The Effects Of Proficiency Pairing On Chinese Efl Learners’ L2 Learning In Collaborative Writing

Though collaborative writing enjoys solid theoretical and empirical support,collaborative writing tasks have been taken inadequate advantage of in our Chinese EFL(English as a Foreign Language) teaching and learning context, and hence their learning advantages have been inadequately researched academically.The present study is intended to investigate the effects of proficiency pairing on Chinese EFL learners’ L2 learning in collaborative writing. Three research questions were addressed:(1) What kind of language learning is involved in Chinese EFL learners’ collaborative writing?(2) To what extent do Chinese EFL learners retain their language learning as measured by dyad-specific posttests?(3) What is the effect of proficiency pairing on Chinese EFL learners’ L2 learning in collaborative writing?To answer the three research questions, we invited 24 Chinese EFL learners to participate in the study. The 24 students formed 12 dyads according to their English proficiency, including 4 higher-higher dyads, 4 higher-lower dyads and 4lower-lower dyads, who formed three groups, named as H-H group, H-L group and L-L group correspondingly. The two members in each dyad were requested to collaboratively complete a writing task. Their interactions were recorded and their transcribed dialogues were analyzed in terms of language-related episodes(LREs).Each participant received a tailor-made dyad-specific posttest(i.e. an error correction task) ten days later.Based on the analysis of the number, type and outcome of the LREs that had occurred in the participants’ interactions as well as participants’ performances on the posttests, three major findings were achieved:(1) Collaborative writing did arouse learners’ language learning for the participants produced 181 LREs in total,including 132 lexical LREs and 49 grammatical LREs. Not only did they discuss language issues, the participants also correctly resolved 69.61% of the linguistic problems.(2) The participants also retained their language learning as measured by the dyad-specific posttests for they retained 59.01% of the correctly resolved LREs.Besides, the participants worked out 41.5% of the unresolved LREs and 33.68% of the incorrectly resolved LREs, respectively, on the posttests.(3) As for the cross-group difference, the L-L dyads produced the most LREs, followed by the H-H dyads and the H-L dyads. The H-H dyads had the highest percentage of lexicalLREs(86.21%) and the lowest percentage of grammatical LREs(13.79%), while the H-L and L-L dyads showed a similar tendency of focusing more on lexical LREs than grammatical LREs. However, no significant difference was found between the three groups’ distributions of the four subcategories of lexical LREs or between the three groups’ distributions of the seven subcategories of grammatical LREs. As to the three groups’ LRE outcomes, the H-H dyads correctly resolved 77.59% of the linguistic problems, followed in a descending order by the L-L dyads(66.18%) and the H-L dyads(65.45%). And a significant difference was discovered among the three groups’ distributions of the three outcomes. With regard to language learning retention, the H-L and H-H dyads performed much better in their retention of correctly resolved LREs than the L-L dyads. As for the incorrectly resolved and unresolved LREs, the three pairing groups were similar in their resolution of the former one with an average accuracy rate around 30%, while the L-L dyads performed worst in solving the unresolved LREs on the posttests.The study can provide at least two pedagogical implications for Chinese EFL teaching and learning. On one hand, the present study proved the applicability of collaborative writing in Chinese EFL context. Therefore, such collaborative tasks,especially writing tasks, could be adopted in our later teaching, in order to help enhance learners’ L2 learning in Chinese EFL context. On the other hand, paring students in collaborative writing is of great importance for their learning results.Generally, pairing higher proficiency learners is always a good choice. Meanwhile,the pairing of lower proficiency learners is not truly a bad trial, because the learners of such proficiency pairing could produce better learning results than we’ve expected. In some cases, the interaction pattern of the learners, whether they are collaborative or non-collaborative in their tasks, might be more influential than the effects of the measured proficiency pairing on learners’ language learning.