Using Rasch analysis to establish the construct validity of rehabilitation assessment tools

Abstract

Background Rasch analysis is a statistical model that can be used in the development of new assessment tools and the evaluation of already existing ones. Rasch analysis can also be used to provide various forms of construct validity evidence. In the revised conceptualization of validity, validity is operationalized as a unitary concept with five types of evidence. This article aims to inform rehabilitation practitioners of the different forms of validity evidence that can be demonstrated using Rasch analysis. Content The history and the more recently accepted conceptualization of validity is briefly outlined. Examples of studies in the field of rehabilitation and health care are given to illustrate the varying forms of Rasch-generated validity evidence. Conclusions An understanding of Rasch analysis application in the validation of assessment tools will equip rehabilitation practitioners with better skills in interpreting the measurement properties of instruments developed using Rasch analysis. Rehabilitation practitioners are also encouraged to use Rasch analysis to evaluate existing instruments, with the aim of generating equivalent measures of underlying constructs.