[R-sig-eco] Question: Is it correct to use HCPC analysis (FactoMineR package) on categorical data

[Adnan Sisic]

Dear Francois,
Thanks for the reply. Really helpful!
Best
Adnan

________________________________
From: francois gillet <francois.gillet using univ-fcomte.fr>
Sent: Monday, July 18, 2022 5:48 PM
To: Adnan Sisic <adnan_sisic using live.com>
Cc: R-sig-ecology using r-project.org <r-sig-ecology using r-project.org>
Subject: Re: [R-sig-eco] Question: Is it correct to use HCPC analysis (FactoMineR package) on categorical data

Dear Adrian,

Your approach seems correct to me. Your disease severity variable is not categorical but ordinal (semi-quantitative) and you can perform PCA (on the covariance matrix: scale.unit = FALSE) and HCPC on the mean scores as suggested. You can also directly perform a hierarchical clustering of the 109 isolates by applying hclust() to the Euclidean distance matrix.

Best wishes,

François

-------------------------------------------------------------------------------
François Gillet , Emeritus Professor
Université Bourgogne Franche-Comté - CNRS
UMR 6249 Chrono-environnement
UFR Sciences et Techniques
16, Route de Gray
F-25030 Besançon cedex
France
[ http://chrono-environnement.univ-fcomte.fr/ | http://chrono-environnement.univ-fcomte.fr/ ]
Phone: +33 (0)3 81 66 62 81
iPhone: +33 (0)7 88 37 07 76
Location: La Bouloie, Bât. Propédeutique, -114L
-------------------------------------------------------------------------------
Associate Editor of Plant Ecology and Evolution
[ http://www.plecevo.eu/ | http://www.plecevo.eu ]
-------------------------------------------------------------------------------
Homepage: [ https://chrono-environnement.univ-fcomte.fr/spip.php?page=perso&nom=GILLET&prenom=Fran%C3%A7ois&lang=en | https://chrono-environnement.univ-fcomte.fr/spip.php?page=perso&nom=GILLET&prenom=Fran%C3%A7ois&lang=en ]
ResearchID : [ http://www.researcherid.com/rid/B-6160-2008 | https://publons.com/researcher/1228828/francois-gillet/ ]
ORCID ID: [ http://orcid.org/0000-0002-3334-1069 | http://orcid.org/0000-0002-3334-1069 ]
Google Scholar : [ http://scholar.google.com.au/citations?user=a5xiIfQAAAAJ | http://scholar.google.com.au/citations?user=a5xiIfQAAAAJ ]
-------------------------------------------------------------------------------
While I may be sending this email outside my normal office hours,
I have no expectation to receive a reply outside of yours.
Il se peut que j'envoie ce courriel en dehors de mes heures de travail normales,
je ne m'attends pas à recevoir une réponse en dehors des vôtres.
-------------------------------------------------------------------------------

----- Mail original -----
De: "Adnan Sisic" <adnan_sisic using live.com>
À: "R-sig-ecology using r-project.org" <r-sig-ecology using r-project.org>
Envoyé: Vendredi 15 Juillet 2022 12:32:40
Objet: [R-sig-eco] Question: Is it correct to use HCPC analysis (FactoMineR package) on categorical data

Dear Sig. Ecology members,

I have categorical disease severity data, disease scale - scores from 1 (healthy plant) to 10 (heavily diseased/dying plant). In total  I have 109 isolates tested, each in 3 replicates. Is it correct to use mean score of 3 replicates for each isolate as inputs, calculate Euclidean distances between isolates and perform Hierarchical Clustering on Principal Components (HCPC) in FactoMineR package?

Thanks and best regards,



----

FG �kologischer Pflanzenschutz
Universit�t Kassel
Nordbahnhofstr. 1a
D-37213 Witzenhausen
Deutschland

Dept. of Ecological Plant Protection
University of Kassel
Nordbahnhofstr. 1a
D-37213 Witzenhausen
Germany

Tel. 05542 98 1569
Fax. 05542 98 1564

www.uni-kassel.de/go/fb11-oekologischer-pflanzenschutz<http://www.uni-kassel.de/go/fb11-oekologischer-pflanzenschutz>

        [[alternative HTML version deleted]]


_______________________________________________
R-sig-ecology mailing list
R-sig-ecology using r-project.org
https://stat.ethz.ch/mailman/listinfo/r-sig-ecology

	[[alternative HTML version deleted]]