A note on a two-stage least squares estimator for higher-order factor analyses

被引:15
作者
Bollen, KA [1 ]
Biesanz, JC
机构
[1] Univ N Carolina, Odum Inst Res Social Sci, Chapel Hill, NC 27514 USA
[2] Univ Wisconsin, Dept Psychol, Madison, WI 53706 USA
关键词
D O I
10.1177/0049124102030004004
中图分类号
O1 [数学]; C [社会科学总论];
学科分类号
03 ; 0303 ; 0701 ; 070101 ;
摘要
This article extends Bollen's two-stage least squares (2SLS) estimator to estimate confirmatory higher-order factor analysis models. This includes estimation of the higher order and lower-order factor loadings and their intercepts. This 2SLS estimator does not require that the observed variables come from normal distributions, and it is less sensitive to specification errors than are the full information estimators. As such, this makes the 2SLS estimator a useful complementary estimator to the other full information estimators.
引用
收藏
页码:568 / 579
页数:12
相关论文
共 18 条
[1]  
[Anonymous], 1997, BEHAVIORMETRIKA, DOI DOI 10.2333/BHMK.24.147
[2]  
[Anonymous], 1989, STRUCTURAL EQUATIONS
[3]  
[Anonymous], 1993, NEW FEATURE LISREL 8
[4]  
BASMANN RL, 1960, J AM STAT ASSOC, V55, P650
[6]   Moderators of self-other agreement: Reconsidering temporal stability in personality [J].
Biesanz, JC ;
West, SG ;
Graziano, WG .
JOURNAL OF PERSONALITY AND SOCIAL PSYCHOLOGY, 1998, 75 (02) :467-477
[7]   An alternative two stage least squares (2SLS) estimator for latent variable equations [J].
Bollen, KA .
PSYCHOMETRIKA, 1996, 61 (01) :109-121
[8]  
Bollen KA., 1996, ADV STRUCTURAL EQUAT, P227
[9]   Interactions of Latent Variables in Structural Equation Models [J].
Bollen, Kenneth A. ;
Paxton, Pamela .
STRUCTURAL EQUATION MODELING-A MULTIDISCIPLINARY JOURNAL, 1998, 5 (03) :267-293
[10]  
Davidson R., 1993, ESTIMATION INFERENCE, DOI DOI 10.1017/S0266466600009452