The goal of the proposed three-year investigation is to identify determinants of alcohol use in predominantly Latino and Euro American youth. The objectives will involve testing the causal model of alcohol use initiation and escalation. This model specifies the operation of key proximal factors in conjunction with theory-based supported general and underlying more distal factors. Guided by the theory of triadic influence (TTI), proximal factors will include alcohol-related intentions, self-efficacy, attitudes, and social noms. Also guided by TTI, distal factors include global expectancies, depression, parental relatedness, parental monitoring, and academic orientation. Finally, potentially vital sociocultural influences will be conceptualized as ultimate factors in the model (most distal to alcohol use within TTI). Global expectancies and cultural orientation in Latinos are two especially promising factors that have been understudied within comprehensive theory-guided models of youths' alcohol involvement. Both mediational and moderational effects will be tested, with special emphasis on the latter for the role of socio-cultural influences.
The aims will be accomplished through secondary data analyses of cohorts from two studies-the Middle School Healthy Kids and Tobacco Survey (N = 1622) and the Add Health Study (N = 12,118); both data sets have Latinos and/or Euro American as the largest groups. The former is led by the Principal Investigator, with data collection (coordinated by ETR Associates in Santa Cruz, CA) to be completed by December 1, 2000; the latter is a national study. Each data set from these studies afford strong tests of the model, including tests of the prediction of changes in alcohol involvement over time. Collectively these data present a unique opportunity to test the model in multiple Latino sub-groups, and include reports of alcohol by the youths and others they know. Structural equation models will be of primary emphasis in the analyses, though other longitudinal models will also be employed depending on the outcome of relevance. Research suggests that programs addressing a range of both proximal and distal determinants may result in stronger, more sustained reductions in adolescent alcohol use over time. However, there is currently insufficient knowledge about the key determinants, especially in high-risk Latino and Euro American youth. By clarifying how causal paths leading to alcohol onset and escalation may be impacted by sociocultural influences, the proposed project will also aid in the development of more potent youth intervention programs for diverse youth.
