The personality structure questionnaire (PSQ): A cross-validation with a large clinical sample

Abstract

In a large sample of adult National Health Service (NHS) patients (initially n = 1296) referred for psychological therapy, the Personality Structure Questionnaire (PSQ) items were found to constitute a unidimensional scale according to maximum likelihood and principal component analyses, and by Cronbach's alpha criteria. The PSQ scores showed a significant small negative association with age, where higher scores indicate pathology; but no sex difference was found. Scores remain stable during time spent on the waiting list, but were reduced significantly following therapy of an equivalent length of time (median = 15 weeks). The intercorrelations between the three clusters of items (sub-concepts, as proposed by the test authors) comprising the full PSQ varied, but were all sizeable and highly statistically significant. Additional information to that of the test authors is provided on convergent and discriminant validity through its correlations with psychological tests and clinicians' ratings. Suggestions are made for the possible further improvement and refinement of the questionnaire. Copyright © 2009 John Wiley & Sons, Ltd.