2044-8317/asset/olbannerleft.png?v=1&s=8856da07bc63124271bd41692effae1b4eb4d01b)
2044-8317/asset/olbannerright.png?v=1&s=0e71c188d8e53b18773ec432a90707d049af0643)
The effect of kurtosis on the power of two test statistics in covariance structure analysis
Article first published online: 19 JAN 2011
DOI: 10.1111/j.2044-8317.2010.02010.x
© 2011 The British Psychological Society
Issue
British Journal of Mathematical and Statistical Psychology
Volume 65, Issue 1, pages 1–18, February 2012
Additional Information
How to Cite
Foldnes, N., Olsson, U. H. and Foss, T. (2012), The effect of kurtosis on the power of two test statistics in covariance structure analysis. British Journal of Mathematical and Statistical Psychology, 65: 1–18. doi: 10.1111/j.2044-8317.2010.02010.x
Publication History
- Issue published online: 10 JAN 2012
- Article first published online: 19 JAN 2011
- Received 20 May 2010; revised version received 10 December 2010
- Abstract
- Article
- References
- Cited By
We study the effect of excess kurtosis on the non-centrality parameters of the rescaled and the residual-based test statistics for covariance structure models. The analysis is based on population matrices and parameters, which eliminates the sampling variability inherent in simulation studies. We show that the non-centrality parameters, and consequently the asymptotic power, decrease as kurtosis in the data increases. Examples are provided to compare this decrease for the two test statistics, and to illustrate how substantial it is.