Multiple Factor Analysis for Contingency Tables in the FactoMineR Package

Abstract

We present multiple factor analysis for contingency tables (MFACT) and its implementation in the FactoMineR package. This method, through an option of the MFA function, allows us to deal with multiple contingency or frequency tables, in addition to the categorical and quantitative multiple tables already considered in previous versions of the package. Thanks to this revised function, either a multiple contingency table or a mixed multiple table integrating quantitative, categorical and frequency data can be tackled.

An original methodology for the analysis and interpretation of word-count based methods: Multiple factor analysis for contingency tables complemented by consensual words

“…The function ''MFA'' of the package columnmargin FactoMineR (Husson, Josse, Lê, & Mazet, 2013;Lê, Josse, & Husson, 2008) was used for multiple factor analysis for contingency tables (Kostov, Bécue-Bertaut, & Husson, 2013). The R function WordCountAna (Word-Count based methods Analysis) was developed by the authors and included into the package SensoMineR (Husson, Lê, & Cadoret, 2013;.…”

The Effects of Storage Temperature on the Aroma of Whole Bean Arabica Coffee Evaluated by Coffee Consumers and HS-SPME-GC-MS

“…For each panelist's PM map, x-and y-coordinates for each sample, as well as descriptors and sample groupings, were recorded. Data were analyzed with multi-factor analysis (MFA), using the x-and y-coordinates, as well as frequencies for descriptors [18]. Confidence ellipses were simulated based on a parametric bootstrapping algorithm [19], and Hotelling's T 2 -test was used to determine whether samples show significant multivariate differences [20].…”

Contributions to assess the reproducibility and the agreement of respondents in CATA tasks

“…Such comparison of configurations can be done by comparing the sample configurations from the two sessions and, additionally, by means of multiblock data analysis, such as Generalized Procrustes Analysis (Arnold, 1986;Arnold & Williams, 1986;Gower, 1975), STATIS (Lavit, Escoufier, Sabatier, & Traissac, 1994), or MFA (Escofier & Pagès, 1994, 1998Pagès, 2013). Since the input tables in CATA tasks are contingency tables (here, one contingency table crossing the products in rows and the terms in columns is obtained for each session), one suitable solution to analyse the data consists in performing MFACT (Bécue-Bertaut & Pagès, 2004;Kostov et al, 2013). This analysis aims at balancing each group in the analysis by performing Correspondence Analysis (CA; Husson, Lê, & Pagès, 2011) on each separate group (a contingency table).…”

“…The methodologies proposed to assess the reproducibility at the panel level and the agreement are based on the McNemar test and Multiple Factor Analysis on Contingency Table (MFACT; Bécue -Bertaut & Pagès, 2004;Bécue-Bertaut, Álvarez-Esteban, & Pagès, 2008;Kostov, Bécue-Bertaut, & Husson, 2013) which includes the first eigenvalue, the partial points' representations, and the RV coefficients measured between the adequate pairs of configurations (Escoufier, 1973;Josse, Pagès, & Husson, 2008;see Tomic et al, 2013).…”