Key points are not available for this paper at this time.
A total of 38 problem learners comprising adolescents with learning disabilities and low achievers participated in this study. Of these, 20 were untrained control participants; the remainder were trained to plan, write, and revise opinion essays. Trainees were randomly divided into pairs and taught to collaboratively use interactive dialogues to plan and revise their essays (but they wrote their essays separately). The results from multivariate analyses of variance (MANOVAs) and effect sizes indicated that trainees improved significantly from pretest to posttest in clarity and cogency of written opinion essays, and maintained such gains. Additional analyses indicated that they surpassed the untrained control group. Questionnaire data indicated trainees improved in self-efficacy of writing only. The same finding held true in comparison with untrained participants. The discussion integrates the trainees' performance data with the questionnaire data and considers the research implications and limitations of the study.
Wong et al. (Fri,) studied this question.