Discriminant Validity as a Criterion for Assessing Scales: Theory or Statistics?

Discriminant validity is one of the usual criterions for evaluating measurement scales that define latent constructs in social sciences. This article shows how different statistical procedures frequently used for accomplishing this aim can yield misleading results. Authors recommend a theoretical ju...

Descripción completa

Autor Principal: Martínez García, José Antonio; Universidad Politécnica de Cartagena
Otros Autores: Martínez Caro, Laura; Universidad Politécnica de Cartagena
Formato: info:eu-repo/semantics/article
Idioma: spa
Publicado: Pontificia Universidad Javeriana 2009
Materias:
Acceso en línea: http://revistas.javeriana.edu.co/index.php/revPsycho/article/view/224
Etiquetas: Agregar Etiqueta
Sin Etiquetas, Sea el primero en etiquetar este registro!
Sumario: Discriminant validity is one of the usual criterions for evaluating measurement scales that define latent constructs in social sciences. This article shows how different statistical procedures frequently used for accomplishing this aim can yield misleading results. Authors recommend a theoretical judgement about divergence among scales that are manifestation of latent concepts. Therefore, content validity represents a robust condition against certain covariance statistical based analysis.