Reliable for Whom? Inferring and Reporting Reliability across Diverse Populations

Abstract

We describe best practices in reporting reliability estimates in counseling research with consideration to precision, generalization, and diverse populations. We provide a historical context to reporting reliability estimates, the limitations of past practices, and new methods to address reliability generalization. We highlight best practices including consideration of reliability estimates as function of scores, not scale, diverse sample, hierarchy of reporting reliability estimates, and methods of reliability generalization.