Question 1	The intraclass correlations (ICCs) indicate the proportion of total variance that is due to between-coalition variance. Between 7 and 8 % of the total variance in goal directedness and opportunities for participation was attributable to coalition-level differences. Twelve percent of the total variance in efficiency and 17 % of the total variance in cohesion was attributable to differences among coalitions. This suggests that CTC coalitions varied in the levels of functioning and that the greatest variance among coalitions was in cohesion.
Question 2	Fixed effect meta-regressions showed that none of the four dimensions were significantly associated with the effect size differences in community leader reports of the adoption of science-based prevention (see Table 2). Each model generated a significant QE statistic, indicating that residual variance in adoption between matched pairs was left to be explained. Because no dimensions of coalition functioning directly explained the variance in the relationship between CTC implementation and the adoption of science-based prevention, no mixed effect models were specified. Mixed effect models tend to produce regression coefficients with larger confidence intervals and are thus insufficiently powered to detect effects that were not detected in fixed effect models (Berkey et al. 1995).
Question 3	Results of the multilevel models indicated strong relationships between coalition member-reported functioning and coalition member-reported capacities. As shown in Table 3, significant (p<.001) positive relationships were found between all of the dimensions of coalition functioning and the two capacities under study, suggesting that better functioning coalitions also had greater capacity in the form of acquired member skills and linkages with diverse community sectors. These particular capacities are associated with reaching higher stages of adoption of science-based prevention (Shapiro et al. 2015), which has in turn been shown to mediate the effect of CTC on the behavioral health of young people (Brown et al. 2014a).