Universität zu Köln

Karay, Yassin, Schauber, Stefan K., Stosch, Christoph and Schuettpelz-Brauns, Katrin (2015). Computer Versus Paper-Does It Make Any Difference in Test Performance? Teach. Learn. Med., 27 (1). S. 57 - 63. ABINGDON: ROUTLEDGE JOURNALS, TAYLOR & FRANCIS LTD. ISSN 1532-8015

Full text not available from this repository.

Abstract

Construct: In this study, we examine the differences in test performance between the paper-based and the computer-based version of the Berlin formative Progress Test. In this context it is the first study that allows controlling for students' prior performance. Background: Computer-based tests make possible a more efficient examination procedure for test administration and review. Although university staff will benefit largely from computer-based tests, the question arises if computer-based tests influence students' test performance. Approach: A total of 266 German students from the 9th and 10th semester of medicine (comparable with the 4th-year North American medical school schedule) participated in the study (paper = 132, computer = 134). The allocation of the test format was conducted as a randomized matched-pair design in which students were first sorted according to their prior test results. The organizational procedure, the examination conditions, the room, and seating arrangements, as well as the order of questions and answers, were identical in both groups. Results: The sociodemographic variables and pretest scores of both groups were comparable. The test results from the paper and computer versions did not differ. The groups remained within the allotted time, but students using the computer version (particularly the high performers) needed significantly less time to complete the test. In addition, we found significant differences in guessing behavior. Low performers using the computer version guess significantly more than low-performing students in the paper-pencil version. Conclusions: Participants in computer-based tests are not at a disadvantage in terms of their test results. The computer-based test required less processing time. The reason for the longer processing time when using the paper-pencil version might be due to the time needed to write the answer down, controlling for transferring the answer correctly. It is still not known why students using the computer version (particularly low-performing students) guess at a higher rate. Further studies are necessary to understand this finding.

Item Type:	Journal Article
Creators:	Creators Email ORCID ORCID Put Code Karay, Yassin UNSPECIFIED UNSPECIFIED UNSPECIFIED Schauber, Stefan K. UNSPECIFIED UNSPECIFIED UNSPECIFIED Stosch, Christoph UNSPECIFIED UNSPECIFIED UNSPECIFIED Schuettpelz-Brauns, Katrin UNSPECIFIED UNSPECIFIED UNSPECIFIED
URN:	urn:nbn:de:hbz:38-414845
DOI:	10.1080/10401334.2014.979175
Journal or Publication Title:	Teach. Learn. Med.
Volume:	27
Number:	1
Page Range:	S. 57 - 63
Date:	2015
Publisher:	ROUTLEDGE JOURNALS, TAYLOR & FRANCIS LTD
Place of Publication:	ABINGDON
ISSN:	1532-8015
Language:	English
Faculty:	Unspecified
Divisions:	Unspecified
Subjects:	no entry
Uncontrolled Keywords:	Keywords Language MULTIPLE-CHOICE TESTS; STUDENTS; EDUCATION Multiple languages Education, Scientific Disciplines; Health Care Sciences & Services Multiple languages
URI:	http://kups.ub.uni-koeln.de/id/eprint/41484