RESUMO
The present study tested the psychometric properties of a new 20-item alcohol protective behavioral strategies (PBS) measure, the Protective Drinking Practices Scale (PDPS). Specifically, we evaluated measurement invariance of the PDPS in a sample of college students recruited nationally, tested the factor structure, evaluated construct validity, and explored potential short-form versions. Participants were 684 college student drinkers from throughout the United States (50.3% male; 65.6% White) obtained using Amazon's Mechanical Turk (MTurk) who completed measures of PBS use, hazardous drinking, alcohol-related negative consequences, alcohol use disorder (AUD) screening criteria, and depression. A differential item functioning analysis found that the PDPS items functioned similarly in the present sample and the original validation sample. Further, a unidimensional item response theory analysis confirmed the single PDPS factor structure identified in the original study. The PDPS demonstrated evidence of convergent validity with measures of hazardous drinking and alcohol-related negative consequences. Finally, five potential three-item PDPS short-form versions were generated, which all had strong correlations with the 20-item PDPS. Overall, these results suggest the PDPS functioned similarly in a United States college student sample with geographical diversity relative to a sample from a single Southeastern United States university, which has implications for future research and clinical work with college students. Future research should continue assessing the psychometric properties of the PDPS among diverse samples of college students and further elucidate an optimal PDPS short form for use in clinical screening and research contexts. (PsycInfo Database Record (c) 2021 APA, all rights reserved).
