We began with preliminary analyses to examine the amount and patterns of missingness of our data, in order to determine the appropriate approach to handle missing data in subsequent analysis, followed by examination of descriptive statistics and correlations among study variables. We then examined the associations between parental ADS, parenting behaviors, and adolescent risky drinking and conduct problems by conducting structural equation models (SEM) using Mplus version 7.31 (Muthén & Muthén, 1998–2012). We conducted a multivariate SEM model to consider adolescents’ risky drinking and conduct problems simultaneously given that they are related constructs (see Figure 1). Fathers’ positive parenting was modeled as a latent variable indicated by paternal involvement, father-adolescent communication, and father-adolescent closeness. Mothers’ positive parenting was modeled as a latent variable indicated by maternal involvement, mother-adolescent communication, and mother-adolescent closeness. To account for method effects, and to take into account the shared variances between fathers’ and mothers’ parenting (so that unique effects of fathers’ and mothers’ parenting can be examined), errors terms were correlated between fathers’ and mothers’ parenting subscales, as were the residual errors for mothers’ and
