Objective: To evaluate the score spectrum of the test, difficulty, and discriminability of questions in the objective multiple-choice tests. Method: Use spectrum analysis for each module. Determining the difficulty of a multiple-choice question is the percentage of students who answered correctly compared to the total number of students participating in answering that question; Discriminability of test questions is the ability of multiple-choice questions to classify students into groups of different levels (good, very good groups and average-weak- poor group). Results: Spectrum of test scores: the average score or higher accounted for 66,93%, of which, good and excellent scores accounted for 30.47%. The score spectrum is also relatively even between classes. There are 35.03% of exam questions with medium difficulty, 41.66% of the exam questions are at the easy level, the questions at the difficult level and the relatively difficult level account for 15.18 and 8.11%, respectively. There are 25.24% of exam questions with low discriminatory level; The levels of good, and average are equivalent with the rate of 15.47% and 15.13% respectively; Questions with a very good level of discrimination account for an overall rate of 43.16%. Conclusion: Determining the score spectrum of the test and assessing the difficulty and discriminability of the questions for the objective multiple-choice test is essential to build a good-quality test, and at the same time help to improve the question bank and eliminate inappropriate multiple-choice questions.
Read full abstract