写作能力行为测试评分员培训研究述评

A Study of Rater Training in Writing Performance Assessment

在大规模语言考试实践中,评分员培训通常被认为是作文人工评分的必经环节,其目的在于提高评分一致性,保证考试公平性。然而,语言测试界对于评分员培训目的以及自身价值的争论颇多,本文将对这些观点进行梳理。同时,还将对评分员培训步骤、评分员培训效果与持续时间、培训中评分员认知发展过程等相关研究进行探讨。

In the practice of large-scale language assessment, rater training is often regarded as an essential part in human essay rating process. It is a measure employed by rating institutions for the aim of improving rater consistence in order to ensure fairness in the exam. However, there are various views on the purpose of rater training and debates over its value in language assessment field. The article first examines these views and then goes on to review studies of rater training procedures, the effects and duration of rater training, and rater＇s cognitive development over the training process. It is obvious that although rater training may not be as ideally efficient as trainers have expected, it is still useful in improving rating validity when carefully designed and administered.