Modelo de ecuaciones estructurales: Una guía para ciencias médicas y ciencias de la salud
Modelo de ecuaciones estructurales: Una guía para ciencias médicas y ciencias de la salud
Authors
Ortíz Parada, Manuel
Fernández Pera, Montserrat
Fernández Pera, Montserrat
Authors
Date
Datos de publicación:
10.4067/s0718-48082017000300047
Keywords
Cuidado de la salud - Precisión de medición - Modelos de ecuaciones estructurales - Literatura científica
Collections
Abstract
El modelo de ecuaciones estructurales (SEM) es una técnica de análisis estadística multivariada, que permite analizar patrones complejos de relaciones entre variables, realizar comparaciones entre e intragrupos, y validar modelos teóricos y empíricos. SEM puede ser utilizado para responder una amplia variedad de preguntas de investigación tanto en diseños experimentales como no experimentales. Pese a sus ventajas sobre técnicas tradicionales como la regresión múltiple o ANOVA, su uso en ciencias médicas y de la salud es poco frecuente. Por tanto, el objetivo de este artículo es introducir esta técnica de análisis a investigadores de las ciencias médicas y de la salud, explicando su aplicación con ejemplos del estudio chileno de predictores psicológicos de obesidad y síndrome metabólico (PPOMS). Se espera contribuir a la comprensión de esta técnica de análisis entre lectores de manuscritos científicos y estimular su uso entre investigadores de las ciencias médicas y de la salud.
Structural equation modeling (SEM) is a multivariate statistical analysis technique, utilized to analyze complex patterns of relationships among a set of variables, conduct between-groups and within-groups comparisons, and validate theoretical and empirical models. SEM can be used to answer several research questions including those formulated in the context of experimental and non-experimental designs. Despite the several advantages that SEM has over traditional procedures, such as multiple regression or ANOVA, it has not been applied frequently in the medical and health science domains. Therefore, the purpose of this article is to present SEM as a robust and comprehensive analytical technique capable of strengthening and increasing the accuracy of the analyses in medical and health research. The functioning and applications of SEM are illustrated through one research example: a Chilean study of psychological predictors of obesity and metabolic syndrome. This article is aimed at contributing to a better understanding of this technique among readers of scientific publications and facilitating its implementation in future research.
Structural equation modeling (SEM) is a multivariate statistical analysis technique, utilized to analyze complex patterns of relationships among a set of variables, conduct between-groups and within-groups comparisons, and validate theoretical and empirical models. SEM can be used to answer several research questions including those formulated in the context of experimental and non-experimental designs. Despite the several advantages that SEM has over traditional procedures, such as multiple regression or ANOVA, it has not been applied frequently in the medical and health science domains. Therefore, the purpose of this article is to present SEM as a robust and comprehensive analytical technique capable of strengthening and increasing the accuracy of the analyses in medical and health research. The functioning and applications of SEM are illustrated through one research example: a Chilean study of psychological predictors of obesity and metabolic syndrome. This article is aimed at contributing to a better understanding of this technique among readers of scientific publications and facilitating its implementation in future research.