Collaborative writing assignments and on-line discussions in an advanced ESL composition class

Abstract

The purpose of this study was to evaluate the effects of collaborative computer-mediated projects on students' writing performance. The subjects of the study were 20 advanced English as a Second Language (ESL) composition students enrolled in an English 107 course at the University of Arizona in the Spring 1999 semester. The class met twice a week: once in a regular classroom, once in the College of Humanities Collaborative Learning Laboratory (COHlab). The study addressed three main research questions: (1) Does student participation in on-line synchronous discussions vary in different configurations of discussions and for different thinking styles? (2) Does writing performance vary between collaborative and independent tasks? (3) Is there a change in students' attitudes to collaborative assignments and to the use of computers in class from the beginning to the end of the semester? To answer the first question, the researcher analyzed the transcripts of on-line discussions and essays written collaboratively and independently. Discussions were conducted in different configurations (whole class vs. small group and anonymous vs. non-anonymous). To collect discussion data, an archive feature was used that provided complete transcripts of discussions, including students' names (or random numbers in anonymous discussions) and time when each comment was written. Repeated measures MANOVA and qualitative analyses were used to examine the data. The second question, investigating the difference in writing performance between independent and collaborative academic writing tasks, was addressed by (1) examining the results of textual analysis performed by the computer, and (2) examining the results of holistic evaluation conducted by ESL raters. Repeated measures MANOVA was used to analyze the data. To answer the third question, pre-semester and post-semester student surveys were used, as well as course evaluations. The students' thinking styles were measured using Sternberg's Thinking Styles Questionnaire. Qualitative analysis was used to examine the data. The results of the study indicated no difference in the discussion content, amount of communication, or interaction dynamics between the discussions of different configurations, or between the students with internal and external thinking styles. In terms of differences in writing performance between collaborative and independent essays, no difference was found by discrete-point computerized text analysis. However, ESL raters evaluated collaborative essays higher than independent ones. Finally, the results of the analysis of the students' attitudes indicated that, in general, the students' attitudes to the use of computers have improved over the course of the semester. This finding is important, as previous research has shown that positive attitudes lead to increased motivation, and increased motivation, in its turn, leads to more favorable learning outcomes in an L2 classroom.