Journal of Studies on Alcohol · 1995 · 130 citations · 34 references
Social SciencesBehavioral SciencesPsychologyDifferential Item FunctioningPsychometricsSubstance Use DisordersAddictionConfigural InvarianceFactor AnalysisItem AnalysisPsychological TestingTrait StructureCross-group Factor ComparisonsGoodness-of-fit IndicesInvariant Factor StructureMulti-group Measurement ModelsStructured Interview Assessment
OBJECTIVE: The alcohol expectancy construct has become prominent in contemporary psychosocial approaches to understanding alcohol use and abuse. In 1980 Brown and colleagues developed the 90-item Alcohol Expectancy Questionnaire (AEQ) to assess experiences. Rohsenow modified this instrument to create the 40-item Alcohol Effects Questionnaire (AEQ-2). In the present study, we replaced the dichotomous response format of the AEQ-2 with a six-point scale to create the AEQ-3. The purpose of the study was to evaluate the AEQ-3 for factor structure confirmation and invariance across gender and race subgroups. METHOD: We administered the AEQ-3 to a large general population sample (N = 1,260). The instrument was self-administered during a structured interview. RESULTS: Confirmatory factor analyses were conducted to test a correlated eight-factor model consisting of six positive expectancies and two negative expectancies. Fit indices revealed that the eight-factor model fit the data moderately well. Furthermore, the fit of the eight-factor model was largely invariant across race and gender subgroups. Nonetheless, factor intercorrelations and modification indices revealed inadequate discriminant validity. CONCLUSIONS: Based on these findings, we cautiously recommend the AEQ-3 as a measure of alcohol expectancies. Specific recommendations and limitations are discussed regarding future use of the AEQ-3.
34
Evaluation of goodness-of-fit indices for structural equation models.
Stanley A. Mulaik, Larry R. James, Judith Van Alstine et al. · Psychological Bulletin · 1989 · 2.9K citations