SEM適配度指標的潛藏問題：最佳模式難求

Potential Problems in the Assessment of SEM Model Fit: The Best Model is Hard to Anchor

李茂能(Mao-Neng Li)

《測驗統計年刊》 16期_下 (2008/12) Pp. 17-29

https://doi.org/10.6773/JRMS.200812.0007

摘要

本文旨在利用模擬實徵資料闡釋SEM適配度指標的潛藏迷思，文中透過幾個理論模式，具體說明「A good fit≠A good model」及「A good fit≠A causal link」，以呼籲研究者切勿過份依賴適配度指標去判定測量或結構模式之優劣及因果關係；文末並說明為何不佳之測量指標會扭曲結構參數之估計，因而測量模式的評鑑應先於結構模式。

關鍵字

SEM ； PLS ； 模式適配度 ； 對等模式 ； 原因指標 ； 效果指標

並列摘要

Several misconceptions or problems concerning the assessment of model fit in SEM are exemplified using simulated data and models. The author strongly argues that ”A good fit ≠ A good model or a good fit ≠ A causal link” and the assessment of measurement models should precede the assessment of structural models to avoid inflated structural parameter estimates.

並列關鍵字

SEM ； PLS ； Model fit ； Equivalent models ； Causal Indicators ； Effect indicators

參考文獻

Anderson, J. C.,Gerbing, D. W.(1988).Structural equation modeling in practice: A review and recommended two-step approach.Psychological Bulletin.103,411-423.

Bollen, K. A.(1989).Structural equations with latent variables.New York:John Wiley & Sons.

Bollen, K. A,Lennox, R.(1991).Coventional wisdom on measurement: A structural equation perspective.Psychological Bulletin.110(2),305-314.

Chin, W. W.(1998).Issues and opinion on structural equation modeling.Management Information System Quarterly.22(1),7-16.

Chin, W. W.,Newsted, P. R.,R. H. Hoyle (Ed.)(1999).Statistical strategies for small sample research.Thousand Oaks, CA:Sage.