The effect of different strategies for combining disordered thresholds on rasch model fit

In the examination of the internal construct validity of the measurement tools used in healthcare field, Rasch analysis is one of the most common used approaches in the context of item response theory. The first stage of this analysis is checking whether there is an item with disordered thresholds. When there are items with multi response category is to examine whether there is any item with disordered thresholds and to combine the categories appropriately in the case of presence of an item with a disordered thresholds. In the later stages of the Rasch analysis, it is assessed whether the items in the measurement tool fit model, whether the items provide invariance property, whether measurement tool is unidimensionality and whether the items have differential item functioning. In this study, a measurement tool which evaluates the disability levels in the field of "self-care, getting around, hold activities" of the patients with Rheumatoid Arthritis in the healthcare field is used. The effect of four different strategies for combining the categories in the items with disordered thresholds is examined on Rasch model fit. These strategies have been obtained by combining the categories with disordered thresholds, "to the left, to the right, to middle-left and to middle-right". In the measurement tools obtained after the category combining, information loss occurred due to the combining of the categories and it has been seen that standard errors of the θ's of individuals have risen. However, it has been seen that the model fit and internal consistency of measurement tools has increased after the combining strategies. Furthermore, in the measurement tools obtained after four different strategies, both the items from each scale and the representation of subsection of each scale are important for evidence of the validity of the scope of these measurement tools. It is thought that this study would be useful for researchers working in similar areas, since there are limited studies examining the effect of different strategies for combining the categories in the items with disordered thresholds on Rasch model fit in the literature.