So sánh phương pháp

	Phân tích trung gian mạnh mẽ	Mô hình phương trình cấu trúc mạnh mẽ
Lĩnh vực	Thống kê	Thống kê
Họ	Latent structure	Latent structure
Năm ra đời	2008–2014	1994
Người khởi xướng	Yuan & MacKinnon (median-regression formulation, 2014); robust bootstrap variants popularised by Hayes (2013) and Preacher & Hayes (2008)	Albert Satorra & Peter M. Bentler
Loại	Causal inference / indirect effects	Latent variable / path model with robust inference
Công trình gốc	Yuan, Y., & MacKinnon, D. P. (2014). Robust mediation analysis based on median regression. Psychological Methods, 19(1), 1–20.	Satorra, A. & Bentler, P. M. (1994). Corrections to test statistics and standard errors in covariance structure analysis. In A. von Eye & C. C. Clogg (Eds.), Latent variables analysis (pp. 399–419). Sage.
Tên gọi khác	robust indirect effects, outlier-resistant mediation, robust causal mediation	Robust SEM, SEM with robust standard errors, Satorra-Bentler SEM, non-normal SEM
Liên quan	5	5
Tóm tắt	Robust mediation analysis estimates the indirect effect of an independent variable on an outcome through one or more mediators using estimators that resist the influence of outliers and non-normal error distributions. By combining robust regression (such as median or M-estimation) with percentile or bias-corrected bootstrap confidence intervals, it yields trustworthy conclusions when standard ordinary-least-squares mediation would be distorted by extreme observations.	Robust structural equation modeling (Robust SEM) applies the full SEM framework — simultaneous estimation of measurement and structural relations among latent variables — while using corrected test statistics and sandwich standard errors that remain valid when observed data depart from multivariate normality. The Satorra-Bentler scaled chi-square is the most widely used correction.