Statistical methods for studying modularity: A reply to mitteroecker and bookstein

Abstract

To summarize, GGMs and related methods are a robust set of statistical techniques appropriate for studying the covariance structure of multivariate data sets. GGMs have found wide application in both biological and nonhiological contexts and have been shown to be a useful tool for both exploratory and confirmatory studies of integration and modularity (e.g., Magwene 2001, Polanski and Franciscus 2006, Pie and Tramello 2007). The Wright-style factor analysis approach favored by Mitteroecker and Bookstein is a related and equally useful statistical tool. In the context of studies of integration, the latter would appear to he best suited to situations where one has strong a priori assumptions about cox ariance structure inherent in the data (Mitteroecker and Bookstein 2007). Copyright © society of Systematic Biologists.