The CALIS Procedure
-
Overview
-
Getting Started
-
Syntax
Classes of Statements in PROC CALIS Single-Group Analysis Syntax Multiple-Group Multiple-Model Analysis Syntax PROC CALIS Statement BOUNDS Statement BY Statement COSAN Statement COV Statement DETERM Statement EFFPART Statement FACTOR Statement FITINDEX Statement FREQ Statement GROUP Statement LINCON Statement LINEQS Statement LISMOD Statement LMTESTS Statement MATRIX Statement MEAN Statement MODEL Statement MSTRUCT Statement NLINCON Statement NLOPTIONS Statement OUTFILES Statement PARAMETERS Statement PARTIAL Statement PATH Statement PCOV Statement PVAR Statement RAM Statement REFMODEL Statement RENAMEPARM Statement SAS Programming Statements SIMTESTS Statement STD Statement STRUCTEQ Statement TESTFUNC Statement VAR Statement VARIANCE Statement VARNAMES Statement WEIGHT Statement
-
Details
Input Data Sets Output Data Sets The COSAN Model The FACTOR Model The LINEQS Model The LISMOD Model and Submodels The MSTRUCT Model The PATH Model The RAM Model Naming Variables and Parameters Setting Constraints on Parameters Automatic Variable Selection Estimation Criteria Relationships among Estimation Criteria Gradient, Hessian, Information Matrix, and Approximate Standard Errors Counting the Degrees of Freedom Assessment of Fit Total, Direct, and Indirect Effects Standardized Solutions Modification Indices Missing Values and the Analysis of Missing Patterns Measures of Multivariate Kurtosis Initial Estimates Use of Optimization Techniques Computational Problems Displayed Output ODS Table Names ODS Graphics
-
Examples
Estimating Covariances and Correlations Estimating Covariances and Means Simultaneously Testing Uncorrelatedness of Variables Testing Covariance Patterns Testing Some Standard Covariance Pattern Hypotheses Linear Regression Model Multivariate Regression Models Measurement Error Models Testing Specific Measurement Error Models Measurement Error Models with Multiple Predictors Measurement Error Models Specified As Linear Equations Confirmatory Factor Models Confirmatory Factor Models: Some Variations The Full Information Maximum Likelihood Method Comparing the ML and FIML Estimation Path Analysis: Stability of Alienation Simultaneous Equations with Mean Structures and Reciprocal Paths Fitting Direct Covariance Structures Confirmatory Factor Analysis: Cognitive Abilities Testing Equality of Two Covariance Matrices Using a Multiple-Group Analysis Testing Equality of Covariance and Mean Matrices between Independent Groups Illustrating Various General Modeling Languages Testing Competing Path Models for the Career Aspiration Data Fitting a Latent Growth Curve Model Higher-Order and Hierarchical Factor Models Linear Relations among Factor Loadings Multiple-Group Model for Purchasing Behavior Fitting the RAM and EQS Models by the COSAN Modeling Language Second-Order Confirmatory Factor Analysis Linear Relations among Factor Loadings: COSAN Model Specification Ordinal Relations among Factor Loadings Longitudinal Factor Analysis
- References
| FITINDEX Statement |
- FITINDEX option <option ...> ;
You can use the FITINDEX statement to set the options for computing and displaying the fit indices, or to output the fit indices. All but the OFF= and ON= options of the FITINDEX statement are also available in the PROC CALIS statement. The options set in the FITINDEX statement will overwrite those set in the PROC CALIS statement.
For the listing of fit indices and their definitions, see the section Overall Model Fit Indices. Note that not all fit indices are available with all estimation methods, which is specified by the METHOD= option of the PROC CALIS statement. See the section Fit Indices and Estimation Methods for more details.
The options of the FITINDEX statement are as follows:
-
ALPHAECV=
specifies a
confidence interval (
) for the Browne and Cudeck (1993) expected cross validation index (ECVI). See the ALPHAECV= option of the PROC CALIS statement. -
ALPHARMS=
specifies a
confidence interval () for the Steiger and Lind (1980) root mean square error of approximation (RMSEA) coefficient. See the ALPHARMS= option of the PROC CALIS statement. - CHICORRECT | CHICORR = name | c
specifies a correction factor
for the chi-square statistics for model fit. See the CHICORRECT= option of the PROC CALIS statement. - CLOSEFIT=p
defines the criterion value p for indicating a close fit. See the CLOSEFIT= option of the PROC CALIS statement.
- DFREDUCE=i
reduces the degrees of freedom of the
test by i. See the DFREDUCE= option of the PROC CALIS statement. - NOADJDF
turns off the automatic adjustment of degrees of freedom when there are active constraints in the analysis. See the NOADJDF option of the PROC CALIS statement.
- NOINDEXTYPE
disables the display of index types in the fit summary table. See the NOINDEXTYPE option of the PROC CALIS statement.
- OFF | OFFLIST= [names] | {names}
turns off the printing of one or more fit indices or modeling information as indicated by names, where a name represents a fit index, a group of fit indices, or modeling information. Names must be specified inside a pair of parentheses and separated by spaces. By default, all fit indices are printed. See the ON= option for the value of names.
- ON | ONLIST < (ONLY) > = [names] | {names}
-
turns on the printing of one or more fit indices or modeling information as indicated by names, where a name represents a fit index, a group of fit indices, or modeling information. Names must be specified inside a pair of parentheses and separated by spaces. Because all fit indices and modeling information are printed by default, using an ON= list alone is redundant. When both ON= and OFF= lists are specified, the ON= list will override the OFF= list for those fit indices or modeling information that appear on both lists. If an ON(ONLY)= list is used, only those fit indices or modeling information specified in the list will be printed. Effectively, an ON(ONLY)= list is the same as the specification with an ON= list with the same selections and an OFF=ALL list in the FITINDEX statement.
Output Control of Fit Index Groups and Modeling Information Group
You can use the following names to refer to the groups of fit indices or modeling information available in PROC CALIS:- ABSOLUTE
Absolute or stand-alone fit indices that measures the model fit without using a baseline model.
- ALL
All fit indices available in PROC CALIS.
- INCREMENTAL
Incremental fit indices that measure model fit by comparing with a baseline model.
- MODELINFO
General modeling information including sample size, number of variables, number of variables, and so on.
- PARSIMONY
Output Control of Modeling Information
You can use the following names to refer to the individual modeling information available in PROC CALIS:- BASECHISQ
- BASEDF
Degrees of freedom of the chi-square statistic for the baseline model.
- BASEFUNC
Baseline model function value.
- BASELOGLIKE
Baseline model
log-likelihood function value for METHOD=FIML. - BASEPROBCHI
P-value of the chi-square statistic for the baseline model fit.
- BASESTATUS
Status of the baseline model fitting for METHOD=FIML.
- NACTCON
- NIOBS
Number of incomplete observations for METHOD=FIML.
- NMOMENTS
Number of elements in the moment matrices being modeled.
- NOBS
Number of observations assumed in the analysis.
- NPARM | NPARMS
Number of independent parameters.
- NVAR
Number of variables.
- SATFUNC
Saturated model function value for METHOD=FIML.
- SATLOGLIKE
Saturated model
log-likelihood function value for METHOD=FIML. - SATSTATUS
Status of the saturated model fitting for METHOD=FIML.
Output Control of Absolute Fit Indices
You can use the following names to refer to the individual absolute fit indices available in PROC CALIS:- CHISQ
Chi-square statistic for model fit.
- CN | CRITICAL_N
Hoelter’s critical N.
- CONTLIKE
Percentage contribution to the Log-likelihood function value of each group in multiple-group analyses with METHOD=FIML.
- CONTRIBUTION | CONTCHI
Percentage contribution to the chi-square value for multiple-group analyses.
- DF
Degrees of freedom for the chi-square test for model fit.
- ELLIPTIC
Elliptical chi-square statistic for ML and GLS methods in single-group analyses without mean structures. This index is computed only when you input the raw data with the KURTOSIS option specified.
- FUNCVAL
Optimized function value.
- GFI
Goodness-of-fit index by Jöreskog and Sörbom.
- LOGLIKE
Fitted model
log-likelihood function value for METHOD=FIML. - PROBCHI
P-value of the chi-square statistic for model fit.
- PROBELLIPTIC
- RMSR
Root mean square residual.
- SRMSR
Standardized root mean square residual.
- ZTEST
Z-test of Wilson and Hilferty.
Output Control of Parsimonious Fit Indices
You can use the following names to refer to the individual parsimonious fit indices available in PROC CALIS:- AGFI
Adjusted GFI.
- AIC
Akaike information criterion.
- CAIC
Bozdogan corrected AIC.
- CENTRALITY
McDonald centrality measure.
- ECVI
Expected cross-validation index.
- ECVI_LL | LL_ECVI
Lower confidence limit for ECVI.
- ECVI_UL | UL_ECVI
Upper confidence limit for ECVI .
- PGFI
Parsimonious GFI.
- PROBCLFIT
Probability of close fit.
- RMSEA
Root mean squares of error approximation.
- RMSEA_LL | LL_RMSEA
Lower confidence limit for RMSEA.
- RMSEA_UL | UL_RMSEA
Upper confidence limit for RMSEA.
- SBC
Schwarz Bayesian criterion.
Output Control of Incremental Fit Indices
You can use the following names to refer to the individual incremental fit indices available in PROC CALIS:- BENTLERCFI | CFI
Bentler comparative fit index.
- BENTLERNFI
Bentler-Bonett normed fit index.
- BENTLERNNFI
Bentler-Bonett nonnormed fit index.
- BOLLENNFI
Bollen normed fit index (Rho1).
- BOLLENNNFI
Bollen nonnormed fit index (Delta2).
- PNFI
James et al. parsimonious normed fit index.
- OUTFIT=SAS-data-set
creates an output data set containing the values of the fit indices. This is the same as the OUTFIT= option of the PROC CALIS statement. See the section OUTFIT= SAS-data-set for details.