听写测试的试验评分法--"部分得分评分制"及初步实验

TEM4听写采用的是较传统的数错扣分法。数错扣分法是负分法，其中存在一些问题。因此我们提出一种实验性的评分方法——部分得分制。实验数据有两组，分别采用TEM4听写评分制和新评分制。数据比较以及部分得分模型(Rasch模型之一)对实验量表效能的分析(如模型与数据拟合值、被试拟合值、信息函数等)说明，实验评分制能较好地测量大多数学生的听写水平。

A Pilot Scoring Scale for Dictation--"Partial Credit Scoring" and Preliminary Experiments

The present scheme used in marking TEM4 dictation is the conventional count-and-deduct approach which is mistake-based and negative. The method is somewhat problematic. The experimental marking scheme suggested is a partial credit scoring method. There are two datasets, one using the TEM4 method and the other the pilot scale. The comparison of the datasets and the further analyses of the data with Masters model (a Rasch model) such as model-data fit, person fit and information function indicate that the experimental scheme can measure most of the subjects' ability accurately.