The CALIS Procedure
-
Overview
-
Getting Started
-
SyntaxClasses of Statements in PROC CALISSingle-Group Analysis SyntaxMultiple-Group Multiple-Model Analysis SyntaxPROC CALIS StatementBOUNDS StatementBY StatementCOSAN StatementCOV StatementDETERM StatementEFFPART StatementFACTOR StatementFITINDEX StatementFREQ StatementGROUP StatementLINCON StatementLINEQS StatementLISMOD StatementLMTESTS StatementMATRIX StatementMEAN StatementMODEL StatementMSTRUCT StatementNLINCON StatementNLOPTIONS StatementOUTFILES StatementPARAMETERS StatementPARTIAL StatementPATH StatementPATHDIAGRAM StatementPCOV StatementPVAR StatementRAM StatementREFMODEL StatementRENAMEPARM StatementSAS Programming StatementsSIMTESTS StatementSTD StatementSTRUCTEQ StatementTESTFUNC StatementVAR StatementVARIANCE StatementVARNAMES StatementWEIGHT Statement
-
DetailsInput Data SetsOutput Data SetsDefault Analysis Type and Default ParameterizationThe COSAN ModelThe FACTOR ModelThe LINEQS ModelThe LISMOD Model and SubmodelsThe MSTRUCT ModelThe PATH ModelThe RAM ModelNaming Variables and ParametersSetting Constraints on ParametersAutomatic Variable SelectionPath Diagrams: Layout Algorithms, Default Settings, and CustomizationEstimation CriteriaRelationships among Estimation CriteriaGradient, Hessian, Information Matrix, and Approximate Standard ErrorsCounting the Degrees of FreedomAssessment of FitCase-Level Residuals, Outliers, Leverage Observations, and Residual DiagnosticsTotal, Direct, and Indirect EffectsStandardized SolutionsModification IndicesMissing Values and the Analysis of Missing PatternsMeasures of Multivariate KurtosisInitial EstimatesUse of Optimization TechniquesComputational ProblemsDisplayed OutputODS Table NamesODS Graphics
-
ExamplesEstimating Covariances and CorrelationsEstimating Covariances and Means SimultaneouslyTesting Uncorrelatedness of VariablesTesting Covariance PatternsTesting Some Standard Covariance Pattern HypothesesLinear Regression ModelMultivariate Regression ModelsMeasurement Error ModelsTesting Specific Measurement Error ModelsMeasurement Error Models with Multiple PredictorsMeasurement Error Models Specified As Linear EquationsConfirmatory Factor ModelsConfirmatory Factor Models: Some VariationsResidual Diagnostics and Robust EstimationThe Full Information Maximum Likelihood MethodComparing the ML and FIML EstimationPath Analysis: Stability of AlienationSimultaneous Equations with Mean Structures and Reciprocal PathsFitting Direct Covariance StructuresConfirmatory Factor Analysis: Cognitive AbilitiesTesting Equality of Two Covariance Matrices Using a Multiple-Group AnalysisTesting Equality of Covariance and Mean Matrices between Independent GroupsIllustrating Various General Modeling LanguagesTesting Competing Path Models for the Career Aspiration DataFitting a Latent Growth Curve ModelHigher-Order and Hierarchical Factor ModelsLinear Relations among Factor LoadingsMultiple-Group Model for Purchasing BehaviorFitting the RAM and EQS Models by the COSAN Modeling LanguageSecond-Order Confirmatory Factor AnalysisLinear Relations among Factor Loadings: COSAN Model SpecificationOrdinal Relations among Factor LoadingsLongitudinal Factor Analysis
- References
Various modeling languages are supported in PROC CALIS because researchers are trained in or adhere to different schools of modeling. Different modeling languages reflect different modeling terminology and philosophies. The statistical and mathematical consequences by using these various modeling languages, however, might indeed be the same. In other words, you can use more than one modeling languages for certain types of models without affecting the statistical analysis. Given the choices, which modeling language is preferred? There are two guidelines for this:
-
Use the modeling language that you are most familiar with.
-
Use the most specialized modeling language whenever it is possible.
The first guideline calls for researchers’ knowledge about a particular modeling language. Use the language you know the best. For example, some researchers might find equation input language like LINEQS the most suitable, while others might feel more comfortable using matrix input language like LISMOD.
The second guideline depends on the nature of the model at hand. For example, to specify a factor analysis model in the CALIS procedure, the specialized FACTOR language, instead of the LISMOD language, is recommended. Using a more specialized the modeling language is less error-prone. In addition, using a specialized language like FACTOR in this case amounts to giving the CALIS procedure additional information about the specific mathematical properties of the model. This additional information is used to enhance computational efficiency and to provide more specialized results. Another example is fitting an equi-covariance model. You can simply use the MSTRUCT model specification, in which you specify the same parameter for all off-diagonal elements of the covariance elements. This is direct and intuitive. Alternatively, you could tweak a LINEQS model that would predict the same covariance for all variables. However, this is indirect and error-prone, especially for novice modelers.
In PROC CALIS, the FACTOR and MSTRUCT modeling languages are considered more specialized, while other languages are more general in applications. Whenever possible, you should use the more specialized languages. However, if your model involves some novel covariance or mean structures that are not covered by the more specialized modeling languages, you can consider the more generalized modeling languages. See Example 29.33 for an application of the generalized COSAN model.