Exploratory and confirmatory factor analysis was carried out in Mplus (Muthen and Muthen 2007). We utilized three fit-indices which reflect the model's balance of explanatory power and parsimony: the Tucker-Lewis Index (TLI) (Tucker and Lewis 1973), the Comparative Fit Index (CFI) (Bentler 1990), and the root mean square error of approximation (RMSEA) (Steiger 1990). For the TLI and CFI, values between 0.90 and 0.95 are considered acceptable and ≥ 0.95 as good. For the RMSEA, good and acceptable models typically have values ≤ 0.05 and 0.10, respectively. The primary association analysis was a linear regression of the alcohol factor score on the number of minor alleles at each SNP (thus, a 1df additive model). Principal components corresponding to population (Shi et al, 2009) were included as covariates in the analysis, to minimize bias due to population stratification. Age and sex were also included as covariates. The effect of possible non-normality of the trait value on the Type I error rate of the linear regression analysis was tested by generating 10,000,000 random sets of genotypes for each of 3 minor allele
