A Longitudinal Study on the Effects of Revision on Essays of Japanese Students

Abstract

The purpose of this study is to examine the long-term effects of revision oh essays of Japanese EFL learners as follow-up of Oikawa and Takayama (2000). Thirty-eight senior high school students are required to write English essays within 10 minutes in class over a nineteen-week period. They submit their essays each time and one week later the essays are given back with grammatical errors underlined. The students are divided into two groups, defending on whether or not they are required to rewrite the essays, as shown below. Group A: required to rewrite the essay Group B: no need to rewrite the essay, but required to write a new essay The pre- and post-tests are evaluated holistically by two EFL writing teachers and analyzed with the following criteria: (1) text length and (2) mean T-uhit length for fluency measures and (3) numbers of error-free T-units, (4) mean error-free T-unit length, (5) error-free T-units / T-units ratio and (6) grammatical error percentage for accuracy measures. As for (3), (4), (5), (6), additional criteria, where local errors are counted out, are used. A questionnaire is used to find the students' attitudes toward the revising practices. The results indicate the grammatical error percentages of both groups get smaller after the long-term writing practices whether or not the essays are revised. Compared with Group B, however, Group A has smaller grammatical error percentage. When local errors are counted out, only Group A shows no significant change in the grammatical error percentages. It may be presumed Group A has come to pay more attention to local errors after the revising practices. Group A also shows the increase in mean T-unit length, suggesting that revising practices may contribute to writing more structurally-complex sentences. The students' attitudes toward the revising practices are found out to be positive.