Physiology Assessment For Biochemistry And Pharmacy: Reliability Analysis

Abstract

Introduction/Problem: Over the past two years, the evaluation of students in Human Physiology at Universidad Nacional del Sur (UNS) has been summative, considering student participation, quiz scores (five quizzes before the midterm exam), and written exam scores (two exams during the course). Although innovations have been implemented to evaluate different aspects of student performance and to establish a continuous assessment approach, no study has been conducted to examine the validity and reliability of these assessments in education. Objectives: The objective was to analyze the reliability of written midterm exams by conducting correlation studies between question types and student scores, in order to determine whether the different question styles effectively discriminate among students with varying levels of knowledge and skills. Methodology: Data were collected from 72 students in 2023 and 97 students in 2024 who completed the physiology exam that included:  Context-based open-ended questions (e.g., clinical cases or physiological regulation scenarios): designed to deeply assess specific knowledge and skills.  Context-free multiple-choice questions: designed to assess general knowledge in a more superficial manner The correlation between scores obtained for each question type and the students’ final exam grades was calculated, categorizing grades into four ranges: 0–25, 25–50, 51–75, and 76–100. Results: The results showed that context-based open-ended questions were statistically significant direct correlation with student grades across all four ranges (p<0,01). However, for context-free multiple-choice questions, a statistically significant direct correlation was observed only in the 0–25 (p<0,01) and 76–100 grade groups (p<0,01), with no correlation found in the 25– 50 and 51–75 groups. Conclusion: The findings suggest that context-based open-ended questions are a more reliable predictor of student performance in physiology than context-free multiple-choice questions. Incorporating context-based open-ended questions can enhance the effectiveness of assessing students’ specific knowledge and skills. It is crucial to review and improve assessment items to ensure their reliability in measuring performance.