Analysis of Error Type and Feedback Efficacy on Genre-Based Writing: Narrative Writing*

Abstract

The current study aims to analyze EFL learners' error types and teachers’ feedback efficacy in English narrative writing. 60 participants divided into two proficiency groups by TOEIC scores were asked to write each of two narrative essays in class and correct each language-form error in their drafts based on two kinds of feedback (explicitly coded and uncoded). Errors in each group’s drafts and revisions were statistically analyzed based on the error types and feedback methods. The findings of the study are as follows: 1) Overall learners’ writing accuracy and fluency analyzed by tokens and error rate after revision show a significant improvement. 2) Detailed error analysis on narrative writing indicates that in the draft, the error types related to verbs (particularly past tense) are more frequent than others, but in the revisions, the correction activity using feedback shows a positive effect on error types relevant to verbs, one of treatable errors. 3) Students with better English proficiency, given more explicit “coded” feedback, can significantly reduce error rates in both drafts and revisions. This suggests teachers to consider troublesome error types as well as feedback methods when giving error feedback to EFL writers.