46 Chapter 2 As shown in Figure A.4 (Appendix), the six effect sizes comparing an applied game to an active condition ranged from –0.15 to 1.34. Notably, Steiner et al. (2011) reported a significant effect favouring the applied game, while Rodrigo-Yanguas et al. (2023) indicated a significant effect favouring the active condition. Both effect sizes comparing an applied game to a casual game were nonsignificant (–0.01 and 0.12, respectively). The four studies comparing an applied game to a passive condition showed effect sizes of –0.30 to 0.86. Van der Oord et al. (2014) and Qian et al. (2018) reported significant effects favouring the applied game. The four effect sizes comparing an active condition to (a) casual game(s) ranged from –0.04 to 0.53, with only the study of Tullo et al. (2018) showing a significant effect in favour of the active condition. Given that the majority of papers also measured ADHD symptoms, this was taken as the second outcome of interest. Effect sizes for ADHD symptoms were calculated on data from 12 papers. Data from a questionnaire comprising all ADHD symptoms (i.e., measuring attention, hyperactivity, impulsivity) were preferred, but if not available, a subscale measuring both hyperactivity and impulsivity was used (Dovis et al., 2015; Shalev et al., 2007; Van der Oord et al., 2014). In addition, parent-reported symptoms were used as much as possible as literature shows this is most reliable (Achenbach et al., 1987; Riley, 2004). Only Weerdmeester et al. (2016) measured ADHD symptoms through teacherreports. Ten papers were not included in the plot, as they did not measure ADHD symptoms (Barkın et al., 2023; Bul et al., 2018; Bul et al., 2016; Ji et al., 2023; Prins et al., 2011; Qian et al., 2018; Tullo et al., 2018), were part of a study that was already included (van Houdt et al., 2019), or data could not be obtained (García-Baos et al., 2019; Kollins et al., 2020). Eight papers included two intervention arms, comparing the applied game to an active condition (Bikic et al., 2018; Smith et al., 2020), to a casual game (Medina et al., 2021; Weerdmeester et al., 2016), or to a passive condition (Van der Oord et al., 2014), comparing (a) casual game(s) to an active condition (Bikic et al., 2017; Shalev et al., 2007) or passive condition (Benzing & Schmidt, 2019). Four papers included three intervention arms. For Rodrigo-Yanguas et al. (2023), Steiner et al. (2011), and Van Houdt et al. (2021) two effect sizes were calculated and for Dovis et al. (2015) one effect size (see the reasoning described in the previous paragraph). Figure A.5 (Appendix) shows that the six effect sizes comparing an applied game to an active condition ranged from –0.58 to 0.29, with no studies showing a significant effect size. The two effect sizes comparing an applied game to a casual game were both nonsignificant (–0.27 and 0.23, respectively). The four effect sizes comparing an applied game to a passive condition were all nonsignificant and ranged from 0.29 to 0.58. The two effect sizes comparing
RkJQdWJsaXNoZXIy MTk4NDMw