论文部分内容阅读
本研究以广东医学院79名医学英语专业学生的5次TEM-8模拟考试为样本,运用Rasch模型,从学生方面对TEM-8模拟测试作文的效度进行详细地研究。结果表明:学生获得的平均分与Rasch模型计算出的平均分对比可见学生实际获得的平均分比预期的略低,学生能力具有显著差异。另外,从加权均方拟合统计量看,2.55%的高能力考生的加权均方拟合统计量大于1.5,7.71%的低能力学生写出较难的句子时发挥较好而意外得分,4.97%的学生的加权均方拟合统计量小于0.5,非拟合学生比例合计15.22%表明这些学生的内部答题行为不太一致。
In this study, five TEM-8 mock tests of 79 medical English majors in Guangdong Medical College were used as samples. The Rasch model was used to study in detail the validity of the TEM-8 mock test essay. The results show that the average score obtained by the students and the average score calculated by the Rasch model shows that the average score actually obtained by the students is slightly lower than expected, and the students’ abilities have significant differences. In addition, weighted mean square fitting statistics, 2.55% of the high-performance candidates weighted mean square fitting statistics greater than 1.5,7.71% of low-ability students to write more difficult sentences to play better and unexpected scores, 4.97 The weighted mean square fitting statistics of less than 0.5% of students and 15.22% of non-fitting students indicate that the internal answering behaviors of these students are not consistent.