This paper examines the importance of the underlying assumption of normality for testing that a bivariate normal correlation equals zero. Both empirical and theoretical evidence suggest that such tests may be employed, more generally, for testing the independence of any two continuous variables.
Get full access to this article
View all access options for this article.
References
1.
Box, G. E. P.Non-normality and tests on variances. Biometrika, 1953, 40, 318-335.
2.
Fisher, R. A.Frequency distribution of the values of the correlation coefficient in samples from an indefinitely large population. Biometrika, 1915, 10, 507-521.
3.
Kendall, M. G. and Stuart, A.The advanced theory of statistics, vol. 2. London: Charles Griffin and Co., 1961.
4.
Levine, H.Robust tests for equality of variances . In Contributions to probability and statistics. Olkin et al. (Eds.), California: Stanford Press, 1960, 270-292.
5.
Norris, R. C. and Hjelm, H. F.Nonnormality and product moment correlation. Journal of Experimental Education, 1961, 29, 261-270.
