Introduction:

In Spain, the regulated procedures of specialized medical training impose the passing of the MIR test. After this, the candidates choose the preferred specialty according to the order number obtained. The order number depends on the score of the applicant in the MIR exam and their academic record.

Subjects and methods:

The database used in the present research contains the responses of a total of 23,809 candidates submitted to the MIR exams since 2009 to 2017. The available information was analyzed using the difficulty index, difficulty index with correction of random effects, discrimination index, biserial point correlation index, as well as the difficulties and discriminations calculated using a two parameters model of the Item Response Theory.

Results:

The results obtained let us to say that the difficulty of the MIR test has decreased in the last three calls. In addition, the discrimination reaches the minimum value of the time series in the 2016 MIR test.

Conclusions:

Given the increase in recent years in the number of recently graduated Spanish doctors who take the MIR exam, it would be desirable to try to improve the psychometric quality of the questions in order to be at least as discriminative as the tests of previous calls. The inclusion by the committee of psychometric criteria of cancellation of questions would help achieve this objective.

        · abstract in Spanish     · text in Spanish     · Spanish ( pdf )