31 Chapter 2 Discussion. We did not find a direct effect of interactivity on impulse urges in Experiment 1. Nevertheless, we performed mediation analysis as a direct effect of X on Y is not required in modern mediation analysis: “Statistical mediation analysis has changed since the publication of Baron and Kenny (1986). The heyday of the causal steps ‘criteria to establish mediation’ approach is over. … Modern mediation analysis emphasizes an explicit estimation of the indirect effect, … and an acknowledgment that evidence of a statistically significant association between X and Y is not necessary to talk about and model intervening variable processes …” (Hayes, 2018, p.146). The mediation analysis shows an indirect effect of interactivity on impulse-buying urges through self-agency. There are no effects found on impulse-visit urges. Since Experiment 1 was partially explorative in nature, we replicate the study in Experiment 2a and Experiment 2b to confirm the results. For Experiment 2a and Experiment 2b, the same stimulus material was used, and largely the same procedure was followed as in Experiment 1. Experiment 1 participants were asked to interact with the screen either for one minute or for as long as they pleased. Since interaction time did not influence the results, we decided to ask all participants in Experiments 2a and 2b to interact with the screen for as long as they pleased to create a more natural situation for all participants. However, we set a minimal interaction time of 40 seconds, to provide the user with enough time to explore the condition. Participants. The samples of Experiments 2a (N = 153) and 2b (N = 190) were similar to those of Experiment 1. The mean ages in Experiment 2a and 2b were 47.69 (SD = 17.79) and 49.37 (SD = 19.12), respectively. All participants were again adult passers-by and female. Seven of the participants in Experiment 2b had already participated in Experiments 1 or 2a and were therefore excluded from further analyses (N = 183). Measurements. Manipulation check. We used the same manipulation check as in Experiment 1 (see Table 1). In Experiment 2a active control had a Cronbach’s alpha of .75, two-way communication of .81, and synchronicity of .75. In Experiment 2b active control had a Cronbach’s alpha of .82, two-way communication of .70, and synchronicity of .83. Dependent variables and mediators. As in Experiment 1, we measured impulse-buying urges (Experiment 2a: explained variance = 70.23%, Cronbach’s alpha = .79; Experiment 2b: explained variance = 74.01%, Cronbach’s alpha = .82), impulse-visit urges (Experiment 2a: explained variance = 73.40%, Cronbach’s alpha = .82; Experiment 2b: explained variance = 73.62%, Cronbach’s alpha = .82), and self-agency (Experiment 2a: ex- Method. Stimulus material & procedure. Experiment 2a & Experiment 2b
RkJQdWJsaXNoZXIy MjY0ODMw