Draw your models with a new SAS application for JMP
Jul 27, 2011 1:11 PM
SAS® Structural Equation Modeling for JMP® is a new application that enables researchers to use SAS and JMP to draw models by using an interface that is built on the SAS/STAT® CALIS procedure.
To create models in SAS Structural Equation Modeling for JMP, you simply drag variables into the diagram area and then use point-and-click features to draw paths among variables, specify variable and path properties, and select other model specifications. SAS Structural Equation Modeling for JMP enables both novice and experienced users of structural equation models to build models easily with no programming involved.
How is structural equation modeling (SEM) used?
Structural equation modeling in academia:
Psychologists have used SEM to examine the properties of personality and depression tests.
Sociologists and criminologists have used SEM to understand what personal and environmental characteristics can be used to predict antisocial behavior.
Public health researchers have used SEM to understand how to improve health communication and health outcomes.
Education researchers have used SEM to model changes in reading and math scores over time.
Marketing researchers have used SEM to understand what factors influence future product purchases.
Business researchers have used SEM to understand what personal and environmental characteristics can be used to predict entrepreneurial intention.
Structural equation modeling in industry:
A large high tech company uses SEM to map customer satisfaction metrics.
A consumer packaged goods company uses SEM to analyze survey responses from consumer product tests.
An insurance company has applied SEM to bring insight into factors that influence customers to buy insurance.
For what models can I use SAS Structural Equation Modeling for JMP?
With SAS Structural Equation Modeling for JMP, you can fit either models that have only observed variables (including linear regression and path analysis models) or models with observed and latent variables (including confirmatory factor analysis and latent growth curve models). Here are some examples of the models you can fit with SAS Structural Equation Modeling for JMP:
This example shows a linear regression model in which the number of employees (N_emp), the advertising spending (Advert), and last year’s sales (LastS) are used to predict the current year’s sales (CurrentS)
This example shows a confirmatory factor analysis (CFA) model with correlated latent variables (Read and Math), which predict observed test scores from three math tests (math1, math2, and math3) and three reading tests (reading1, reading2, and reading3)
Typically, you have several models you want to examine. With SAS Structural Equation Modeling for JMP, you can fit several models to the same data set using the Single Group Analysis option. You can easily compare multiple models using the Comparisons tab.
Example of the Comparisons tab
What are the software requirements for SAS Structural Equation Modeling for JMP?
SAS Structural Equation Modeling for JMP requires:
Base SAS 9.2 or later.
SAS/STAT 9.22 or later.
Windows versions of JMP 9.0.2 or later; JMP Pro 9.0.2 or later.
Macintosh support is coming soon.
How do I acquire SAS Structural Equation Modeling for JMP?
You can acquire SAS Structural Equation Modeling for JMP in one of the following ways: