57 Chapter 3 Results. A one-way ANOVA shows no direct effect of message frame on impulse buying urge (F (1, 525) = 3.30, p = .07). Therefore, Hypothesis 1 is rejected again. However, Process Model 4 demonstrates a mediation effect of message frame on impulse buying urges through moral justification (b = .68, SE = .11, 95% CI: [.48 to .89]), meaning that an other-benefit frame positively affects impulse buying urges through moral justification (H3). Message framing does not affect impulse buying urges through deservingness justification (b = .09, SE = .12 95% CI: [-.16 to .33]). In line with Experiment 2, Experiment 3 also finds a direct positive effect of message frame on impulsive buying behavior. People who were exposed to an other-benefit frame chose to buy the cookies more often (17.3%, with points they earned by participating in this study) than those who were exposed to the self-benefit frame (8.6%, Chi2 = 8.97, f2 = .16, p = .003). Therefore, Hypothesis 2 is again accepted. Again, Process Model 4 shows that moral justification functions as a mediator in the effect of message frame on impulse buying (b = .47, SE = .12, 95% CI: [.27 to .74]), and deservingness justification does not (b = 03, SE = .04, 95% CI: [-.06 to .12]). Thus, the effect of the message frame on impulse buying is explained only by moral justification. Hence, Hypothesis 3 is also accepted. Moreover, in both conditions, people again admitted that they (at least partially) wanted to buy the cookies for selfish reasons (Mself = 3.08, SD = 1.70; Mother = 4.28, SD = 1.92). This again indicated that people who were exposed to the other-benefit frame morally justified why they were allowed to buy the cookies on impulse and did not buy the cookies purely for, for example, altruistic reasons. Finally, a linear regression shows that Hypothesis 4 (impulse buying urges lead to impulse buying behaviors) is also supported (F [1, 525] = 17.14, p = < .001). The impulse buying urge is indeed a predictor of impulse buying behavior (b = .18, t [526] = 4.14, p = < .001). An overview of the accepted and rejected hypotheses based on the results of Experiments 1 to 3 is presented in Figure 4. Discussion. Experiment 3 replicates the effects found in Experiment 2, gaining confidence in the results (McEwan et al., 2018). Furthermore, it shows that impulse buying behavior is predicted by impulse buying urges. These results strengthen the proposed idea that other-benefit advertising frames are used by consumers to morally justify the purchase of hedonic goods, which results in impulse buying behavior. Again, we performed a sensitivity analysis using G*Power (Faul et al., 2007). The smallest effect size to be detected was f2 = .02. Because the found effect size in Experiment 3 is considerably larger (f2 = .16), it can be concluded that this study is well powered.
RkJQdWJsaXNoZXIy MjY0ODMw