Anne Fleur Kortekaas-Rijlaarsdam

CHAPTER 3 64 and quickly as possible in 1 minute. Reliability of the OMR is high ( r =.90) and OMR performance correlates with reading comprehension ( r =.55-.70) as well as with vocabulary ( r =.55-.76), indicating good validity (Brus & Voeten, 1999). Dependent variable was productivity, as accuracy is on average >95% (>60% of participants scored >95% correct, leaving no room for improvement), also for the most difficult words. PI-dictation (Geelhoed & Reitsma, 2004) was used to assess spelling performance. During this test, blocks of 15 sentences, increasing in difficulty, are dictated during 15 minutes. The child was instructed to write down specific words. Entrance-level was predetermined and the test was discontinued when a child made more than seven errors in one block. Reliability of the PI-dictation is high (KR-20 ≥.90) (Geelhoed & Reitsma, 2004). Dependent variable was the number of words spelled correctly multiplied by achieved level (1-9). Moderating and mediating variables: The mediating role of ADHD symptom improvement was measured with the Strength and Weakness of ADHD-symptoms and Normal Behavior Rating Scale (SWAN; (Hay et al., 2007; Swanson et al., 1994)). This widely used questionnaire contains two subscales: the Inattention scale and the Hyperactivity/Impulsivity scale, each comprising 9 items, and a Total scale. Items are scored on a 7-point Likert scale, with more negative scores indicating worse symptoms. The items are based on the DSM-IV symptoms of ADHD, but reflect both ends (strong and weak) of the behavior described in each symptom. Reliability of the SWAN is high (Cronbach’s alpha .95) and adequate validity is indicated by moderate correlations ( r =.54) with other inattention/hyperactivity subscales and low correlation ( r =.11) with subscales measuring emotion (Lakes, Swanson, Riggs, 2012). Both the parent- and teacher versions of the SWAN were used. Symptom improvement was measured with the Total scores of the SWAN, depicting the full range of ADHD symptoms. To identify possible clinical subgroups profiting more or less from methylphenidate, we tested the influence of several moderators: ADHD severity (number of symptoms) was determined based on DSM-IV guidelines (DISC-P interview). Validity of the DISC-P is adequate, as indicated by agreement between DISC-P and clinician ratings ( κ =.72). Further, symptom and criterion scales of the DISC-P show higher reliability than the categorical diagnosis of ADHD (Schaffer, Fisher, Lucas, Dulcan, & Schab-Stone, 2000). Learning abilities were defined as deficit scores based on the academic performance measures described above: For all academic measures, z-scores were calculated based on the mean and standard deviation of the TD group, calculated separately for each grade level. These z-scores were based on the raw scores (total correct) on the relevant academic task, thereby combining productivity and accuracy.