NotesFAQContact Us
Search Tips
Peer reviewed Peer reviewed
Direct linkDirect link
ERIC Number: EJ940430
Record Type: Journal
Publication Date: 2004
Pages: 8
Abstractor: As Provided
Reference Count: 3
ISSN: ISSN-0304-3797
An Automated Grade Prediction System
Verhoeven, Chris; Van Staveren, Arie
European Journal of Engineering Education, v29 n2 p251-258 2004
At the end of a lecturing period examinations are used to take a snapshot of the level of knowledge of a student. It is essential that this snapshot be taken at the right moment. Particularly when a student fails, it is important to make sure that it is because his level of knowledge is not adequate, not because, for example, he had a severe headache, was blocked by examination stress or misread the questions. Even when a student does pass, it may be that the grade is too low because of unjust causes. The system presented in this paper is designed to prevent this situation. An automated Internet-based system makes multiple-choice tests available for students of which the contents depend on the date. It tracks the actual course content. When a student does a test, the system will grade the student and predict the official grade that can be expected at the final examination. The results of the final examination are compared with the prediction. When the prediction is higher than the actual grade, the cause is investigated and possibly an extra assessment is done. The system is completely automated and is available for students 24h per day. This makes sure that the snapshot is taken at the right time, since the student himself can decide when to take a test. The system does not officially grade the students and is therefore not very sensitive to fraud, but it efficiently removes the "snapshot risk" of the official exam, resulting in a much better quality of the grading process. (Contains 1 table and 6 figures.)
Taylor & Francis, Ltd. 325 Chestnut Street Suite 800, Philadelphia, PA 19106. Tel: 800-354-1420; Fax: 215-625-2940; Web site:
Publication Type: Journal Articles; Reports - Descriptive
Education Level: Higher Education
Audience: N/A
Language: English
Sponsor: N/A
Authoring Institution: N/A