Modeling Marijuana Involvement and Linguistic Acculturation among Hispanic Subgroups using Generalized Linear Mixed Models.
Shujin Zhu
Master's Candidate, Department of Statistics, Colorado State University

Friday, June 22, 2007
2:00 p.m.
006 Statistics


The Hispanic population is one of the largest ethnic groups in the U.S. There is evidence that lifetime drug use rates for Hispanic Americans are increasing, and that marijuana use and related health consequences pose a significant public health problem for Hispanics. However, research on substance use and acculturation in Hispanic population tends to look at Hispanics as a monolithic ethnicity. A better understanding of the homogeneity and heterogeneity of marijuana use patterns by Hispanic subgroups may assist in the development of more effective prevention programs.

To fill the research gap, this project examined the associations between acculturation and marijuana opportunity and between acculturation and marijuana use given opportunity, respectively, among the Hispanic subgroups ( e.g. Puerto Rican, Mexican, Cuban, and Others) , using data from National Household Survey on Drug Abuse 1991-1994. A series of generalized linear mixed models was applied to model these associations and their dependence on age, gender, and Hispanic population proportion in the community, accommodating a random effect for geographic area.

A number of important similarities and differences were identified in marijuana involvement patterns among the four Hispanic subgroups. About 45% of the participants had opportunity to try marijuana, and about 58% of the participants who had opportunity have used marijuana. Acculturation, indicated by language preference, Hispanic subgroups and gender were found to be significantly associated with marijuana involvement, controlling for other covariates such as age at assessment, social economic status, and Hispanic population proportion. The odds of marijuana opportunity and use given opportunity among Hispanics with high acculturation are nearly three times higher than those with low acculturation. Compared to females, males have two times higher odds of having marijuana opportunity but about the same level of odds using marijuana.

Though the general trends for marijuana involvements are consistent among the four Hispanic subgroups, three-way interactions among acculturation, Hispanic subgroup and gender reveal different effects of acculturation and gender on the odds of marijuana involvement across Hispanic subgroups. Compared to the low-acculturated Other females, the odds ratios (OR) of marijuana involvement in each group are estimated to indicate the differences. Among Puerto Rican males, high acculturation (OR=7.9) and low acculturation (OR=7.6) have almost the same effects on marijuana opportunity; while among Mexican females, high acculturation (OR=4.4) and low acculturation (OR=0.7) have very different effects on marijuana opportunity. After having an opportunity to try marijuana, the odds ratios of use marijuana among low acculturated participants are small (OR<2.0) in all Hispanic subgroups except for Puerto Rican (OR=3.1 for females and OR=5.1 for males). The effects of acculturation are much larger among Mexicans (male: OR=7.1 for high acculturation vs. OR=1.7 for low acculturation; female: OR=6.2 vs. OR=1.4) and Other females (OR=5.0 vs. OR=1.0) than that among other subgroups.

Our findings may have important implications for the design of prevention and intervention strategies for the marijuana use among Hispanic subgroups.




  home for studentspeoplecoursesdistance degreeresearchconsulting seminars resources
PRIMESSTARMAPemail us equal opportunitydisclaimerweb master • last modified: August 11 2003