Reformulation as a form of response in an ESL academic writing programme

Abstract

Feedback plays an important role in the teaching of second language writing. There has, however, been relatively little research carried out to examine the exact nature of the relationship between teacher response and written outcomes in second language classroom settings. One area of interest is the relationship between the use of reformulation - a technique for providing feedback to advanced writers - and the way ESL students revise their drafts during an academic writing task. It has been claimed that reformulation is relatively more successful in promoting discourse level revision of text than more conventional forms, such as feedback using comments and an error coding scheme. A study was therefore designed to assess this claim. The subjects in the study were 34 adult ESL learners, from two classes on an English Proficiency Programme. The majority of the students were preparing for English medium tertiary study in New Zealand. A ten week writing programme was planned, providing practice in the types of expository and argumentative writing tasks required of university level students. The subjects were randomly assigned to an experimental group and a comparison group for the purpose of writing instruction. The programme for the two groups was identical, except for the nature of the feedback each group received on the first drafts of each writing task, and the person giving the feedback. The teacher of the experimental group used reformulation, while the teacher of the comparison group used a conventional responding method. The experimental design enabled the collection of both product and process data at three points in the writing programme. The results showed that subjects as a whole improved in their ability to write more fluent, coherent and accurate texts over the ten week period. Against prediction, however, the comparison group wrote significantly longer and more coherent texts than the reformulation group on their second drafts. Compared to the use of a conventional method of response, reformulation could not therefore be said to play a greater enabling role in improving the quality of final text in this study. Process data gathered from the two groups, and from four case studies, indicates that there are factors beyond the particular responding method that are equally important in encouraging students to restructure text. These include, for teachers, the nature of the pre- and post writing support given, and for students the motivation to engage with the task more deeply. The results of the study therefore suggest that teachers may need to look at all factors associated with a responding technique if they are to help students to improve the quality of their writing. The study thus underlines the value of observing the effects of techniques in classroom contexts. It also raises questions about the best ways of approaching the design of academic writing tasks and of enhancing revising practices. A problem-solving rather than an a priori framework for both tasks and responding practices is recommended.