A testing load: comparing cognitive load in computer and paper-based testing

James Pengelley

The aim of this thesis was to extend the literature on the differences between computer-based and paper-based assessment in school-aged students. Year nine students (aged 13-14 years) at two independent co-educational schools in Western Australia completed equivalent forms of a paper and computer-based quiz which served as the end-of-topic revision quiz for their regular timetabled teaching and learning programme. Applying a post-positivist paradigm, the research project outlined here explored the effect of assessment modality by measuring test scores, subjective cognitive load, working memory capacity, scratch paper use, and preferences for assessment modality. Chapter 1 presents a background and rationale for this thesis, including research questions and hypotheses. In Chapter 2, a detailed overview of the literature on cognitive load theory and assessment modality is presented. Chapter 3 presents the main research project that informed this thesis. In Chapter 4, an extended discussion of the literature and analysis of additional measures taken during the data collection process. These include the effect of assessment mode on cognitive efficiency, gender differences, assessment modality preferences and confidence in learning science. Finally, Chapter 5 presents and overview of the project with key findings, limitations, and recommendations for future research. This thesis provides novel findings relating to the use of computer-based assessment in school aged children. The main findings from the hierarchical linear models used in this analysis suggest that while students report lower levels of cognitive load during computer-based assessments, they tend to invest more effort in, and perform better on equivalent paper-based questions. Additionally, scratch paper used decreased during computer-based assessment as question difficulty increased. These findings indicate that using computer-based assessments in place of paper-based ones with school-aged children is unlikely to produce equivalent test scores. Likewise, the use of computer-based assessments was found to reduce adaptive test-taking behaviours, such as scratch paper use, which were associated with better test outcomes. Furthermore, while no gender difference in test scores was found, digital forms of assessment may be inequitably disadvantageous to students with lower working memory capacity.

A testing load: comparing cognitive load in computer and paper-based testing

Files and links (1)

Abstract

Details

Metrics