An examination of the convergent validity of three measures of patient satisfaction in an outpatient treatment center

Assessing the convergent validity of three common measures in a sample of 97 outpatients in a comprehensive Multiple Sclerosis treatment center suggests that scores on the three types of measures—global, facet/Likert and facet/ semantic differential—are strongly interrelated but the global measure appears to largely reflect a personalized vs general referent.

Abstract

Patient satisfaction is a concept frequently used by health service researchers to either predict certain behavior (e.g. compliance) or reflect upon patient reactions to health care encounters. An implicit assumption is that such findings are comparable despite the wide variety of measures that have been developed and the possibility of measurement error. This study examines one part of this general question by assessing the convergent validity of three common measures in a sample of 97 outpatients in a comprehensive Multiple Sclerosis treatment center. Findings suggest that scores on the three types of measures—global, facet/Likert and facet/ semantic differential—are strongly interrelated but the global measure appears to largely reflect a personalized vs general referent. This evidence suggests that measurement bias may partially delimit the comparability of patient attitude studies that employ different types of referent and assessment procedures.