A cluster-based factor rotation

Michio Yamamoto, Robert I. Jennrich

Research output: Contribution to journalArticle

1 Citation (Scopus)

Abstract

A new oblique factor rotation method is proposed, the aim of which is to identify a simple and well-clustered structure in a factor loading matrix. A criterion consisting of the complexity of a factor loading matrix and a between-cluster dissimilarity is optimized using the gradient projection algorithm and the k-means algorithm. It is shown that if there is an oblique rotation of an initial loading matrix that has a perfect simple structure, then the proposed method with Kaiser's normalization will produce the perfect simple structure. Although many rotation methods can also recover a perfect simple structure, they perform poorly when a perfect simple structure is not possible. In this case, the new method tends to perform better because it clusters the loadings without requiring the clusters to be perfect. Artificial and real data analyses demonstrate that the proposed method can give a simple structure, which the other methods cannot produce, and provides a more interpretable result than those of widely known rotation techniques.

Original languageEnglish
Pages (from-to)488-502
Number of pages15
JournalBritish Journal of Mathematical and Statistical Psychology
Volume66
Issue number3
DOIs
Publication statusPublished - Nov 2013
Externally publishedYes

Fingerprint

Oblique
Gradient Projection
Projection Algorithm
K-means Algorithm
Gradient Algorithm
Dissimilarity
Normalization
Tend
Demonstrate
Artificial

ASJC Scopus subject areas

  • Psychology(all)
  • Statistics and Probability
  • Arts and Humanities (miscellaneous)
  • Medicine(all)

Cite this

A cluster-based factor rotation. / Yamamoto, Michio; Jennrich, Robert I.

In: British Journal of Mathematical and Statistical Psychology, Vol. 66, No. 3, 11.2013, p. 488-502.

Research output: Contribution to journalArticle

@article{820c1850d9b84fe8b2d88759032170aa,
title = "A cluster-based factor rotation",
abstract = "A new oblique factor rotation method is proposed, the aim of which is to identify a simple and well-clustered structure in a factor loading matrix. A criterion consisting of the complexity of a factor loading matrix and a between-cluster dissimilarity is optimized using the gradient projection algorithm and the k-means algorithm. It is shown that if there is an oblique rotation of an initial loading matrix that has a perfect simple structure, then the proposed method with Kaiser's normalization will produce the perfect simple structure. Although many rotation methods can also recover a perfect simple structure, they perform poorly when a perfect simple structure is not possible. In this case, the new method tends to perform better because it clusters the loadings without requiring the clusters to be perfect. Artificial and real data analyses demonstrate that the proposed method can give a simple structure, which the other methods cannot produce, and provides a more interpretable result than those of widely known rotation techniques.",
author = "Michio Yamamoto and Jennrich, {Robert I.}",
year = "2013",
month = "11",
doi = "10.1111/bmsp.12007",
language = "English",
volume = "66",
pages = "488--502",
journal = "British Journal of Mathematical and Statistical Psychology",
issn = "0007-1102",
publisher = "Wiley-Blackwell",
number = "3",

}

TY - JOUR

T1 - A cluster-based factor rotation

AU - Yamamoto, Michio

AU - Jennrich, Robert I.

PY - 2013/11

Y1 - 2013/11

N2 - A new oblique factor rotation method is proposed, the aim of which is to identify a simple and well-clustered structure in a factor loading matrix. A criterion consisting of the complexity of a factor loading matrix and a between-cluster dissimilarity is optimized using the gradient projection algorithm and the k-means algorithm. It is shown that if there is an oblique rotation of an initial loading matrix that has a perfect simple structure, then the proposed method with Kaiser's normalization will produce the perfect simple structure. Although many rotation methods can also recover a perfect simple structure, they perform poorly when a perfect simple structure is not possible. In this case, the new method tends to perform better because it clusters the loadings without requiring the clusters to be perfect. Artificial and real data analyses demonstrate that the proposed method can give a simple structure, which the other methods cannot produce, and provides a more interpretable result than those of widely known rotation techniques.

AB - A new oblique factor rotation method is proposed, the aim of which is to identify a simple and well-clustered structure in a factor loading matrix. A criterion consisting of the complexity of a factor loading matrix and a between-cluster dissimilarity is optimized using the gradient projection algorithm and the k-means algorithm. It is shown that if there is an oblique rotation of an initial loading matrix that has a perfect simple structure, then the proposed method with Kaiser's normalization will produce the perfect simple structure. Although many rotation methods can also recover a perfect simple structure, they perform poorly when a perfect simple structure is not possible. In this case, the new method tends to perform better because it clusters the loadings without requiring the clusters to be perfect. Artificial and real data analyses demonstrate that the proposed method can give a simple structure, which the other methods cannot produce, and provides a more interpretable result than those of widely known rotation techniques.

UR - http://www.scopus.com/inward/record.url?scp=84885406859&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84885406859&partnerID=8YFLogxK

U2 - 10.1111/bmsp.12007

DO - 10.1111/bmsp.12007

M3 - Article

C2 - 23398639

AN - SCOPUS:84885406859

VL - 66

SP - 488

EP - 502

JO - British Journal of Mathematical and Statistical Psychology

JF - British Journal of Mathematical and Statistical Psychology

SN - 0007-1102

IS - 3

ER -