In predicting children’s scores, to take into account within-group correlations, generalized estimating equations (GEE) were used to model the effects of risk while handling familial correlations between subjects resulting from the inclusion of siblings in the sample. Ancestral proportion scores were entered as a covariate in all models and retained regardless of significance given its relevance in interpreting Study results (34). Effects of age and sex were also examined, but these variables were not significant as expected, because the outcome measures are normed by age and sex. Before conducting GEE, distribution of outcome measures (TRF scores) was examined for normality using the Shapiro-Wilks test. Because the measures were non-normally distributed and standard transformations failed to correct distributions, rank transformed scores were used in subsequent analyses. Main effects of MAOA genotype (e.g., “low activity” vs. “high activity”) and the continuous trauma exposure measure (TTES), as well as the interaction of these two factors, were explored, and Bonferroni tests were performed to control for multiple comparisons in examining outcomes on the three primary measures. Analyses conducted using the categorical classification of subjects
