Polychoric/Tetrachoric Matrix or Pearson Matrix? A methodological study.

Mostrar todas las versiones(2)

The use of product-moment correlation of Pearson is common in most studies in factor analysis in psychology, but it is known that this statistic is only applicable when the variables related are in interval scale and normally distributed, and when are used in ordinal data may to produce a distorted...

Descripción completa

Guardado en:
Detalles Bibliográficos
Autor principal: Dominguez Lara, Sergio Alexis
Formato: Artículo revista
Lenguaje:Español
Publicado: Universidad Nacional de Córdoba 2014
Materias:
análisis factorial
matrices tetracóricas
matrices policóricas
validez
Psicología
Psicometría
factor analysis
polychoric matrix
tetrachoric matrix
validity
Acceso en línea:https://revistas.unc.edu.ar/index.php/racc/article/view/6357
Aporte de:
Revistas de la UNC de Universidad Nacional de Córdoba
id I10-R10-article-6357
record_format ojs
institution Universidad Nacional de Córdoba
institution_str I-10
repository_str R-10
container_title_str Revistas de la UNC
language Español
format Artículo revista
topic análisis factorial
matrices tetracóricas
matrices policóricas
validez
Psicología
Psicometría
factor analysis
polychoric matrix
tetrachoric matrix
validity
spellingShingle análisis factorial
matrices tetracóricas
matrices policóricas
validez
Psicología
Psicometría
factor analysis
polychoric matrix
tetrachoric matrix
validity
Dominguez Lara, Sergio Alexis
Polychoric/Tetrachoric Matrix or Pearson Matrix? A methodological study.
topic_facet análisis factorial
matrices tetracóricas
matrices policóricas
validez
Psicología
Psicometría
factor analysis
polychoric matrix
tetrachoric matrix
validity
author Dominguez Lara, Sergio Alexis
author_facet Dominguez Lara, Sergio Alexis
author_sort Dominguez Lara, Sergio Alexis
title Polychoric/Tetrachoric Matrix or Pearson Matrix? A methodological study.
title_short Polychoric/Tetrachoric Matrix or Pearson Matrix? A methodological study.
title_full Polychoric/Tetrachoric Matrix or Pearson Matrix? A methodological study.
title_fullStr Polychoric/Tetrachoric Matrix or Pearson Matrix? A methodological study.
title_full_unstemmed Polychoric/Tetrachoric Matrix or Pearson Matrix? A methodological study.
title_sort polychoric/tetrachoric matrix or pearson matrix? a methodological study.
description The use of product-moment correlation of Pearson is common in most studies in factor analysis in psychology, but it is known that this statistic is only applicable when the variables related are in interval scale and normally distributed, and when are used in ordinal data may to produce a distorted correlation matrix . Thus is a suitable option using polychoric/tetrachoric matrices in item-level factor analysis when the items are in level measurement nominal or ordinal. The aim of this study was to show the differences in the KMO, Bartlett`s Test and Determinant of the Matrix, percentage of variance explained and factor loadings in depression trait scale of Depression Inventory Trait - State and the Neuroticism dimension of the short form of the Eysenck Personality Questionnaire -Revised, regarding the use of matrices polychoric/tetrachoric matrices and Pearson. These instruments was analyzed with different extraction methods (Maximum Likelihood, Minimum Rank Factor Analysis, Unweighted Least Squares and Principal Components), keeping constant the rotation method Promin were analyzed. Were observed differences regarding sample adequacy measures, as well as with respect to the explained variance and the factor loadings, for solutions having as polychoric/tetrachoric matrix. So it can be concluded that the polychoric / tetrachoric matrix give better results than Pearson matrices when it comes to item-level factor analysis using different methods.
publisher Universidad Nacional de Córdoba
publishDate 2014
url https://revistas.unc.edu.ar/index.php/racc/article/view/6357
work_keys_str_mv AT dominguezlarasergioalexis polychorictetrachoricmatrixorpearsonmatrixamethodologicalstudy
AT dominguezlarasergioalexis matricespolicoricastetracoricasomatricespearsonunestudiometodologico
first_indexed 2022-08-20T00:49:23Z
last_indexed 2022-08-20T00:49:23Z
_version_ 1770716104829698048
spelling I10-R10-article-63572019-05-27T16:17:36Z Polychoric/Tetrachoric Matrix or Pearson Matrix? A methodological study. ¿Matrices Policóricas/Tetracóricas o Matrices Pearson? Un estudio metodológico Dominguez Lara, Sergio Alexis análisis factorial matrices tetracóricas matrices policóricas validez Psicología Psicometría factor analysis polychoric matrix tetrachoric matrix validity The use of product-moment correlation of Pearson is common in most studies in factor analysis in psychology, but it is known that this statistic is only applicable when the variables related are in interval scale and normally distributed, and when are used in ordinal data may to produce a distorted correlation matrix . Thus is a suitable option using polychoric/tetrachoric matrices in item-level factor analysis when the items are in level measurement nominal or ordinal. The aim of this study was to show the differences in the KMO, Bartlett`s Test and Determinant of the Matrix, percentage of variance explained and factor loadings in depression trait scale of Depression Inventory Trait - State and the Neuroticism dimension of the short form of the Eysenck Personality Questionnaire -Revised, regarding the use of matrices polychoric/tetrachoric matrices and Pearson. These instruments was analyzed with different extraction methods (Maximum Likelihood, Minimum Rank Factor Analysis, Unweighted Least Squares and Principal Components), keeping constant the rotation method Promin were analyzed. Were observed differences regarding sample adequacy measures, as well as with respect to the explained variance and the factor loadings, for solutions having as polychoric/tetrachoric matrix. So it can be concluded that the polychoric / tetrachoric matrix give better results than Pearson matrices when it comes to item-level factor analysis using different methods. El uso de la correlación producto-momento de Pearson es frecuente en la mayoría de estudios en análisis factorial en psicología, mas es conocido que este estadístico solo es aplicable cuando las variables que se relacionan están al menos en escala de intervalo y distribuidas normalmente; su utilización en datos ordinales puede resultar en una matriz de correlaciones distorsionada. Es así que se plantea como una opción adecuada el uso de matrices policóricas/tetracóricas a nivel de ítem en análisis factorial cuando los reactivos están en un nivel de medición nominal u ordinal. El objetivo de este trabajo fue mostrar las diferencias en cuanto a las medidas de adecuación muestral (KMO, Test de Esfericidad de Bartlett y Determinante de la Matriz), porcentaje de varianza explicada y saturaciones factoriales tanto en la escala Depresión Rasgo del Inventario de Depresión Rasgo-Estado (Spielberger, Agudelo & Buela Casal, 2008) como en la dimensión Neuroticismo de la forma corta del Cuestionario de Personalidad de Eysenck-Revisado (Eysenck & Eysenck, 2008) con relación al uso de matrices Policóricas/Tetracóricas y matrices Pearson. Se analizaron dichos instrumentos con diversos métodos de extracción (Máxima verosimilitud, Minimum Rank Factor Analysis, Mínimos Cuadrados No Ponderados y Componentes Principales), manteniendo constante el método de rotación Promin. Se apreciaron diferencias con relación a las medidas de adecuación muestral, así como con  respecto a la varianza explicada y las cargas factoriales, a favor de las soluciones que tienen como base matrices policóricas/tetracóricas. Se concluye que las matrices policóricas/tetracóricas permiten obtener mejores resultados que las matrices Pearson cuando se trata de análisis factorial a nivel de ítem utilizando distintos métodos. Universidad Nacional de Córdoba 2014-04-02 info:eu-repo/semantics/article info:eu-repo/semantics/publishedVersion application/pdf https://revistas.unc.edu.ar/index.php/racc/article/view/6357 Argentinean Journal of Behavioral Sciences; Vol. 6 No. 1 (2014): Revista Argentina de Ciencias del Comportamiento; 39-48 Revista Argentina de Ciencias del Comportamiento; Vol. 6 Núm. 1 (2014): Revista Argentina de Ciencias del Comportamiento; 39-48 1852-4206 10.32348/1852.4206.v6.n1 spa https://revistas.unc.edu.ar/index.php/racc/article/view/6357/Lara2014 Derechos de autor 2014 Sergio Alexis Dominguez Lara

Ejemplares similares