Instructor feedback on free writing and automated corrective feedback in drills: Intensity and efficacy

Abstract

This study investigated the intensity and efficacy of automated corrective feedback (CF) in a tutorial CALL (computer-assisted language learning) environment during form-focused drills as compared with those of instructor CF on free writing in the classroom. The English simple past tense, a previously learned target structure, was selected as the target structure. A hundred and fifty-two Chinese intermediate learners of English as a foreign language (EFL) were randomly assigned to six groups and completed three writing tasks: a pretest, an immediate posttest, and a delayed posttest. The six groups included two automated feedback groups (direct correction on an error-correction test and metalinguistic feedback on an error-correction test), two feedback-on-writing groups (direct correction and metalinguistic feedback), a group who completed the error-correction test with no feedback, and a control group who participated in none of the interventions. The results showed that the error-correction test provided the participants with more intensive CF exposure than the writing task. However, the groups that received CF in the writing task were the only groups that performed better than the control in subsequent written production. Post hoc analyses of the production data revealed that the metalinguistic-feedback-on-writing group’s improvement could be attributable to error avoidance.