Confidence Intervals for Reliability 1 Running head: CONFIDENCE INTERVALS FOR RELIABILITY COEFFICIENTS Estimation of and Confidence Interval Formation for Reliability Coefficients of Homogeneous Measurement Instruments

The reliability of a composite score is a fundamental and important topic in psychology and related disciplines. The most commonly used reliability estimate of a composite score is coefficient alpha. However, under regularity conditions, the population value of coefficient alpha is only a lower bound on the population reliability, unless the items are essentially tau-equivalent, an assumption that is likely violated in most applications. A generalization of coefficient alpha, termed omega, is discussed and generally recommended. Furthermore, a point estimate itself almost certainly differs from the population value. Therefore, it is important to provide confidence interval limits so as not to over-interpret the point estimate. We detail analytic and bootstrap methods for confidence interval construction for omega. We recommend the bias-corrected bootstrap approach for omega and provide open source and freely available R functions via the MBESS package to implement the methods discussed.