The present study aimed to investigate the participants’ overall English proficiency through T-unit analysis, the use of English connectors and learners’ perceptions regarding their usage. The participants of the present study included students enrolled in English composition classes in successive semesters.
As a research instrument for the study, three students in the pre-intermediate English composition class participated in semi-structured interviews and 17 students in the low-intermediate English composition class participated in Zoom interviews. The qualitative data was analyzed and coded using NVivo R1. Overall, English writing skill was analyzed using SCA (Syntactic Complexity Analyzer) and the status of five participants were individually analyzed their status for one year. 126 essays written by students in pre-intermediate English composition and 136 essays produced by students in low-intermediate English composition were collected. Students’ writing data was used to identify overuse and underuse of connectors compared to LOCNESS.
The findings of the study were as follows: First, the average length of T-units was 14.023 and the average length of clauses was 9.580 in pre-intermediate English composition. These results are in line with the comparison studies. However, the average length of clauses in low-intermediate English composition was slightly higher compared to LOCNESS of Lu and Ai (2015).
Second, students in both English composition courses overused connectors in comparison to LOCNESS. Overused connectors both in pre-intermediate English composition and low-intermediate English composition were ‘so, when, finally, first, in addition, in conclusion, to sum up’. In LOCNESS, ‘because’ was identified as an overused connector compared to low-intermediate English composition results.
Next, students learned English connectors by using their own notes, checking the use of connectors through examples, making lists of connectors, and from examples presented in PPT’s, and multiple writing assignments opportunities to learn connectors. Participants were more likely to have learned connectors when studying abroad than in Korea.
The participants recognized the importance of connectors and the roles of connectors in writing. They said that they needed to learn connectors due to incorrect use of connectors and repetitive use of the same connectors rather than the importance of connectors. In addition, a student in pre-intermediate mentioned the problems of repetitive connector use. When they were asked about the difficulty of learning connectors, the participants did not seem too difficult, but they did point out the difficulty of using appropriate connectors and the problem of using the same connectors repeatedly.
Finally, the study revealed that participants had trouble distinguishing between contrastive and additive connectors in context, and it was found that ‘moreover’ and ‘furthermore’ were used without distinction. Also they could not distinguish between ‘however’ and ‘nevertheless’.
Based on these findings of this study, the pedagogical implications of this study are as follows. First, while many studies have focused on analyzing learners’ overuse and underuse of connectors in a one-time period, this study has the advantage of examining connector use over a period of time from pre-intermediate to low-intermediate composition. Second, it is significant that the improvement of the five learners from pre-intermediate to low-intermediate composition was analyzed, so it is possible to determine the approximate improvement of the learners after two semesters of English composition classes. Also this study investigated the perceptions related to learning connectors to understand the difficulties learners face in learning connectors and their preferred methods of learning connectors.