Qualitative issues in developing educational diagnostic instruments and assessment procedures for diabetic patients.

Abstract

With one exception, psychometric analyses of the quality of instrumentation and educational assessment procedures for diabetic patients have not been reported in the literature. Following an extensive internal review process, a pilot test of 56 diabetic patients found that the newly developed instrument had a high degree of internal consistency for the major indexes, 0.89 and 0.85, respectively. An item analysis found individual questions to be of an acceptable quality. An analysis of interrater reliability of patient assessment procedures produced an r = 0.93. Some support for content and concurrent validity was noted. Using an external review by an expert panel, a revised instrument and protocol was used to conduct a formal field test of 100 diabetic patients. Levels of internal consistency similar to the pilot, 0.87 adn 0.86, were found. An item analysis produced similar positive results. While some support for concurrent validity of the data was found, little support for discriminant validity was evident. The instrument and assessment procedures need to undergo more extensive and rigorous examination of its psychometric characteristics, particularly stability and predictive validity.