CO‐INERTIA ANALYSIS AND THE LINKING OF ECOLOGICAL DATA TABLES

By Stéphane Dray, Daniel Chessel, Jean Thioulouse. 2003

643 citations recorded by the source corpus.

Ecological studies often require studying the common structure of a pair of data tables. Co-inertia analysis is a multivariate method for coupling two tables. It is often neglected by ecologists who prefer the widely used methods of redundancy analysis and canonical correspondence analysis. We present the co-inertia criterion for measuring the adequacy between two data sets. Co-inertia analysis is based on this criterion as are canonical correspondence analysis or canonical correlation analysis, but the latter two have additional constraints. Co-inertia analysis is very flexible and allows many possibilities for coupling. Co-inertia analysis is suitable for quantitative and/or qualitative or fuzzy environmental variables. Moreover, various weighting of sites and various transformations and/or centering of species data are available for this method. Hence, more ecological considerations can be taken into account in the statistical procedures. Moreover, the principle of this method is very general and can be easily extended to the case of distance matrices or to the case of more than two tables. Simulated ecological data are used to compare the co-inertia approach with other available methods.

Subfields: Ecology, Evolution, Behavior and Systematics, Food Science, Statistics and Probability

View Paper: https://www.semanticscholar.org/paper/77898adf13dfd073efddcf044e6efb49294cf402

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Ecology, Evolution, Behavior and Systematics

In Food Science

In Statistics and Probability

Citations in the corpus

Earthworm indicators as tools for soil monitoring, characterization and risk assessment. An example from the national Bioindicator programme (France) (2011, positive): “Co-inertia analysis was subsequently carried out to test for significant co-variation among the two data sets (Dray et al. 2003).”
Trait-Based Approaches in the Analysis of Stream Fish Communities (2010, methodology): “Trait-based modeling and applications continue to grow concomitantly with the development of more sophisticated multivariate analysis techniques (Diniz-Filho et al. 1998; Dray et al. 2003; Dray and Legendre 2008; Jackson et al. 2010, this volume).”
CARNIVORE OCCURRENCE ALONG AN URBAN–RURAL GRADIENT: A LANDSCAPE-LEVEL ANALYSIS (2006, methodology): “…ordination technique designed for gradient studies and is broadly similar to canonical correspondence analysis in that variance in species occurrence is maximized along ordination axes derived from the input of environmental data (Dolédec et al. 2000; Dray et al. 2003; Ter Braak 1986).”
The mycorrhizal fungus Glomus intraradices and rock phosphate amendment influence plant growth and microbial activity in the rhizosphere of Acacia holosericea (2005, methodology): “When the number of rows is low, CANCOR cannot be used, and CCA is reduced to a plain CA (see for example Dray et al., 2003).”
Comparison of structural and functional approaches to determining landuse effects on grassland stream invertebrate communities (2006, methodology): “Last, we used coinertia analyses (Dolédec and Chessel 1994, see a recent review in Dray et al. 2003) to address the relationships between each of the 4 types of biological information and the 3 measures describing the landuse gradient (DIN, DRP, and %FINES).”
Spatial ordination of vegetation data using a generalization of Wartenberg's multivariate spatial correlation (2008, methodology): “For a review in ecology, the reader should consult Dray et al. (2003a).”