Psychometric qualities of the RAND 36-item health survey 1.0: A multidimensional measure of general health status

Abstract

The reliability and validity of the RAND 36-Item Health Survey 1.0 were investigated in a population sample of 1,063 inhabitants of a Dutch township, all age 17 or older. Confirmatory factor analysis only partly supported the internal structure of the RAND 36-Item Health Survey 1.0. The internal consistency of the instrument was high. Pointing to high convergent validity, a multitrait-multimethod matrix revealed that the RAND-36 scales showed higher correlations with corresponding scales from other instruments than with noncorresponding scales. However, indicating low discriminant validity, some of these correlations did not exceed the intercorrelations among the RAND-36 scales. Multivariate analysis of variance (MANOVA) showed significant effects of age for physical functioning, role limitations (physical problem), general health perception and pain, and significant effects of education on physical functioning and general health perception. Significant sex differences were found for mental health only. The results of this study on the psychometric properties of the RAND 36-Item Health Survey 1.0 seem promising. There is a need for further studies investigating its factor structure and cross-cultural equivalence.