Toth, Krisztina R.;
Examining media-effect among subgroups of students with different ability levels
In: European Association for Research on Learning and Instruction (Hrsg.): Book of abstracts and extended summaries
URL des Volltextes:
4. Beiträge in Sammelwerken; Tagungsband/Konferenzbeitrag/Proceedings
Due to the innovative possibilities provided by computers and the cost-efficiency of the system, technology-based assessment should displace paper-based testing (Csapo, Latour, Bennett, Ainley, & Law, 2010). If computer-based assessment is applied to replace traditional testing a number of questions arise regarding media-effect (see e.g. Clariana, & Wallance, 2002). In the past several media-effect studies were carried out focusing on different student and item-related variables, some of these studies attempted to identify the effect of background variables (gender, race/ethnicity, and ICT related factors) for the differences between paper- and computer-based test performance, however, less attention has been paid to the examination of paper-and-pencil test performance of different sample subgroups. The purpose of this paper is to investigate a comparison of data derived from two test delivery among subgroups of students based on paper-and-pencil performance. This paper argues that the media-effect is related to the ability level of test takers in low-stakes testing. Various age groups, measured context and research design validate our findings. Results indicate that different ability level students behave differently in traditional and computerized assessment. Low ability students performed significantly better in computerized environment than in traditional testing. The performance of students with average ability level was media independent, or they achieved higher scores in the computerized test delivery. High ability students were disadvantaged in the online assessment. These findings support that the comparison of paper-based and computer-based performance should consider the students ability level realized in printed medium.