概化理论方差分量及其变异量估计:Jackknife方法与Traditional方法比较被引量：1

Generalizabilty Theory Variance Component and Its Variance Estimation:Comparison Between Jackknife Method and Traditional Method

在线阅读下载PDF

导出

摘要文章生成概化理论p×i、p×i×h、p×(i:h)三种不同设计下的正态数据、多项数据和二项数据,用Jackknife方法和Traditional方法估计数据的方差分量、标准误和置信区间,并比较这两种方法的性能。结果表明:(1)Jackknife方法在方差分量估计和标准误估计上都较为准确;(2)相较于Traditional方法,Jackknife方法在方差分量置信区间估计上略有不足。(3)相较于Traditional方法,Jackknife方法估计的准确性不随数据类型、研究设计和方差分量的不同而产生波动,具有更强的稳健性。 This paper generates generalizability theory,normal data,multinomial data and binomial data under three different designs of p×I,p×i×h and p×(i:h).And then the paper uses jackknife method and traditional method to estimate variance component,standard error and confidence interval of data,and also compares the performance of these two methods.The results show that(1)the jackknife method is accurate in variance component estimation and standard error estimation;(2)compared with the traditional method,the jackknife method is slightly inadequate in the estimation of variance component confidence interval;(3)compared with the traditional method,the accuracy of the jackknife method does not fluctuate with the difference of data types,research design and variance component,so it has stronger robustness.

作者黎光明黄梓龙 Li Guangming;Huang Zilong(School of Psychology,South China Normal University,Guangzhou 510631,China;Center for Studies of Psychological Application,South China Normal University,Guangzhou 510631,China)

机构地区华南师范大学心理学院华南师范大学心理应用研究中心

出处《统计与决策》 CSSCI 北大核心 2020年第6期10-14,共5页 Statistics & Decision

基金国家自然科学基金面上项目(31470050) 教育部人文社会科学研究规划基金项目(18YJA190006) 广东省哲学社会科学“十三五”规划一般项目(GD17CXL01) 广州市哲学社会科学“十三五”规划一般项目(2017GZYB111)。

关键词概化理论 JACKKNIFE Traditional方法方差分量 generalizability theory Jackknife method traditional method variance component

分类号 O212.1 [理学—概率论与数理统计]

作者简介黎光明(1977-),男,江西广昌人,博士,副教授,研究方向:心理统计与测量。

引文网络
相关文献

参考文献1

1黎光明,张敏强.基于概化理论的方差分量变异量估计[J].心理学报,2009,41(9):889-901. 被引量：16

二级参考文献38

1杨志明,张雷.用多元概化理论对普通话的测试[J].心理学报,2002,34(1):50-55. 被引量：21
2American Educational Research Association, American Psychological Association, National Council on Measurementin Education (1985). Standards for educational and psychological testing. Washington, DC: Author.
3American Educational Research Association, American Psychological Association, National Council on Measurement in Education (1999). Standards for educational and psychological testing (Rev. ed. ). Washington, DC: Author.
4Brennan, R. L. (1992). Elements of generalizability theory (Rev. ed. ). Iowa City, IA: ACT.
5Brennan, R. L. (2001). Generalizability theory. New York: Springer-Verlag.
6Brennan, R. L. (2006). Unbiased estimates of variance components with bootstrap procedures: Detailed results (CASMA Research Report No 21). Iowa City, IA: Center for Advanced Studies in Measurement and Assessment, University of Iowa. Available from http://www.education. uiowa.edu/casma.
7Brennan, R. L., Harris, D. J., & Hanson, B. A. (1987). The bootstrap and other procedures for examining the variability of estimated variance components in testing contexts (ACT Research Report Series87-7). Iowa City, IA: American College Testing Program.
8Briggs, D. C., & Wilson, M. (2007). Generalizability in item response modeling. Journal of Educational Measurement, 44(2), 131-155.
9Clauser, B. E., Harik, E, & Clyman, S. G. (2000). The generalizability of scores for a performance assessment scored with a computer-automated scoring system. Journal of Educational Measurement, 36(2), 245-262.
10Clauser, B. E., Harik, P., & Margolis, M. J. (2006). A multivariate generalizability analysis of data from a performance assessment of physicaians' clinical skill. Journal of Educational Measurement, 43(3), 173-191.