| Cohesive devices, as the mechanism to realize cohesion, refer to words or phrases that signal the connections between T-units or parts of a text. There have been a great number of empirical studies on cohesive devices in writing, but the results are too mixed to yield consistent conclusions about how the use of cohesive devices indicates learners' proficiency level or the quality of their writing. Meanwhile, few empirical studies attempted to address the ESL learners in China due to the difficulties in data collection and analysis.The present quantitative study was designed to investigate the use of cohesive devices in Chinese ESL learners' English writing from a quasi-developmental perspective. The specific questions are: a) to examine where there are significant changes in the use of cohesive devices between students at different grade levels; b) to determine whether there are significant differences between different writing topics; and c) to examine the relationship between the use of cohesive devices and writing quality. The subjects were 120 randomly selected undergraduate English majors with 30 in each of the four grade levels, from whom 240 compositions were collected on two separate topics in writing classes towards the end of the second semester. Fourteen cohesive devices were adopted in the analyses of the compositions, whose frequencies were counted and then controlled for length by being transformed into the frequencies per T-unit. In addition, all the compositions were holistically rated based on TOFEL scoring schemes by three independent raters.ANOVA analyses and/or non-parametric tests were performed to compare the use of cohesive devices by the four grade levels, from which significant linear increase was found on lexical cohesive devices, but a significant decline from low grade levels to high grade levels emerged on several grammatical cohesive devices. The increase in lexical cohesive devices supports the prior finding that learners tended to employ more lexical means to maintain the coherence of their writing as their proficiency levels improved. The decline in several grammatical cohesive devices was suggested related to the properties of the devices. These devices have been proved in previous research more characteristic of oral discourse than written discourse. Since theories on writing development suggest a transition from oral to written language forms in L2 writing development, it is likely that these devices appear more prevalently in the writing of lower-proficiency writing, as evidenced by the result of the study. The use of cohesive devices between the two writing topics was also compared through MANOVA analyses with repeated measures, from which significant differences emerged on devices for reference, demonstrative noun phrases, additive conjunctions, causal conjunctions and synonyms. Factors including demands of writing tasks and learners' background knowledge were suggested contributing to the differences.Besides, correlation analyses were performed for each writing topic between the frequencies of cohesive devices per T-unit and the writing scores, from which lexical cohesive devices showed the strongest positive correlation with the writing scores, and significant negative correlation was found on several grammatical cohesive devices. However, all the correlation is rather low, which indicates the frequency of cohesive devices has little power for discriminating writing quality. The present study discussed developmental trends in the use of cohesive devices in English majors' L2 writing in China. It is suggested that language teachers and learners should be cautious about not to make overabundant use of inappropriate cohesive devices. Special attention should be drawn to the characteristics of the devices and which kinds of discourses they should be appropriate for. More studies are needed addressing more language groups and other measures of cohesive devices to deepen the study of L2 writing development.
|
PDF Full Text Request