Testing Structural Equation Models, Volume 154

Capa
Kenneth A. Bollen, J. Scott Long
SAGE, 1993 - 320 páginas
What is the role of fit measures when respecifying a model? Should the means of the sampling distributions of a fit index be unrelated to the size of the sample? Is it better to estimate the statistical power of the chi-square test than to turn to fit indices? Exploring these and related questions, well-known scholars examine the methods of testing structural equation models (SEMS) with and without measurement error, as estimated by such programs as EQS, LISREL and CALIS.
 

Páginas selecionadas

Conteúdo

Introduction
1
Multifaceted Conceptions of Fit in Structural
10
Monte Carlo Evaluations of GoodnessofFit Indices
40
Some Specification Tests for
66
Bootstrapping GoodnessofFit Measures
111
Alternative Ways of Assessing Model
136
Direitos autorais

Termos e frases comuns

1K SAMPLE AGFI alternative approximation assessment assumptions asymptotic Bentler Bollen Bonett bootstrap Browne categorical chi-square test coefficients components computed correctly specified correlation covariance structure analysis covariance structure models Cudeck degrees of freedom DELTA2 discrepancy function discussed ECVI eigenvalues evaluate example expected value factor analysis fit index fit indices fit measures fitting function fixed parameters free parameters goodness-of-fit Hall version heteroscedasticity kurtosis latent variables least squares linear LISREL LM statistic maximum likelihood estimation misspecification model fit modified bootstrap multivariate Muthén noncentrality parameter nonnormal null hypothesis null model observed variables obtained overall parameter estimates PERCENT REJECTED polychoric positive definite probit procedure Psychological Psychometrika regression residuals restrictions robust sample covariance matrix sample size Saris Satorra saturated model skewness specification tests specified model standard errors Steiger structural equation models substantive Table Tanaka test statistic tetrachoric tion variance vector zero

Sobre o autor (1993)

Scott Long is Distinguished Professor and Chancellor′s Professor of Sociology and Statistics at Indiana University, Bloomington. He teaches quantitative methods both at Indiana University and at the ICSPR Summer Program. His earlier research examined gender differences in the scientific career. In recent years, he has collaborated with Eliza Pavalko, Bernice Pescsolido, John Bancroft, Julia Heiman and others in studies of health and aging, stigma and mental health, and human sexuality.

Informações bibliográficas

TítuloTesting Structural Equation Models, Volume 154
Volume 154 de SAGE Focus Editions
Testing Structural Equation Models, Kenneth A. Bollen
EditoresKenneth A. Bollen, J. Scott Long
Ediçãoilustrada
EditoraSAGE, 1993
ISBN0803945078, 9780803945074
Num. págs.320 páginas
  
Exportar citaçãoBiBTeX EndNote RefMan