Search Results
Title: PROC GLM or PROC CALIS
Resulting in 1 citation.
| 1. |
Suhr, Diana D. |
|
PROC GLM or PROC CALIS Presented: Long Beach, CA, SAS Users Group International Conference, April 2001. Also: http://www.sas.com/usergroups/sugi/sugi26/grid.wedam.html Cohort(s): Children of the NLSY79 Publisher: SAS Institute Inc. Keyword(s): Children, Academic Development; Children, School-Age; Cognitive Development; Gender Differences; Longitudinal Data Sets; Longitudinal Surveys; Methods/Methodology; Modeling; NLS Description; Peabody Individual Achievement Test (PIAT- Reading); Statistical Analysis; Statistics Permission to reprint the abstract has not been received from the publisher. Traditional statistical approaches to data analysis use PROC GLM, whereas Structural Equation Modeling (SEM) techniques use PROC CALIS. Regression, analysis of variance (anova), or repeated measures anova are traditional methods using PROC GLM. SEM with PROC CALIS is a comprehensive and flexible approach to multivariate analysis using observed (measured) and unobserved (latent) variables (Hoyle, 1995). Research hypothesis typically tested by traditional methods may be tested using SEM techniques. Similarities and differences between traditional and SEM approaches will be discussed. A repeated measures analysis of variances example illustrates traditional and SEM methodologies. The repeated measures anova hypothesizes differences between the means and linear trends. The SEM analysis estimates nonlinear trends, variances of measured and latent variables, relationship between latent variable variances, means of latent variables (initial level and rate of change). The audience could be beginner through advanced SAS programmers. |
|
|
Bibliography Citation
Suhr, Diana D. "PROC GLM or PROC CALIS." Presented: Long Beach, CA, SAS Users Group International Conference, April 2001. |