100 Chapter 4 During off-job time, they may actively engage in sports activities or read a book with the intention to change their mood state. Moreover, our construct validity research (masked, in press) indicates that employees may engage in a range of activities that raise their levels of energy, positive affect, and concentration. Second, managers may facilitate the process of proactive vitality management, by creating a work-family culture that prevents overwork and offers opportunities to choose for flexible working times (Thompson, Beauvais, & Lyness, 1999). Such a culture would also offer employees more autonomy during work and leisure time for activities that foster resources. Third, the findings indicate that learning goal orientation is crucial for the relationship between proactive vitality management and work engagement and creativity. Organizations that are highly dependent on creative performance may want to select employees on the basis of their goal orientations, or could offer their employees trainings in which they learn to focus on learning from mistakes (Noordzij, van Hooft, van Mierlo, Dam, & Born, 2013). Strengths and Limitations The present study has some limitations that should be acknowledged. First, we used one source of information, which may raise concerns regarding the validity of the findings. However, we combined the survey method with repeated measurements using a weekly questionnaire, and created a theoretically valid cross-level interaction model. Specifically, we integrated two different methods and analyzed relationships of proactive vitality management with work engagement and creative performance after controlling for previous (week) levels of work engagement and creative performance – alleviating problems of commonmethod bias. Although we do think that future researchmay profit from using objective outcomes of the creativity process, we are confident that we used a robust test of our hypotheses. It seems highly unlikely that our participants would have been able to produce exactly the theoretically predicted interaction patterns if they would have been inclined to please the researchers and would have tried to produce consistent answers. Thus, we believe that demand characteristics are not a major threat to the validity of our findings. A second possible limitation is that students were asked to collect data, which resulted in a convenience sample. This means that our sample may not be representative of the working population. However, Demerouti and Rispens (2014) have argued that student-recruited samples may actually enhance external validity, because
RkJQdWJsaXNoZXIy MTk4NDMw