Advances in Medical Education and Practice (Jan 2019)
Effect of blueprinting methods on test difficulty, discrimination, and reliability indices: cross-sectional study in an integrated learning program
Abstract
Hussein Abdellatif,1,2 Abdullah M Al-Shahrani3 1Anatomy and Embryology Department, Faculty of Medicine, University of Mansoura, Mansoura, Egypt; 2Department of Anatomy, College of Medicine, University of Bisha, Bisha, Saudi Arabia; 3Department of Family Medicine, College of Medicine, University of Bisha, Bisha, Saudi Arabia Purpose: Exam blueprinting achieves valid assessment of students by defining exactly what is intended to be measured in which learning domain and defines what level of competence is required. We aimed to detect the impact of newly applied method for blueprinting that depends on total course credit hours and relate the results with item analysis reports for students’ performance. Participants and methods: A new method for blueprint construction was created. This method utilizes course credit hours for blueprint creation. Survey analysis was conducted for two groups of students (n=80); one utilized our new method (credit hours based) for blueprinting and the other used traditional method depending on exam duration and time for individual test items. Results: Results of both methods were related to item analysis of students’ achievements. No significant difference was found between both groups in terms related to test difficulty, discrimination, or reliability indices. Both achieved close degrees of test validity and reliability measures. Conclusion: We concluded that our method using credit hours system for blueprinting could be considered easy and feasible and may eventually be utilized for blueprint construction and implementation. Keywords: learning, students, reproducibility, competency, curriculum, blueprint
