COA RISK IN YOUNG ADULTS--A LONGITUDINAL SIBLING DESIGN

点击翻译按钮获取中文摘要

Secondary data analyses of the National Longitudinal Survey of Youth (NLSY) are proposed to address several research questions related to the prevalence, correlates, and consequences of alcohol use and abuse among a nationally representative sample of young adult COAs. The NLSY is a multiwave dataset that includes approximately 12,686 subjects, including 5,863 sibling pairs. The retention rate across seven annual waves of data collection has been 95%. The large number of sibling pairs in the NLSY is of importance for addressing substantive issues and for determining interrater agreement regarding the validity reports of familial alcoholism. The five principal objectives of the proposal are: (1) to conduct family resemblance analysis to examine the prevalence of alcoholism for three racial/ethnic groups--whites, blacks, and Hispanics and two gender groups; (2) to use alternative categorical schemes (e.g., FHP/FHN; multigenerational FHP, unigenerational FHP, FHN) for family history of alcoholism and to compare findings of these categorization schemes for group differences in alcohol use/abuse and other young adult problem behaviors (e.g. , delinquent activity, marital and occupational stability; (3) to assess sibling similarity with regard to alcohol use and alcohol problems and to use regression-based statistical models to assess older sibling influences on younger sibs drinking behavior; (4) to identify variables that discriminate those sibs who eventually develop drinking problems from their matched sib who has yet to have drinking problems; and (5) to examine the extent of assortative mating for daughters of male alcoholics and to evaluate the drinking problems of those female sibs who married an alcoholic or problem drinker, and those that did not. Assortative mating for sons of male alcoholics and differences in drinking problems among male COA sibs who did and who did not marry alcoholic or problem drinker spouses will also be investigated for exploratory purposes. Data analyses will include intraclass correlations, hierarchical multiple regression models, discriminant function analysis, and covariance structure models.