This page is a stub
This page is under construction. If you would like to contribute to this endpoint, please let us know through our user forum!
== incomplete ==

The main idea...

The BIOENV procedure (Clarke and Ainsworth, 1993) is a dissimilarity-based and exploratory method concerned with identifying that subset of a set of explanatory variables whose Euclidean distance matrix correlates maximally with the Bray-Curtis dissimilarity matrix derived from count data (e.g. abundance data) obtained from the same sites or samples (i.e. objects). This procedure generally uses Spearman rank-based correlations and has some similarities with the Mantel test and Procrustes analysis.

  • If the explanatory variables selected through BIOENV are used in further analysis, there is a risk of data dredging.
  • Note, that several properties of (dis)similarity-based approaches have been criticised (Warton et al., 2012).
  • Different implementations may have key differences that can result in incomparable results.
  • This technique is intended as an exploratory approach, and should be followed with more rigourous tests of association (e.g. constrained analyses)
Walkthroughs featuring BIOENV
  • R
    • The bioenv() function in the vegan package. Note, however, that there are key differences between this and the PRIMER implementation.