SNPs within CHRM2, all the more striking. In addition, it is important to note that the rather modest association that we observe is not at odds with extant evidence supporting a more robust association between antisocial peers and externalizing behavior, as those findings generally derive from simple linear regressions with continuous measures of both the dependent and independent variables, whereas our results derive from a multinomial logistic regression assessing the predictive ability of peer group antisocial behavior with respect to probability of membership within discrete externalizing trajectories.
