The cross-lagged model including the full sample (Figure 1) showed acceptable fit [χ2(df)= 785.05(437); p<.001; χ2/df ratio = 1.79; CFI= .92; RMSEA=.06]. An inspection of residuals showed no points of ill fit. There were significant cross-lagged relations between neuropsychological functioning and ADHD severity between F1 and F2, and between F2 and F3. However, BL ADHD severity and neuropsychological functioning were not significantly associated with F1 neuropsychological functioning and ADHD severity, respectively. Therefore, we trimmed the model to exclude BL ADHD severity and BL NEPSY scores. This model [χ2 (df)= 356.93(241); p<.001; χ2/df ratio = 1.48; CFI= .96; RMSEA=.05] showed significant cross-paths between ADHD symptom severity and neuropsychological functioning, and vice versa at each time-point.
