Most significance tests require assumptions about how data are distributed. Multivariate procedures and covariance modelling procedures such as LISREL require that data be multivariate normal. Program MULTINOR provides a graphical test of whether data are multivariate normally distributed. Significance tests of multivariate normality are not available in commonly-used statistics packages, and graphical tests are tedious to implement with statistical packages and require access to statistics not readily available for some research problems.
Get full access to this article
View all access options for this article.
References
1.
Baldwin, B. (1989). A primer in the use and interpretation of structural equation models. Measurement and Evaluation in Counseling and Development, 22, 100-112.
2.
Cliff, N. (1987). Analyzing multivariate data. San Diego: Harcourt Brace Jovanovich.
3.
Gnanedesikan, R. (1977). Methods for statistical analysis of multi-variate observations. New York: Wiley.
4.
Mardia, K. V. (1974). The effect of non-normality on some multi-variate tests and robustness to non-normality in the linear model. Biometrika, 58, 105-121.
5.
Stevens, J. (1986). Applied multivariate statistics for the social sciences. Hillsdale, NJ: Erlbaum.
6.
Thompson, B. (1984). Canonical correlation analysis: Uses and interpretation. Newbury Park, CA: SAGE.
7.
Thompson, B. (1988, April). Canonical correlation analysis: An explanation with comments on correct practice. Paper presented at the annual meeting of the American Educational Research Association, New Orleans. (ERIC Document Reproduction Service No. ED 295 957)
8.
Thompson, B. (1990, April). Variable importance in multiple and canonical correlationl regression. Paper presented at the annual meeting of the American Educational Research Association, Boston. (ERIC Document Reproduction Service No. ED forthcoming)
