Construct Validity of a New Rating Scale for Self-Reported Impairments in Persons With Late Effects of Polio

Abstract

To evaluate the construct validity of a new rating scale for self-reported impairments in persons with late effects of polio. Psychometric analysis of data on self-perceived impairments in persons with prior polio. Two hundred seventy-three persons with prior polio (119 men and 154 women; mean age, 63.5 years). Rasch analysis of a 13-item rating scale with 5 response categories, in which the participants rated how much they have been bothered by various post-polio-related impairments during the past 2 weeks. The initial analysis showed disordered categories, misfit with some of the items, multidimensionality, and local dependency. After adjustment of the categories, which resulted in a 4-category rating scale, fit to the model was achieved, but the scale still showed signs of multidimensionality. Analyses of local dependency revealed correlations among some of the items, which resulted in a 5 testlet solution, which gave fit to the model and unidimensionality. After adjustment of the categories and local dependency, this new rating scale, Self-Reported Impairments in Persons With Late Effects of Polio, can be considered as unidimensional. The good psychometric properties implies that the Self-Reported Impairments in Persons With Late Effects of Polio scale could be a useful rating scale that would increase our understanding of the impairments that persons with late effects of polio can experience. With further refinements, this scale may assist in the planning and evaluation of appropriate rehabilitation interventions.