Univariate indices of normality revealed no significant skewness or kurtosis except for scores on the CPC for TD children (skewness and kurtosis ranged from 2.13–10.30). This variable was log transformed for subsequent analysis. Given the significant differences between groups on SES, this variable was included as a covariate in all analysis. Although IQ differed between the TD and H/I children, this was not included in the primary analyses because cognitive deficits may be a part of the profile of a neurodevelopmental disorder such as ADHD (Frazier et al., 2004) and adding it as a covariate may produce overcorrected and inconsistent findings (Dennis et al., 2009). After examining correlations we conducted structural equation modeling (SEM) fitting latent variables of the NEPSY (five domain scores) with latent variables of ADHD severity (parent report on K-SADS and teacher reports on BASC-2 and CPC) in a cross-lagged framework using the program AMOS 18. A non-significant chi square, Comparative Fit Index (CFI) >.95, and Root Mean Square Error of Approximation (RMSEA) <.08 were considered indices of good fit (Hu & Bentler, 1999; Kline, 2005). As
