Department of Physiology, College of Medicine, Imam Abdulrahman Bin Faisal University, Dammam, Saudi Arabia.
Postgrad Med J. 2023 Aug 22;99(1175):1020-1026. doi: 10.1093/postmj/qgad005.
Multiple assessment tools are used to assess future doctors' knowledge, clinical skills, and professional attitudes. In the present research, the difficulty level and discriminating ability of different types of written and performance-based assessments designed to measure the knowledge and competency of medical students were compared.
The assessment data of 2nd & 3rd-year medical students (the academic year 2020-2021) in the College of Medicine at Imam Abdulrahman Bin Faisal University (IAU) were retrospectively reviewed. Based on end-of-the-year overall grades, students were divided into high and low scorers. Both groups were compared by independent sample t-test for their mean scores achieved in each type of assessment. Difficulty level and discriminating ability of the assessments were also explored. MS Excel and Statistical Package for Social Sciences (SPSS version 27) were used for analysis. Area under the curve was calculated through ROC analysis. A p-value <0.05 was believed significant.
In each type of written assessment, the high scorer group achieved significantly higher scores compared to the low scorers. Among performance-based assignments (except the PBLs), scores did not differ significantly between high and low scorers. The difficulty level of performance-based assessments was "easy" whereas it was "moderate" for written assessments (except the OSCE). The discriminating ability of performance-based assessments was "poor" whereas it was "moderate/excellent" for written assessments (except the OSCE).
Our study results indicate that written assessments have excellent discriminatory ability. However, performance-based assessments are not as difficult and discriminatory as written assessments. The PBLs are relatively discriminatory among all performance-based assessments. Key messages What is already known on this topic At Imam Abdulrahman Bin Faisal University, written and performance-based assessments both are graded on criterion-referenced scales. The student's grades at the end of the year are an aggregate of his/her scores in written and performance-based assessments. What this study adds Our study results show that performance-based assessments are not as difficult and discriminatory in differentiating between high and low scorers as written assessments. How this study might affect research, practice or policy Performance-based assessments should be made a hurdle exam (pass or fail) for the students to move to the next level, or students must pass each assessment component (written and performance-based) separately.
多种评估工具用于评估未来医生的知识、临床技能和职业态度。在本研究中，比较了不同类型的笔试和操作考试评估工具，旨在衡量医学生的知识和能力，评估其难度水平和区分能力。
回顾了 2020-2021 学年阿卜杜勒阿齐兹国王大学医学院二、三年级医学生的评估数据。根据期末总成绩，将学生分为高分组和低分组。通过独立样本 t 检验比较两组在每种评估类型中的平均得分。还探讨了评估的难度水平和区分能力。使用 Excel 和统计软件包（SPSS 版本 27）进行分析。通过 ROC 分析计算曲线下面积。置信区间为 95%，p 值<0.05 被认为有统计学意义。
在每种笔试评估中，高分组的得分均显著高于低分组。在操作考试任务中（PBL 除外），高分组和低分组的得分无显著差异。操作考试的难度水平为“容易”，而笔试评估的难度水平为“中等”（OSCE 除外）。操作考试的区分能力为“差”，而笔试评估的区分能力为“中等/优秀”（OSCE 除外）。
本研究结果表明，笔试评估具有良好的区分能力。然而，操作考试的难度和区分能力不如笔试评估。在所有操作考试任务中，PBL 相对具有区分性。
在阿卜杜勒阿齐兹国王大学，笔试和操作考试均采用基于标准的等级制进行评分。
学生的期末成绩是其笔试和操作考试成绩的总和。
对研究、实践或政策的影响：
