Answers are grounded in this paper and cite their sources.
An empirical study of the distribution of the Spearman rank correlation coefficient under certain general assumptions demonstrates that its distribution is better approximated by using the T-transformation and the t-distribution than by using the z-transformation and the normal distribution. Furthermore, use of the T-transformation facilitates calculation of critical values for tests, confidence intervals and the power of tests.
No takes yet. Share an insight, caveat, or question.
Helena C. Kraemer (1974) studied this question.
Synapse has enriched 10 related papers. Consider them for comparative context: