Instrumentos de avaliaÃÃo da aprendizagem: uma questÃo em aberto?




This work requests the need for tests in the process of learning evaluation.Research the reliability and validity of results obtained from multiple-choice questions compared to the results from open-ended questions at university level.Raise the hypothesis of what the results of tests with these types of questions, exploring the same content, with the same framework of specifications, they present a positive correlation. Using data from Exame Nacional de Cursos (ENC) - Medicina-2003 provided by the Instituto Nacional de Pesquisas PedagÃgicas (INEP) (National Institute of Pedagogical Research). The open-ended test was used as criteria to check the concurrent validity of the multiple-choice test. The multiple-choice test was submitted to the exam of the determination of the concurrent validity in relation to the open-ended test and the content validity and the construct validity, in relation to the tables of the specification of the contents and abilities to evaluate, with positive results. The treatment of the statistical data available revealed having a coefficient of validity, which was moderately positive to the order of 0.652 between the results obtained from the open-ended questions test and the multiple-choice questions test. In conclusion such results permit us to infer that the multiple-choice questions are efficient to evaluate the learning of a large contingency of students, at university level, the utilization of this in the classroom depends on the evaluator and the context where he works.


learning evaluation grandes contingentes large scale proficiency tests fidedignidade avaliaÃÃo de aprendizagem aprendizagem; avaliaÃÃo;avaliaÃÃo educacional avaliacao da aprendizagem validade large contingent testes de proficiÃncia em larga escala validity reliability

Documentos Relacionados