语用能力测试的评卷对比研究

Comparing native and nonnative speakers' scoring in an interlanguage pragmatics test,

本研究利用多层面Rasch模型,对中外老师在语用能力测试评卷中的表现进行了对比研究。结果表明,所有中外老师的评卷都具有较好的内部一致性,但评卷的严厉度有显著差异,中国老师比外国老师要明显宽松。总体上,中外老师都能较好地使用评分量表,但在评分量表的某些方面出现了分阶无序的现象。中国老师在言语行为、信息量和得体性等方面都出现了分阶无序现象,而外国老师仅在言语行为方面有此现象。

This article reports on a comparative study that looks into native and nonnative English speakers' scoring in an interlanguage pragmatics test via the many-facet Rasch model. Eight raters were invited to rate the responses from 38 students in a university in China according to the analytic rating rubrics developed by Hudson et al. (1995). Results revealed that both the native and nonnative English-speaking scorers, though fairly consistent in their overall ratings, differed strongly in the severity with which they rated the students. The NNSs were found to be more lenient but less consistent than the NSs. However, both types of scorers demonstrated some inconsistency in relation to the rating dimensions which included speech act, amount of information, expression, and appropriateness. Specifically, step disordering occurred for both the NSs and the NNSs in the speech act, but only for the NNSs in the appropriateness and in the amount of information.