We hypothesized that higher alc-GPS would be associated with higher levels of alcohol use. Further, we hypothesized that the effect of alc-GPS would, at least in part, manifest indirectly by influencing sensation seeking and social support. We also hypothesized that social support would buffer risks associated with alcohol use, such that the association between alc-GPS and alcohol use, as well as the association between sensation seeking and alcohol use, would be attenuated by high levels of social support. We considered social support from family and friends separately in order to examine whether patterns of associations vary as a function of the source of social support given prior evidence of differential effects (Salvatore et al., 2015). In addition, we tested sex as a moderator of the hypothesized pathways given evidence of sex differences in rates of alcohol use and levels of sensation seeking and social support (Cross et al., 2013; Grant et al., 2015). Since sensation seeking and sources of social support tend to change across the adulthood years, we capitalized on the rich dataset from the Collaborative Study on the
