Esther Mertens

| 147 General Discussion of effective interventions for this group of students and highlights the importance of the current dissertation. In general, universal school-based interventions showed some positive effects on improving competencies and decreasing problems in both the intra- and interpersonal domain. Overall, the general intervention effects were small, with somewhat stronger effects in the intrapersonal domain (Cohen’s d = .19, 95%CI [.13; .25]) than in the interpersonal domain (Cohen’s d = .15, 95%CI [.10; .19]). In the RCT evaluating R&W, I found a comparable pattern. R&W is roughly equally effective as other universal school-based interventions in the intrapersonal (Average Cohen’s d = .23, range Cohen’s d = .03 - .38) and interpersonal (Average Cohen’s d = .15, range Cohen’s d = .01 - .31) domain. Additionally, also R&W was more effective in the intrapersonal domain (significantly improving psychological wellbeing, sexual autonomy, and internalizing behavior, but not resilience) than in the interpersonal domain (significantly improving aggression, promising changes (but not significant) in interpersonal relations in the class and bullying, but no effect on externalizing behavior and victimization). Although most intervention effects were found in the intrapersonal domain, it is interesting to point out that in the interpersonal domain the intervention effect of R&W on aggression was relatively strong compared to the general intervention effect on aggression (R&W Cohen’s d = .31 vs. meta-analysis Cohen’s d = .10). Taken together, it appears a priority to examine how universal school-based interventions can be improved to meaningfully affect a broad range of competencies and problems in the interpersonal domain. This is especially important for interventions targeting prevocational students, as these students seemingly experience most problems in the interpersonal domain. Notwithstanding the positive effects in the intrapersonal domain, it is noticeable that school-based interventions appear to be ineffective in stimulating students’ resilience. Specifically, no intervention effects on resilience were found either for R&W or in the meta-analysis, even though these interventions often aim to improve resilience and many are described as “resilience-focused” (e.g., Dray et al., 2017). Improving students’ resilience is important as it enables them to thrive and deal with challenges, stress, and adversity they experience in life (Campbell-Sills & Stein, 2007). More research is needed to study how interventions can improve resilience in order to help students to recover from stressful events and achieve their full potential. Regarding secondary outcomes, R&W showed small positive effects on self-control and emotional self-regulation, showed promising (but not significant) changes in self- esteem, but showed no effect on students’ self-reflection. I tested these four additional outcomes because the theory of R&W indicates these as important competencies that the intervention also aims to improve (Ykema 2002; 2018). The intervention effects of R&W on self-control, self-esteem, and emotional self-regulation were comparable in magnitude to those found in the meta-analysis (R&W Cohen’s d = .22 - .29; Meta- analysis Cohen’s d = .21 - .25). Besides these positive effects, R&W was not able to improve students’ self-reflection. The correlations in the present study (not previously 7