«上一篇/Previous Article|本期目录/Table of Contents|下一篇/Next Article»

《心理学探新》[ISSN:1003-5184/CN:36-1228/B]

期数:

 2020年03期

页码:

 261-268

栏目:

 心理统计与测量

出版日期:

 2020-06-25

文章信息/Info

Title:

 The Impact of the Representativeness of Difficulty Level of Anchor Test on Vertical Scaling

文章编号:

1003-5184(2020)03-0261-08

作者:

 叶 萌¹; 2; 辛 涛²; 胡卫平¹; 2; 孙小坚³

 (1.陕西师范大学现代教学技术教育部重点实验室,西安 710062; 2.中国基础教育质量监测协同创新中心,北京 100875; 3.西南大学数学与统计学院,重庆 400715)

Author(s):

 Ye Meng¹; 2; Xin Tao²; Hu Weiping¹; 2; Sun Xiaojian³

 (1.Key Laboratory of Modern Teaching Technology,Ministry of Education,Shaanxi Normal University,Xi'an 710062; 2.China Basic Education Quality Monitoring and Cooperative Innovation Center,Beijing 100875; 3.School of Mathematics and Statistics,Southwest University,Chongqing 400715)

关键词:

 垂直量尺化;  锚测验;  难度水平;  锚题代表性

Keywords:

 vertical scale;  anchor test;  difficulty level;  representativeness of anchor test

分类号:

 B841.2

DOI:

 -

文献标识码:

 A

摘要:

 考察了锚测验难度水平对其来源测验水平的代表性对垂直量尺化的影响。采用模拟研究的方法,比较了锚测验难度等于来源测验、位于高低年级测验水平难度区间的第25百分位处及区间第50百分位处时,年级能力分布和垂直量尺特性上的参数返真结果,发现锚题难度水平高于其来源测验非但不会导致垂直量尺化结果变差,在有的情境下反而可能会提高其准确性。研究揭示人们构建垂直量尺时,可以根据内容和其他统计特征的需要对锚测验的难度水平做出适当调整

Abstract:

 The influence of the representativeness of difficulty level of anchor test to its source test level on vertical scaling was investigated.Using the method of simulation,it compared recovery of the parameters in grade ability distribution and features of the vertical scale among conditions of difficulty level of anchor test equal to the source test,at the 25th percentile of the difficulty interval composed of the upper and lower test levels and at the 50th percentile of the interval.The results indicated that rather than deteriorating the results of vertical scaling,anchor test with difficulty level higher than that of its source test may improve its accuracy in some situations.The research revealed that the difficulty level of anchor test can be adjusted appropriately according to the needs of content and other statistical characteristics when vertical scale is constructed

参考文献/References

 叶萌,辛涛.(2015).测验链接中的锚题代表性研究.心理科学,38(1),209-215.
Cao,Y.(2008).Mixed-format test equating effects of test dimensionality and common-item sets(Unpublished doctorial dissertation).University of Maryland,Maryland,US.
Gao,X-H.,Hanson,B.A.,& Harris,D.J.(1999).Effect of using different common item sets under the common item non-equivalent groups design.Paper presented at the Annual Meeting of the American Educational Research Association,Montreal,Canada.In Y.Cao(Ed.),Mixed-format test equating effects of test dimensionality and common-item sets(Unpublished doctoral dissertation).University of Maryland,Maryland,US.
Chin,T.Y.,Kim,W.,& Nering,M.L.(2006).Five statistical factors that influence IRT vertical scaling.Paper presented at the Annual Meeting of the National Council on Measurement in Education,San Francisco,CA.
Fitzpatrick,J.,& Skorupski,W.P.(2016).Equating with miditests using IRT.Journal of Educational Measurement,53(2),172-189.
Humphrey,S.N.(2018).The impact of levels of discrimination on vertical equating in the Rasch model.Journal of Applied Measurement,19(3),216-228.
Liu,J.S.,Sinharay,S.,Holland,P.,Feigenbaum,M.,& Curley,E.(2011).Observed score equating using a mini-version anchor and an anchor with less spread of difficulty:A comparison study.Educational and Psychological Measurement,71(2),346-361.
Marco,G.L.,Petersen,N.S.,& Stewart,E.E.(1983).A large-scale evaluation of linear and curvilinear score equating models(ETS
Research Memorandum No.RM-83-2).Princeton,NJ:Educational Testing Service.Cited in J.S.Liu,S.Sinharay,P.Holland,M.Feigenbaum,& E.Curley(Eds.),Observed score equating using a mini-version anchor and an anchor with less spread of difficulty:A comparison study.Educational and Psychological Measurement,71(2),346-361.
O'Neil,T.P.(2010).Maintenance of vertical scales under conditions of item parameter drift and Rasch model-data misfit(Unpublished doctorial dissertation).University of Massachusetts-Amherst,Massachusetts,US.
Sinharay,S.,& Holland,P.W.(2006).The correlation between the scores of a test and an anchor test (ETS RR-06-04).Princeton,NJ:Educational Testing Service.
Sinharay,S.,& Holland,P.W.(2007).Is it necessary to make anchor tests mini-versions of the tests being equated or can some restrictions be relaxed?Journal of Educational Measurement,44(3),249-275.
Sinharay,S.(2018).On the choice of anchor tests in equating.Educational Measurement-Issues and Practice,37(2),64-69.
Trierweiler,T.J.,Lewis,C.,& Smith,R.L.(2016).Further study of the choice of anchor tests in equating.Journal of Educational Measurement,53(4),498-518.
Wei,H.(2010).Impact of non-representative anchor items on scale stability.Paper presented at the Annual Meeting of the National Council on Measurement in Education,Colorado,US.

备注/Memo

备注/Memo:

 基金项目:教育部人文社会科学研究西部和边疆地区青年基金项目(15XJC190003),国家社会科学基金青年项目(17CTJ014)。 通讯作者:辛涛,E-mail:xintao@bnu.edu.cn。

常用功能

更新日期/Last Update:  2020-06-30