True or false: is that the question?/Falso o verdadero, ¿es esa la pregunta?/Falso ou verdadeiro, e essa a pergunta?

Objectives: Standard tests designed for medical students frequently include objective True/ False (T/F) questions, in which there are only two possible answers that can be successfully reached at random. The purpose of this worh was to demonstrate that ashing for support of the answers diminishes the number of randomly correct ones, so improving the ability of questions to verify learnings. Methods: 8.188 T/F questions whose answers had to be supported; they were applied in 28 written tests of Pathology, at the undergraduate medical program (La Sabana University, Chia, Colombia). Answers were classified as correct or incorrect, and their support, as correct, partially correct and incorrect. In order to analyze the data, the happa agreement coefficient between T/F questions and T/F questions with support was calculated, as well as per term and per test. Results: Out of 8.188 T/F questions, 6.112 (74.6%) were correctly answered as true or false, 3.655 (44.6%) had a correct support, and 2.336 (28.5%) were not correctly supported, exhibiting a global agreement index of 0.378 and low agreement indexes per term. Conclusion: T/F questions could be accompanied by support of the answers in order to reduce the probability of randomly correct ones, to enhance the development of superior cognitive processes, and communicative essential competences. KEYWORDS Education, Medical; Educational Measurement; Undergraduate Medical Education Objetivos: en los examenes para estudiantes de medicina se emplean preguntas objetivas de la modalidad Falso o Verdadero (F/V), que solo tienen dos opciones de respuesta con alta probabilidad de ser contestadas correctamente por azar. El proposito de este trabajo fue demostrar que solicitar la argumentacion de las respuestas disminuye el numero de las correctas mejorando la capacidad de las preguntas para verificar los aprendizajes. Metodos: se revisaron 8.188 preguntas de F/V que requirieron sustentacion de la respuesta, aplicadas en 28 evaluaciones de Patologia de la carrera de Medicina de la Universidad de la Sabana durante cuatro semestres. Las respuestas se clasificaron en correctas e incorrectas y las sustentaciones, en correctas, parcialmente correctas o incorrectas. Se calculo el coeficiente de concordancia feappa entre las preguntas F/V y las preguntas F/V con sustentacion, asi como por periodo y por evaluacion. Resultados: de un total de 8.188 preguntas, 6.112 (74,6%) fueron contestadas correctamente, 3.655 (44,6%) tuvieron suste