Psychometric Properties of the AUDIT-C within an Amazon Mechanical Turk Sample.

Objectives: Amazon's Mechanical Turk (MTurk) has become a popular data collection tool in the addiction sciences. We sought to examine the psychometric properties of the AUDIT-C in an MTurk sample. Methods: Data collection was facilitated via MTurk (N=309; 52.8% female), where an online survey assessed demographic data, alcohol use behaviors (AUDIT-C), and alcohol-related consequences (CAPS-r). Responses to the AUDIT-C were subjected to a principal component analysis to evaluate the structure of the 3-item measure. Alcohol-related consequences were used as a measure of convergent validity. Results: Results provided evidence for a single-factor structure. Pearson's product-moment correlation coefficients between AUDIT-C scores and CAPS-r scores produced statistically significant results (r = 0.51, p < .001). Using biological sex-based suggested cut-off scores for the AUDIT-C, hazardous drinkers (M = 19.15, SD = 8.27) demonstrated statistically significantly higher levels of alcohol-related consequences than non-hazardous drinkers (M = 12.56, SD = 5.35; t(295) = -8.34, p < .001). Reliability and stability statistics demonstrated strong internal consistency. Conclusions: Results demonstrate the sound psychometric properties of the AUDIT-C for an MTurk sample and provide evidence supporting the use of AUDIT-C as a screening tool to be employed with digitally accessed populations to identify and reach hazardous drinkers.