[Data quality in surveys on alcohol consumption among university students]

Different survey modalities have been developed to assess alcohol consumption and related problems. Research that compares data quality between survey modalities is scarce in Latin America. The aim of this study was to assess data quality in three survey modalities on alcohol consumption: self-administered online, self-administered hard-copy, and face-to-face interviews. Data were obtained from three probabilistic samples of students (n = 60 each) from the National University of Mar del Plata, Argentina, using the same questionnaire. Data quality was measured for each modality by overall response rate, item response rate, and accuracy. Data accuracy was evaluated as the percentage of self-reported binge drinking, positive results on AUDIT, and internal consistency of AUDIT for each modality. The overall and item response rates were lower in the online modality and similar between the other two. No differences were found between modalities in the accuracy of responses.