Psychometric properties of an observation instrument to assess ADHD symptoms in the classroom using a continuous sampling approach

Objective: This study investigates the psychometric properties of the Dutch version of the Ghent University Classroom Coding Inventory (GUCCI-NL) by comparing children with (subthreshold) ADHD to typically developing peers. The advantage of the GUCCI is that it measures the proportion of time ADHD behavior was displayed and length of such intervals in a continuous manner that allows for more detailed assessment of behavior, such as on-task span, compared to traditional time sampling methods. Method: We coded the scales Attention Problems, Motor and Verbal Hyperactivity, and Oppositional Behavior in a sample of 57 children with (subthreshold) ADHD and 34 controls (aged 6–12). Construct validity (i.e., group differences, convergent and divergent validity), predictive validity (including ROC curves), and inter-observer reliability were investigated. Outcome measures were percentage of time and interval span. Results: The GUCCI-NL discriminated between the ADHD and control group for all scales (small to very large effects), with largest effects on percentage of time. Convergent validity against questionnaire and interview scores was confirmed for Verbal Hyperactivity and Oppositional Behavior, although divergent validity was weak. The Attention Problems scale was most sensitive to group membership (AUC = .79). Inter-observer reliability was sufficient (r >.82). Group differences on span length were found for Attention Problems but not for Hyperactivity, confirming the additional value of continuous coding. Conclusion: The GUCCI-NL is a valid objective measure for ADHD symptoms in the classroom, which may be of added value in scientific practice, although for diagnostic purposes additional instruments should be used.