THE EFFICACY OF WRITTEN CORRECTIVE FEEDBACK ON KFL LEARNERS’ WRITING

Abstract

The main purpose of this study was to investigate the effectiveness of written corrective feedback (WCF) in a Korean as a foreign language (KFL) context. The assumption that indirect feedback (IF) is more effective for L2 learners with higher-proficiency levels has yet to be fully corroborated, which necessitates substantial research-based evidence to validate it. In a KFL context where direct feedback (DF) is the most dominant type of WCF, this study implemented dynamic written corrective feedback (DWCF) that provided two types of feedback (DF and IF) depending on error types and explored its effect when compared to DF. Also, this study examined the impact of different types of WCF on students’ progress in acquiring three targeted grammatical structures (particles, clausal connectives, and final endings). Students (N = 77) enrolled in intermediate, high-intermediate, and advanced Korean classes were asked to write nine short texts during classes. DF (n = 28) and DWCF (n = 38) were provided for the two treatment groups respectively while only encouraging comments were provided for the control group (n = 11). To examine the relative effectiveness of WCF, accuracy, complexity, and fluency scores measured in the pretest writing were compared to those in the posttest writing at the end of the semester (RQ1). After the accuracy scores of the three targeted grammatical features were obtained in week 4, 7, 11, and 14, statistical analysis was performed to examine whether certain categories of errors were amenable to WCF (RQ2). Although the differences among the three groups did not reach levels of statistical significance (p > .05), the results of a two-way repeated measures analysis of variance (ANOVA) revealed that the two treatment groups outperformed the control group in terms of mean accuracy scores over time. No significant difference in either complexity or fluency scores among the three groups was observed (RQ1). The ANOVA and the generalized estimating equations (GEE) results indicated that there was no differential effect for WCF on the three targeted grammatical errors (particles, clausal connectives, final endings) (RQ2). The findings of this study suggest that the results of WCF studies conducted within an English as a second language (ESL) context should not be blindly applied to determine more effective instructional methods for KFL learners’ writing development. Empirical evidence from well-designed studies on the longitudinal effects of WCF on KFL learners’ writing can provide significant implications for how to best use WCF in a KFL context. Keywords: direct and indirect feedback, dynamic written corrective feedback