Factor classification

Factor classification is a key task of WP 1.1, and a task on which WP 2 partially depends for input. A preliminary factor classification, to be continuously refined, is already available and presented here.

The underlying data for the factor classification is available here (this link opens in googledocs). Criteria were chosen so that they could be coded in binary answers (0 or 1). More criteria can be easily added to further describe environmental factors. From this exercise, also a possible criterion for factor individuation emerged: when two closely related environmental factors cannot be separated by additional criteria, then they should be regarded as one factor (and thus joined).

The following preliminary dendrogram (Jaccard dissimilarities) shows a factor classification based on the available data file:

Below are links (opens pdf) to an NMDS analysis of the factors, a PCA, and a cluster dendrogram based on Jaccard dissimilarities. Distance matrices can be used to assemble sets of factors of various dissimilarity for use in experiments as described in WP 2.