Anne Fleur Kortekaas-Rijlaarsdam

67 MPH EFFECTS ON MATH PERFORMANCE: INFLUENCE OF BEHAVIOR 3 group, as well as the group comparisons. GLMM analyses showed that methylphenidate treatment resulted in large-sized reductions in ADHD symptoms (SWAN) according to parents ( t (62) = 5.90, β = 14.11, p = <.001, d = .96) and medium-sized reductions according to teachers ( t (61) = 4.74, β = 10.58, p = <.001, d = .62), confirming that methylphenidate acts as a powerful agent to reduce ADHD symptoms. For the academic performance measures, GLMM analysis showed significant effects of methylphenidate with small to medium-sized effects on mathematical productivity on the SAT, ( t (62) = 3.88, β = 1.95, p = <.001, d = .19), and accuracy on mathematical word problems ( t (58.6) = 3.29, β = 10.07, p = .002, d = .31). On average, children with ADHD were 2% more productive on the SAT and 9% more accurate on mathematical word problems with medication compared to placebo. There were no significant effects of methylphenidate on SAT accuracy, mathematical word problems productivity, technical reading, or spelling (all p between .171 and .997). Comparison of Academic Performance of ADHD and TD group Table 3.2 displays mean accuracy and productivity on the academic measures for the ADHD group and TD group with accompanying test statistics and effect sizes. On average, on all academic tasks, academic accuracy and productivity of TD children was higher than accuracy and productivity of children with ADHD taking placebo. However, only for mathematical accuracy and productivity, these differences were significant. Children with ADHD were less productive and less accurate on the SAT than TD children with small to medium effects ( t (122.2) = 3.80, p = <.001, d = .67 and t (128) = 2.02, p = .045, d = .36, respectively). Furthermore, children with ADHD were less accurate on mathematical word problems than TD children ( t (121) = 2.19, p =.030, d = .42). Because our ADHD and TD group significantly differed on SES, we repeated the analyses with SES as a covariate using ANCOVA. After adjusting for SES, children with ADHD were still less productive on the SAT ( F (1,124) = 12.20, p = .001, d = .62). However, the differences between our ADHD and TD groups in math accuracy just escaped conventional levels of significance: F (1,124) = 3.24, p = .074, d = .32 and F (1,119) = 3.34, p = .07, d = .33 for SAT accuracy and mathematical word problem accuracy, respectively. Although productivity on the SAT improved with medication in children with ADHD, performance did not normalize, as productivity was still significantly worse in children with ADHD taking methylphenidate than in TD children ( t (117.9) = 3.0, p = .003, d = .52). Calculating z-scores based on performance (items completed correctly) of the TD group, children with ADHD on placebo lagged behind by .59 standard deviation