Criterion-Referenced Measurement in Medicine and Nursery: Different Understandings of Teachers and Students at a Public College of Brasília, Brazil

ABSTRACT Introduction Criterion-referenced tests measure academic performance without comparing one student to another, therefore differing from the norm-referenced tests, which are designed to determine the highest and the lowest scores. The Escola Superior de Ciências da Saúde (ESCS) of Federal District, Brazil, uses criterion-referenced measurement as its only evaluative method, attempting to assess students in a more integrative way and to avoid stimulating competitiveness among students. Objective To analyze the interpretation students and teachers of Medicine and Nursery courses about this measurement. Material and Methods Cross-sectional, descriptive, quali-quantitative study, performed by application of semi-structured questionnaires to teachers and students from first to fourth year of Medicine and Nursery courses. We compared courses, students and professors using the Chi-Square test. Results and Discussion: The sample size was of 413 subjects (n = 55.9%), composing 54 teachers and 344 students. We observed fragility in the concept of criterion-referenced measurement, as well as dissatisfaction regarding the method of evaluation adopted by the Institution (323 or 79,6%). Most answers point subjectivity in the correction of exams and difficulties in competing for residency programs after graduation as the major problems. Other authors state that criterion-referenced measurement tends to inflate grades of students, although it also evaluates them in a more integrative way. They recommend the adoption of both measurements (criterion and norm-referenced) in order to respectively: obtain best assessment and to compare students’ performances. Conclusion Comparison of academic performance is necessary for selective processes inside and outside the studied institution. Therefore, criterion-referenced measurement alone cannot meet the needs of the institution, generating problems during and after graduation. We suggest the adoption of both methods (criterion and norm-referenced measurement) as well as the application of standard setting concepts for classification of students. Evaluative instruments also need objective criteria for correction of tests, in order to provide a uniform and fairer correction. For that, we propose an increase in teacher training by adding new discussions about elaboration of tests, like Item Response Theory.