Objective tests and their discriminating power in business courses: a case study

Abstract

Evaluating students' learning experiences outcomes cannot be considered a simple task. This paper aims at investigating students' overall performance and the discriminating power of particular tests' items in the context of business courses. The purpose of this paper is to contribute with this issue while analyzing it, with scientific approach, from an accounting information systems standpoint: two experiments based on a database management system (DBMS) undergraduate course, involving 66 and 62 students (experiments E 1 and E 2 , respectively). The discriminant analysis generated discriminant functions with high canonical correlations (E 1 = 0.898 and E 2 = 0.789). As a result, high percentages of original grouped cases were correctly classified (E 1 = 98.5% and E 2 = 95.2%) based on a relatively small number of items: 7 out of 22 items from E 1 (multiple-choice), and 3 out of 6 from E 2 (short-answer). So, with only a few items from the analyzed instruments it is possible to discriminate "good" or "bad" academic performance, and this is a measure of quality of the observed testing instruments. According to these findings, especially in business area, instructors and institutions, together, are able to analyze and act towards improving their assessment methods, to be of minimum influence while evaluating students' performance.