The Effects Of Pair Writing In L2 Classroom

Writing is generally thought to be an individual activity, however, many advantages have been found in collaborative writing in the second language classroom (Stoch,2005; Swain & Lapkin,1998). Previous research suggests that pair writing helps resolve language-related problems and produces more accurate written texts than individual writing.This study is a basic step in investigating the use of collaborative writing in second language classroom of non-English major students in a university in Shanxi Province. It compares the performance of two groups of second language learners:one group (N=21) worked individually, and the other group (N=21) worked in pairs. They worked on the same writing task in the same amount of time about the topic of online shopping. The first research question concerns whether paired writers perform better than individual writers in content, organization, grammar, vocabulary and mechanics. By using the analytic scoring method and T-test, this study concluded that paired writers performed better than individual writers in the aspect of content, organization and vocabulary. However, there was no significant difference in the aspect of grammar and mechanics. The possible explanations are the poor performance of those students who participated in the study and the time allocation for the two groups. This is the first research question of this essay.The second research question is based mainly on the recordings of the paired writers. According to the transcription of their recordings, language-related episodes (LREs) produced by the pair writers were analyzed. It tries to figure out how many language-related episodes (LREs) are produced in form, lexis and mechanics and whether the LREs are correctly resolved, unresolved or incorrectly resolved? The results showed that of the 714 LREs produced,60% were lexical-LREs, and form-focused and mechanical-LREs accounted for 25% and 15% respectively. We also investigated the outcome of LREs and found that 60% of them had been resolved correctly. The remaining was resolved inappropriately, with 17% unresolved and 23% resolved incorrectly.This study makes an effort to compare the performance of paired writers and individual writers. Although the outcomes of the present research may not be readily applicable for all writing tests, the effect is both predictable and significant within this context. What’s more, when the process of pair writing is studied clearly by focusing on LREs, it will be helpful for test design and execution in classroom writing practice and it will be likely that students can improve their writing ability when working in pairs.