It is often useful to analyze a covariance matrix pooled over multiple groups. This note shows that the pooled covariance matrix cannot in general be calculated from the weighted average of the separate covariance matrices for the multiple groups. The correct equation for calculating the pooled covariance matrix adjusts for group differences in sample means.
Get full access to this article
View all access options for this article.
References
1.
Bentler, P. M. and Weeks, D. G. (1979). Interrelations among models for the analysis of moment structures. Multivariate Behavioral Research, 14, 169-185.
2.
Jöreskog, K. G. (1971). Simultaneous factor analysis in several populations. Psychometrika, 36, 409-426.
3.
Jöreskog, K. G. and Sorbom, D. (1979). Advances in factor analysis and structural equation models. Cambridge, MA: Abt Associates.
4.
Meredith, W. (1964). Notes on factorial invariance. Psychometrika, 29, 177-185.
