| Summary: | 碩士 === 國立高雄師範大學 === 英語學系 === 95 === This study investigated a key issue in the implementation of the Common European Framework of Reference for Languages (CEFR) in Taiwan. Specifically, this study researched the accuracy and reliability of self-assessments by young Taiwanese EFL students (12-17 years old) when compared against a test-based criterion assessment. The 322 subjects in this study were junior high and senior school students from four regular public schools in Taiwan. The instruments in this study included two versions of a self-assessment instrument comprised of the CEFR global descriptors for levels A1 to B1 (junior high version) and levels A1 to B2 (senior high version), translated into Chinese, and the practice version of an online English proficiency test reciprocal to the CEFR (NETPAW). Three levels of the NETPAW exam were utilized: Basic, Elementary and Intermediate, reciprocal to CEFR levels A1, A2 and B1, respectively. The scores of the NETPAW Phase One exams (receptive language skills) were analyzed statistically to ascertain trends in English test performance. The self-assessment data were also analyzed statistically to ascertain trends in self-perceived English abilities. Both types of assessment were then compared statistically to derive validity coefficients for the self-assessments. In addition, the reliability of the self-assessment instruments was analyzed statistically. In total, four phases of analysis were employed to understand how well the subjects discern distinct constructs of English proficiency inherent in the CEFR descriptors, and how well the subjects apply their understanding of these constructs to themselves. Based on the analyses herein, the findings are as follows: 1. The CEFR global descriptors in translation can accurately convey distinct constructs of language proficiency to young Taiwanese EFL students. 2. Young Taiwanese EFL learners experience difficulty applying the constructs of language proficiency to themselves accurately. 3. The descriptors comprising CEFR A1 work well, as they produce consistent responses from all subjects. 4. The descriptors comprising CEFR levels A2 and B1 are the focal points of response effects that impede the subjects’ self-assessments. 5. Young Taiwanese EFL learners exhibit a salient tendency to accurately perform self-assessments using the CEFR global descriptors in translation, although the correlation with test-based assessment is modest. 6. CEFR A1 is a suitable proficiency level for junior high first and second graders, whereas CEFR A2 is suitable for junior high third graders and senior high first graders. In addition to these findings, pedagogical implications and suggestions for future research into the implementation of the CEFR guidelines are discussed. Teachers should consult the CEFR in order to critically evaluate their classroom instruction in terms of content and quality. Teachers should reflect on the type and amount of task-based instruction they utilize and also the frequency of use. Students should consult the CEFR to familiarize themselves with the process of self-assessment, which in turn, will provide an impetus for self-directed learning. Future research should focus on introspective methods in order to gather in-depth qualitative information about the psychological and affective factors influencing the accuracy of self-assessment by young Taiwanese EFL learners.
|
|---|
