方法对比

	多群体信度分析	概化理论（G-Theory）
领域	心理测量学	心理测量学
方法族	Latent structure	Latent structure
起源年份	1990s–2000s	1963–1972
提出者	Classical test theory traditions; synthesized in modern practice by Vandenberg & Lance (2000) and Sijtsma (2009)	Lee J. Cronbach, Goldine Gleser, Harinder Nanda, Nageswari Rajaratnam
类型	Reliability estimation and comparison	Variance-components reliability model
开创性文献	Vandenberg, R. J. & Lance, C. E. (2000). A review and synthesis of the measurement invariance literature: Suggestions, practices, and recommendations for organizational research. Organizational Research Methods, 3(1), 4–70.	Cronbach, L. J., Gleser, G. C., Nanda, H. & Rajaratnam, N. (1972). The Dependability of Behavioral Measurements: Theory of Generalizability for Scores and Profiles. Wiley.
别名	reliability comparison across groups, group-specific reliability estimation, multi-sample reliability analysis, cross-group internal consistency	G-theory, G-study / D-study framework, variance components reliability
相关	4	4
摘要	Multi-group reliability analysis estimates internal consistency or stability coefficients separately within each group and then formally compares them to determine whether a scale functions with equal precision across populations. It is a foundational step in cross-group measurement research, typically carried out alongside or prior to measurement invariance testing.	Generalizability Theory is a psychometric framework that decomposes observed score variance into multiple sources — persons, items, raters, occasions, and their interactions — using analysis of variance. It replaces the single reliability coefficient of classical test theory with a family of coefficients that tell researchers how well scores generalize across different measurement conditions.