An investigation of moderate skewing of latent and error variables in longitudinal structural equation models /
The purpose of the present research was to investigate the effect of non-normality, specifically moderate skewing of latent, error and both latent and error variables in longitudinal bivariate structural equation model. Three models (i.e., the quasi-simplex, one-factor, and autoregressive) were exam...
| Main Author: | |
|---|---|
| Format: | Thesis Book |
| Language: | English |
| Published: |
[Place of publication not identified] :
[publisher not identified] ;
1999.
|
| Subjects: | |
| Online Access: | http://proxy.library.tamu.edu/login?url=http://proquest.umi.com/pqdweb?did=730340131&sid=1&Fmt=2&clientId=2945&RQT=309&VName=PQD |
| Summary: | The purpose of the present research was to investigate the effect of non-normality, specifically moderate skewing of latent, error and both latent and error variables in longitudinal bivariate structural equation model. Three models (i.e., the quasi-simplex, one-factor, and autoregressive) were examined through simulation study with various sample sizes (20, 50, 100, and 200). The conclusions included 1) the quasi-simplex data was fitted well by quasi-simplex and AR models because these two models have similar autoregressive structure; 2) the one-factor data was only supported well by the one-factor model; 3) AR data was supported well by both quasi-simplex and AR models. Generally, the Comparative Fit Index (CFI) exhibited very stable patterns and was less influenced by sample size or generated data types compared with other fit indices. In addition, the Tucker-Lewis Index (TLI) and Normed Fit Index (NFI) had very similar values with CFI for some cases. |
|---|---|
| Item Description: | Vita. "Major Subject: Educational Psychology". |
| Physical Description: | ix, 84 leaves : illustrations ; 28 cm. Issued also on microfiche from University Microfilm Inc. |
| Bibliography: | Includes bibliographical references (leaves 68-73). |