DIGITAL LIBRARY
CREATION OF A QUESTION BANK FOR AN INTRODUCTORY LOGIC MODULE
University of Debrecen (HUNGARY)
About this paper:
Appears in: ICERI2019 Proceedings
Publication year: 2019
Pages: 2958-2963
ISBN: 978-84-09-14755-7
ISSN: 2340-1095
doi: 10.21125/iceri.2019.0754
Conference name: 12th annual International Conference of Education, Research and Innovation
Dates: 11-13 November, 2019
Location: Seville, Spain
Abstract:
The steady growth in student numbers, increased availability of computer networks, and the proliferation of LMS systems have led to the increased use of multiple-choice questions (MCQs) as an addition or even a substitute for traditional assessment methods. Moodle was introduced in 2015 at the Faculty of Informatics of the University of Debrecen. With the inclusion of this LMS, we have the opportunity to use MCQs for practice tests and exams. Our goal was to create a question bank, which could be used to evaluate learning even at higher cognitive levels, and to give students options to practice their knowledge, while reducing the probability of getting the same questions during the exam.

The module Introduction to Logic and Computer Science typically has the worst performance indicator for the first semester, and since it is a foundation module, not completing it significantly slows student progress. Moodle and the new 150 seat computer lab on campus provided an ample opportunity to switch to online examinations. Since students adapt their learning style for the exam, in order to achieve the best results we made the exam exercises available in advance so they could be used for preparation. Our goal of course was not for students to simply memorize the correct answers. To avoid this happening, we created an extensive database, with a large number of questions for each question types. The manual compilation of all these questions and corresponding answers requires significant work. Since this module introduces many algorithms, and these algorithms have to be used to solve the questions, it has become partially possible to generate the questions and solutions automatically. For this, we implemented all the algorithms that were introduced in the lectures and practices, and applied them to randomly chosen logical formulas. This gave us questions along with the correct answers. Constructing the incorrect answers was a more complicated task. In some cases the correct answers for other questions could be used as incorrect answers, and so could be generated. However there were question types where the incorrect answers had to be created manually.

In this article, we describe how this question bank was created, and how we carried out the generation of questions and corresponding incorrect answers, which helped us to create a large number of test question of the same types. With this we produced such a large test bank that our students could use it to practice the same type of questions that would be asked at the exams, but with different data.
Keywords:
Logic, MCQ, Moodle.