6533b86ffe1ef96bd12ce600

RESEARCH PRODUCT

Factorial Validity of the General Health Questionnaire 12 in an Angolan Sample

subject

description

Abstract. Most evidence available has found multidimensional solutions for the General Health Questionnaire’s (GHQ12) structure, contrary to the authors’ original one-factor solution. Recently, the study of method effects associated to negatively worded items has been included in the study of GHQ12’s factor structure, and the most common conclusion has been that the best-fitting structure is a single factor confounded by response bias. Current study analyzes the factor structure of the Portuguese version of the GHQ12. The sample consisted of 1,332 young Angolan adults. Previously-reported structures, including one-factor and multidimensional substantive ones, as well as bifactor structures, and models positing method effects among negatively worded items were tested via Confirmatory Factor Analyses. Results suggested that a three substantive factor solution adequately represented the scale’s underlying structure. The emerging factors were dysphoria, social dysfunction, and loss of confidence. Method effects did not play a clear role in explaining the multidimensionality observed in the factor structures tested for the GHQ12.