In order to familiarize question-specific grading comparison of a board examination, the study analyzed primary score of 447 first semester examinees to the Principles of Economics course in five grading: fail, pass, second division, first division, and distinction rank for each question. Relationships in grading performance were analyzed by using the chi-square test, independent t-test, and ordered logit regression. Among the examinees, 12.98%, 38.48%, 34.0%, and 8.28% secured distinction, first, second, and pass division, respectively and 94% got success with their average of 25.48 score. Compared to boys, the girls outperformed in terms of distinction, first division, and success rate. Fairly superior grading performance was analyzed for the constituent campus examinees over private to community campuses. Significant variations in grading analyzed due to gender, inter-campus and affiliation-type; however, no ethnicity effect was assessed in overall grading. Both coefficient and odds ratio of ordered logit regression determined the likelihood of positive contribution in grade upgrading. Marginal effect of grading scale=3 likely increased grades by 60%, 20%, 24%, 22%, 18%, 22%,23%, 20%, 22%, 20%, 21%, 19% if those examinees solve first to last question, respectively. Estimated predicting probability of each examinee were closer to grading performance. The Examination Board is suggested using Bloom’s Taxonomy in questionset preparation and moderation, introduce cumulative grade point average, and coach potential examiners preceding to question setting and answer-book evaluation. All colleges are suggested to follow quality assurance and accreditation system in order to improve academic environment or establish, at least, advisory support to the poor-scored students.
Read full abstract