ON SIMULTANEOUS ANALYSIS

- Arguments for and against combined analysis of multiple data sets in phylogenetic inference are reviewed. Simultaneous analysis of combined data better maximizes cladistic parsimony than separate analyses, hence is to be preferred. Simultaneous analysis can allow "secondary signals" to emerge because it measures strength of evidence supporting disparate results. Separate analyses are useful and of interest to understanding the differences among data sets, but simultaneous analysis provides the greatest possible explanatory power, and should always be evaluated when possible. The mechanics of simultaneous analysis are discussed.