Abstract/Summary

This study explores the complexity behind vocabulary learning in the ESL classroom in Saudi Arabia. It has been suggested that vocabulary learning, particularly when it includes words and collocations which are not semantically transparent, presents a particular challenge for learners. The primary aim of the study was to explore the impact of two instructional methods - contrastive analysis and translation (CAT) and CAT+Corpus - on the short-term recall of semantically non-transparent vocabulary in receptive and productive knowledge. Nineteen vocabulary target items, which included collocations and semantically non-transparent words, were selected for this study. A total of 56 students agreed to take part. All participants were female first-year university students enrolled in a compulsory ESL course run by the English Language Institute (ELI) at a university in Jeddah, Saudi Arabia. Participants were divided into three different groups: CAT, CAT+Corpus, and control. There were two separate teaching interventions that took place over the course of two weeks. The CAT only group received interventions consisting of reading a passage in English, which included the target items, followed by a reading comprehension task. Afterwards, the teacher explained the similarities and differences between the target items in both the L1 and L2 where the words and lexical system were briefly but explicitly explained. Students also received two worksheets (for the receptive task and the productive task) which required them to match the target items to their equivalent translations. The intervention in the CAT+Corpus group was similar, however participants also worked with a parallel corpus when completing both receptive and productive tasks. The control group did not receive any treatment. The study adopted a mixed-methods approach comprising pre- and post-tests in receptive and productive areas, with all three groups. Additionally, 11 students (four from the CAT+Corpus group, three from the CAT group and four from the control group) also took part in think-aloud (TA) interviews while completing the fill-in-the-gaps task. Among these, seven students from the two treatment groups were also asked about their views on the treatment they received. The findings revealed that the two treatment groups who received CAT and CAT+Corpus instruction made significantly greater vocabulary gains on their post-tests for both receptive and productive short-term recall compared to the control group, which only experienced incidental learning. However, the findings of the immediate post-tests in receptive and productive knowledge suggest that learners who received CAT instruction led by the teacher learned more vocabulary items compared to those who received CAT+Corpus instruction, although the differences were not significant between the two groups. While both groups received explicit CAT for vocabulary instruction, the results suggest that CAT teaching alone could be more effective compared to supplementing it with parallel corpus tasks. The thesis concludes by discussing these findings in relation to the noticing hypothesis, L1 transfer, and the involvement load hypothesis (ILH), as well as their implications for ESL pedagogy.