Examination of the MIR exam. An approach to the structural validity through the classical test theory Introduction. In Spain, to practice as a medical specialist, it is required to have the certification of the appropriate medical specialty. In order to have access to a medical specialist training programme, it is mandatory to overcome the MIR test. Passed the test, MIR training programs accessed in different hospitals and teaching units are relatively homogeneous. Aim. To approach to the structural validity of the MIR examination of the last call (2015), held on February 6, 2016, with particular emphasis on those measurable aspects of it. Subjects and methods. The database used in this study corresponds to the answers to the questions of the MIR exam of 2015 of a total of 3,712 examinees. Results. The average rate of difficulty of all questions was 0.6882, while the corrected index of difficulty was 0.5422, the discrimination of 0.2492 and 0.2954 the value of the point biserial correlation. The formula number 21 of Kuder-Richardson and the Cronbach's alpha were also applied giving as results 0.9459 and 0.9579 respectively. These values were compared with those obtained for the MIR test in the range from the calls 1989 to 1993. Conclusions. In view of the psychometric results, it can be said that the examination MIR is an objective, with an adequate level of difficulty and discrimination and also structurally valid.
Background and Objectives: The aim of the present research is to study the questions used in the 2018 MIR exam (a test that allows access to specialized medical training in Spain), describe their psychometric properties, and evaluate their quality. Materials and Methods: This analysis is performed with the help of classical test theory (CTT) and item response theory (IRT). The answers given to the test questions by a total of 3868 physicians are analyzed. Results: According to CTT, the average difficulty index for all of the test questions was 0.629, which falls into the acceptable category. The average difficulty index with correction for random effects was 0.515, which corresponds to a value within the optimal range. The mean discrimination index was 0.277, which is in the good category, while the mean point biserial correlation coefficient, with a value of 0.275 fits in the regular category. The values of difficulty and discrimination calculated according to the model of two parameters of the IRT seem adequate with average values of −0.389 and 0.677. The Cronbach alpha score obtained for the overall test was 0.944. This value is considered as very good. Conclusions: A decrease was observed in the average values of discrimination in the last three calls, which may be related to the greater proportion of Spanish graduates that take the exam in the same year of finalization of their studies in Medicine.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.