Examination of the scoring structure of the psychopathology instrument for mentally retarded adults

"The purpose of this study was to assess the validity of an instrument designed to assist in the diagnosis of mental illness in individuals diagnosed with mental retardation titled "The Psychopathology Inventory for Mentally Retarded Adults (PIMRA)." Procedures included conducting an exploratory factor analysis (EFA) to identify a more parsimonious model and a series of confirmatory factor analyses (CFA) to test the hypotheses of factorial invariance, first, with two random samples, and then with three groups based on level of mental retardation. A series of logistic regression analyses were conducted to assess the ability of each scoring model to predict the "true" mental health diagnosis. Results of CFA of the PIMRA found the model to be ill fitting. Examination of the factor correlations, item correlations and item R2 values found significant problems such that the scoring model of the PIMRA was found to be unsupported. Results of the EFA identified an interpretable six factor solution. A confirmatory factor analysis of the six factor solution revealed a model that approached adequacy after deleting ten items. The hypothesis of factorial invariance was not supported in two random samples and three groups based on level of mental retardation. Results of the logistic regressions revealed that both models were better predictors of schizophrenia, affective disorder and psychosexual disorder than other mental health disorders. Both models are better predictors of lack of diagnosis rather than diagnosis. The six factor model was only slightly better than the PIMRA. These results suggest that neither the PIMRA nor the six-factor scoring model provide any diagnostic in determining the mental health status of an individual with mental retardation."--Abstract from author supplied metadata.