Análisis confirmatorio para la prueba AEQ-III en Bucaramanga, Colombia

• OSCAR F HERRÁN ^[1] ; MARÍA F ARDILA ^[1]
  1. [1] Universidad Industrial de Santander

     Universidad Industrial de Santander

     Colombia

     
• Localización: Revista Médica de Chile, ISSN-e 0034-9887, Vol. 138, Nº. 1, 2010, págs. 88-94
• Idioma: español
• Títulos paralelos:
  • Confirmatory analysis for the test AEQ-III in Bucaramanga, Colombia
• Enlaces
  • Texto completo
• Resumen

  • Background: The Alcohol Expectancy Questionnaire-III (AEQ-III) test evaluates positive expectations towards alcohol consumption. Aim: To perform a confirmatory factorial analysis of the AEQ-III test. Material and Methods: During 2006-2008, 1199 adults between 18 and 60years old (588 males) answered a survey that included socio-economic variables and the AEQ-III test to determine alcohol consumption and its consequences. A factorial analysis was performed and the goodness off it of the final model was assessed. Results: "Increase in expressivity and sexuality", "decrease in physical tension", "lack of inhibition and feelings of power" and "decrease in psychological tension" were the four confirmed dimensions. This model differs of others applied in the region, has 22 variables and adjusts well. Values for median quadratic approximation error, Tucker- Lewis Index, normal adjustment index and goodness off it were 0.07, 0.93 and 0.92, respectively. The variance explained by the model increased from 62% in the exploratory to 76% in the confirmatory analysis. Infernal consistency in the exploratory and confirmatory models was 0.85 and 0.87. Conclusions: Two confirmed dimensions correspond to group behaviors and the others, to individual behaviors. All indices of relevance of grouping variables, goodness off it and parsimony were maintained and even improved and are generally optimal.