114) = 44.55, p < 0.001, h2 = 0.26]. More importantly, however, the expected interaction between model, product relevance, and elaboration was significant [F(1,110) = 10.75,p < 0.001, h2 = 0.05].