Analyses were conducted using Stata (Stata Statistical Software, 2017) to accommodate sampling and non-response weights adjusting for survey design. Preliminary analyses found FH was not significantly associated with the area-level social control variables.8 Associations of alcohol outcomes with FH and area-level social control were assessed using adjusted logistic regression models that included all demographic controls and indicators for survey year (using 2000 as referent). Because the national samples were selected through RDD methods, multilevel modeling was not necessary. The multiplicative interactions were tested simultaneously, as correlations between the two area-level social control variables were modest.9 Interactive effects were assessed using multiple-degrees-of-freedom contrasts of the joint effects of the interaction (adjusted Wald tests), as well as tests of marginal effects of FH across different levels of the area-level variables. Due to generally limited power for tests of interactions (Frazier, Tix, & Barron, 2004), those interactions that did not reach p < 0.10 were removed to improve interpretability of the main effects (Aiken, West, & G., 1991). Preliminary analyses (see Supplemental Tables S1-S2) showed interactions of neighborhood alcohol availability with sex in
