32 The Urge to Splurge plained variance = 95.32%, Cronbach’s alpha = .91; Experiment 2b: explained variance = 87.79%, Cronbach’s alpha = .86; see Table 2). In Experiment 1, we also asked about some additional variables for exploratory reasons. Since these variables were not affected by interactivity, we did not measure them again in Experiment 2a and Experiment 2b. Therefore, these experiments can be considered as more confirmative in nature than Experiment 1.5 Results. Participants in Experiment 2a scored significantly higher on active control in the high-interactivity condition (M = 3.56) than in the low-interactivity condition (M = 2.92; t = -3.63, p < .001). They also scored significantly higher on two-way communication (M = 3.64) in the high-interactivity condition than in the low interactivity condition (M = 2.33; t = -4.84, p < .001), on synchronicity (M = 3.68 versus M = 2.63, t = -4.05, p < .001), and on overall perceived interactivity (M = 4.31 versus M = 2.79, t = -5.22, p < .001). Also in Experiment 2b participants scored significantly higher on all manipulation-check variables in the high-interactivity condition than in the low-interactivity condition (active control: M = 4.45 versus M = 2.69, t = -6.35, p < .001; two-way communication: M = 3.58 versus M = 2.59, t = -3.81, p < .001; synchronicity: M = 3.98 versus M = 2.98, t = -3.47, p = .001; overall perceived interactivity: M = 4.31 versus M = 2.84, t = -5.76, p < .001). The manipulation in Experiments 2a and 2b therefore succeeded. In line with the results of Experiment 1, a one-way ANOVA showed that there is no direct effect of interactivity on impulse-visit urges (F (1, 151) = .92, p = .338) and impulse-buying urges (H1b, F (1, 151) = .69, p = .406) in Experiment 2a. Similarly, also in Experiment 2b no direct effects of interactivity on impulse-visit urges (F (1, 188) = 1.12, p = .292) and impulse-buying urges (F (1, 188) = 3.18, p = .076) were found. Therefore, hypotheses 1a and 1b are rejected. The mediation hypotheses were again tested in PROCESS. Experiment 2a shows a positive significant effect of interactivity on self-agency (b = 1.79, SE = .26, 95% CI: [1.28 to 2.30]), and of self-agency on impulse-visit urges (b = .17, SE = .07, 95% CI: [.03 to .32]). Subsequent to these results, but opposite to the results of Experiment 1, we found a significant positive indirect effect of interactivity on impulse-visit urges through self-agency in Experiment 2a (b = .31, SE = .15, 95% CI: [.03 to .62]). Experiment 2b shows similar results. There is a positive effect of interactivity on self-agency (b = 1.38, SE = .24, 95% CI: [.92 to 1.85]) and a positive effect of self-agency on impulse-visit urges (b = .37, SE = .07, 95% CI: [.24 to .51]). Subsequently, we also find in Experiment 2b a significant positive indirect effect of interactivity on impulse visit urges through self-agency (b = .52, SE = .14, 95% CI: [.28 to .82]. Based on Experiment 2a and 2b, hypothesis 2a, self-agency mediates the effect of interactivity (high versus low) on impulse-visit urges, is therefore accepted. Hypothesis 2b, self-agency mediates the effect of interactivity (high versus low) on impulse-buying urges, is -in line with the outcomes of Experiment 1- also accepted based on the results from Experiments 2a and 2b. Experiment 2a shows 5. For exploratory purposes, we also included state mindfulness and private self-awareness in Experiment 2b to explore if these variables would explain a possible negative effect of interactivity on impulse-visit urges and on impulse-buying urges. Process’ model nr 4 shows that both mindfulness and self-awareness did not function as a mediator in these effects.
RkJQdWJsaXNoZXIy MjY0ODMw