Comparative Judgement (CJ) is a pairwise comparison evaluation method, typically conducted online. Multiple judges each compare the quality of a series of paired performances and, from their decisions, a rank order is constructed and scores calculated. Research across different educational contexts supports CJ’s reliability for evaluating written performances, permitting more precise scoring of scripts and for dimension-focused evaluation. However, scant insights are available about the basis of judges’ evaluations. This issue is important because argument-based approaches to validation (common in the field of language testing and adopted in this study) require evidence to support claims about how scores are appropriate for test purpose. Therefore, we investigate the scoring validity of CJ, both when used holistically (the standard application of CJ) and when evaluating scripts by individual criteria (termed dimensions in the research context). Twenty-seven judges evaluated 300 scripts addressing two writing task types in a national English as a Foreign Language examination for young learners in Austria. Judges reported via questionnaires what they had focused on while judging. Subsequently, eight judges provided think-aloud data while evaluating 157 scripts, offering further insight into the writing features they considered and their decision-making during CJ. Findings showed that while most judges adapted a decision-making process similar to traditional rating methods, some adapted their method to accommodate the nature of CJ evaluation. Furthermore, results indicated that the judges considered construct-relevant criteria when using CJ, both holistically and by dimension, thus offering support to an argument for the appropriateness of using CJ in this context. • Comparative Judgement can offer an alternative to analytic rating of EFL writing. • Judges with teaching or rating experience largely focus on relevant text features. • Some judges adopt a decision-making process that appears well suited to CJ. • Dimension-based CJ has the potential to provide richer feedback than holistic CJ.

In response to changing stakeholder needs, large-scale language test providers have increasingly considered the feasibility of delivering paper-based examinations online. Evidence is required, however, to determine whether online delivery of writing tests results in changes to writing performance reflected in differential test scores across delivery modes, and whether test-takers hold favourable perceptions of online delivery. The current study aimed to determine the effect of delivery mode on the two writing tasks (reading-into-writing and extended writing) within the Trinity College London Integrated Skills in English (ISE) test suite across three proficiency levels (CEFR B1-C1). 283 test-takers (107 at ISE I/B1, 109 at ISE II/B2, and 67 at ISE III/C1) completed both writing tasks in paper-based and online mode. Test-takers also completed a questionnaire to gauge perceptions of the impact, usability and fairness of the delivery modes. Many-facet Rasch measurement (MFRM) analysis of scores revealed that delivery mode had no discernible effect, apart from the reading-into-writing task at ISE I, where the paper-based mode was slightly easier. Test-takers generally held more positive perceptions of the online delivery mode, although technical problems were reported. Findings are discussed with reference to the need for further research into interactions between delivery mode, task and level.