Abstract
This analysis of the National Household Survey on Drug Abuse (NHSDA) focuses on correlates of substance use among Hispanic groups. Using 1994-B and 1996 NHSDA surveys (U.S. Department of Health and Human Services [USDHHS], 1996, 1998), these analyses use sampling areas to control for intracluster correlation and its subsequent effect on empirical conclusions. Models are specified for alcohol, cigarettes, marijuana, cocaine, and any illicit drug, controlling for national origin, gender, age, socioeconomic status, language, and family structure. Results indicate that models failing to account for intracluster correlation may lead to erroneous ethnic contrasts of substance use. Also, substance use varies widely by national origin—and Hispanic men and women have highly divergent patterns of use, more so than non-Hispanic White men and women. Finally, controlling for neighborhood does not wholly diminish ethnic contrasts in substance use.
Get full access to this article
View all access options for this article.
