Measures of simplicity and factorial fit: An approach for the evaluation of factorially derived scales

• Autores: Fleming, James S., Merino Soto, César
• Formato: artículo
• Fecha de Publicación: 2005
• Institución: Pontificia Universidad Católica del Perú
• Repositorio: Revistas - Pontificia Universidad Católica del Perú
• Lenguaje: español
• OAI Identifier: oai:ojs.pkp.sfu.ca:article/2150
• Enlace del recurso: http://revistas.pucp.edu.pe/index.php/psicologia/article/view/2150
• Nivel de acceso: acceso abierto
• Materia: factor analysis; scales; psychometric research; análisis factorial; escalas; investigación psicométrica

A very simple structure is sought when using factor analysis to develop measurement scales. The present article is about the SIMLOAD program; it computes measures of factor simplicity for rows and columns of loading matrices (usually the factor pattern) as well assome overall measures. These include Kaiser’s (1974) index of factor simplicity for variables (rows), Fleming’s scale fit index for factors (columns), Bentler’s (1977) scale-free matrix measure, and hyperplane counts. Routine use of these measures is recommended for multifactor scale development. The measures may also be useful in more general factorapplications, and in confirmatory as well as  exploratory analysis. SIMLOAD additionally computes factor scale intercorrelations, scale alpha coefficients, including alpha when item removed, and sorted loadings for ease of interpretation.