Self-reported drinking and alcohol-related problems among early adolescents: dimensionality and validity over 24 months.

OBJECTIVE: Researchers rely on adolescents' self-reports of alcohol consumption and alcohol-related problems, despite little evidence concerning their validity. We assessed the reliability and validity of adolescents' self-reports, employing collateral validation and focusing on the understudied transitional years of early adolescence. METHOD: Subjects were 214 boys and 247 girls who participated in school-wide surveys that assessed drinking, drunkenness and alcohol-related problems each year for 3 years. These measures were validated by collateral (peer) reports and by separate, 7-day drinking calenders. Internal consistency and test-retest reliability were also assessed. RESULTS: Results replicated findings with older adolescents that drinking/drunkenness and alcohol-related problems fall on two partially overlapping dimensions. Scales assessing each dimension had moderate to high internal consistency and high test-retest stability. Correlations with collateral reports were relatively strong for the drinking/drunkenness scale, moderate for a dichotomous variable reflecting the presence or absence of alcohol-related problems, and more modest for the alcohol-related problems scale. Correlations with diary reports of drinking behavior were strong for drinking/drunkenness. Results generally replicated across gender and over time. CONCLUSIONS: Researchers can have some confidence in the reliability and validity of early adolescents' survey self-reports, particularly of alcohol consumption (alcohol-related problems occurred with low base rates, perhaps limiting validity coefficients). Because drinking/drunkenness and alcohol-related problems shared 30% of their variance, factors other than consumption (e.g., personality factors) apparently influenced the experience of alcohol-related problems.