英语专业八级考试汉译英项目评分员研究——以2015年TEM8为例

An Examination of Rater Performance on the Chinese-to-English Translation Section of TEM8 in 2015

文章通过对所有56位评分员的评分实验以及对其中四位评分员的深入实验和评分回溯,间接地考察了2015年全国高校英语专业八级考试汉译英项目的评分。研究得出:本次评分的评分员间信度较好,对高质量文本和低质量文本评分都具有较好的相对一致性,但在对低质量译文评分时自由度较大;同时,译词的选用、译句的形式变化以及译文的语法正确性是评分员实际整体评分时的主要考虑因素,虽然这些方面可能存在交互作用,但其中对译句结构的考量要略重于对译词选用的考量;此外,评分员对译文中出现的连逗句这一语法失误还关注不够。文章最后对提高八级考试汉译英的评分效度提出了建议。

This study, integrating both quantitative and qualitative approach, indirectly examined rater performance on the section of Chinese-to-English Translation of TEM8 in 2015 by elieiting simulated-experimental ratings from all 56 raters as well as exercising an in-depth probing of the data from four representative raters. The results showed that the Chinese-to-English translation of TEM8 in 2015 exhibited acceptable inter-rater reliability. In particular, high relative agreement was achieved on either end of quality spectrum, although the absolute agreement on low-quality texts was yet to be raised. Meanwhile, it is believed that the raters＇ holistic scorings were largely motivated by their considerations for three factors -- rendering of lexis, transformation of sentence structures, spelling and grammatical accuracy. It is also believed that slightly greater weight was placed on the sentential level than on the other two, even though the effect of three-way interactions cannot be ruled out. In addition, raters were found to have paid insufficient attention to run-on sentences in translated versions. The study finally put forward some suggestions regarding the improvement of rating validity of TEM8＇s Chinese-to-English translation section.