Multiple Correspondence Analysis

Overview

Multiple Correspondence Analysis is a monograph presenting the geometric foundations of MCA, its application to the analysis of questionnaires and its place within the framework of Geometric Data Analysis (GDA). The book also develops structured data analysis and an introduction to inductive data analysis (typicality tests, homogeneity tests, confidence ellipses).

Authors and Publisher

Brigitte Le Roux, Henry Rouanet — Sage Publications, Thousand Oaks (CA), 2010.
Series: Quantitative Applications in the Social Sciences, no. 163.
Publisher’s page

	Chapter	Page
	About the Authors	vii
	Series Editor’s Introduction	viii
	Acknowledgements	x
1	Introduction	1
	1.1 MCA as a geometric method	1
	1.2 Historical landmarks	2
	1.3 Bourdieu and statistical data analysis	4
	1.4 The Taste example	5
	1.5 Methodological points	10
	1.6 Organisation of the monograph	12
2	The Geometry of a Cloud of Points	14
	2.1 Basic geometric notions	14
	2.2 Cloud of points	16
	2.3 Subclouds and partitions of a cloud	20
	2.4 Contributions	22
	2.5 Principal axes of a cloud	24
	2.6 From two-dimensional to higher-dimensional clouds	30
	2.7 Computation formulas for a weighted cloud in a plane	32
3	The Method of Multiple Correspondence Analysis	34
	3.1 Principles of MCA	34
	3.2 MCA of the Taste example	46
	3.3 Two variants of MCA	61
4	Structured Data Analysis	68
	4.1 From supplementary variables to structuring factors	68
	4.2 From experimental to observational data	69
	4.3 Concentration ellipses	69
	4.4 Taste example: study of gender and age	71
5	Inductive Data Analysis	81
	5.1 Typicality tests	82
	5.2 Homogeneity tests	85
	5.3 Confidence ellipses	89
6	Full-Scale Research Studies	91
	6.1 The field of publishers in France	91
	6.2 The Norwegian field of power	97
	Appendix	103
	References	110
	Index	113

Data and Software

Data associated with the monograph are available on the software and data page.

Multiple Correspondence Analysis

Overview

Authors and Publisher

Table of Contents

Data and Software

See Also