Citation

Kahler CW, Strong DR, Hayaki J, Ramsey SE, Brown RA. J. Stud. Alcohol 2003; 64(1): 127-136.

Affiliation

Center for Alcohol and Addiction Studies, Brown University, Box G-BH, Providence, Rhode Island 02912, USA. Christopher_Kahler@brown.edu

Copyright

(Copyright © 2003, Rutgers Center of Alcohol Studies)

DOI

unavailable

PMID

12608493

Abstract

OBJECTIVE: In this study we use methods based on Item Response Theory to examine in depth the psychometric properties of the Alcohol Dependence Scale (ADS). In particular, we examine the ability of each ADS item to discriminate among individuals across the continuum of alcohol dependence severity and also examine the extent to which item-response options provide useful and reliable information about the level of alcohol dependence. METHOD: Participants were 166 alcohol-dependent patients with elevated depressive symptoms. We conducted a maximum likelihood common factors analysis on the ADS, and then used a nonparametric kernel smoothing method to create Item Characteristic Curves (ICC) and Option Characteristic Curves (OCC) for each ADS item. On the basis of these curves, we identified items showing at least fair discrimination and modified the scoring of response options where indicated. We then created an empirically derived ADS score and correlated it with the original ADS and with other measures of alcohol involvement. RESULTS: Replicating previous studies, our results indicated a primarily unidimensional factor structure. A total of 12 of the 25 ADS items showed good discrimination, and examination of the OCC indicated that dichotomous scoring was most appropriate for these items. This 12-item abbreviation of the ADS was highly correlated with the original scale (r = 0.91), and showed similar patterns of correlations with other measures of alcohol involvement. CONCLUSIONS: Results suggest potential gains in measurement efficiency using methods based on Item Response Theory and indicate potential ordering of dependence symptoms based on item severities.

Language: en