44 Chapter 2 ‘break’ (answer: ‘coffee’). The six items varied in difficulty level, and participants were given two minutes to (try to) complete the test. We randomly assigned participants to ‘conditions’ that determined whether they received the RAT items at the beginning, halfway, or at the end of the day-level questionnaire. RESULTS AND DISCUSSION Prior to testing our hypotheses, we calculated the Average Variance Extracted (AVE) for each of the variables to examine whether PVM can be distinguished from the other variables in the nomological network. To establish this, the AVE estimates of two variables both have to be greater than their shared variance (i.e., squared correlation) (Farrell, 2010; Fornell & Larcker, 1981). This was the case for every combination of PVM and each of the variables in the nomological network. After that, in order to establish convergent, discriminant, and criterion validity, we calculated correlations between all the variables in the nomological network. An overview of all the general-level results, including the AVE estimates, can be found in Table 2, and the day-level results are displayed in Table 3. In the general sample, we found positive relationships between PVM and proactive personality (r = .36, p < .001) and self-insight (r = .33, p < .001), supporting hypothesis 1a and 1b. In addition, we found positive relationships between PVM and job crafting (i.e., increasing social job resources: r = .21, p < .001, increasing structural job resources: r = .47, p < .001, increasing challenging job demands: r = .34, p < .001), and between PVM and relaxation (r = .18, p < .001), providing initial support for hypothesis 2a and 2b (Table 2). In the day-level study, we only measured increasing social job resources and relaxation, which were both positively related to day-level PVM (r = .25, p < .001 and r = .24, p < .001, respectively), further supporting hypothesis 2a and 2b (Table 3). In hypothesis 3, we tested the null-hypothesis that PVM would be unrelated to (a) decreasing hindering job demands and (b) psychological detachment. PVM did not significantly correlate with decreasing hindering job demands, both on general and day-level (r = -.06, p = .114 and r = -.02, p = .816, respectively), supporting hypothesis 3a (Table 2 and 3). Furthermore, PVM was not significantly related to psychological
RkJQdWJsaXNoZXIy MTk4NDMw