영어 발음평가에서 한국인 평가자의 영어 숙련도에 따른 타당성 및 신뢰도 (Korean Raters' Viability and Reliability on Pronunciation Evaluation Depending on Raters' English Proficiency)

서지정보

· 발행기관 : 한국외국어대학교 언어연구소
· 수록지 정보 : 언어와언어학 / 51호 / 71 ~ 94페이지
· 저자명 : 윤원희

초록

Accuracy judgment of non-native speakers' English pronunciation may vary depending not only on what criteria raters set up in the pronunciation evaluation but also on whether or not raters are native speakers. As non-native raters, raters' English proficiency could be a factor disrupting the consistency of the evaluation. Through Cronbach's alpha and corrected item-total correlations as well as open correlations this study aims to see if there is any difference in consistency and evaluation patterns between three types of raters grouped by English proficiency. The results show the highest proficiency group composed of teachers is more reliable than the other two groups, suggesting a better consistency can be achieved by evaluation training and consistency is not much related to the raters' English proficiency. We also found that the evaluation pattern of the teacher raters is similar to the higher proficiency student raters, suggesting high proficiency raters may share the implicit and intuitive pronunciation norm referred when raters set up their evaluation criteria. This implies that the different proficiency raters may have different pronunciation norms.

영어초록

Accuracy judgment of non-native speakers' English pronunciation may vary depending not only on what criteria raters set up in the pronunciation evaluation but also on whether or not raters are native speakers. As non-native raters, raters' English proficiency could be a factor disrupting the consistency of the evaluation. Through Cronbach's alpha and corrected item-total correlations as well as open correlations this study aims to see if there is any difference in consistency and evaluation patterns between three types of raters grouped by English proficiency. The results show the highest proficiency group composed of teachers is more reliable than the other two groups, suggesting a better consistency can be achieved by evaluation training and consistency is not much related to the raters' English proficiency. We also found that the evaluation pattern of the teacher raters is similar to the higher proficiency student raters, suggesting high proficiency raters may share the implicit and intuitive pronunciation norm referred when raters set up their evaluation criteria. This implies that the different proficiency raters may have different pronunciation norms.

참고자료

· 없음