Multiple Factor
Analysis: Main Features and Application to Sensory Data
Jérôme Pagès
Abstract
Data table in which a single set of individuals is described by several
groups of variables are frequently encountered. In the factor analysis
framework, taking into account different groups of variables in a unique
analysis firstly raises the problem of balancing the different group. This
problem being solved, beyond classical outputs from factor analysis, it is
necessary to have at one’s disposal specific tools in order to compare the
structure upon individuals induced by the different groups of variables. That
is the aim of Multiple Factor Analysis (MFA), factor analysis devoted to such
data table. This paper presents the method, its main properties and an
application to sensory data.
Key words: Factor analysis, Principal
components analysis, Canonical analysis.
PDF (English)