Applying the Del Technique to the Analysis of Cross-Classification Data: A Test of Ceo Succession and Top Management Team Development

Abstract

This article introduces to the organizational sciences a relatively new cross-classification analysis technique based on a statistic called Hildebrand's del. Del is shown to be superior to the other means available for analyzing cross-classification data for several reasons. Its most notable features are that it allows a researcher to develop and test explicit a priori predictions where other procedures do not, provides a strength-of-association measure, and is independent of sample size. The del procedure is compared to traditional approaches to analyzing cross-classification data by testing a theory of chief executive succession and top management team development in technology-based new ventures.