輔助變項對全訊息最大概似法表現之影響：非隨機遺漏情形之結構方程模型

Effects of Adding Auxiliary Variables to FIML in Structural Equation Models When Data Are Not Missing at Random

鄭中平(Chung-Ping Cheng) ； 翁儷禎(Li-Jen Weng)

《調查研究-方法與應用》 17期 (2005/04) Pp. 151-173

https://doi.org/10.7014/SRMA.2005040004

摘要

本研究目的在以蒙地卡羅法，瞭解加入輔助變項對結構方程模型全訊息最大概似法之參數估計與卡方值之影響。結果發現，加入輔助變項後，全訊息最大概似法參數估計更佳，僅當輔助變項與遺漏原因無關時，可能產生較大之估計偏誤。研究者應判斷輔助變項與遺漏原因之關係，以決定是否於分析時納入。

關鍵字

全訊息最大概似法 ； 輔助變項 ； 結構方程模型 ； 非隨機遺漏

並列摘要

The aim of this study is to explore the effect of adding auxiliary variables on parameter estimation and chi-squares statistic of full-information maximum likelihood method (FIML) in structural equation models. Results showed that FIML with auxiliary variables performed better in parameter estimation except when the auxiliary variables were uncorrelated with the causes of missingness. Researchers should consider the relationship between auxiliary variables and the cause of missingness before adding auxiliary variables in missing data treatment.

並列關鍵字

Full-Information Maximum Likelihood Method ； Auxiliary Variable ； Structural Equation Models ； Nonignorable Missingness

參考文獻

Allison, P. D.(2003).Missing data techniques for structural equation modeling.Journal of Abnormal Psychology.112,545-557.

Allison, P. D.(2003).Quantitative Applications in the social Science, series no. 07-136.Newbury Park, CA:Sage.

Arbuckle, J. L.(1996).Advanced structural equation modeling: Issues and techniques.Mahwah, NJ:Lawrence Eribaum Associates, Inc..

Bentler, P. M.(2001).EQS 6 structural equations program manual.Encino, CA:Multivariate Software.

Cohen, J.(1988).Statistical power analysis for the behavioral sciences.Mahwah, NJ:Lawrence Eribaum Associates, Inc..