Abstract
A number of research questions (e.g., differential validity, moderator variables) require comparisons of correlation coefficients across independent populations. In the two population case, the test statistic used in these comparisons is not a simple function of the difference between the sample correlations. Hence, this difference is not a direct indicator of significance. The formula can be rewritten to give the value of the larger correlation (r L ) required to reach significance at a specified alpha level, given the value of the smaller correlation (r s ) and the sample sizes. Quick-reference tables are presented giving critical values of r L for a two-tailed critical region at an alpha level of .05. Tabled values of r s range from .10 to .95, and sample sizes range from 15 to 200. The tables allow quick assessment of significance without requiring calculation of the test statistic.
Get full access to this article
View all access options for this article.