Journal of Agricultural, Biological, and Environmental Statistics

, Volume 7, Issue 2, pp 222–232

Multifactorial analysis of distance in studies of ecological community structure


DOI: 10.1198/10857110260141256

Cite this article as:
Krzanowski, W.J. JABES (2002) 7: 222. doi:10.1198/10857110260141256


A framework is described for partitioning the matrix of between-sample distances that forms a starting point for many ecological studies into contributions attributable to different factors in a structured data set. This partitioning enables a series of ordinations to be produced that better enable an insight to be gained into the effects of the factors. More detailed application of the same partitioning provides a decomposition of each factorial effect into single degree-of-freedom contrasts, which enables reasons for observed trends to be investigated. The methods are illustrated by application to data from a study of marine community structure.

Key Words

Analysis of varianceFactorial effectsOrdinationOrthogonal contrastsSums of squares

Copyright information

© International Biometric Society 2002

Authors and Affiliations

  1. 1.School of Mathematical SciencesUniversity of ExeterExeterU.K.