普通话水平测试第四题“人机”测试模式信度研究  

Study on the Reliability of Man-Machine Test Mode in the Fourth Question of Putonghua Proficiency Test

作　　者：杨惠麟 张俊 YANG Huilin;ZHANG Jun(Southwest Medical University,Luzhou,Sichuan 646000;Luzhou Language Working Committee,Luzhou,Sichuan 646000)

机构地区：[1]西南医科大学,四川泸州646000 [2]泸州市语言文字工作委员会,四川泸州646000

出　　处：《绵阳师范学院学报》2023年第1期40-46,共7页Journal of Mianyang Teachers＇ College

基　　金：四川省第二批“三全育人”综合改革试点院(系)项目“‘三破、三拓、四提升’,创建伞形立体育人模式”(2021SC05)。

摘　　要：普通话水平测试是一项国家级语言测试,目前此项工作已大部分由计算机完成,只一名测试员参与第四大题的评分,原来的计算机“辅助”测试正在逐渐变成测试员“辅助”测试。现以2018、2019、2020年泸州市语言文字工作委员会测试站的测试数据为依据,筛选出偏差复审的成绩,提取第四题“命题说话”的评分细节进行分析。分别进行了三年复审率比对,2019、2020年复审数据计算机与测试员评分差、计算机与复审员评分差、复审成绩等级变化比对,异常偏差复审语料50例语音面貌描述和第三次测评比对、讨论,并辅以测试员评分问卷调查表,多维度地对“人机”模式测试第四题的信度进行了研究,得出了这一测试模式测试成绩基本可信的结论,并提出了相关建议。Putonghua Proficiency Test is a national language test.At present,most of the work is done by computer,and only one tester takes part in the scoring of the fourth question.The former computer-aided test is gradually becoming tester-aided test.Based on the test data of Luzhou Language Working Committee test station in 2018,2019 and 2020,this paper screened out the scores of subjects in bias review and extracted the scoring details of"Speaking"item for analysis.The review rate was compared for three years respectively.In 2019 and 2020,the comparison of the score difference between the computer and the tester,the score difference between the computer and the tester,and the change of the score level of the review.At the same time,50 cases of abnormal deviation review corpus were selected for voice description and third evaluation comparison.The reliability of man-machine test mode in the fourth question of Putonghua Proficiency Test is explored in multiple dimensions.It is concluded that the results of this test mode are basically credible,and some relevant suggestions are put forward.

关 键 词：普通话水平测试 “人机”测试模式 信度 偏差复审 

分 类 号：H102[语言文字—汉语]