56 Chapter 2 internalising symptoms, and general psychological difficulties for the purpose of this review. Some papers fit into more than one group. Well-Being Eleven studies (described in 13 papers) measured well-being. Participants’ mean age ranged from 12 to 22 years, and the proportion of male participants varied from 0% to 100%. Eight studies used a regular RCT design, while three used cluster randomisation (C. Y. Hsieh & Chen, 2019; Kuosmanen et al., 2017; Sun et al., 2022). See Table A.12 (Appendix) for the characteristics and findings of these studies. Effect sizes were calculated on data from eight papers. Outcome measures used were self-reported happiness (Nguyen et al., 2018), positive emotions (David et al., 2019b), quality of life (Kato et al., 2008; Schakel et al., 2020; Staiano et al., 2018), and well-being (Kuosmanen et al., 2017; Sun et al., 2022; Walsh et al., 2019). Data from six papers were not included in the plot, because the paper used the same data as an already included paper (David et al., 2019a; David & Fodor, 2022), focused on effects of one intervention arm (David & Magurean, 2022; David, Magurean, et al., 2022) or data could not be obtained (C. Y. Hsieh & Chen, 2019; Ruiz-Ariza et al., 2018). Of the papers included in the forest plot, two compared an applied game to a casual game (Kato et al., 2008; Walsh et al., 2019), three compared an applied game to a passive condition (Kuosmanen et al., 2017; Schakel et al., 2020; Sun et al., 2022), and two compared casual game(s) to a passive condition (Nguyen et al., 2018; Staiano et al., 2018). David et al. (2019b) compared an applied game to both an active and a passive condition and therefore two effect sizes were calculated. Finally, Kato et al. (2008) used different measures on minors and adults to measure quality of life and is therefore included twice in the plot. The forest plot comprises ten effect sizes (see Figure A.14 in Appendix). The effect size for the study by David et al. (2019b), comparing an applied game to active condition, was 0.18 and nonsignificant. The three effect sizes comparing an applied game to casual game ranged from 0.17 to 0.57, with only Walsh et al. (2019) showing a significant effect size in favour of the applied game. The four effect sizes comparing an applied game to a passive condition ranged from 0.09 to 0.35 and were all nonsignificant. The two effect sizes comparing (a) casual game(s) to a passive condition were –0.13 and 0.18, both nonsignificant.
RkJQdWJsaXNoZXIy MTk4NDMw