Correspondence analysis

from Wikipedia, the free encyclopedia

The correspondence analysis , even simple correspondence analysis abbreviated, CA ( correspondence analysis ) is a multivariate statistical technique , with the relations between the variables of a contingency table are represented graphically. The column and row profiles of a matrix are represented by points in a space, the coordinate axes of which are formed by the respective features . It is also known as principal component analysis with categorical data.

The Canonical Correspondence Analysis (CCA) is an extension of the CA by Ter Braak . The derived ordination axes are linear combinations of the environmental variables , which are generated using a multiple linear regression model.

application areas

Correspondence analysis was made popular in empirical social research largely through Pierre Bourdieu's study “ The Subtle Differences ”. Correspondence analysis is also widespread in market research . In archeology , it can be used to organize tables that record the combination of types in closed finds . According to such an order - also known as seriation - the types and find complexes often show a chronological sequence.

Individual evidence

  1. Ilona Leyer, Karsten Wesche: Multivariate Statistics in Ecology: An Introduction , 1st, Edition 2007. Corr. Reprint. Edition, Springer, Berlin, June 2009, ISBN 3540377050 , p. 91.

literature

Web links