Scholarlink

<p>At the construct-articulation stage, current ESL writing rubrics quietly pull rhetorical preference into the operational definition of language proficiency. The interpretive consequences reach beyond measurement alone: once scores cannot tell organisational strategy apart from language control, teacher interpretation, feedback, and gatekeeping decisions all lose precision. This framework-development study traces the problem through critical discourse analysis of four major frameworks (TOEFL iBT, IELTS Academic Writing, CEFR, and classroom assessment principles). It shows how descriptor language governing organisation, coherence, and development encodes linear, thesis-first norms. Coding-scheme stability is established through intra-analyst (Cohen’s κ = 1.00) and inter-analyst (81.7 per cent agreement) checks on stratified subsets matched in size and scheme (12 of 34; 35.3 per cent). The Critical Extension Framework (CEF) names this diagnosis explicitly as pattern recognition folded into rhetorical effectiveness at the descriptor level. It responds by inserting a pattern-recognition step before scoring, separating the three inferential layers that current rubrics conflate: pattern recognition, rhetorical effectiveness, and Language Resources. Situated within established validity architecture at the domain-description inference, the framework makes organisational assumptions explicit before they enter later scoring, generalisation, and use claims. A preliminary two-rater pilot suggests that Language Resources scores may converge when raters work from the scoring descriptors and scoring protocol in Appendix B together with the essay texts. Rhetorical Effectiveness ratings on the reader-responsible implicit essay show pronounced, construct-sensitive disagreement. That disagreement locates the framework’s heaviest validation burden. The point holds with particular force for multilingual writing assessment in Asian EFL/ESL contexts.</p>

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

The critical extension framework: making organisational assumptions explicit in multilingual writing assessment

  • 期刊论文

  • 出版模式：

    开放获取

  • 发表日期：

    2026年7月3日

查看全文
  • Projnya Mojumdar

摘要

At the construct-articulation stage, current ESL writing rubrics quietly pull rhetorical preference into the operational definition of language proficiency. The interpretive consequences reach beyond measurement alone: once scores cannot tell organisational strategy apart from language control, teacher interpretation, feedback, and gatekeeping decisions all lose precision. This framework-development study traces the problem through critical discourse analysis of four major frameworks (TOEFL iBT, IELTS Academic Writing, CEFR, and classroom assessment principles). It shows how descriptor language governing organisation, coherence, and development encodes linear, thesis-first norms. Coding-scheme stability is established through intra-analyst (Cohen’s κ = 1.00) and inter-analyst (81.7 per cent agreement) checks on stratified subsets matched in size and scheme (12 of 34; 35.3 per cent). The Critical Extension Framework (CEF) names this diagnosis explicitly as pattern recognition folded into rhetorical effectiveness at the descriptor level. It responds by inserting a pattern-recognition step before scoring, separating the three inferential layers that current rubrics conflate: pattern recognition, rhetorical effectiveness, and Language Resources. Situated within established validity architecture at the domain-description inference, the framework makes organisational assumptions explicit before they enter later scoring, generalisation, and use claims. A preliminary two-rater pilot suggests that Language Resources scores may converge when raters work from the scoring descriptors and scoring protocol in Appendix B together with the essay texts. Rhetorical Effectiveness ratings on the reader-responsible implicit essay show pronounced, construct-sensitive disagreement. That disagreement locates the framework’s heaviest validation burden. The point holds with particular force for multilingual writing assessment in Asian EFL/ESL contexts.