Approximating Pearson Product-Moment Correlations from Kendall's Tau and Spearman's Rho

Abstract

This article used Monte Carlo methods to investigate the accuracy of formulas presented by Kendall for estimating Pearson's r from X and by Pearson for estimating Pearson's r from rs. Results indicated that the formula for approximating r from r is somewhat more accurate than the formula for approximating r from rs. However, both formulas were found to be quite accurate. The largest asymptotic difference between a converted and actual r was -.005. In addition, the standard errors of the transformed X and rs coefficients were compared to the standard error of r, and the empirically derived percentage increase in standard error was compared to theoretically derived estimates. Implications of these findings for meta-analyses of correlations are discussed.

Get full access to this article

View all access options for this article.

References

Esscher, F. (1924). On a method of determining correlation from the ranks of variates. Skandinavisk Aktuarietidskrift, 7, 201-219.

Greiner, R. (1909). Über das Fehlersystem der Kollektivmasslehre. Zeitschift für Mathematik und Physik, 121-158.

Kendall, M. G. (1949). Rank and product-moment correlation. Biometrika, 36, 177-193.

Kendall, M. G. (1962). Rank correlation methods (3rd ed.). New York: Hafner.

Kendall, M. G. , & Gibbons, J. D. (1990). Rank correlation methods (5th ed.). New York: Oxford University Press.

Hunter, J. E. , & Schmidt, F. L. (1990). Methods of meta-analysis: Correcting error and bias in research findings. Newbury Park, CA: Sage.

Marsaglia, G. , Ananthanarayanan, K. , & Paul, N. J. (1976). Improvements on fast methods for generating normal random variables. Information Processing Letters, 5, 27-30.

Pearson, K. (1907). Mathematical contributions to the theory of evolution. XVI. On further methods of determining correlation. Drapers' Company Research Memoirs (Biometric Series 4). Cambridge: Cambridge University Press.