한국어 학습자의 등급별 오류 분석

An Analysis of Korean Learners’ Errors by Proficiency: Focus on Statistical Analysis using the Multinominal Logistic Regression Model

This study attempted to grasp the characteristics of errors according to the proficiency level of Korean learners by statistically analyzing the corpus in large-scale learners. This study examined the characteristics of proficiency-level errors in the location, pattern, and level of errors in the learner’s corpus using the multinominal logistic regression model along with observing the occurrence rate. As a result, some errors showed that the frequency or probability of occurrence decreased significantly as the proficiency level increased, but there were some types of errors that appeared steadily. Through this study, the characteristic types of proficiency-level errors were presented. This study is significant in that it statistically searched for significant variables at each stage of learning by comparing and analyzing the overall errors using large-scale learner balanced corpus data.