中国少数民族汉语水平考试(MHK)三级口试效标证据的Toulmin模型论证

Criterion Evidence of Toulmin's Argument Model for MHK Level 3 Oral Test

建立小型描述语语料库并参考MHK三级口试评分标准,对达到MHK三级水平的考生进行"能做什么"的描述分析,建立MHK三级口语能力评分量表。教师给出本班每位参加MHK三级口试学生的口语能力等级,以此为效标计算考生MHK口试成绩和教师评价间的相关等。将分析结果作为论证支撑(backing),以Toulmin效度论证模型为理论框架,从实证角度对少数民族汉语水平考试(MHK)的外推(extrapolation)阶段进行效度论证。结果显示:MHK三级口试效度论证的外推(extrapolation)阶段基本成立,考生的MHK三级口试成绩可以反映考生在实际汉语交际环境中口语能力的表现。

A small descriptive corpus is built based on 10 pieces of language ability rating scale. Depended on the standard scoring of MHK level 3 oral test and intuitive method,we will make a descriptive analysis of ＂what can they do＂the examinees whose mandarin ability reach to MHK Level 3. In the third step,we will train the front- line teachers from a certain school of the scoring method and standard. After that,the trained teachers are required to give a rank（ both whole ability rank and each single item rank） to every student in their classes of their oral ability based on the questionnaire. The assessment from teachers will be the criterion for calculating the correlation between examinees＇ score of MHK oral test and teacher＇s assessment. In the last step,the test validation of extrapolation state of MHK from empirical perspective would be done with the correlation result for backing and Toulmin ＇s Argument Model for framework. The results reveal that the test validation of extrapolation state of MHK level 3 is basically reasonable,the scores of examinees＇ in MHK level 3 oral test can reflect their oral performance in practical mandarin communication.