Aplicación del análisis factorial confirmatorio a un modelo de medición del rendimiento académico en lectura
Fecha
Tipo
artículo original
Autores
Fernández Aráuz, Andrés
Título de la revista
ISSN de la revista
Título del volumen
Editor
Universidad de Costa Rica
Resumen
Descripción
El objetivo de este trabajo es mostrar mediante un ejemplo empírico las diferencias entre las técnicas estadísticas de análisis factorial exploratorio y confirmatorio. Los resultados obtenidos con el análisis factorial confirmatorio muestran que tanto el nivel educativo como el estatus socioeconómico, así como las posesiones del hogar son buenos indicadores para la medición de la variable latente llamada Extracción Socioeconómica. El gusto de los estudiantes por la lectura y la diversidad de lecturas son buenos indicadores de medición para la variable latente de Actitud Personal del estudiante, mientras que la actitud del estudiante hacia el centro educativo no es una buena medición para la variable latente de Actitud Personal. Estos resultados sugieren que los factores inobservables de Extracción Socioeconómica, Estrategias de Lectura y Actitud Personal podrían ser utilizados para evaluar la hipótesis causal del efecto de estas variables sobre el rendimiento educativo de los estudiantes en la prueba de lectura.
The aim of this paper is to show the differences between exploratory and confirmatory factor analysis. The results obtained with confirmatory factor analysis show that both education and socioeconomic status of the parents of students, as well as material possessions at home are good indicators to measure the latent variable called Socioeconomic Extraction. Moreover, the enjoyment of reading and diversity of reading are good indicators of measurement for the latent variable Personal Attitude of the student, while the student's attitude towards school, which measure attitudes to educational process as a whole and not specifically to the reading is not a good measure for the latent variable of personal attitude. These results suggest that unobservable factors Socioeconomic Extraction, Reading Strategies and personal attitude could be used to assess the causal hypothesis of the effect of these variables on the educational performance of students in the reading test.
The aim of this paper is to show the differences between exploratory and confirmatory factor analysis. The results obtained with confirmatory factor analysis show that both education and socioeconomic status of the parents of students, as well as material possessions at home are good indicators to measure the latent variable called Socioeconomic Extraction. Moreover, the enjoyment of reading and diversity of reading are good indicators of measurement for the latent variable Personal Attitude of the student, while the student's attitude towards school, which measure attitudes to educational process as a whole and not specifically to the reading is not a good measure for the latent variable of personal attitude. These results suggest that unobservable factors Socioeconomic Extraction, Reading Strategies and personal attitude could be used to assess the causal hypothesis of the effect of these variables on the educational performance of students in the reading test.
Palabras clave
Análisis causal, MODELOS ESTADÍSTICO, PISA, FACTOR ANALYSIS, VARIANCE-COVARIANCE MATRIX