匠人府分享 http://blog.sciencenet.cn/u/meiweipingg

博文

源文献引用系列(1): 主成分分析中要求 KMO>0.7 的文献依据

已有 9708 次阅读 2017-5-31 11:36 |个人分类:数据处理与统计分析|系统分类:论文交流

原始文献引用为

Kaiser, H. F., & Rice, J. (1974). Little Jiffy, Mark IV. Educational and PsychologicalMeasurement, 34 (1), 111-117.


如何确定是否可以做主成分分析。要看KMO的统计值,如果值>0.7,才表示适合做主成分分析!http://blog.sciencenet.cn/blog-74956-1058026.html  




完整pdf下载如下


Kaiser_1974.pdf

Hair et al_2010.pdf

Dziuban_1974.pdf


-------------------------------------------------

相关论述①

In other words, KMO indicates the amount of variance  shared among  the items designed  to  measure  a  latent  variable   when compared  to that shared with the error.  Kaiser (1974)  recommends  accepting  values  greater  than  0.5 as acceptable.  More specifically,  values between 0.5 and 0.7  are  considered    mediocre,  values between 0.7 and 0.8 are considered good,  values between 0.8 and 0.9  are deemed    great  and  values  above  0.9  are  superb   (Hutcheson  and Sofroniou,  1999).  A v alue  more   than  0.7  is  the common threshold  for  confirmatory  analysis  (Hair et al., 2010).

http://qrclinic.blogspot.jp/2013/06/kmo-and-bartlets-test-in-factor-analysis.html

更多文献引用:

Dziuban, C. D., & Shirkey, E. C. (1974).When is a correlation matrix appropriate for factor analysis? PsychologicalBulletin, 81, 358-361.

Multivariate Data Analysis, 7/e. J. F. Hair Jr., W. C. Black, B. J. Babin, R. E. Anderson. 2010, Pearson Prentice Hall.

-------------------------------------------------

相关论述②

The KMO statistic is a Measure of SamplingAdequacy, both overall and for each variable (Kaiser 1970; Cerny and Kaiser1977; Dziuban & Shirkey, 1974).

The SPSS program code sets KMO to .5 when the correlation matrix  is  an identity  matrix,  avoiding  the  division-by-0  problem.

http://www-01.ibm.com/support/docview.wss?uid=swg21479963

更多文献引用:

Cerny, C.A., & Kaiser, H.F. (1977). A studyof a measure of sampling adequacy for factor-analytic correlation matrices.Multivariate Behavioral Research, 12(1), 43-47.
Dziuban, C. D., & Shirkey, E. C. (1974). When is acorrelation matrix appropriate for factor analysis? Psychological Bulletin, 81,358-361.
Kaiser, H.F. (1970). A second generation Little Jiffy.Psychometrika, 35, 401-415.


如有纰漏或错误之处,欢迎批评指正!



https://blog.sciencenet.cn/blog-651374-1058170.html

上一篇:SCI 等收录期刊检索便捷入口
下一篇:源文献引用系列(2): MDS分析要求应力系数 stress<0.2 的文献依据

2 梅志平 杨正瓴

该博文允许注册用户评论 请点击登录 评论 (0 个评论)

数据加载中...
扫一扫,分享此博文

Archiver|手机版|科学网 ( 京ICP备07017567号-12 )

GMT+8, 2022-1-29 07:05

Powered by ScienceNet.cn

Copyright © 2007- 中国科学报社

返回顶部