Balanced Scoring Method for Multiple-mark Questions

Darya Tarasowa, Sören Auer

2013

Abstract

Advantages and disadvantages of a learning assessment based on multiple-choice questions (MCQs) are a long and widely discussed issue in the scientific community. However, in practice this type of questions is very popular due to the possibility of automatic evaluation and scoring. Consequently, an important research question is to exploiting the strengths and mitigate the weaknesses of MCQs. In this work we discuss one particularly important issue of MCQs, namely methods for scoring results in the case, when the MCQ has several correct alternatives (multiple-mark questions, MMQs). We propose a general approach and mathematical model to score MMQs, that aims at recognizing guessing while at the same time resulting in a balanced score. In our approach conventional MCQs are viewed as a particular case of multiple-mark questions, thus, the formulas can be applied to tests mixing MCQs and MMQs. The rational of our approach is that scoring should be based on the guessing level of the question. Our approach can be added as an option, or even as a replacement for manual penalization. We show that our scoring method outperforms existing methods and demonstrate that with synthetic and real experiments.

Download


Paper Citation


in Harvard Style

Tarasowa D. and Auer S. (2013). Balanced Scoring Method for Multiple-mark Questions . In Proceedings of the 5th International Conference on Computer Supported Education - Volume 1: CSEDU, ISBN 978-989-8565-53-2, pages 411-416. DOI: 10.5220/0004384304110416

in Bibtex Style

@conference{csedu13,
author={Darya Tarasowa and Sören Auer},
title={Balanced Scoring Method for Multiple-mark Questions},
booktitle={Proceedings of the 5th International Conference on Computer Supported Education - Volume 1: CSEDU,},
year={2013},
pages={411-416},
publisher={SciTePress},
organization={INSTICC},
doi={10.5220/0004384304110416},
isbn={978-989-8565-53-2},
}


in EndNote Style

TY - CONF
JO - Proceedings of the 5th International Conference on Computer Supported Education - Volume 1: CSEDU,
TI - Balanced Scoring Method for Multiple-mark Questions
SN - 978-989-8565-53-2
AU - Tarasowa D.
AU - Auer S.
PY - 2013
SP - 411
EP - 416
DO - 10.5220/0004384304110416