Teaching Cohesive Devices In EFL Writing For Non-English College Students

The primary objective of this study is to apply cohesion theory in EFL writing instruction for college students. It also investigates the influence of teaching cohesive devices to other aspects in writing measurement including grammar, vocabulary, content and mechanics; the different learning strategies in the cohesive devices training and the possible causes of cohesion errors.55 non-English major freshmen of two intact classes participate in this study. The 55 subjects are of a similar age, ranging from 17 to 20 years. They are Chinese students of English as a foreign language. At the time of the research they had been studying English for at least 6 years before entering university. The two classes, equivalent in terms of age group, previous exposure to EFL and language proficiency, were chosen as experimental group (EG) and control group (CG) respectively. A quasi-experimental design was adopted in the duration of 16 weeks. EG received cohesive devices training and was required to perform writing practices carefully designed to encourage the use of cohesive devices. CG received writing instruction without the treatment for cohesive devices. Several data collection methods were employed in the study, namely, pretest and posttest, interviews and questionnaires. The results of pretest indicate students from both groups did not use cohesive devices adequately when they wrote a composition in English. But there is a significant difference between CG and EG in posttest aroused by the treatment, which indicates students who had cohesive devices trainingwrite more logically and coherent than those students without it Meanwhile, the teaching of cohesive devices has positive impact upon other aspects in the writing measurement, including content, grammar, vocabulary and mechanics.It is discovered that cohesion errors by subjects in this study mainly are of three kinds: 1) misuse of conjunctive type, such as overuse of and, underuse of the inferential, transitional and summative subtypes, etc.; 2) improper use of reference type, for instance, misselection of this and that, personal reference in disaccord; 3) cohesion errors by substitution type. Moreover it indicates the possible causes of cohesion errors rest with incorrect application of cohesion rules, language transfer and insufficient language resources available.The investigation findings imply that 1) the teaching of cohesive devices can direct learners' attention to both meaning and form, 2) the teaching of cohesive devices should aim to combine the meaning and the form instead of focusing on the form of cohesion solely, and 3) the teaching of cohesive devices does not deny the necessary instruction for grammar and vocabulary.In a word, writing instruction should attach weight to cohesive devices training and the application of cohesion theory to teaching is very feasible.