Evaluating the psychometric properties of the AUDIT-C among college students

Abstract

Aims: Despite the various settings and diverse populations in which the utility of the AUDIT-C has been examined, college students have remained conspicuously absent from the vast majority of these studies. We evaluated the psychometric properties of the AUDIT-C with a sample of college student bar patrons.Methods: Intercept interviews were used to recruit currently enrolled college students (n = 347) exiting bars. A brief face-to-face interview assessed demographic data, alcohol-related behaviors (AUDIT-C), and breath alcohol concentration (BrAC). AUDIT-C responses were subjected to a principal component analysis. BrAC was used as a measure of concurrent validity against two distinct gender-based AUDIT-C cut-off scores.Results: All AUDIT-C items loaded on a single factor, accounting for 68% of total variance. Pearson’s product–moment correlation coefficients between BrAC and both hazardous drinking AUDIT-C cut-offs (r = 0.32; r = 0.38) were statistically significant (p < 0.001). Moreover, those meeting each hazardous drinking gender-based criteria exhibited significantly higher (p < 0.001) BrAC readings than those not meeting these criteria. Effect sizes for these group differences were considerable (Cohen’s d > 0.70). Reliability and stability statistics indicated strong internal consistency.Conclusions: Results reported herein highlight the sound psychometric qualities of the AUDIT-C and its ability to produce valid and reliable scores among college students.