CONFIRMATORY ANALYSES OF 3 SCORING SYSTEMS FOR THE 90-ITEM ALCOHOL EXPECTANCY QUESTIONNAIRE

Objective: The Alcohol Expectancy Questionnaire (AEQ) is a widely used instrument in the field of alcohol research, and several different ve rsions and scoring systems for the AEQ are in current use. Two previou s confirmatory factor analytic studies failed to provide support for t wo of the present scoring systems for the AEQ, but these studies' resu lts may have ken an artifact due to the methods employed. In this meth odological study, three factor structures for the AEQ are tested, base d on three different models: (I)the original multidimensional model, ( 2) a unidimensional model, and (3) a second-order or hierarchical mode l. Method: Data from a previous study provided an effective N of 507 c ollege men and women, who completed a self-administered version of the AEP. Results: Nearly all of the models tested clearly did not fit the data. One model, a three-scale scoring system, may be cautiously view ed as having demonstrated a marginal fit. Conclusions: Within the cont ext of the present study's limitations, the validity of current scorin g systems for the AEQ is questionable.