PLAB能测量中国学生的外语学能吗?——一项基于Rasch模型的研究  

Can PLAB measure Foreign Language Aptitude of Chinese students?——A Rasch-based study

作　　者：李兰荣[1] 

机构地区：[1]中国教育科学研究院,北京100088

出　　处：《外语测试与教学》2015年第2期11-20,共10页Foreign Language Testing and Teaching

基　　金：中国教育科学研究院2014年度基本科研业务专项基金"高中生个体因素对英语成绩的影响研究"(课题批准号:GY2014XDY24)的研究成果之一

摘　　要：本研究的目的是使用Rasch模型检验Pimsleur语言学能组题(Pimsleur Language Aptitude Battery,简称PLAB)的三项分测验在中国外语学习者中的构念效度,并在Rasch分析结果的基础上进一步检验PLAB的预测效度。来自国内某所重点大学的65名英语专业二年级学生参与了测试。Rasch分析表明PLAB的最后三部分测试的绝大部分的题目与模型拟合良好,题目信度较好,较能有效测出学生的外语学能。同时,研究也发现该试题对本研究中的被试群体过易,导致部分试题被试信度较低。分层回归分析表明在控制了英语学习年限以后,外语学能显著解释了学生的英语水平总变异的20.3%,这为试题的预测效度提供了证据。This study aims to use the Rasch model to examine the construct validity of three subtests of Pimsleur Language Aptitude Battery （PLAB） and to further explore the predictive validity of the subtests on the basis of Rasch analyses. 65 second-year English majors from a key university in China took the test. Rasch analyses showed that the vast majority of the items displayed good fit to the Rasch model, that the subtests had acceptable or good item reliability and could measure the students＇ foreign language aptitude validly. However, it was found that the subtests were too easy for the students, which resulted in low reliability of some subtests. Hierarchical regression analysis revealed that after controlling the length of English learning, foreign language aptitude explained 20.3 % of the total variance of the students＇ English proficiency, thus lending support for the predictive validity of these subtests.

关 键 词：外语学能 PLAB 效度 RASCH模型 

分 类 号：H319[语言文字—英语]