A corpus-based study of Malaysian ESL students' use of discourse connectors in upper and post-secondary argumentative writing

Discourse connectors (DCs) are one of the elements of cohesive devices that bring about cohesion to a piece of writing or speech. They are potentially useful means for writers, particularly in ESL and EFL writing pedagogic settings. DCs usefulness is two-pronged. First, they help and guide readers through the text, and then they are tools for writers to engage with their readers. It has been well-documented that appropriate and efficient use of DCs will create a coherent flow of the text. However, second/foreign language learners have some difficulties to use them efficiently and systematically in their writing. Literature review shows that the Malaysian ESL students are also suffering from improper and efficient use of DCs which leads them in failing to produce a cohesive text. Surprisingly, no single study was found in the context of Malaysia to investigate Malaysian ESL students’ understanding and use of DCs. Hence, this study attempted to investigate and understand the nature and the use of the DCs in the Malaysian student writing compared with Native speakers writing. The study also was set to examine the correlation between the frequency use of the DCs and the quality of writing. The final goal of this research study was to find out to what extent Malaysian ESL students are committing errors while using DCs. A corpus-based approach was adopted to meet the objectives of the study. To this end, an argumentative topic was assigned to the Form 4, Form 5 (upper-secondary) and the first year college students (post-secondary) and they were asked to write about the given topic in the classroom and submit their works to the instructors. They were required to write 250 words within 45 minutes. Upon compilation of the essays, the Malaysian Corpus of Students' Argumentative Writing (MCSAW) was built with ≃ 600,000 tokens. To compare and find out a vivid picture of Malaysian ESL students use of DCs with Native English Speakers, the Louvain Corpus of Native Essay Writing [LOCNESS] corpus was used. Oxford Wordsmith Tools (5) was employed to extract data from corpus for analysis, by using frequency count and concordance functions. Aiming to identify what type of DCs is used by Malaysian ESL students, Discourse Connector List developed by Rezvani Kalajahi and Neufeld (2014) was used. To be able to examine the relationship between the quality of writing and the frequency of the use of the DCs, ESL composition profile offered by Jacobs et al. (1981) was utilized. Finally, a framework of identification of DCs error type was developed by the researcher to explore the errors that students commit while using DCs. Findings of this study entail three phases. First, it was observed that Malaysian students tend to use DCs more frequently than native students. The overall frequency of the use of the DCs between Malaysian and native students was statistically significant at p < .05. However, the native students used more variety of DCs types than Malaysian students (398vs.328). It was also found that Malaysian students use DCs in some categories frequently and infrequently. Based on the findings in the native students writing (LOCNESS Corpus), the most frequent DCs in written English were offered. Second, there was a very weak negative but insignificant correlation between writing quality and the frequency of the use of the DCs in the writing of Malaysian ESL students. Finally, the qualitative analysis revealed that the erroneous use of DCs made by Malaysian ESL student writers mainly manifested in eight different categories. They had problems with the use of these devices which involved semantic, syntactic, stylistic, positional and mechanical errors. They also appeared to have tendency for unnecessary addition, omission, and redundant repetition of the DCs. In conclusion, this study demonstrated that Malaysian ESL students’ use of DCs was still at an evolving level. It is vitally important that the accurate use of DCs in writing among Malaysian students be further highlighted in the classrooms through using concordance lines and adopting explicit instruction technique. Besides, material developers may take the outcome of the research into consideration and could find out possible ways to distribute and introduce DCs systematically across the educational levels