Anne Fleur Kortekaas-Rijlaarsdam

85 MPH EFFECTS ON MATH PERFORMANCE: INFLUENCE OF COGNITION, MOTIVATION AND PERCEIVED COMPETENCE 4 (SPPC: Veerman, Straathof, Treffers, van den Bergh, & ten Brink, 2004). Six statements were (self-)rated by children, their parents and teachers on a 4-point Likert scale. For all questionnaires, raw scale cores were used as dependent variables. Analyses Participation of 63 children was sufficient to guarantee .80 power (α=.05) to detect MPH effects (Twisk, 2013). Performance of the ADHD group was compared to the TD group using independent sample t-tests or, in case of non-normality, Wilcoxon Signed- Rank Tests, using SPSS version 21 (IBM, 2012). Outliers were rescaled to the nearest observation plus one unit. MPH effects on cognition, motivation and competence were estimated using generalized linear mixed model (GLMM) analysis with a fixed effect for treatment and a random intercept to control for within subject dependency of measurements. For cognitive, motivational and competence measures that improved significantly with MPH, the possible mediating effect of this variable on academic improvement with MPH was estimated. Elsewhere, we demonstrated positive effects of MPH on mathematical productivity and accuracy (for details see Kortekaas-Rijlaarsdam et al., 2017b), thus variables were selected as possible mediators if (1) there was a significant effect of MPH on the mediator, and (2) there was a significant effect of the mediator on math performance when controlled for treatment condition. One case was missing for the CAIMI measuring intrinsic motivation, 3 cases were missing for parent-rated academic competence and 5 cases were missing for teacher-rated academic competence. Using GLMM with a random intercept, we calculated the regression coefficients of the effects of MPH on the mediator (β a ) and of the mediator on math performance (β b ) when fixed effects of both MPH and the mediator were entered in the model. To quantify the significance of the mediator, the product term β a β b of the regression coefficients with accompanying confidence interval was calculated using the asymptotic normal distribution method (Krull & MacKinnon, 2001; Tofighi & Mackinnon, 2011), thereby assuming a covariance of zero between β a and β b (Tofighi et al., 2009). If the 95% CI did not include zero, the effect of the mediator was significant. Procedure The study was approved by the local ethics committee. All parents as well as children above the age of 11 provided written informed consent prior to enrollment in the study. For children with ADHD, after the screening procedure confirming the ADHD diagnosis and establishing sufficient intelligence (IQ>70), testing took place on the last day of each treatment week and started 60 to 90 minutes after medication intake.