Answers are grounded in this paper and cite their sources.
The method and the associated graphical representations proposed in this paper facilitate the interpretation of the correspondence analysis (CA) of a samples/species data table. By using the "reciprocal discrimination" model of CA, this method introduces directly and on the same graph the notions of niche width and within—sample diversity. We propose two Macintosh programs for computing conditional means, variances, and covariances of factor scores for species and samples, and for drawing ellipses and Gaussian curves from these values. Two examples of ecological interest show how resulting graphics help in understanding the results of multivariate analyses.
No takes yet. Share an insight, caveat, or question.
Thioulouse et al. (1992) studied this question.
Synapse has enriched 10 related papers. Consider them for comparative context: