A multigroup confirmatory factor analysis of adolescent alcohol-related behaviors across ethnic groups

Includes bibliographical references (pages 17-19).Alcohol-related behaviors are often examined using national and state surveys assessing participants' self report of attitudes and actions. However, few studies have examined the psychometric properties and the factorial invariance of various alcohol-related behaviors across ethnic group. Past research has revealed ethnic differences in scale construction and evaluation. Also, surveys that evaluate alcohol behaviors have been found to lack invariance across different ethnic groups. These dissimilarities may be due to deep rooted differences in ethnic classification of unhealthy substance use behaviors. Because the conclusions made from state and national surveys can affect various prevention, policy, and intervention programs, valid conclusions should be made concerning all ethnic groups. The present study examined the factor structure of 3 alcohol-related items: "Have you ever, even once, had a drink of any type of alcoholic beverage?", "During the past 30 days, on how many days did you drink one or more drinks of an alcoholic beverage?", and "On the days that you drank, how many drinks did you usually have?" administered as part of the 2011 National Survey on Drug Use and Health (NSDUH). Factorial invariance was examined across 12 to 17 year old Caucasian Americans (N = 11,235; 62.0%), African Americans (N = 2,640; 14.6%), Asian Americans (N = 598; 3.3%), and Latin Americans (N = 3535; 19.6%). A multigroup confirmatory factor analysis was conducted to statistically determine whether the factor structure was invariant across all four ethnic groups. All of the baseline models were good fits to the NSDUH data (all factor loadings were significant: p < .001). However, only the Asian American and Latin American metric invariance model was a good fit to the data (p > .05). The analyses revealed the alcohol scale administered as part of the NSDUH to lack invariance across all ethnic groups. The findings from this study accentuate the significance of evaluating psychometric properties of scales assessing alcohol-related behaviors generalized across ethnic groups