Test-retest reliability of alcohol measures: Is there a difference between Internet-based assessment and traditional methods?

• DOI: 10.1037/0893-164x.16.1.56
• Publication Date: 2005-10-07T19:17:19Z
• Authors: Elizabeth T. Miller; Dan J. Neal; Lisa J. Roberts; John S. Baer; Sally O. Cressler; Jane Metrik; G. Alan Marlatt
• Keywords: Adult; Male; Analysis of Variance; Internet; Psychological Tests; Adolescent; Alcohol Drinking; Reproducibility of Results; United States; Alcoholism; 03 medical and health sciences; 0302 clinical medicine; Humans; Mass Screening; Female

This study compared Web-based assessment techniques with traditional paper-based methods of commonly used measures of alcohol use. Test-retest reliabilities were obtained, and tests of validity were conducted. A total of 255 participants were randomly assigned to 1 of 3 conditions: paper-based (P&P), Web-based (Web), or Web-based with interruption (Web-I). Follow-up assessments 1 week later indicated reliabilities ranging from .59 to .93 within all measures and across all assessment methods. Significantly high test-retest reliability coefficients support the use of these measures for research and clinical applications. Furthermore, no significant differences were found between assessment techniques, suggesting that Web-based methods are a suitable alternative to more traditional methods. This cost-efficient alternative has the advantage of minimizing data collection and entry errors while increasing survey accessibility.