• Version
• Download 817
• File Size 594.48 KB
• File Count 1
• Create Date October 22, 2019
• Last Updated October 22, 2019

Redefining Students Success

Creating successful learners is a challenging task for educational institutions. As a rule, the measurement of student learning performance is carried out using tests and exams. However, test and exam data is often subjected to inadequate analysis, which leads to incorrect conclusions about the progress of student learning and therefore misleading recommendations on how to improve the learning process. Traditional assessment of students’ proficiency is done by summing up or averaging raw scores of an exam. However, this approach does not consider differences in the difficulty of the exam questions, interdependencies of the questions as well as differences in the ability of the students. In this paper we show that the traditional assessment approach produces misleading results. In this paper, we show how to analyze exam data using the Polytomous Rasch Measurement Model combined with the Relational Bayesian Networks methodology. We demonstrate that assessment of students’ proficiency using these methods is realistic, accurate and reliable. Such assessment is instrumental in creating the Student Success Profile for each course. These profiles help to create actionable recommendations for addressing gaps in student education, and eventually help educational institutions to develop better and more successful learners, identify and handle issues in the educational process before they become problems, and, ultimately, significantly reduce students’ attrition.

Attached Files