Abstract
Abstract With the rise of Internet technology, e-mail exchange has become a popular pedagogical task and is regarded as one of the efficient communicative means, and thus adopted by lots of language instructors and researchers. Many studies mentioned the possible effect of e-mail exchanges on students’ writing skills, but rare support was provided by empirical investigation. Further, the findings from most previous studies could only be generalized in the college context. Therefore, this study was conducted to provide evidence for the effectiveness of e-mail exchanges on writing of EFL senior high school students through empirical textual analyses. In the present study, 28 senior high school students were paired randomly with a group of college English majors. From September to December 2002, the 14 dyads exchanged weekly e-mail based on assigned topics for 8 times in a 14-week period. Besides e-mail writing, pre- and post-project in-class compositions were collected. To investigate the writing development in senior high school students’ e-mail and in-class writing, two types of measurements were used to analyze students’ 2nd and 8th e-mail entries, and the two in-class compositions. The revised ESL Composition Profile (Jacobs, et al, 1981) analyzed students’ writing in content (30%), organization (20%), vocabulary (20%), language use (20%), and mechanics (10%). The t-unit analyses (Hunt, 1965) analyzed the writing data in terms of length, accuracy, and syntactic complexity. T-tests were used to compare the mean scores in each subcategory. In addition, to investigate the background information of the subjects, and whether both of the less capable learners, the senior high school students, and the more capable learner, the college English majors, felt positive about the e-mail exchanges project, a Background and an Evaluation Questionnaires were used on both groups before and after the project. The results showed that e-mail exchanges as a main component of writing training facilitated senior high school students’ overall writing ability, especially in the in-class writing context. Students’ in-class writing progressed obviously in total grades, content, organization, language use, and syntactic complexity, while the decrease only occurred in the total length by words. And their e-mail writing performance was also slightly enhanced in content, organization, language use, and mechanics. Furthermore, the senior high school students felt positively toward the project and the effects on their in-class writing ability. They also showed slightly more self-confidence in their own writing ability and less writing apprehension after the project. As for the college key pals, although they were less interested in the e-mail project than the senior high school students did, they perceived some positive effects of the e-mail exchanges project. According to our study, the instructors are recommended to take some steps in advance to make an e-mail project more efficiently. For example, holding a training session for students who are not familiar with computer and e-mail, selecting appropriate key pals for students, manifesting the schedule in advance, leaving the e-mail address of the instructor/researcher for the students, and keeping connection with the instructor of the key pals. Besides, to prevent writing block, it was suggested to survey students’ favorite topics, and then provide more than one topic for e-mail writing every week, or the instructor could leave some weeks for students to select their topics freely. For future studies, four suggestions are provided. First, effects of e-mail beyond the writing skill should be investigated. Second, the effect of scaffolding can be analyzed further. Third, an e-mail project can be incorporated with some other tasks to increase subjects’ motivation. Fourth, the connection between individual learner variables like learning strategies and the preference of e-mail exchanges should be emphasized. At last, to link students’ actual performance and their perceived improvement, future studies were suggested to collect data from both the self-report questionnaires and students’ actual performance.