Corpus-Assisted Approach: The Effectiveness in Learning English Preposition Collocations of Undergraduate EFL Students and Their Learning Perceptions

Abstract

This quasi-experimental study implemented a corpus-assisted approach with data-driven learning (DDL) to examine its effectiveness in learning English preposition collocations of Thai undergraduate students, investigate the extent to which the students apply their collocational knowledge obtained through concordances in their written productions, and explore the students’ perceptions of learning the collocations through concordances. Forty students from one intact class at a private university in southern Thailand participated in this study. The experiment period took 10 weeks in the English for Communication course, which aims to develop students’ communicative abilities. Pre- and post-writing tests, a questionnaire, a stimulated-recall interview, and a semi-structured interview were employed as the data collection instruments. The test results indicated that after the experiment, the participants’ accuracy scores on their collocational knowledge increased significantly in all targeted patterns (p = 0.00). The effect size was found to have a large overall practical significance (d = 1.26). The stimulated-recall interview results revealed that most participants could correctly: 1) classify the types of patterns being learned; 2) identify the usage of the content words with different prepositions; and 3) explain some key considerations when using the collocations. It was also found that the participants had acquired a number of collocations other than the targeted ones. The results from the questionnaire and the semi-structured interview showed the participants’ positive perceptions toward the approach. It can be inferred from the results that the corpus-assisted approach was effective for preposition collocation learning; nevertheless, learners who are novel to this approach require more time to become acquainted with the learning procedures and the concordancing tool. The results of the study also have pedagogical implications for similar teaching contexts, as teachers can deliberate whether to adopt a guided DDL, guiding the learners through each step of learning, or an autonomous DDL, where the learners can learn on their own without direct assistance from the teachers. To facilitate their learning especially low-proficiency learners, it is also recommended that more teacher monitoring and intervention are necessary during computer-based concordance learning process