Introduction:

When the future professional competence of medical students is decided based on results of an objective structured clinical examination (OSCE), the reliability of this test should be adequate to this purpose.

Aim:

To calculate the reliability (Cronbach’s alpha) of each one of OSCEs we performed and its relationship with the duration, number of participants, stations, items and evaluators.

Subjects and methods:

Fourteen OSCE tests performed to 2995 medical students of 4th and 5th year of the Faculty of Medicine of Granada between 2004 to 2013 were analyzed.

Results:

The reliability was ≥ 0.70 in 92.84% of the OSCEs. It was also significant ≥ 0.70 with a total duration ≥ 60 minutes (p=0.042), and a number of stations ≥ 10 (p=0.019), a number of items ≥ 50 (p=0.018) and a number of evaluators ≥ 6 (p=0.018). No differences with the number of students, neither with the options to the item were observed.

Conclusions:

The OSCEs carried out in centers which results are used to approve subjects of the medical career, must have a reliability ≥ 0.70. To achieve this reliability or greater, the format should consist of at least: 10 stations, a duration ≥ 60 minutes, and having ≥ 50 items and ≥ 6 evaluators.

        · resumo em Espanhol     · texto em Espanhol     · Espanhol ( pdf )