Comparing the effects of direct and indirect synchronous written corrective feedback: Learning outcomes and students' perceptions

Abstract

From a sociocultural perspective, collaborative writing tasks offer opportunities to negotiate in decision-making processes while also sharing responsibility for the production of a single text (Storch, 2013, Collaborative writing in L2 classrooms. Bristol: Multilingual Matters). Although research has found benefits for such tasks (Storch, 2019, Lang Teach, 52, 40-59; Taguchi & Kim, 2016, Appl Linguist, 37, 416-437), variation in how different types of synchronous written corrective feedback (SWCF) in such a setting impacts students' language learning and their perception of SWCF has not been examined. This study compares the role of direct and indirect SWCF during collaborative writing tasks on the learning of Korean among high beginning-level students. The study was conducted during an existing beginning level Korean course and focused on three textbook units over 6 weeks. Fifty-three learners of Korean were assigned to one of three conditions: no feedback, indirect SWCF, or direct SWCF. All students completed a pretest, two collaborative writing tasks, a posttest, and a survey for each unit. At the end of the third unit, students in the feedback groups also participated in an exit interview session. The study targeted a total of 12 linguistic features (e.g., auxiliary verbs), which were the focus of the three textbook units. The amount of feedback and students' responses to the feedback were analyzed in terms of the accuracy of resolution on their writing. Students' learning of these linguistic features was measured using written production tests, and their responses were coded for suppliance of target forms and accurate production of them. Our findings indicate that direct SWCF was more useful in helping students produce accurate writing, but both feedback types were effective in promoting learning of new linguistic features through collaborative writing. In terms of students' perceptions, there seemed to be no difference between the two feedback conditions. The findings of the study are discussed in light of effective ways to implement collaborative writing tasks as well as the provision of SWCF.