Cognitive validity evidence of computer- and paper-based writing tests and differences in the impact on EFL test-takers in classroom assessment

María Eugenia Guapacha Chamorro

DOI: https://doi.org/10.1016/j.asw.2021.100594

IF: 3.9

2022-01-01

Assessing Writing

Abstract:This mixed-methods study collected evidence of the cognitive validity of CB and PB writing tests and the impact on EFL university students’ cognitive processes (e.g., macro-planning, micro-planning) as part of the validity and fairness of classroom writing assessment. To this purpose, two sets of data were collected from 38 Colombian EFL undergraduates from two intact classes. Two retrospective writing process questionnaires were analysed qualitatively and quantitatively to examine students’ cognitive processing across CB and PB writing tests. Three perception questionnaires were analysed qualitatively to identify students’ perceptions of the impact of the writing mode on their cognitive processes. The findings suggest that both CB and PB tests are cognitively valid, as they activated cognitive processes necessary to perform a writing assessment task. However, each test mode impacted EFL students’ cognitive processes differently. The PB mode triggered greater detailed macro-planning of content and text organisation. In contrast, the CB mode triggered micro-planning of text organisation and after-writing revisions at both high and low levels. There were no significant differences in task representation and translating processes. The findings have implications for EFL writing teaching and classroom assessment validity and fairness.

education & educational research,linguistics