Estudio psicométrico de la prueba MIR de 2022

  1. José Curbelo 1
  2. Alberto García-Guerrero 2
  3. Fernando Sánchez-Lasheras 3
  4. Carmen Peñalver-San Cristóbal 2
  5. Tomás Villacampa 4
  6. Paula Jiménez-Fonseca 5
  1. 1 Facultad de Medicina. Universidad Francisco de Vitoria. Pozuelo de Alarcón
  2. 2 Cursos Intensivos MIR Asturias
  3. 3 Departamento de Matemáticas. Facultad de Ciencias. Universidad de Oviedo
  4. 4 Clínica Oftalmológica Villacampa. Avilés, España
  5. 5 Servicio de Oncología Médica. Hospital Universitario Central de Asturias. Oviedo
Journal:
FEM. Revista de la Fundación Educación Médica

ISSN: 2014-9832 2014-9840

Year of publication: 2023

Volume: 26

Issue: 5

Pages: 219-226

Type: Article

More publications in: FEM. Revista de la Fundación Educación Médica

Abstract

Introduction. In Spain, to practice as a specialist in any medical specialty, it is necessary to have the corresponding qualification. In order to access training as a specialist, it is mandatory to pass the MIR test. In this article, a psychometric analysis of the 2022 MIR test, held in January 2023, has been carried out using both classical test theory and item response theory. Materials and methods. A subset made up of 3,229 doctors who took the MIR test in 2022 and who entered their response template in a computer application made available for this purpose. Data were analyzed using classical test theory and item response theory. Specifically, Cronbach’s alpha, Kuder and Richardson’s formula number 21, the difficulty index, the difficulty index corrected for chance effects, the discrimination index, the biserial-point correlation index, as well as the difficulty and discrimination indices according to item response theory were calculated. Results. The 2022 MIR test, like those of previous years, shows high reliability. Both the overall results of the test and those detailed by subjects and subject blocks are analyzed. Conclusion. The results obtained in this study allow us to affirm that the MIR exam is an objective exam, of adequate difficulty and discrimination, as well as relatively stable from one call to another.