For these analyses, the alcohol outcome (ALCOUT) was determined as a latent variable from the age13/14 questions extracted from the Semi-Structured Assessment for the Genetics of Alcoholism (SSAGA) instrument developed by the Collaborative Study on the Genetics of Alcoholism investigation [37, 38]. The alcohol questions have both retest reliabilities and validities compared to another standardized instrument >.80 [38]. The three indicators used here included the maximum number of drinks consumed in the prior six months, the usual number of drinks per week over the prior six months, and the number of 12 possible alcohol problems in the last 12 months. The quantity-based outcomes were chosen as the most direct measure of the hypothesized higher levels of alcohol intake related to the low effect per drink [2, 6], while alcohol problems have been projected as an expected consequence of heavier drinking at even a young age. The problems queried here included the need for more and more alcohol to get effects (tolerance), exceeding set alcohol limits, feeling the need to stop or cut down on drinking, skipping school to drink, going
