A Cognitive Diagnostic Model of Reading Ability based on China’s Standards of English Language Ability
Year: 2024
Type: article
Abstract: Traditional English reading assessments in China typically provide a single summative score, without revealing the extent to which students have mastered different aspects of reading comprehension. Cognitive Diagnostic Models (CDMs), as a type of multidimensional latent trait model, offer diagnostic insights by presenting examinees' proficiency levels across a set of predefined skills. Cognitive diagnostic assessments based on these models can deliver multidimensional, fine-grained evaluation results, thus overcoming the limitations of traditional unidimensional score-based assessments. Based on the China's Standards of English Language Ability (CSE), this study constructs a cognitive diagnostic assessment model for English reading. Using college general English reading test papers and the G-DINA model, the English reading abilities of 200 students were diagnosed, and personalized performance reports were generated. A questionnaire survey was conducted to gather students' evaluations and suggestions regarding the feedback reports. The results indicate that the cognitive diagnostic model for English reading, constructed according to the CSE, exhibits good model fit and high diagnostic reliability. The personalized performance reports visually and comprehensively present the strengths and weaknesses of the test takers' reading abilities, providing valuable guidance for students' future remedial learning. This study utilizes a multidimensional assessment approach based on cognitive diagnostic models to evaluate the English reading abilities of Chinese university students, effectively addressing the limitations of traditional assessment methods that typically offer only a single score.
Source: Forum for Linguistic Studies
Institution: National University of Malaysia
Language: English
Field: Decision Sciences
Domain: Social Sciences
SDG: Quality education
Open Access status: diamond
OpenAlex ID: W4405559078
