Testing for predicted patterns: When interest in the whole is greater than in some of its parts

This article provides a general rationale and step-by-step procedures for statistically assessing a theory-based or predicted pattern in a set of K > 2 means. A coherent hypothesis-testing strategy that incorporates Type I error and power considerations is an integral part of the process. Some specific advantages that are conferred by predicted pattern testing, relative to commonly adopted alternatives, include its (a) scientific parsimony; (b) precision, with respect to getting to the heart of a researcher's hypothesis; (c) sensitivity, as reflected by greater statistical power and which includes applicability to experimental factors associated with small "piecewise" effects; and (d) ability to lend varying degrees of statistical support to a researcher's hypothesis. Both a hypothetical example and a published application of predicted pattern testing are provided to demonstrate its use and usefulness.