Effectiveness and inclusiveness of locally adapted human-delivered synchronous written corrective feedback for English referential articles.

The present paper reports on the effectiveness and inclusiveness of human-delivered synchronous written corrective feedback (SWCF) in paired writing tasks. Replicating Yamashita, Study 2 and Study 3 each conducted a classroom-based quasi-experimental study in an English-as-a-Second-Language (ESL) writing program at an American university. In Study 2, 50 learners were assigned to either of the two experimental groups (direct, indirect) or the control group based on their language analytical ability (LAA). All the three groups worked on two in-class computer-mediated animation description tasks in pairs for 50 minutes. The two experimental groups received either direct or indirect SWCF on their uses of the referential articles during the tasks, while the control group did not. Learners' article use was assessed with an animation description test and a sentence rewriting test on a pretest (one week prior to the treatment), posttest (a few days after), and delayed posttest (two weeks after). Inclusiveness of SWCF was explored by examining changes in performance of high LAA and low LAA learners separately. Study 3 recruited another cohort of 51 learners from the same writing program and replicated Study 2 with a modified test in place of the sentence rewriting test. Findings suggested that only the direct group improved to a greater extent than the control group, and this finding was found only in Study 2, potentially questioning the effectiveness of SWCF in this context. However, SWCF improved low LAA learners' article use to a comparable extent to high LAA learners, contributing to inclusive classroom teaching. [ABSTRACT FROM AUTHOR]