Phrasal Complexity Measures as Predictors of EFL University Students’ English Academic Writing Proficiency
Krittaya Thongyoi, Kornwipa Poonpon · rEFLections · 2020
The study aims to investigate phrasal complexity measures that can predict EFL students’ academic writing proficiency. Academic English written test responses were derived from written responses from the Khon Kaen University Academic English Language Test (KKU-‐AELT). Five hundred and thirty written responses were separated into groups based on their writing scores. Sixty-‐six phrasal complexity measures (Kyle, 2016) were analyzed for this study. The Tool for the Automatic Analysis of Syntactic Sophistication and Complexity (TAASSC), a computational tool for phrasal complexity analysis, was used to calculate the average numbers of occurring measures in written responses. Phrasal complexity measures occurring in written responses were analyzed with the independent t-‐test. Then, 11 significant phrasal complexity measures, derived from the independent t-‐test, were entered into Binary logistic regression in order to examine potential phrasal complexity measures that can predict proficiency levels. The results revealed three phrasal complexity measures that can predict academic writing for higher proficiency level students.