Practical Issues in Sample Size Determination for Correlation Coefficient Inference

Abstract

Determination of the appropriate sample size to use when performing inference for a single Pearson correlation coefficient is usually based on achieving sufficient power for the test of H 0 : = 0. However, sample sizes found using this method can yield confidence intervals that are so wide that they provide very little useful information about the magnitude of the population correlation. Alternative approaches for determining the appropriate sample size are proposed and compared to the "usual" method.