Corpus-Based Instruction for Developing Colligational Competence in Arab EFL Translation Students
Publication date: 2026-07-08
The present research was conducted to examine the efficiency of corpus-based teaching in influencing the capacity of Arab English as a Foreign Language (EFL) university learners to recognize and use English colligational grammatical patterns in translation. A convergent mixed-methods design combined with a quasiexperimental one-group pre-test–post-test approach was used. Twenty-one first-year undergraduate translation students at the Arab Open University, Egypt, were involved in a 12-week Data-Driven Learning (DDL) intervention utilizing the British National Corpus (BNC), Sketch Engine, and BootCaT. Both quantitative and qualitative data were collected, the former through a translation achievement test administered before and after the intervention, and the latter through a post-intervention questionnaire and reflective audio recordings. Quantitative analyses involved descriptive statistics, the Kolmogorov–Smirnov test, paired-samples t-tests, and Cohen's d, while the qualitative data underwent thematic analysis. The results showed statistically significant changes in students' recognition and use of English colligational grammatical patterns, with both skills demonstrating large effect sizes. However, no major difference emerged in the case of longer translation activities, which implies that discourse-level translation competence needs a more sustained period of instructional support. Participants also reported improved grammatical awareness, increased confidence in making translation decisions, and greater reliance on authentic corpus evidence. These results affirm the need to incorporate corpus-based teaching into higher-education translation courses to strengthen grammatical knowledge and evidence-based translation practices among Arab EFL students.
