Exam security strategies that are frequently used to stop cheating in a range of situations include using numerous copies of the test. Large, commercial testing firms sometimes employ psychometric procedures to guarantee test version equivalency, but these methods are usually too expensive and time-consuming for individual schools. Because of this, there is a practical conflict with test versions between exam security (improved by versioning) and fairness (arising from variations in difficulty among versions). In this study, we collected data from 2 sets of exams of students to see the trade-off between test security and fairness on a versioned programming exam. The results indicate that certain populations place a higher importance on one feature than the other. Nevertheless, developing equivalent versions, which will have equal levels of difficulty and require the same level of conceptual understanding, remains a significant task. The present analysis project involved data obtained from the latest exam trials the students performed in the introductory programming course, which delivers various classes over many days. Through thorough scrutiny, this article aims to reveal the complex factors involved in choosing exam versions, which can lead to student assessment inequality but also provides security measures for the system.
- Citation du texte
- Mangeti Wesonga (Auteur), 2025, Exams Security and Fairness Trade-Offs, Munich, GRIN Verlag, https://www.grin.com/document/1707792
