Sem in Sas Jsm2010 - Yung

Introduction to Structural
Equation Modeling Using

the CALIS Procedure in
SAS/STAT Software
Yiu-Fai Yung
Senior Research Statistician
SAS Institute Inc.
Cary, NC 27513 USA
Computer technology workshop (CE_25T) presented at the JSM 2010 on
August 4, 2010, Vancouver, Canada. Email: Yiu-Fai.Yung@sas.com
SAS and all other SAS Institute Inc. product or service names are registered trademarks or trademarks of SAS
Institute Inc. in the USA and other countries. indicates USA registration.
Other brand and product names are registered trademarks or trademarks of their respective companies.
Copyright 2010 SAS Institute Inc. All rights reserved.
Abstract
TheCALISprocedureinSAS/STATisageneralstructuralequation modeling(SEM)tool.
ThisworkshopintroducesthegeneralmethodologyofSEMandtheapplicationsofthe
CALISprocedure.Historicaltopicssuchascasualmodels,pathdiagram,confirmatory
factoranalysis,measurementerrormodel,andlinearstructuralrelations(LISREL)are
reviewed.ApplicationsoftheCALISproceduretoSEMaredemonstratedwithexamples
insocial,educational,behavioral,andmarketingresearch.Specifically,thefollowing
howtotechniquesoftheCALISprocedure(SAS/STAT9.22)arecovered:(1)Specifying
structuralequationmodelswithlatentvariablesbyusingthePATHmodelinglanguage;
(2)Interpretingthemodelfitstatisticsandestimationresults;(3)Testingmodelswith
multiplegroupsandmultiplemodels;(4)Analyzingdirectandindirecteffects;(5)
Modifyingstructuralequationmodels.
Thisworkshopisdesignedforstatisticiansanddataanalystswhowanttooverviewthe
applicationsoftheSEMbytheCALISprocedure.Attendeesshould haveabasic
understandingofregressionanalysisandexperienceusingtheSASlanguage.Previous
exposuretoSEMisuseful,butnotrequired.
Citationofthisworkshop:
Yung,Y.F.(2010).IntroductiontoStructuralEquationModelingUsingtheCALIS
ProcedureinSAS/STAT Software.Computertechnologyworkshoppresentedatthe
JointStatisticalMeetingonAugust4,2010,Vancouver,Canada.
SAS/STAT 9.22 or later

is assumed
for this workshop
Copyright 2010, SAS Institute Inc. All rights reserved.
Inthisworkshop,SAS/STAT9.22(TS2M3)orlaterisassumedfortheCALISprocedure.
SomeofthecodemightworkwithPROCTCALIS(anexperimentalprocedure)in
SAS/STAT9.2(TS2M2).However,thereisamajorsyntacticaldifferencebetweenPROC
TCALISandPROCCALIS.InPROCTCALIS,theparameterspecificationforeachpathin
thePATHstatementmustnot beprecededbyanequalsign.Butthisequalsignis
requiredinPROCCALISwhenyouspecifyparameters.Also,PROCTCALISdoesnot
supportthegeneralizedpathspecifications(forvariances,covariances,means,and
intercepts)andmultiplepathspecificationswhenyouusethePATHmodelinglanguage,
whichisthemainfocusoftodaystalk.
Causal Model, Prediction, and Path Diagram

X causes Y
X predicts Y
Linear regression equation
Y = b X + ey
Path diagram
b
ey
This is the essential representation.

Often the error term is omitted.
3
Thecentralideaofstructuralequationmodelingisthestudyof causalrelationship
betweenvariables.Forexample,youhaveanXandanYvariable. XisthecauseofY,or
doingXresultsinY.Togiveamorerealisticexample:eatingmorevegetables(X)brings
downyourcholesterollevel(Y).However,thiscausalstructureisonlyanidealized
framework.Inmakingcausalinferences,youmusthaveisolatedallotherbackground
variablesandestablishtemporalsequenceofthevariables.Becauseofthecomplicated
philosophicalissuesinvolvedinmakingcausalinferences,ingeneralSEMwouldavoid
claimingcausalinferences.
Apredictoroutcomeframeworkmightbemoreappropriatephilosophically.The
semanticisnow XpredictsY.Mathematicallyandstatistically,thisideaisrepresented
inthesimplelinearregressionanalysis,asshowninthelinear regressionequation:
Y=b*X+e.
Thepathdiagramforthisrepresentationisshownintheslide,wherebiscalledthe
effect,regressioncoefficient,orpathcoefficient.Noticethatanerrortermisaddedto
showthatthepredictionofYfromXisnotperfect.Butessentially,thepredictor
outcomeframeworkisrepresentedbytheYXpathinthepathdiagram.
Structural Equation Modeling versus

Regression Analysis
Group 1
More variables
More equations
LV
Correlated errors
Direct and indirect effects

Latent variables
Parametric constraints
Group 2
Y
Multiple-group analysis
LV
Z
W
4
WhatarethedifferencesbetweenSEMandregressionanalysis?WhatmorecanSEM
offerthanthelinearregressionanalysis?
YoucanviewSEMasamuchmorecomplicatedsystemformultiplepredictoroutcome
relationships.SEMcanhandlethefollowingsituationswherelinearregressionanalysis
isoflimitedusefulness:
1.Morevariables(notjustXandY,butyoucanalsoaddWandZintothepathdiagram).
2.Moreequationsorfunctionalrelationships(notjustXY,butyoucanalsoanalyze
WZsimultaneously).
3.Correlatederrors,systemofequationscanhavecorrelatederrors.Forexample,the
doubleheadedarrowbetweenYandZ.
4.Directandindirecteffects:XhasadirecteffectonZandanindirecteffectonZviaits
effectonW.Thatis,XZandXWZaredirectandindirecteffects,respectively.
5.Latentvariables.Forexample,LVinthepathdiagramhaseffectsonXandW.
6.Parametricconstraints.Forexample,theconstraintsonthepath coefficientsor
effectslabeledasa intheupperpathdiagrams.
7.Multiplegroupanalysis.Fordifferentgroupsofpopulations,theoverall structureof
themodelarethesame,butthepathconstraintscouldbedifferentwhilethe
constrainedeffectinGroup1isdenotedasa, theconstrainedeffectinGroup2is
denotedasb, whichwillhaveadifferentestimatethanthatfora inGroup1.
Other Names for Structural Equation Modeling (SEM)

Path analysis
LISREL model (Jreskog 1973, Keesling 1972, Wiley
1973)
Covariance structures analysis
Analysis of moment structures
Confirmatory factor analysis
Causal modeling
CALIS: Covariance Analysis of Linear Structural
Equations
SEMhasalotofsynonymsinthefield:Pathanalysis(attributedtoSewallWright),
LISRELmodel(JKWmodel),covariancestructuresanalysis,analysisofmoment
structures,confirmatoryfactoranalysis,causalmodeling,andetc.Intermsofthe
statisticalmethodologyinvolved,allthesenamesaremoreorlessthesame.
PROCCALIS,whichstandsforcovarianceanalysisoflinearstructuralequations,isa
softwarethatwasdesignedtohandlealltheseanalysesundertheumbrellatermSEM.
Hopefully,onedayPROCCALISwouldalsoberememberedasasynonymofSEM.
SEM Software
AMOS, EQS, LISREL, MPLUS,
Why PROC CALIS?
Which is best?
ThereareseveralwellknownsoftwareinthefieldfordoingSEM:AMOS,EQS,LISREL,
andMPLUS,andmaybemore.WhyPROCCALIS?Whichisbest?Althoughtheseare
veryinterestingquestions,asthecurrentdeveloperofPROCCALISprocedureIamnot
atlibertytojudgeotherSEMsoftware.Thisworkshopgivesyouanintroductorytourof
SEMwiththeuseofPROCCALIS.Therefore,youmightcomparePROCCALISwithother
softwareonyourownafterlearningsomeofthefeaturesofPROC CALIS.
A Very Brief History of PROC CALIS

Original developer: Wolfgang Hartmann (80s)
Influences
o Statistical/mathematical: COSAN (McDonald 1978, 1980)
o Syntax: EQS (Bentler 1985, 1995)
TCALIS (SAS 9.2, 2008): experimental version

New CALIS (SAS 9.22, 2010): PATH modeling
language, multiple-group analysis, mean structures,
name-free approach to parameter specifications, and
much more
LetusstartwithabriefhistoryofPROCCALIS.
Ineighties,WolfgangHartmanndesignedanddevelopedthefirstversionofPROCCALIS.
Thestatisticalandmathematicalmodelwasgreatlyinfluencedby theCOSANmodel
proposedbyR.P.McDonald.Infact,therewasevidencethatCosan,insteadofCalis,
mighthavebeenproposedasthenameoftheprocedure.Themostpopularsyntaxin
PROCCALIS,however,wasundertheinfluenceoftheEQSprogrambyPeterBentler.
TheLINEQSsyntaxinPROCCALISformodelspecificationisbasicallyatwinbrotherof
thesyntaxoftheEQSprogram.
Ipickedupthedevelopmentofthesoftwarearound2000.Iactuallyrewrotethe
mathematicalfoundationsofthesoftware.Ikepttheoptimizationtechniquesandinitial
estimationtechniquessothatthenewCALISiscompatiblewiththeoldCALIS.
In2008,anexperimentalversioncalledTCALISwasreleased.Sincethen,Ihavemodified
thesyntaxalittlemoreandfixedsomemajorbugs.
ThenewCALIS(SAS9.22)hasbeenreleasedthisyear.IfyouhaveusedPROCCALIS
before,youwillnoticeonemajorchange:theemphasisonthePATHmodelinglanguage.
YoucanseeexamplesusingthePATHstatementeverywhereinthePROCCALIS
documentation.Othernoteworthynewfeaturesare:multiplegroupmodeling,
redesignedmeanstructureanalysis,andthenamefreeapproachtoparameter
specifications.Certainly,therearemanymorenewfeaturesthan these,asyouwilllearn
fromthisworkshopandelsewhere.
Structure of the Workshop

First Part: Basic Modeling
1.
A brief description of the process of SEM
2.
The PATH modeling language in PROC CALIS
3.
Specifying models and interpreting results
4.
LISMOD a language tailored to LISREL users
Second Part: Advanced Modeling

1.
Multiple-group analysis
2.
Analyzing direct and indirect effects
3.
Testing specific hypotheses
4.
Model modifications
ThefirstpartoftheworkshopisaboutthebasicSEMmodelingusingPROCCALIS.Iwill
describetheresearchprocessofSEMbriefly.ThenIwillintroducethePATHmodeling
languageinPROCCALISbyusingasimplelinearregressionexample.Next,Iwillmove
ontomorecomplicatedexamplesthatanalyzeconfirmatoryfactormodels.Iwilluse
PROCCALISintheseexamplestoshowhowyoucanspecifySEMmodelsbythePATH
modelinglanguage,inrelationtothepathdiagramrepresentations.Iwillshowyouhow
tointerprettheresultsgeneratedbyPROCCALIS.Iwillendthe firstpartbyshowingyou
howaLISRELmodelcanbespecifiedbytheLISMODstatementinPROCCALIS.
Thesecondpartoftheworkshopisaboutadvanced modelingrelativelyspeaking.I
willshowhowmultiplegroupanalysiscanbedoneinPROCCALIS.Otherimportant
topicssuchasdirectandindirecteffectanalysis,testingspecifichypotheses,andmodel
modificationsarediscussed.
Emphases of the Workshop
Introducing the structural equation methodology

and applications through examples What is SEM?
Analyzing structural equation models with PROC
CALIS How to do SEM?
Therearetwoemphasesofthistalk.
One,IwanttoshowyouanoverallpictureofSEM.ThisaddressesthewhatisSEM?
question.Iwillnotgiveyouatechnicaldefinition,butIwill showyouSEMexamplesso
thatyouwillhaveareal feelingabouttheapplicationsofSEM.
Two,IwanttoshowyouhowtousePROCCALIS.ThisaddressestheHowtodoSEM?
question.Ihopethatintheendoftheworkshop,youwillfindthatPROCCALISisvery
usefulforSEM.
Illustrating the Process of

Structural Equation Modeling
10
10
A Structural Equation Model of Web-Surfing

Behavior (Novak, Hoffman, & Yung 2000)
.42
Start Use
.19
.38
Skill
Positive
Affect
Control
.46
Time Use
-.32
.42
Interactive
Speed
.19
.31
Playfulness
.33
Challenge
.24
.51
.14
Focused
Attention
Future Use
Exploratory
Behavior
Arousal
.70
.51
.19
.45
.47
.25
Importance
.78
.67
.24
Time
Distortion
.13
Stimulation
Level
11
Thisisastructuralequationmodelaboutwebsurfingbehavior.Theresearchers
hypothesizethatthePlayfulness ofawebsitewouldenhancethefutureuse(Future
Use)ofthesamewebsite.However,thetheorydoesnotendthere.Theresearchers
thenhypothesizewhatwouldmakeawebsitetobeperceivedasplayful.Three
additionalconstructsarehypothesizedinthepathdiagram:Control (ofthewebpage),
Arousal (ofinterest),andFocusedAttention aredeterminantsofPlayfulness. In
fact,theresearchershypothesizedevenfurther.Forexample,theyuseStartUse
(whentheusersstartedtousecomputers)andTimeUse (howoftentheyuse
computers)asremotecauses ofalotoflatentconstructsinthepathdiagram.Insum,
thisisarelativelylargeSEMthattheorizescomplicatedrelationshipsamongconstructs.
Inthispathdiagram,theovalshapesrepresentlatentvariables,whicharenotmeasured
butserveasusefulconstructsinthemodel(e.g.,Playfulness).Therectanglesrepresent
measuredorobservedvariables(e.g.,StartUse,TimeUse,FutureUse).Inorderto
analyzethelatentconstructs,somemeasuredvariables(orindicators)areneeded.Inthe
pathdiagram,thosesmallunlabeledrectanglesaremeasuredindicatorsforthelatent
constructs.Inthisresearch,thesemeasuredindicatorsareratingresponsesona
questionnaire.Forexample,Ilosttrackofthetimewhenusingthiswebsite (thisis
notanexactitemfromtheactualresearch)couldbeanitemfor theTimeDistortion
construct.
Giventhispathdiagramforthetheoryaboutwebsurfingbehavior,anSEMsoftwarefits
themodelbasedontheobserveddataandinformsyouthemodelfitandtheestimates
oftheeffects(pathcoefficients)inthepathdiagram.TheSEMsoftwarealsotellsyouthe
significanceoftheseestimates.Ifthemodeldoesnotfitthedatawell,theSEMsoftware
wouldsuggestwaystoimprovethemodel.
11
Key Features of SEM

Analyzing complicated relationships among variables
Path diagram representations for models
Ability to handle latent and observed variables
simultaneously
Testing the model fit and significance of the parameters
Suggesting ways to improve the model
12
HereisalistofthekeyfeaturesofSEM:
Analyzingcomplicatedrelationshipsamongvariables
Pathdiagramrepresentationsformodels
Abilitytohandlelatentandobservedvariablessimultaneously
Testingthemodelfitandsignificanceoftheparameters
Suggestingwaystoimprovethemodel
12
Basics: A Simple Regression

Model and the PATH
Modeling Language
13
13
A Simple Linear Regression Model
y = b x + ey
y: outcome variable
x: predictor variable
ey: error term
b: effect or regression coefficient
Assumption: Variables are centered.
14
TointroducethePATHmodelinglanguageinPROCCALIS,asimple linearregression
modelisused.Intheregressionequation,yistheoutcomevariable,xisthepredictor
variable,e_y istheerrortermandbistheeffectorregressioncoefficient.The
regressionmodelwritteninthisformassumesthatxandyarecenteredwithmeans
zero.Butthisassumptionwillnotaffectthegeneralitytouncenteredvariables.
14
Measures of the Number of Hen Pheasants

Fuller (1987) p.34
y : average of the number of birds in August
x : average of the number of birds in Spring (April/May)
Averages were based on the number of birds sighted by
15 trained observers
Goal: How many birds will survive 3 months?
15
Onp.34ofFullersbookMeasurementErrorModels,hedescribesadatasetabout
thecountingofhenpheasantsinAprilandAugust.Fifteentrainedobserverscounted
thenumberofbirdsinthetwooccasions.YisthenumberofbirdsinAugustandXisthe
numberofbirdsinApril.Thegoalofthelinearregressionistopredictthenumberof
birdsinAugust(Fall)bythenumberofbirdsinApril(Spring).
15
Regression Analysis by PROC REG
data hens;
input y x @@;
datalines;
8
6.6
9.8
12.3
10.8
11.9
9.7
11.9
9.3
12
9.2
9.6
6.9
7.5
8.1
10.9
8.7
10.4
8.7
10.2
7.4
7.4
10.1
11
10
11.8
7.3
8.2
;
proc reg data=hens;
model y = x;
run;
16
Toconductalinearregressionanalysis,youcanuseaSASprocedurecalledPROCREG.
Thesyntaxisquitesimple.First,defineyourdataset.Second, callPROCREGwiththe
interesteddatasetspecifiedinthePROCREGstatement.Then,themodelstatement
specifiesthaty=x,whichmeansyispredictedbyx.Noerrortermneedstobespecified,
althoughPROCREGdoesassumethatpredictionisnotperfectsothattheerrordoes
existwithnonzerovarianceintheregression.
16
Results Obtained from PROC REG
Parameter Estimates
Parameter
Standard
DF
Estimate
Error
t Value
Intercept
2.14227
0.84513
2.53
0.64941
0.08275
7.85
Variable
Given a base survival of 2.14 birds, every additional bird in Spring predicts a
0.65 bird surviving in August.
17
ThistableshowstheessentialresultsfromPROCREG.Theoutput showsanestimateof
0.65fortheregressioncoefficientb.Theinterceptestimateis2.14.PROCREGalso
showsthestandarderrorestimatesandthetvaluesforjudgingstatisticalsignificance.
Bothestimatesarestatisticallysignificant.
Aninterpretationabouttheseregressionestimatesisthis:Givenabasesurvivalof2.14
birds,everyadditionalbirdinSpringpredictsa0.65birdsurvivinginAugust(Fall).
17
Regression Equation, Path Diagram, and the PATH

Modeling Language
Outcome
Path
Statement
Parameter
<---
ey
Predictor
PATH
b x
Path Relation
b;
Parameter (optional)
18
Asshownpreviously,youcanrepresentthelinearregressionmodelbythepathdiagram,
whichisalsoarepresentationschemeforSEM.Hence,regression modelscouldbe
specifiedasSEMbyusingthepathdiagram.
HereiswhatyoudotospecifyasimplelinearregressionmodelinPROCCALIS.Youuse
thePATHstatementtospecifythepathintheregressionmodel.Inthiscase,itisjustY<
XinthePATHstatement.Optionally,youcandenotethecorrespondingpath
coefficientparameter.Forexample,youcanput=b atthebackofthepathtodenote
theparameterlabelorname.
18
Regression Model Specified by PROC CALIS
proc calis data=hens;

path
y <--- x;
run;
19
ThisistheentirePROCCALISsyntaxforthesimplelinearregressionmodel.Isntthat
easyandsimple?
19
Results from PROC CALIS for the Pheasant Data

PATH List
Standard
--------Path-------y
<---
Parameter
_Parm1
Estimate
Error
t Value
0.64941
0.07974
8.14445
The same estimate

of b by PROC REG
Variance Parameters
Default variance
parameters
Variance
Type
Variable
Parameter
Exogenous
Error
Standard
Estimate
Error
t Value
_Add1
3.62124
1.36870
2.64575
_Add2
0.32233
0.12183
2.64575
20
ThisslideshowstheresultsfromPROCCALIS.
Theestimatedeffectofxony,denotedasy< xintheoutput,is0.65,whichisthe
sameasthatinthePROCREGresults.Becauseyoudidnotnamethisregression
coefficientparameter(butyouspecifythepathnonetheless),PROCCALISgeneratesa
uniqueparameternamecalled_Parm1forit.Thestandarderrorestimateandthet
valuearealittlebitdifferentfromthatofthePROCREGresults.Thisisbecause
differentdegreesoffreedomforcomputingthestandarderrorsareusedinthetwo
approaches.
InPROCCALIS,italsoincludesresultsfortwomoreparametersintheSEM.Thevariance
ofxandtheerrorvarianceofyaretreatedasmodelparameters.Theirestimatesare
alsoshowninthePROCCALISresults.NotethatPROCCALIScreatesdefaultparameter
namesforthesedefaultvarianceseventhoughyoudidnotspecifythem.Inthis
example,thesevarianceparametersarenamed_Add1 and_Add2,respectively.
20
Specification with Parameter Names
Without an explicit error term

path
errv_y
y <--- x
= b;
b
y
var_x
pvar
x
= var_x,
= errv_y;
Equivalent representations
run;
b
y
Use the PVAR statement to specify

variance or error variance parameters.
You can also define parameters
explicitly in PROC CALIS.
errv_y
var_x
ey
With an explicit error
term
21
YoucouldnamealltheparametersinPROCCALISbyputtingyourpreferrednames.
Inthepathdiagramatthetoprightcorner,theparametersareshowninred.Inthe
regressionmodel,bistheregressioncoefficient,var_x isthevarianceforthepredictor
variablex,anderrv_y istheerrorvarianceofy.Thispathdiagramrepresentationis
equivalenttotheoneshownatthebottomrightcorner,wherean expliciterrortermis
attachedtoY.Theerrortermisrepresentedbyanovalshapebecauseitistreatedasa
latentvariable.Thisrepresentationhasthesamesetofparameters,onlythaterrv_y is
nowattachedtotheerrorvariabledirectly.
YoucanspecifytheseparametersexplicitlyinPROCCALIS.Intheleftpaneloftheslide,
theparameterbisspecifiedafterthey< xpath,separatedbyanequalsign.To
specifythevariancesorerrorvariancesinthemodel,youcanusethePVARstatement.
Forexample,x=var_x meansthatthevarianceofxisaparametercalledvar_x.
Noticethatnamingparametersisentirelyoptional.Forthisexample,naming
parametersappearstoserveonlyasanillustration.Laterinthistalk,youwillfind
situationswheretheuseofparameternamesisnotonlyuseful,butalsonecessary.
21
Hen Pheasants Results with Parameter

Names Specified
PATH List
Standard
--------Path--------
Parameter
<---
Estimate
Error
t Value
0.64941
0.07974
8.14445
Parameter name specified
Variance Parameters
Variance
Standard
Type
Variable
Parameter
Estimate
Error
t Value
Exogenous
Error
var_x
3.62124
1.36870
2.64575
errv_y
0.32233
0.12183
2.64575
Parameter names specified

22
Asshowninthisslide,thenumericalresultsfromPROCCALISwithexplicitparameter
namesspecifiedarethesameasthosewithoutusingparameternames.Theonly
differenceisthatnowyoucanusetheseparameternamestolocatethecorresponding
resultsdirectly.
22
Keys to the PATH Modeling Language

As easy as drawing a path diagram
PATH statement specifies the functional relationships
required specification
PROC CALIS sets variances and error variances by
default optional specification (most of the time)
Naming free parameters is optional
23
Sofar,Ihaveshownyouthat:
1.ThePATHmodelinglanguageisaseasyasdrawingapathdiagram.
2.YoucanusethePATHstatementtospecifythepathsinpathdiagram,withorwithout
specifytheparameternamesforthepathcoefficients.
3.Youcanalsospecifythevarianceorerrorvarianceparametersexplicitly.Inmost
practicalapplications,variancesanderrorvarianceshavealreadybeensetbydefault
andyoudonotneedtoworryaboutspecifyingthem.TheessentialpartofSEMis
specifiedinthePATHstatement.
4.NamingparametersisoptionalinPROCCALIS.
23
Measurement Errors in Predictors
Bird counting might involve measurement errors in x

x = fx + ex
fx : true score, but not observed
x : observed, but with measurement error ex
24
LetusmakealittlestepforwardtoshowaspecialSEMfeaturethatlinearregression
cannothandleeasily.
Inthebirdcountingexample,wedidnottakeintoaccountthatbirdcountingcould
involvemeasurementerrors.Inthecurrentcontext,themeasurementerrorinbird
countingcouldbeduetotheenvironmentfactorsintheforest:obstructionfromthe
treebranches,biased anglesfromthebirdobservers,andetc.
Mathematically,youcanhypothesizeavariablecalledfx torepresentthetrue counts
obtainedfromthebirdobserved.Theobservednumberofbirdsxisthesumoffx,the
truescore,andex,anerrorterm.
Whatyougotfromthedataisx,theobservedfalliblescore.However,ideally,you
wouldwanttousefx,thetruescoreinyourregressionanalysis.
24
A Measurement Error Model for the Pheasant Data

Structural Equation
y = b fx + ey
Measurement equation
x = fx + ex
Can you estimate b?
Problem: The measurement equation introduces an
additional parameter: Var(ex) (variance of ex or error
variance of x)
25
TheprecedingideaisformalizedasthefollowingSEMwithalatentvariablefx.
Inthesocalledstructuralmodel,yispredictedfromfx,thetruescore,inthelinear
regressionmodel.Thissocalledstructuralequationtakestheroleoftheoriginallinear
regressionequationonlynowyouaresupposedtohaveabettermodelbyusingthe
measurementerrorfreefx asthepredictor.
Inthesocalledmeasurementmodel,youhypothesizethattheobservedvariablexis
obtainedasthesumoffx andanmeasurementerrorterme_x.
Canyouestimatebwiththelatentvariablefx inthestructuralequation?
Thisanswerisyes.
Butthetechnicalproblemencounteredhereisthatthemeasurementequation
introducesoneadditionalparametersinvar(ex)errorvarianceofx.Thisproblemwill
maketheSEMunidentified.Inaveryloosesense,thismeansthatyourmodelestimates
moreparametersthanwouldbeallowedbythegiveninformationofthedataset.
Consequently,theparametersinthemodelarenotestimable.Iwilldescribeamethod
todealwiththisidentificationproblemlater.
25
Path Diagram Representations
Linear Regression Model
Measurement Error Model
fx
errv_y
var_x
errv_y
var_x
errv_x
One more
parameter
26
Thisslidecomparesthelinearregressionmodelwiththemeasurementerrormodelby
theuseofpathdiagram.Itdemonstrateswhythemeasurementerrormodelhasone
moreparametertoestimate.
Intheleftpanel,thepathdiagramforthesimplelinearregressionanalysisisshown.
Intherightpanelforthemeasurementerrormodel,westillhavexandyasthe
observedvariables.Butnowwehavealatentvariablefx thattakestheroleofthe
predictorofy.Var_x inthismodelnowrepresentsthetruevarianceofthepredictor fx.
Thenewparameterinthemeasurementerrormodeliserrv_x (errorvarianceofx).With
thisadditionalparameter,weneedtomakeadditionalassumption toestimatethe
modelparameters.
26
Constraining the Error Variances
Bird counting is more accurate in fall (y) than in spring (x)

In an independent study, error variance (for x) in spring is
six times as much as that (for y) in fall
Fullers recommendation: Var(ex) = 6 Var(ey)
errv_x = 6* errv_y
27
Fortunately,wehaveareasonableassumptionabouttherelativesizeoftheerror
variancesinthemodel.
ThisassumptionisbasedonthefactthatbirdcountinginFallismoreaccuratethanthat
inspring.ThereasonisthatthefallenleavesinFallmakesthecountingofbirdsless
obstructive.
Theassumptionwearegoingtomakeisbasedonanindependentstudyaboutthe
relativeerrorvariancesinxandiny.InFullersbook,theratioofthesevariancesis
about6.Mathematically,Var(ex)=6*Var(ey).Thatis,errorvarianceforxissixtimesas
muchastheerrorvarianceofy.Or,inthePROCCALISspecification,youwanttostate
thefollowingparametricconstraintinthemodeling:errv_x=6*errv_y.
27
A Measurement Error Model with a Constraint

for the Pheasant Data

path
y
<--- fx
fx ---> x
= b,
= 1;
pvar
y
fx
x
b
y
fx
errv_y
var_x
= errv_y,
= var_x,
= errv_x;
errv_x = 6 * errv_y;
errv_x
errv_x = 6* errv_y
run;
The required constraint is specified
as a SAS programming statement.
28
ItturnsoutthatitisprettystraightforwardtospecifythisparametricconstraintinPROC
CALIS.Youjustsimplyaddonemorelineofcodetorepresentthisrelationship,as
shownintheSAScodeoftheslide.IntheSASliterature,thislineofcodeiscalledaSAS
programmingstatement,whichisusedextensivelyintheDATAstepofSAS.Youcanuse
asmanySASprogrammingstatementsasyouwanttodescribetherelationshipsofthe
parametersinthemodel.
28
Measurement Error Model Results for the Pheasant Data

PATH List
Standard
--------Path--------
Parameter
<---
fx
fx
--->
Estimate
Error
t Value
0.75158
0.09228
8.14427
1.00000
A larger estimated effect than the one
estimated without taking the
measurement error into account (0.649)
Variance Parameters
Variance
Standard
Type
Variable
Parameter
Estimate
Error
t Value
Error
errv_y
0.08205
0.03101
2.64575
Exogenous
Error
fx
var_x
3.12893
1.36180
2.29765
errv_x
0.49231
0.18608
2.64575
Six times as much as the estimate for errv_y
29
Afteryoutakethemeasurementerrorintoaccount,theregressioncoefficientbisnow
0.75,whichisalargereffectthan0.649,whichyouobtainedfromthelinearregression
modelwithouttakingthemeasurementerrorinxintoaccount.Therefore,theprevious
regressionanalysisunderestimatedthiseffectbecauseitfailed toincorporatethe
measurementerrorintothemodel.However,withSEM,youcaneasilyincorporatethe
measurementerrorsintotheanalysis.
Estimatesofthevariancesanderrorvariancesareshowninthenexttable.Youcansee
thattheconstraintspecifiedinthePROCCALISishonoredintheestimation.Theerror
varianceestimateofxis0.49,whichissixtimesasmuchastheerrorvarianceestimate
ofy,whichis0.08.
29
Some Features of the PATH Modeling Language

Specifying paths in the PATH statement is straightforward
Deals with latent variables easily variables are latent if
they are not present in the data set
PVAR statement for specifying variances and error
variances
PCOV statement for specifying covariances and error
covariances (to be shown)
Parameter dependency can be specified by the SAS
programming statements. For example,
parm1 = 4 * parm2 + exp(parm4) ** parm6;
30
ThisslidesummarizessomefeaturesofthePATHmodelinglanguage.
1.Itisasstraightforwardasdrawingthepaths.
2.Itcandealwithlatentvariableseasily.
3.YoucanusethePVARstatementtospecifyvariancesorerrorvariances(double
headedarrowsattachedtoindividualvariablesinthepathdiagram).
4.YoucanusethePCOVstatementtospecifycovariances orerrorcovariances (double
headedarrowsattachedtopairsofvariablesinthepathdiagram).
5.YoucanspecifyparameterdependencybyusingtheSASprogrammingstatements
directly.Indeed,evenverystrangeandcomplicated(continuous) parametricfunctions
aresupportedinPROCCALIS.
30
A Confirmatory Factor Model
31
Wenowmoveontoamorecomplicatedtypeofstructuralequation modelscalled
confirmatoryfactormodels.
31
Political Democracy Data

Bollen (1989) Chapter 7
Two latent factors: political democracy in 75 developing
countries in 1960 and 1965
Four indicator measures for the latent factors in each year:
o
Freedom of press (Press60, Press65)
Freedom of group oppositions (Freop60, Freop65)
Fairness of elections (Fair60, Fair65)
Elective nature of the legislative body (Legis60, Legis65)
Purpose of the confirmatory factor analysis: Validate the

measurement indicators
32
ThisexampleisbasedonanexampleinChapter7ofBollens classictextbook:Structural
EquationModeling.
Inthisexample,twolatentfactorsformeasuringpoliticaldemocracyin75developing
countriesin1960and1965werehypothesized.
Thesetwolatentfactorsarenotobserved,buttheyhavesomerelatedobserved
variablesthatserveasindicators.Ineachyear,youmeasurefourvariablestogaugethe
politicaldemocracy:freedomofpress,freedomofgroupoppositions,fairnessof
elections,andelectivenatureofthelegislativebody.
Thepurposeoftheconfirmatoryfactoranalysisistovalidatethesemeasurement
indicatorsstatistically.
32
A Confirmatory Factor Model for the Political

Democracy Data (Implicit Error Representation)
1.
Press60
Freop60
Dem60
Fair60
Legis60
Press65
Freop65
Dem65
Fair65
1.
Legis65
33
Thispathdiagramshowsthehypothesizedconfirmatoryfactormodel.
Inthepathdiagram,twolatentfactorsarerepresentedbytwoovals.Dem60isthe
politicaldemocracyin1960andDem65isthepoliticaldemocracy in1965.Theyare
linkedtotherespectivemeasuredvariables,asshowninthepathdiagram.Thesesingle
headedpathsrepresentthetypicalfactorobservedvariablerelationships.
ThedoubleheadedarrowthatconnectsDem60andDem65representsthecovariance
parameterbetweenthetwofactors.Itmeansthatthetwofactorsarecorrelated.
DoubleheadedarrowsthatareattachedtoDem60andDem65individuallyrepresent
thevarianceparametersofthetwofactors.Inthemodel,youfixthesevariancesto1so
thatthescalesofthefactorsareidentified.Thisisconventionallydonebecausethe
scaleoflatentfactorsisarbitrary(youdonotmeasurelatentvariablesdirectlysothat
theycouldbedefinedonanyunitofmeasurement).
Thedoubleheadedarrowsthatareattachedtotheobservedvariablesrepresentthe
errorvariances.Theysignifythefactthatthefactorsinthemodeldonotaccountfor
100%ofthevariancesoftheobservedvariables.Theerrorvariancesaretheunique
partofthevariancesinthevariablesthatarenotduetotheir relationshipswiththe
factorsinthemodel.
33
A Confirmatory Factor Model for the Political

Democracy Data (with Error Terms)
1.
Press60
Freop60
Dem60
Fair60
Legis60
Press65
Freop65
Dem65
Fair65
1.
Legis65
1.
1.
1.
1.
1.
1.
1.
1.
e1
e2
e3
e4
e5
e6
e7
e8
34
Thisisanalternativepathdiagramrepresentationwiththeuseofexpliciterrorterms.
Noticethatthedoubleheadedarrowsfortheobservedvariablesnowshifttotheerror
terms.Thispathdiagramrepresentationisshownhereonlyforillustrationpurposes.In
thisworkshop,Irelyonthepathdiagramrepresentationthatdoesnotuseexpliciterror
terms.
34
Basic Confirmatory Factor Model for the Political

Democracy Data
proc calis data=polidem;
path
Dem60 ---> Press60,
Dem60 ---> Freop60,
Dem60 ---> Fair60,
Dem60 ---> Legis60,
Dem65 ---> Press65,
Dem65 ---> Freop65,
Dem65 ---> Fair65,
Dem65 ---> Legis65;
pvar
Dem60 = 1, Dem65 = 1,
Press60 Freop60 Fair60
Legis60 Press65 Freop60
Fair65 Legis65;
pcov
Dem60 Dem65;
run;
1.
Press60
Freop60
Dem60
Fair60
Legis60
Press65
Freop65
Dem65
Fair65
1.
Legis65
35
SpecifyingtheCFAmodelisnotmuchharderthanthepreviousmeasurementerror
model.Basically,youonlyneedtospecifymorepathsfortheCFAmodel.
InthePATHstatement,youspecifyallthesingleheadedpaths(arrows)inthepath
diagram.
InthePVARstatement,youspecifyalldoubleheadedarrowsthatareattachedto
individualvariables.PVARactuallystandsforpartialvarianceyoucanspecifythe
variancesanderrorvariancesinthisstatement.Dem60=1 meansthatthevarianceof
Dem60isfixedtoone.SimilarlyforDem65=1.Theeightobservedvariableare
specifiedinthePVARstatementtosignifythattheirerrorvariancesarefreeparameters
inthemodel.
InthePCOVstatement,youspecifypairsofvariablesthathavecovariances orerror
covariances asparametersinthemodel.Inthecurrentpathdiagram,Dem60and
Dem65arecorrelated.
35
Error Variances, Variances, and Exogenous

Covariances Are Free Parameters by Default
Specifying All Variance and
Covariance Parameters
Using Default Variance and Covariance

Parameters

path
Dem60 ---> Press60,
Dem60 ---> Freop60,
Dem60 ---> Fair60,
Dem60 ---> Legis60,
Dem65 ---> Press65,
Dem65 ---> Freop65,
Dem65 ---> Fair65,
Dem65 ---> Legis65;
pvar
Dem60 = 1, Dem65 = 1,
Press60 Freop60 Fair60
Legis60 Press65 Freop65
Fair65 Legis65;
pcov
Dem60 Dem65;
run;

path
Dem60 ---> Press60,
Dem60 ---> Freop60,
Dem60 ---> Fair60,
Dem60 ---> Legis60,
Dem65 ---> Press65,
Dem65 ---> Freop65,
Dem65 ---> Fair65,
Dem65 ---> Legis65;
pvar
Dem60 = 1, Dem65 = 1;
run;
These could have been set

automatically by default.
36
Tomakemodelspecificationmoreefficientanderrorfree,PROCCALISemploysdefault
freeparametersinthemodel.Thesedefaultfreeparametersaresetbecausetheyare
commonlyemployedinpractice.
Forexample,becausepredictionsofoutcomevariablesareusuallynotperfect,theerror
variancesarefreeparametersbydefault.ThismeansthatallthePVARspecificationsfor
theobservedvariablesarenotnecessarybecausePROCCALISwouldhavetreatedthem
asfreeparametersbydefault.
Similarly,thevariancesofDem60andDem65andtheircovariance aredefaultfree
parametersbecausetheyareassumedinmostpracticalapplications.Inthecurrent
example,thismeansthatthePCOVstatementspecificationforthecovariancebetween
Dem60andDem65isnotnecessary.
However,becausethevariancesofDem60andDem65arefixedto1 (foridentification
ofthelatentvariablescales),theymustbespecifiedexplicitlyinthePVARstatement.
Otherwise,thesevarianceswouldhavebeenfreeparametersbydefault.
36
Estimates of Path Coefficients (Loadings) for the

PATH List
Standard
----------Path----------
Parameter
Estimate
Error
t Value
Dem60
--->
Press60
Dem60
--->
Freop60
_Parm1
2.20567
0.25122
8.77998
_Parm2
3.00132
0.39735
7.55335
Dem60
--->
Dem60
--->
Fair60
_Parm3
2.31033
0.34026
6.78989
Legis60
_Parm4
2.89483
0.31582
Dem65
9.16619
--->
Press65
_Parm5
2.04790
0.25930
7.89771
Dem65
--->
Freop65
_Parm6
2.68003
0.33258
8.05834
Dem65
--->
Fair65
_Parm7
2.70879
0.31804
8.51711
Dem65
--->
Legis65
_Parm8
2.76604
0.30830
8.97190
All path estimates are significant (t > 1.96).

37
ThistableshowstheestimatesofpathcoefficientsfromPROCCALIS.
Inthefactoranalysisliterature,thesepathcoefficientsarealsocalledloadings.To
validatetherelationshipsbetweenthedemocracyfactorsandthe observedvariables,
thetvaluesmustbeexaminedforstatisticalsignificance.Usingnormalapproximation,
tvalueswiththeirabsolutevaluesbiggerthan1.96aresignificantlydifferentfromzero.
Inatypicalfactoranalysisstudy,youwouldwantallthesetvaluestobesignificantin
ordertoclaimnonzerofactorvariablerelationships.Aninsignificanttvaluemeansthat
thecorrespondingvariableisnotanindicatorforthepurported factor.Insignificantt
valuesforpathcoefficientswouldchallengethevalidityofyourfactormodel.
37
Estimates of Variances for the Political

Democracy Data
Variance Parameters
Variance
Standard
Type
Variable
Exogenous
Dem60
1.00000
Dem65
1.00000
Error
Parameter
Estimate
Error
t Value
Press60
_Add1
2.01359
0.41048
4.90549
Freop60
_Add2
6.57189
1.20964
5.43294
Fair60
_Add3
5.42661
0.96546
5.62076
Legis60
_Add4
2.83887
0.61417
4.62229
Press65
_Add5
2.63180
0.49311
5.33709
Freop65
_Add6
4.19276
0.79422
5.27906
Fair65
_Add7
3.46180
0.68155
5.07928
Legis65
_Add8
2.88292
0.59927
4.81068
All error variance estimates are significant (t > 1.96).
38
Estimatesofvariancesanderrorvariancesareshowninthistable.Thevariancesof
Dem60andDem65arefixedto1andthereforetherearenosignificancetestsforthese
variances.Allothererrorvarianceestimatesaresignificantlylargerthanzeros.Thisalso
meansthatthefactorsdonotaccountforallthevariancesoftheobservedvariables.
Thisisnaturalbecausedeterministicrelationshipsbetweenfactorsandobserved
variablesarerare.
38
Estimate of Covariance for the Political

Democracy Data
Covariances Among Exogenous Variables
Standard
Var1
Var2
Parameter
Dem65
Dem60
_Add9
Estimate
Error
t Value
0.97528
0.02656
36.72321
High and significant correlation between the Democracy factors in 1960 and 1965.
39
ThistableshowsthecovariancebetweenDem60andDem65.Thisestimateisalsothe
estimatedcorrelationbetweenthetwolatentfactorsbecausethe variancesofthe
factorsarefixedtoone.Thiscorrelationisextremelyhigh,possiblybecausethepolitical
democracystatusdonotchangemuchduringthose5years.
39
How Is the Model Fit?

Fit Summary
Modeling Info
Absolute Index
Parsimony Index
Incremental Index
N Observations
N Variables
N Moments
N Parameters
N Active Constraints
Baseline Model Function Value
Baseline Model Chi-Square
Baseline Model Chi-Square DF
Pr > Baseline Model Chi-Square
Fit Function
Chi-Square
Chi-Square DF
Pr > Chi-Square
Z-Test of Wilson & Hilferty
Hoelter Critical N
Root Mean Square Residual (RMSR)
Standardized RMSR (SRMSR)
Goodness of Fit Index (GFI)
Adjusted GFI (AGFI)
Parsimonious GFI
RMSEA Estimate
RMSEA Lower 90% Confidence Limit
RMSEA Upper 90% Confidence Limit
Probability of Close Fit
ECVI Estimate
ECVI Lower 90% Confidence Limit
ECVI Upper 90% Confidence Limit
Akaike Information Criterion
Bozdogan CAIC
Schwarz Bayesian Criterion
McDonald Centrality
Bentler Comparative Fit Index
Bentler-Bonett NFI
Bentler-Bonett Non-normed Index
Bollen Normed Index Rho1
Bollen Non-normed Index Delta2
James et al. Parsimonious NFI
75
8
36
17
0
6.1482
454.9633
28
<.0001
0.6009
44.4686
19
0.0008
3.1383
51
0.5388
0.0494
0.8658
0.7457
0.5875
0.1346
0.0833
0.1865
0.0062
1.1240
0.9065
1.4608
78.4686
134.8659
117.8659
0.8438
0.9403
0.9023
0.9121
0.8560
0.9416
0.6122
A lot of fit indices, but researchers

usually report just a few of them.
40
Wehavelookedattheestimatesandconcludedthattherelationshipsbetweenthe
factorsandthevariablesarestrongandsignificant.Thoseresultsvalidatedthe
individualfactorvariablerelationships.
Togainsupportfortheoverallconfirmatoryfactormodel,youwouldwanttoexamine
themodelfitstatistics.ThistableshowsvariousfitindicescomputedbyPROCCALIS.In
theSEMfield,alargenumberoffitindiceshavebeenproposed. Thereisnoconsensus
astowhichindicesarebesttoreportintheresearch.Butresearcherstendtoreport
someofthemostpopularonesintheirrespectivefields.
Becausealargenumberofindicesmightbeconfusing,PROCCALIS providesawayto
customizethisfitsummarytable.
40
Using the FITINDEX Statement to Customize the

Fit Summary Output
path
Dem60 ---> Press60,
Dem60 ---> Freop60,
Dem60 ---> Fair60,
Dem60 ---> Legis60,
Dem65 ---> Press65,
Dem65 ---> Freop65,
Dem65 ---> Fair65,
ON(ONLY)= selects the set of fit
indices to display.
Dem65 ---> Legis65;
NOINDEXTYPE suppresses the
pvar
printing of index types.
Dem60 = 1, Dem65 = 1;
fitindex on(only) = [chisq df probchi rmsea cn srmsr
bentlercfi agfi] noindextype;
run;
41
YoucanusetheFITINDEXstatementtocustomizeyourfitsummary table.
UsetheON(ONLY)=optiontoselectyourfavorite fitindices.
UsetheNOINDEXTYPEoptiontosuppresstheprintingofthefitindextypes.
41
Customized Fit Summary Table
Fit Summary
Chi-Square
Chi-Square DF
Pr > Chi-Square
Hoelter Critical N
Adjusted GFI (AGFI)
RMSEA Estimate
44.4686
19
0.0008
51
0.0494
0.7457
0.1346
0.9403
Good SRMSR and Bentlers CFI. Bad chi-square,

AGFI, RMSEA.
42
ThisisthecustomizedfitsummaryoutputbyusingthepreviousFITINDEXstatement
specifications.
Thistablecontainsthemorepopularfitindicesreportedinresearch(asrecognizedby
theauthor).
Inpractice,themodelfitchisquaremodelstatistic,itsdf,andthecorrespondingp
valueareroutinelyreportedeventhoughveryfewresearchersin thefieldwoulduse
themodelfitchisquarealonetojudgemodelfit.Asshowninthistable,thepvalueis
verysmallsothatstatisticallyitmeansthatthehypothesizedmodelshouldberejected.
However,itisaknownissueinSEMthatevenveryusefulSEMmodelswithminimum
departuresfromthedatawouldberejectedstatistically.Therefore,researchersinthe
SEMfieldtendtofocusmoreonotherfitindicestojudgemodel fit.
TheSRMSR,AGFI,RMSEA,andCFIarefourofthemostpopularfit indicesintheSEM
field.Seetheglossarypageforthedescriptionsofthesefitindices.FortheSRMSRand
RMSEA,thesmallerthevaluesthebetterthefit.Usually,valuesunder0.05indicate
goodmodelfit.Therefore,theSRMSRsaysthatthecurrentmodel isgood,butthe
RMSEAsaysthatthecurrentmodelisbad.FortheAGFIandBentlers CFI,thelargerthe
valuesthebetterthemodelfit.Therefore,theAGFIsaysthatthecurrentmodelisbad,
buttheCFIsaysthatitisgood.Becausetheseindicesdonotconsistentlyindicatea
goodmodelfit,itissafetosaythatthecurrentCFAmodelispromising,butitneeds
furtherconfirmation.
42

with Loading Constraints
43
43
Constraining the Path Coefficients (Loadings)

1.
Press60
lam1
Dem60
Freop60
lam2
lam3
Fair60
lam4
Legis60
Press65
Dem65
1.
lam1
lam2
Freop65
lam3
Fair65
lam4
Legis65
Note: Error variances

are not represented
because they are
default parameters.
44
Inadditiontofittingabasicconfirmatoryfactormodel,PROCCALISenablesyoutoset
upparameterconstraintseasily.Themaintoolistouseparameternamesinthe
specification.
Forthepoliticaldemocracyexample,theresearcherwantstoconstraintthefactor
loadings(pathcoefficients)acrosstime.Thetheoreticalreason isthatbasicallythe
measuredvariablesarethesameinthetwoyears.Inthepathdiagram,youcan
representequalityconstraintsbyputtingthesameparameternamesorlabelstothe
pairsoftherelatedpaths.Forexample,lam1istheloadingofPress60onDem60.Itis
alsotheloadingofPress65onDem65.Similarly,youcansetthe other3setsof
constraintsinthepathdiagram.
44
Fitting a CFA Model with Constraints on the Loadings

path
Dem60 ---> Press60
= lam1,
These constrain the path coefficients.
Dem60 ---> Freop60
= lam2,
Dem60 ---> Fair60
= lam3,
Dem60 ---> Legis60
= lam4,
Dem65 ---> Press65
= lam1,
Dem65 ---> Freop65
= lam2,
Dem65 ---> Fair65
= lam3,
Dem65 ---> Legis65
= lam4;
pvar
Dem60 = 1, Dem65 = 1;
run;
45
InthePATHmodelinglanguage,theconstraintscouldbehandledsimilarly.Thecode
showninthisslideismodifiedfromthepreviouscodebyadding theparameternames
inthepaths.Thesyntaxistoaddanequalsignandthentheparameternamesafterthe
pathspecificationsinthePATHstatement.Withthesameparameternamesforthe
pairsoftherelatedpaths,theestimateswouldbeexactlythesame.
45
Fit Summary Table for the Political Democracy Data

with Loading Constraints
Fit Summary
Chi-Square
Chi-Square DF
Pr > Chi-Square
Hoelter Critical N
Adjusted GFI (AGFI)
RMSEA Estimate
46.8893
23
0.0023
56
0.0714
0.7844
0.1185
0.9440
Only Bentlers CFI indicates a good model fit.
46
Thistableshowsthefitsummaryofthemodelwiththeloadingconstraints.Becauseof
theconstraints,thismodeldoesnotfitaswellastheprevious model.TheSRMSRis
largerthan0.05.TheAGFIismuchsmallerthan0.9.TheRMSEAismuchlargerthan0.05.
Alltheseshowabadmodelfit.However,Bentlers CFI(0.94)showsagoodmodelfit.
46
Estimates of the Constrained Loadings for the

PATH List
----------Path----------
Parameter
Dem60
Dem60
Dem60
Dem60
Dem65
Dem65
Dem65
Dem65
lam1
lam2
lam3
lam4
lam1
lam2
lam3
lam4
--->
--->
--->
--->
--->
--->
--->
--->
Press60
Freop60
Fair60
Legis60
Press65
Freop65
Fair65
Legis65
Estimate
Standard
Error
t Value
2.13970
2.80116
2.54987
2.82969
2.13970
2.80116
2.54987
2.82969
0.21716
0.29976
0.27316
0.27285
0.21716
0.29976
0.27316
0.27285
9.85319
9.34473
9.33459
10.37075
9.85319
9.34473
9.33459
10.37075
All path coefficients are significant.
47
Asrequiredfromthemodel,pathswiththesameloadingparameterhavethesame
estimates.Forexample,bothDem60>Press60andDem65>Press65havealoading
estimateof2.14(lam1).Allloadingestimates,again,arestatisticallysignificant.This
showsthatallthepurportedfactorvariablerelationshipsaresupported.
47
Estimates of Variances and Covariances for the Political

Democracy Data with Loading Constraints
Variance Parameters
Variance
Type
Exogenous
Error
Variable
Dem60
Dem65
Press60
Freop60
Legis60
Press65
Freop65
Legis65
Fair60
Fair65
Parameter
_Add1
_Add2
_Add3
_Add4
_Add5
_Add6
_Add7
_Add8
Estimate
Standard
Error
t Value
1.00000
1.00000
2.01017
6.72037
2.88468
2.61966
4.16958
2.85029
5.40824
3.55382
0.40312
1.21196
0.60956
0.49456
0.79818
0.59556
0.97833
0.67700
4.98647
5.54503
4.73243
5.29699
5.22383
4.78593
5.52803
5.24935
Var1
Var2
Parameter
Dem65
Dem60
_Add9
Estimate
Standard
Error
t Value
0.97480
0.02682
36.34662
48
Alltheerrorvarianceestimatesarealsosignificant.ThecorrelationbetweenDem60and
Dem60isveryhighandsignificant.
48

with Correlated Errors
49
49
Adding Error Covariances

1.
Press60
lam1
Dem60
Freop60
lam2
lam3
Fair60
lam4
Legis60
Press65
Dem65
1.
lam1
lam2
Freop65
lam3
Fair65
lam4
Legis65
50
Withtheloadingconstraints,youobservedaworsemodelfit.
Thefourequalityconstraintsontheloadingsyoubasicallyreducethenumberofmodel
parametersby4.Thisnaturallyleadstoaworsemodelfitthanifyouwouldallowallthe
loadingstobefreelyestimated.
Nowyouconsideranoppositedirection.Insteadofreducingparametersbyputtingequality
constraints,youwanttoaddmoreparameterstothemodel.Addingmoreparameterstoyour
modelwouldimprovethemodelfit.Butthedrawbackofaddingmoreparametersisthatit
makesyourmodelmorecomplicated,whichisusuallyjudgedasan undesirablepropertyfora
scientifictheory.Itdoesnotmeanthatyoucannotaddparameters.Itonlymeansthatyou
shouldaddonlythoseparametersthatcouldbejustifiedbytheoreticalorsubstantivereasons.
Inthisexample,ithasbeenarguedthatfreedomofgroupoppositionandtheelectivenatureof
thelegislativebodyhaveapartoftheircorrelationthatisbeyondtheircommonlatentfactors
couldexplain(seeBollen).InSEM,thisextra correlationisconceptualizedasacorrelation(or
covariance)betweentheerrorsofthetwovariables.Inthepath diagram,thiserrorcovarianceis
representedbyadoubleheadedarrowconnectingthetwovariables.Thatis,Freop60and
Legis60areconnectedbyadoubleheadedarrowin1960.Bythesameargument,Freop65and
Legis65arealsoconnectedbyadoubleheadedarrowtorepresenterrorcovariance.
Inaddition,itisarguedinBollen thateachofthevariablepairsthatwereofthesamenaturebut
weremeasuredatdifferenttimeshaveapartofcorrelationthat isbeyondtheircommonlatent
factorscouldexplain.Forexample,Press60andPress65areconnectedbyadoubleheaded
arrowtorepresenttheirerrorcovariance,whichexplainsthepartofthecovariancebetween
thetwovariablesthatisbeyondtheexplanationbythecovariancebetweenDem60andDem65.
Similarly,theFreop,Fair,andLegis pairsareallconnectedbydoubleheadedarrowsto
representerrorcovariances.
50
Adding Error Covariances (with Error Terms

Displayed)
1.
Press60
lam1
Dem60
Freop60
lam2
lam3
lam4
Fair60
Legis60
Press65
Dem65
1.
lam1
lam2
Freop65
lam3
Fair65
lam4
Legis65
1.
1.
1.
1.
1.
1.
1.
1.
e1
e2
e3
e4
e5
e6
e7
e8
51
Thepathdiagraminthisslideisequivalenttothepreviousrepresentationthatdoesnot
useexpliciterrorvariables.
Inthispathdiagram,errortermsforthemeasuredvariablesare shown.Thedouble
headedarrowsareshiftedtotheerrorterms.Thismakesitobviousthatthosedouble
headedarrowsarecovariances betweentheerrorvariables(butnotaspartial
covariances betweentheobservedvariables,asshowninthepreviousslide).
Therefore,thispathdiagramrepresentationisconceptuallycleareraboutwhatare
reallybeingcorrelatedinthemodel.However,theadditionoftheerrortermsmakes
thepathdiagrammorecluttered.Inthisworkshop,Iwouldstick withthepathdiagram
representationthatdoesnotuseexpliciterrorterms.
51
Fitting a CFA Model with Loading Constraints

and Correlated Errors
path
Dem60 ---> Press60
= lam1,
Dem60 ---> Freop60
= lam2,
Dem60 ---> Fair60
= lam3,
Dem60 ---> Legis60
= lam4,
Dem65 ---> Press65
= lam1,
Dem65 ---> Freop65
= lam2,
Dem65 ---> Fair65
= lam3,
Dem65 ---> Legis65
= lam4;
Use the PCOV statement to
pvar
specify error covariances.
Dem60 = 1, Dem65 = 1;
pcov
Freop60 Legis60, Freop65 Legis65,
Press60 Press65, Freop60 Freop65,
Fair60 Fair65, Legis60 Legis65;
run;
52
Withthesixadditionalpairsofcorrelatederrors,youhavesix moreerrorcovariance
parametersinthemodel.
InthePATHmodelinglanguage,youcanspecifythesecovarianceparametersinthe
PCOVstatement.Inthisexample,thismeansthatyouenumeratethesixpairsof
measuredvariablesinthePCOVstatement.Forexample,thefirst pairisFreop60and
Legis60,whichrepresentacovarianceparameterbetweentheirerrorterms.
52
Fit Summary Table for the CFA Model with Loading

Constraints and Correlated Errors
Fit Summary
Chi-Square
Chi-Square DF
Pr > Chi-Square
Hoelter Critical N
Adjusted GFI (AGFI)
RMSEA Estimate
15.1946
17
0.5815
135
0.0590
0.9043
0.0000
1.0000
All indices indicate a good model fit.
53
Thismodelissupposedtofitbetterbecauseoftheaddedparametersfortheerror
covariances.
Infact,themodelfitchisquareisnotstatisticallysignificant.Thissupportsthe
hypothesizedmodelinthepopulation.
Allotherfitindicesshowgoodorexcellentfit.TheSRMSRis0.059,whichisonlyslightly
largerthanthe0.05criterion.TheAGFIis0.90,whichisanindicationofgoodmodelfit
byconvention.TheRMSEAisessentiallyzero,whichisthesmallestRMSEAyoucould
everget.TheCFIis1,whichisalsothelargestCFIyoucouldeverget.
53
Estimates of the Loadings for the CFA Model with

Constrained Loadings and Correlated Errors
PATH List
----------Path----------
Parameter
Dem60
Dem60
Dem60
Dem60
Dem65
Dem65
Dem65
Dem65
lam1
lam2
lam3
lam4
lam1
lam2
lam3
lam4
--->
--->
--->
--->
--->
--->
--->
--->
Press60
Freop60
Fair60
Legis60
Press65
Freop65
Fair65
Legis65
Estimate
Standard
Error
t Value
2.16450
2.61630
2.61693
2.75291
2.16450
2.61630
2.61693
2.75291
0.23009
0.32500
0.28700
0.28312
0.23009
0.32500
0.28700
0.28312
9.40738
8.05017
9.11832
9.72356
9.40738
8.05017
9.11832
9.72356
All path coefficients are significant.
54
Allloading(pathcoefficients)estimatesarestatisticallysignificant,supportingthe
relationshipsbetweenthelatentfactorsandthemeasuredvariables.
54
Estimates of the Variances for the CFA Model with

Variance Parameters
Variance
Type
Exogenous
Error
Variable
Dem60
Dem65
Press60
Freop60
Legis60
Press65
Freop65
Legis65
Fair60
Fair65
Parameter
_Add1
_Add2
_Add3
_Add4
_Add5
_Add6
_Add7
_Add8
Estimate
Standard
Error
t Value
1.00000
1.00000
1.91664
7.65544
3.27028
2.52969
4.87208
3.25392
5.03798
3.32508
0.43982
1.39023
0.73387
0.52882
0.94384
0.73319
0.98299
0.71220
4.35780
5.50661
4.45621
4.78360
5.16199
4.43805
5.12514
4.66875
All error variance estimates are significant.

55
Allerrorvarianceestimatesaresignificant.
55
Estimates of the Covariances for the CFA Model with

Var1
Var2
Parameter
Dem65
Dem60
_Add9
Estimate
Standard
Error
t Value
0.96603
0.02928
32.99044
Covariances Among Errors

Error
of
Error
of
Parameter
Freop60
Freop65
Press60
Freop60
Fair60
Legis60
Legis60
Legis65
Press65
Freop65
Fair65
Legis65
_Parm1
_Parm2
_Parm3
_Parm4
_Parm5
_Parm6
Estimate
Standard
Error
t Value
1.42826
1.26677
0.58548
2.09624
0.74805
0.47686
0.69666
0.59365
0.37178
0.74763
0.62336
0.46214
2.05017
2.13389
1.57478
2.80386
1.20003
1.03186
Bad news: Some error covariance estimates are not significant.

56
Thefirsttableshowsthecorrelationbetweenthetwolatentfactors.Again,the
correlationisveryhighandsignificant.
Thesecondtableshowstheestimatesforthenewlyaddedcovariances betweenerrors.
Threeofthesecovariances aresignificant,whiletheothersarenot.Forexample,
Freop60andLegis60,Freop65andLegis65,andFreop60andFreop65arethreeerror
covariances thathavetvalueslargerthan1.96.Theotherthreepairshave insignificant
tvalues.Thismeansthataddingthesethreecovariances mightbesomewhat
undesirablebecausetheirestimatesareactuallynotsignificantlydifferentfromzero,
castingdoubtsabouttheirpresenceinthemodel.
Thelessonhereisthateventhoughaddingerrorcorrelations(orcovariances)might
improvethemodelfit,youshouldnotroutinelyadderrorcovariances onlytoboostthe
modelfit.Addingunjustifiederrorcovariances makesyourmodelmorecomplicatedand
hardertointerpret,especiallywhensomeerrorvarianceestimatesturnouttobe
insignificant.
56
Political Democracy and

Industrialization:
A Full Structural Equation Model
57
57
Political Democracy and Industrialization

Bollen (1989) Chapter 8
A full structural equation model (a full LISREL model)
Additional variables for measuring industrialization (Indust)
in 1960
o
Gross national product per capita (Gnppc60)
Energy consumption per capita (Enpc60)
Percent of labor force in industrial occupations (Indlf60)
Purposes: Validate the measurement model and the

structural relationships
58
Wecontinuewiththepreviousmodelandaddonemorelatentfactoranditsindicators
intothemodel.
Thisexampleillustratesafullstructuralequationmodel(orafullLISREL)model.
Essentially,thismeansthatourfocusisnotonlyonvalidating therelationshipsbetween
thelatentfactorsandthemeasuredvariables(thatis,themeasurementmodel),but
alsoonvalidatingthefunctionalrelationshipsamonglatentvariables(thatis,the
structuralmodel).
Forexample,youhavealatentfactorcalledindustrialization(Induct)thatissupposedto
bereflectedbythreeobservedvariables:grossnationalproduct percapita(Gnppc60),
energyconsumptionpercapita(enpc60),andpercentoflaborforceinindustrial
occupations(Indlf60).Allthesevariablesweremeasuredin1960.
Theindustrialization(Induct)latentvariableservesasapredictorofthetwodemocracy
factors(Dem60andDem65).Thiskindoffunctionalrelationships betweenlatent
variableshasnotbeenexploredintheconfirmatoryfactormodelsdiscussedpreviously.
58
Political Democracy and Industrialization:

The Path Diagram
Press60
1.
Dem60
Gnppc60
Enpc60
1.
Freop60
lam2
lam3
Fair60
lam4
Legis60
Indust
Press65
Indlf60
Dem65
1.
lam2
Freop65
lam3
Fair65
lam4
Legis65
59
TheentireSEMmodelisdepictedinthepathdiagramofthecurrentslide.Themost
notableadditionisthepathsfromIndust toDem60andDem65 industrializationin
1960servesasapredictorofdemocracyinboth1960and1965.Threeobserved
variablesserveasindicatorsoftheindustrialization:Gnppc60, Enpc60,andIndlf60.
Therearetwomainmodificationsfromtheprecedingconfirmatory factormodel.
First,insteadofallowingDem60andDem65tofreelycovary intheCFA,thecurrent
modeltreatsDem60asapredictorofDem65.
Second,adifferentmethodforidentifyingthelatentfactorscalesisusedinthecurrent
model.IntheprecedingCFAmodel,variancesofDem60andDem65arefixedtoone.
Butbecausetheybecomeendogenousinthecurrentmodel,youcan nolongerusethis
typeofscaleidentificationmethod.Instead,oneoftheirobservedindicatorvariables
(thatis,Press60andPress65)nowhasafixedpathcoefficientatone.Similarly,thepath
coefficientfromIndust toGnppc60isfixedtooneforscaleidentification.
59
Fitting the Structural Equation Model for the Political

Democracy and Industrialization Data
Multiple-path specifications

path
Dem60
---> Press60 Freop60 Fair60 Legis60
1.
lam2
lam3
lam4,
Dem65
1.
lam2
lam3
lam4,
1.,
Indust ---> Gnppc60 Enpc60 Indlf60

Indust ---> Dem60 Dem65,
Dem60
---> Dem65;
pcov
run;
60
YoucanusePROCCALIStospecifythisstructuralequationmodel easily.
InthePATHstatement,Iuseamultiplepathspecificationsyntax.Inthefirst
specification,Dem60isapredictorof4outcomevariables:Press60,Freop60,Fair60,
andLegis60.Thisspecifiesfourpathsinasinglepathspecification.Afterusinganequal
sign,Ispecifyfourparametersforthefourpaths.Thefirstoneisafixedconstant1,
whichisappliedtotheDem60>Press60path.Thesecondoneisafreeparameter
lam2,whichisappliedtotheDem60>Freop60path,andsoon.
Inthenext3pathspecifications,Ialsousethemultiplepathspecificationsyntax.The
secondmultiplepathsyntaxspecifiesthatDem65isafactorwithfourindicators.The
pathcoefficients(loadings)arealsospecifiedexplicitly.Thethirdmultiplepathsyntax
specifiesthatIndust isafactorofthreeobservedindicators,withafixedoneforthe
effectofIndust onGnppc60.ThepathcoefficientsforthepathsIndust>Enpc60and
Indust>Indlf60areunnamedfreeparameters(withtheemptyspecifications).The
fourthmultiplepathsyntaxspecifiesthatIndust isapredictorofbothDem60and
Dem65.Thecorrespondingpathcoefficientsare(unnamed)freeparametersinthe
model.
ThelastpathinthePATHstatementspecifiesDem60asapredictorofDem65.Notice
thatnoPVARstatementisusedbecausefixingtheDem60andDem65variancestoone
isnotusedinthecurrentmodel.Thescalesofthelatentfactorsareidentifiedbyfixing
somepathcoefficientsto1.
60

Fit Summary Table
Fit Summary
Chi-Square
Chi-Square DF
Pr > Chi-Square
Hoelter Critical N
Adjusted GFI (AGFI)
RMSEA Estimate
39.6438
38
0.3966
100
0.0558
0.8606
0.0242
0.9975
Not a bad fit for the data.
61
Thefitofthestructuralmodelisacceptable,ifnotexceptionallygood.
Themodelfitchisquareisnotsignificant,supportingthehypothesizedmodel.The
SRMSRiscloseto0.05.TheAGFIis0.86,whichshowsareasonablefit.TheRMSEA
indicatesaverygoodmodelfit,asthevalue(.0242)ismuchlowerthan0.05.TheCFIis
almost1,whichshowsaperfectmodelfit.
61

Estimates of Path Coefficients
PATH List
----------Path----------Dem60
Dem60
Dem60
Dem60
Dem65
Dem65
Dem65
Dem65
Indust
Indust
Indust
Indust
Indust
Dem60
--->
--->
--->
--->
--->
--->
--->
--->
--->
--->
--->
--->
--->
--->
Press60
Freop60
Fair60
Legis60
Press65
Freop65
Fair65
Legis65
Gnppc60
Enpc60
Indlf60
Dem60
Dem65
Dem65
Parameter
lam2
lam3
lam4
lam2
lam3
lam4
_Parm01
_Parm02
_Parm03
_Parm04
_Parm05
Estimate
1.00000
1.19079
1.17454
1.25099
1.00000
1.19079
1.17454
1.25099
1.00000
2.17966
1.81821
1.47133
0.60046
0.86504
Standard
Error
t Value
0.14020
0.12121
0.11757
8.49336
9.68988
10.64006
0.14020
0.12121
0.11757
8.49336
9.68988
10.64006
0.13932
0.15290
0.39496
0.22722
0.07538
15.64530
11.89126
3.72529
2.64267
11.47648
62
Allpathcoefficientsaresignificantaprettygoodsign.
62

Estimates of Variances
Variance Parameters
Variance
Type
Exogenous
Error
Variable
Parameter
Indust
Press60
Freop60
Legis60
Press65
Freop65
Legis65
Gnppc60
Enpc60
Indlf60
Fair60
Fair65
Dem60
Dem65
_Add01
_Add02
_Add03
_Add04
_Add05
_Add06
_Add07
_Add08
_Add09
_Add10
_Add11
_Add12
_Add13
_Add14
Estimate
Standard
Error
t Value
0.45466
1.87973
7.68378
3.26801
2.34432
5.03534
3.35236
0.08249
0.12206
0.47297
5.02270
3.60813
3.92767
0.16668
0.08846
0.44229
1.39404
0.73807
0.48851
0.93993
0.71788
0.01986
0.07105
0.09197
0.97587
0.72394
0.88311
0.23158
5.13991
4.25001
5.51189
4.42779
4.79895
5.35716
4.66983
4.15376
1.71776
5.14268
5.14691
4.98402
4.44753
0.71975
63
Someerrorvariancesarenotsignificant:Enpc60andDem65.Enpc60isanindicatorof
theIndustrializationin1960.ThisinsignificanterrorvariancemeansthattheIndust
factorpredictEnpc60perfectly.However,thecorrespondingtvalueis1.71,whichcould
bejudgedasmarginallysignificant.
TheerrorvarianceforDem65isalsonotsignificant,asevidentbythenonsignificantt
valueof0.72.ThismeansthatgivenIndust andDem60,Dem65canbepredictedalmost
perfectly.
63

Estimates of Covariances
Covariances Among Errors
Error
of
Error
of
Parameter
Freop60
Freop65
Press60
Freop60
Fair60
Legis60
Legis60
Legis65
Press65
Freop65
Fair65
Legis65
_Parm06
_Parm07
_Parm08
_Parm09
_Parm10
_Parm11
Estimate
Standard
Error
t Value
1.45956
1.39032
0.59042
2.21252
0.72123
0.36769
0.70251
0.58859
0.36307
0.75242
0.62333
0.45324
2.07764
2.36212
1.62619
2.94054
1.15706
0.81125
64
Again,therearesomeinsignificanterrorcovariances.Thisresultchallengestheir
presenceinthemodel.
64
Political Democracy and Industrialization: Squared

Multiple Correlations
Squared Multiple Correlations
Variable
Enpc60
Fair60
Fair65
Freop60
Freop65
Gnppc60
Indlf60
Legis60
Legis65
Press60
Press65
Dem60
Dem65
Error
Variance
Total
Variance
R-Square
0.12206
5.02270
3.60813
7.68378
5.03534
0.08249
0.47297
3.26801
3.35236
1.87973
2.34432
3.92767
0.16668
2.28211
11.79895
10.09361
14.64875
11.70144
0.53715
1.97602
10.95502
10.70953
6.79166
7.04548
4.91193
4.70116
0.9465
0.5743
0.6425
0.4755
0.5697
0.8464
0.7606
0.7017
0.6870
0.7232
0.6673
0.2004
0.9645
65
Thesquaremultiplecorrelationsareusuallyusedtomeasurethe percentageof
overlappingvariancebetweenthepredictorsandtheoutcomevariables.Inthecurrent
example,Rsquaresrangefrom0.2toextremehighvaluessuchas0.95and0.96.
ThesmallestRsquareistheoneforpredictingDem60,whichis0.2.Thisactuallyisnot
thatsmallanRsquarevalueforsocialsciencedata.
ButtheRsquare(0.96)forDem65isextremelyhigh.ThismeansthatDem65 isalmost
perfectlypredictedfromdemocracyandindustrializationin1960.
65
A LISREL Model for the Political Democracy and

Industrialization Data
Structural
model
1.
Measurement
model
Gnppc60
Enpc60
Measurement
model
Dem60
1.
Press60
Freop60
lam2
lam3
Fair60
lam4
Legis60
Indust
Press65
Indlf60
Dem65
Measurement
model
1.
lam2
Freop65
lam3
Fair65
lam4
Legis65
66
ThisfullSEMmodelisalsoagoodillustrationoftheLISRELmodel.
Thepathdiagramfortheprecedingmodelremainsunchangedhere. Inordertocallthis
pathdiagramaLISRELmodel,youhavetoidentifytheLISRELcomponentsinthispath
diagram.ThetwomaincomponentsinLISRELarethemeasurementmodelsandthe
structuralmodel.
First,themeasurementmodelsareidentified.Ameasurementmodelisabouthow
observedvariablesarerelatedtothelatentvariablesorconstructsinthemodel.
Specifically,themeasurementmodelinvolvingindustrializationisthemeasurement
modelofxbecauseIndust servesasanexogenous(independent)variableinthepath
diagram.ThemeasurementmodelinvolvingDem60andDem65isthemeasurement
modelofybecauseDem60andDem65areendogenous(dependent)variablesinthe
pathdiagram.
Second,thestructuralmodelisidentifiedandhighlightedinthecenterofthepath
diagram.Thestructuralmodeldescribesthefunctionalrelationshipsamongthelatent
variables(constructs)inthepathdiagram.
Therefore,alltheessentialcomponentoftheLISRELmodelisidentifiedinthecurrent
pathdiagram.
66
Fitting the Structural Equation Model for the Political

Democracy and Industrialization Data
Measurement model: Relationships between
latent and observed indicators

path
Dem60
1.
lam2
lam3
lam4,
Dem65
1.
lam2
lam3
lam4,
1.,
Structural model: Relationships

among latent constructs

Dem60
---> Dem65;
pcov
Measurement model
run;
67
InthePATHmodelinglanguage,youcanalsoidentifythecodeforthemeasurement
modelsandthestructuralmodel.Theprecedingcodeisrecitedhereforillustrations.
InthePATHstatement,thefirstthreemultiplepathspecificationsareconcernedwith
themeasurementofthelatentconstructs.Inaddition,allspecificationsinthePCOV
statementareforthecovariances ofthemeasurementerrors.
ThelasttwospecificationsinthePATHstatementareforthestructuralmodel.They
describethefunctionalrelationshipsbetweenIndust,Dem60,andDem65.
AfteridentifyingtheLISRELcomponentsinthepathdiagramandintheSAScode,you
nowhaveacluetospecifytheLISRELmodelinPROCCALIS.Essentially,itshouldbe
clearthatthesamepathdiagramisbeingusedbythePATHmodelinglanguageandthe
LISRELmodel.TheonlytaskistotranscribethecodeinthePATHmodelinglanguageto
thelanguagefortheLISRELmodel.
67
A LISREL Model Specified by the LISMOD

Modeling Language of PROC CALIS
proc calis data=polidem nose noparmname;
lismod
xvar = Gnppc60 Enpc60 Indlf60,
yvar = Press60 Freop60 Fair60 Legis60 Press65
xi
= Indust,
eta = Dem60 Dem65;
matrix
_LambdaY_ [ 1, @1] = 1. lam2 lam3 lam4,
[ 5, @2] = 1. lam2 lam3 lam4;
matrix
_ThetaY_ [ 4, 2], [ 8, 6],
[ 5, 1], [ 6, 2], [ 7, 3], [ 8, 4];
matrix
_LambdaX_ [ 1, 1] = 1.,
[ 2 to 3, 1];
matrix
_Gamma_
matrix
_Beta_
run;
[ 1 to 2, 1];
NOSE: No standard errors

NOPARMNAME: No parameter names in the
output
Freop65 Fair65 Legis65,
Measurement model for
y
/* Paths from Dem60 and Dem65 to yvar */
/* pcov statement in the path model */
/* Path from Indust to xvar*/

Measurement model for
x
Structural
model
[2,1];
68
PROCCALISsupportsthesocalledLISMODmodelinglanguage.Inordertofullyunderstandthe
PROCCALIScodefortheLISRELmodel,knowledgeaboutmatrixalgebraisneeded.ButIwillonly
describethecodeinaconceptualway.
IntheLISMODstatement,youfirstclassifyyourvariablesintooneofthefourcategories:
1.xvariables:observedindicatorsfortheexogenous(independent)latentfactorsinthemodel.
2.yvariables:observedindicatorsfortheendogenous(dependent)latentfactorsinthemodel.
3.xivariables:exogenous(independent)latentfactorsinthemodel.
4.etavariables:endogenous(dependent)latentfactorsinthemodel.
TheLISRELmodelorLISRELprogramhadbeendevelopedasamatrixbasedlanguage.
Parametersinthemodelsarespecifiedasmatrixelementsinsomespecificmodelmatriceswith
Greeknames.PROCCALISsupportsthematrixinputoftheseLISRELmodelmatrices.For
example,inthemeasurementmodelfory,_LambdaY_isthematrixthatrelatestheyvariables
totheetavariables.InsteadofspecifyingthepathsasinthePATHstatement,theMATRIX
statementfor_LambdaY_servesthesamepurposeintheLISMODmodelinglanguage.The
MATRIXstatementfor_ThetaY_specifiestheerrorvariancesandcovariances oftheyvariables,
muchlikethespecificationsofthePCOVstatementinthePATHmodelinglanguage.Inother
words,thePATHmodelspecificationsaretranscribedintotheLISMODmodelspecificationsfor
theyvariables.
Similarly,theMATRIXstatementfor_LAMBDAX_specifiestheparametersinthemeasurement
modelforthexvariables.
Finally,thestructuralrelationshipsorthepathrelationshipsamongthelatentfactorsare
specifiedintheMATRIXstatementsforthe_GAMMA_and_BETA_matrices.
Tosimplifytheoutput,IusedtwooptionsinPROCCALISstatement.TheNOSEoption
suppressestheprintingofstandarderrorsandtheNOPARMNAMEoptionsuppressesthe
printingoftheparameternames.
68
LISMOD Output for the Political Democracy and

Industrialization Data (Measurement for y)
_LAMBDAY_ Matrix
Dem60
Dem65
Press60
1.0000
Freop60
1.1908
Fair60
1.1745
Legis60
1.2510
Press65
1.0000
Freop65
1.1908
Fair65
1.1745
Legis65
1.2510
Note: The NOSE and NOPARMNAME options suppress the printing of

the standard error estimates and parameter names.
69
ThefollowingfewslidesshowtheoutputfromPROCCALISforthe LISRELmodel.Allthe
resultsarematrixoriented.DetailsfortheseresultshavebeendiscussedforthePATH
modeloutputandwillnotberepeatedhere.Ingeneral,youcanfindcorrespondence
betweentheLISMODandthePATHresults.
Thisslideshowsthemeasurementmodelfortheyvariables.
69

Industrialization Data (Measurement for y)
_THETAY_ Matrix
Press60
Freop60
Fair60
Legis60
Press65
Freop65
Fair65
Legis65
Press60
1.8797
0.5904
Freop60
7.6838
1.4596
2.2125
Fair60
5.0227
0.7212
Legis60
1.4596
3.2680
0.3677
Press65
0.5904
2.3443
Freop65
2.2125
5.0353
1.3903
Fair65
0.7212
3.6081
Legis65
0.3677
1.3903
3.3524
Note: Error variances (diagonal elements) were set by default.
70
Thisslideshowsthemeasurementerrorvariancesandcovariances fortheyvariables.
70

Industrialization Data (Measurement for x)
_LAMBDAX_ Matrix
Indust
Gnppc60
1.0000
Enpc60
2.1797
Indlf60
1.8182
_THETAX_ Matrix
Gnppc60
Gnppc60
Enpc60
Indlf60
0.0825
Enpc60
0.1221
Indlf60
0.4730
Note: Error variances (diagonal elements in _THETAX_) were set by default.

71
Thisslideshowstheresultsofthemeasurementmodelforthexvariables,includingthe
pathcoefficientsandtheerrorvariances.
71

Industrialization Data (Structural Model)
_BETA_ Matrix
Dem60
Dem65
Dem60
Dem65
0.8650
_GAMMA_ Matrix
Indust
Dem60
1.4713
Dem65
0.6005
72
Thisslideshowsthefunctionalrelationshipsbetweenlatentconstructs.
72

Industrialization Data (Structural Covariances)
_PSI_ Matrix
Dem60
Dem65
Dem60
3.9277
Dem65
0.1667
_PHI_ Matrix
Indust
Indust
0.4547
Note: All variances (diagonal elements in _PSI_ and _PHI_) were set by default.
73
Thisslideshowstheerrorvariancesoftheetavariableandthevarianceofthexi
variable.
73
Features of the LISMOD Modeling Language

Supports the JKW (LISREL) models (not the LISREL
program)
Supports mean structure analysis
Users input:
The ordered lists of x, y, , and variables
MATRIX statements to define free and fixed parameters
Names for parameters (not required for free parameters)
Default covariance structure parameters of the LISMOD

language:
Diagonal elements of all covariance matrices (all variances)
Lower triangular elements of the _PHI_ matrix (covariances of the variables)
74
Insum,theLISMODmodelinglanguageinPROCCALISsupportstheLISRELmodelbyproviding
syntaxtospecifytheessentialcomponentsoftheLISRELmodel.However,LISMODitselfdoesnot
interpretaLISRELprogram.
TheLISMODmodelinglanguageinPROCCALISalsosupportsthemeanstructureanalysis.Thisis
donebyprovidingadditionalMATRIXstatementinputforthemean modelmatricesintheLISREL
model.
IfyouunderstandtheLISRELmodel,herearethreethingsyouinputbyusingtheLISMODlanguage:
1.Theorderedlistsofx,y,,and variables
2.MATRIXstatementstodefinefreeandfixedparameters
3.Namesforparameters(notrequiredforfreeparameters)
(Note:Thisslidewasupdatedaftertheprintingofthehandout)
TheDefaultcovariancestructureparametersintheLISMODlanguageare:
1.Diagonalelementsofallcovariancematrices(allvariances)
2.Lowertriangularelementsofthe_PHI_matrix(covariances ofthe variables)
Specifyingthedefaultparametersexplicitlyiscertainlyallowed,especiallywhenyouneedtoset
constraintsontheseparameters.
Inaddition,whenthemeanstructuresaremodeled,theinterceptsofthex andy variablesare
defaultfreeparameters,whiletheinterceptsofthe variablesandthemeansofthe variables
arefixedzerosbydefault.
74
The Generalized
PATH Modeling Language
75
75
Political Democracy and Industrialization: Variances

and Covariances Are Paths
Press60
1.
Dem60
Gnppc60
Enpc60
1.
Freop60
lam2
lam3
Fair60
lam4
Legis60
Indust
Press65
Indlf60
Dem65
1.
lam2
Freop65
lam3
Fair65
lam4
Legis65
Covariances are represented as double-headed

arrows.
76
TogeneralizethePATHmodelinglanguage,errorcovariances inthepathdiagramcould
alsobespecifiedaspaths inPROCCALIS.Infact,covariances inthepathdiagramare
alreadyrepresentedasdoubleheadedarrows,asshowninthepoliticaldemocracyand
industrializationexample.
76
An Example of the Generalized

PATH Modeling Language
PROC CALIS DATA=polidem;
PATH
Dem60 ---> Press60 Freop60 Fair60 Legis60
Dem65 ---> Press65 Freop65 Fair65 Legis65
Dem60 ---> Dem65;
PCOV
path
Dem60 ---> Press60 Freop60 Fair60
Dem65 ---> Press65 Freop65 Fair65
Dem60 ---> Dem65,
Freop60 <--> Legis60, Freop65 <-->
Press60 <--> Press65, Freop60 <-->
Fair60 <--> Fair65, Legis60 <-->
Legis60
Legis65
=
=
=
1. lam2
1. lam2
1.,
lam3
lam3
lam4,
lam4,
Use the PCOV statement to specify

the covariances
=
=
=
1. lam2
1. lam2
1.,
Legis65,
Freop65,
Legis65;
lam3
lam3
lam4,
lam4,
Use the generalized path to

specify covariances (and
variances)
77
ThetoppanelshowstheuseofPCOVstatementtospecifythecovariances.Thebottom
panelshowsthatthesecovariances arespecifiedasdoubleheadedpaths,which
resembletheirrepresentationsinthepathdiagram.
ThetwoPROCCALISspecificationsshownaboveareequivalent.
77
Political Democracy and Industrialization: Output with

Generalized Paths
PATH List
-----------Path----------Dem60
Dem60
Dem60
Dem60
Dem65
Dem65
Dem65
Dem65
Indust
Indust
Indust
Indust
Indust
Dem60
Freop60
Freop65
Press60
Freop60
Fair60
Legis60
--->
--->
--->
--->
--->
--->
--->
--->
--->
--->
--->
--->
--->
--->
<-->
<-->
<-->
<-->
<-->
<-->
Press60
Freop60
Fair60
Legis60
Press65
Freop65
Fair65
Legis65
Gnppc60
Enpc60
Indlf60
Dem60
Dem65
Dem65
Legis60
Legis65
Press65
Freop65
Fair65
Legis65
Parameter
lam2
lam3
lam4
lam2
lam3
lam4
_Parm01
_Parm02
_Parm03
_Parm04
_Parm05
_Parm06
_Parm07
_Parm08
_Parm09
_Parm10
_Parm11
Estimate
1.00000
1.19079
1.17454
1.25099
1.00000
1.19079
1.17454
1.25099
1.00000
2.17966
1.81821
1.47133
0.60046
0.86504
1.45956
1.39032
0.59042
2.21252
0.72123
0.36769
Standard
Error
t Value
0.14020
0.12121
0.11757
8.49336
9.68988
10.64006
0.14020
0.12121
0.11757
8.49336
9.68988
10.64006
0.13932
0.15290
0.39496
0.22722
0.07538
0.70251
0.58859
0.36307
0.75242
0.62333
0.45324
15.64530
11.89126
3.72529
2.64267
11.47648
2.07764
2.36212
1.62619
2.94054
1.15706
0.81125
78
TheresultsobtainedfromPROCCALISnowshowsthecovarianceestimatesaspaths
inthePATHlist.Thistablecouldbeuseddirectlyinresearchpaperforreportingthe
SEMestimationresults.
78
Features of the Generalized PATH Modeling

Language
Extension of the PATH modeling language
Represents all generalized paths in the PATH statement
Variance-path:
Y <--> Y
Covariance-path: X <--> Y
Mean or intercept (one-path): 1 ---> Y
79
Insum,thegeneralizedpathmodelinglanguageenablesyoutospecifyalltypesof
arrowsinthepathdiagramaspaths, includingthevariance,covariance,intercept,and
meanparameters.
VarianceofYisapathlikeY<>Y.
CovariancebetweenXandYisapathlikeX<>Y
MeanorinterceptforYisonepathlike1>Y.
79
Default Free and Fixed Parameters in PROC CALIS

Default free parameters
o Variances of and covariances among all exogenous
(independent) variables (observed or latent, except for error terms)
o Error variances of all endogenous (dependent) variables
o Means or intercepts of all observed variables
Default fixed zeros

o Unspecified paths and error covariances
o Means or intercepts of all latent variables
The main purpose of setting default parameters is to enable you to specify only the
functional relationships among variables in most practical applications.
80
KnowingthedefaultfreeandfixedparametersinPROCCALISareusefulbecauseit
enhancesthecodingefficiencyandaccuracy.Hereisalistofdefaultfreeparameters
andfixedzerosusedinPROCCALIS:
(Thisslidehasbeenchangedslightlyaftertheprintingofthehandout)
Defaultfreeparameters
o Variancesofandcovariances amongallexogenous (independent)variables
(observedorlatent,exceptforerrorterms)
o Errorvariancesofallendogenous (dependent)variables
o Meansorinterceptsofallobserved variables
Defaultfixedzeros
o Unspecifiedpathsanderrorcovariances
o Meansorinterceptsofalllatent variables
Atthefirstglance,itmightseemtobetediousanddemandingthatmodelersmust
rememberallthesedefaultparameterrulestospecifyanSEMaccurately.However,the
defaultparameterizationusedinPROCCALISmatchesthatofregressionanalysisandit
isdesignedwiththefollowingmainpurposeinmind:Inmostpracticalapplications,you
wouldonlyneedtospecifythefunctionalrelationshipsamongvariables(thatis,the
singleheadedpathsinthepathdiagram)andthefixedvariancesofthe latentvariables.
80
Multiple-Group Analysis
81
MultiplegroupanalysisrepresentsanimportantclassofSEMapplications.
81
PROC CALIS Syntax for Specifying a

Multiple-Group Analysis
proc calis;
group 1 / data=ds1;
group 2 / data=ds2;
model 1 / group=1;
/* ds1: data set for Group 1 */

/* ds2: data set for Group 2 */
/* Group 1 is fitted by Model 1 */
/* Insert the Model Specification for Group 1 */

model 2 / group=2;
/* Group 2 is fitted by Model 2 */
/* Insert the Model Specification for Group 2 */

run;
82
TheskeletonofthemultiplegroupsyntaxofPROCCALISisshowninthisslide.
Inthisspecification,youhavetwoindependentgroups,whicharestoredinseparate
datasetsds1 andds2. Youdefinetwomodels:Model1andModel2,respectively,for
fittingtothetwogroups.
UndereachMODELstatement,youspecifythemodelbyoneofthemodelinglanguages
supportedbyPROCCALIS.Forexample,aPATHstatementoraLISMODstatement.
82
Group-Model Mapping and Model Referencing

proc calis;
group 1 / data=ds1;
group 2 / data=ds2;
Both Groups 1 and 2 are fitted by
Model 1.
group 3 / data=ds3;
model 1 / group = 1,2;
path
x1 ---> x2-x5,
x2 ---> x3-x5,
x3 ---> x5
= effect1;
Model 2 makes reference to Model 1,
model 2 / group = 3;
but with a unique parameter for the
path from x3 to x5. All other paths
refmodel 1;
are constrained between the two
path
models.
x3 ---> x5
= effect2;
run;
Note: REFMODEL does not constrain default free parameters in the two models.
It constrains only those parameters that are specified explicitly.
83
Thisexampleshowsthreegroupsofdata.Model1isfittedtoGroups1and2bythe
PATHmodelspecifiedimmediatelyafter.Model2isfittedtoGroup3.TodefineModel
2,aREFMODELstatementisused.REFMODEL1meansthatthespecificationsinModel
1isreferencedhere.Thismeansthatallexplicit specificationsinModel1arecopied
intothecurrentmodel.Forexample,allthePATHstatementspecificationsinModel1
arecopiedintoModel2.However,REFMODELallowsmodificationsfromthereference
model.Inthisexample,thex3>x5pathisrespecifiedwithanewpathcoefficient
calledeffect2. Thiseffect2 inModel2isadifferentparameterthaneffect1 inModel
1,eventhoughbothmodelhavethisx3>x5path.Exceptforthisx3>x5path,Models
1and2sharealltheremainingspecifiedpathswiththesamesetofpathcoefficients.
NoticethatbyusingtheREMODELstatementtoconnectmodels,all(andonly)the
explicit specificationsinthemodelsareconstrained(unlessbeingmodifiedorre
specified).Theconstraints,however,willnotapplytodefaultparameters(assumingthat
theyarenotexplicitlyspecified).Toconstrainthedefaultparameters,theywillhaveto
bespecifiedexplicitlyintheMODELdefinitions.
83
Completely Constrained CFA Models for Male

and Female Groups
Males
Females
v1
1.
Spatial
a2
Visual
ev1
Cubes
ev2
v1
1.
Spatial
a3
a2
Visual
ev1
Cubes
ev2
Lozenges
ev3
Paragraph
ev4
Sentence
ev5
WordMean
ev6
a3
Lozenges
ev3
cv
cv
1.
Verbal
a5
Paragraph
ev4
Sentence
ev5
1.
a6
a5
a6
WordMean
v2
Verbal
ev6
v2
Data source: Holzinger and Swineford1939 (an example used in Arbuckle 2008)
84
Letususeanexampletoillustratethemultiplegroupanalysis.TheHolzinger and
Swineford dataaresued.ThisdatasetisalsousedinArbucklesAMOSmanual.
Inthisresearch,visualandverbaltestscoreswereobserved.Visual,Cubes,and
Lozengesarespatialtests.Paragraph,Sentence,andWordMean areverbaltests.CFA
modelswerehypothesizedforthetwogroups:onegroupisformalesandtheotherfor
females.Thebasicfactorstructuresforthegroupsarethesame.Twolatentfactors
Spatial andVerbal areassumedforthemeasuredvariables.Allparameters(including
thosedefaultparametersinPROCCALIS)inthetwomodelsforthegroupsarelabeledin
red.
Thisisacompletelyconstrainedmultiplegroupmodelbecausethetwopathdiagrams
formalesandfemalesareexactlythesame(thatis,bothhavethesamepathstructures
andthesamesetofparameters).
84
Different Methods to Specify the Completely

Constrained Models
1. Male and female groups fit by two models, but with all
parameters (including all default parameters) being
constrained in the models
2. Male and female groups fit by a single model
definition
3. Male and female groups fit by two models that are
constrained through the REFMODEL specification
85
InPROCCALIS,youcanuseoneofthefollowingthreewaystospecifythepreceding
completelyconstrainedmodel:
1.Maleandfemalegroupsfittedbytwomodels,butwithallparameters(includingall
defaultparameters)beingconstrainedinthemodels
2.Maleandfemalegroupsfittedbyasinglemodeldefinition
3.Maleandfemalegroupsfittedbytwomodelsthatareconstrained throughthe
REFMODELspecification
Iwilldescribeeachofthesemethods.
85
Method 1: Two Groups Fitted by the Two Completely

Constrained Models
proc calis;
Models are
group 1 / label='Males' data=males;
constrained through
group 2 / label='Females' data=females;
the use of the same
parameter names.
path
Spatial ---> Visual Cubes Lozenges
= 1. a2 a3,
Verbal
---> Paragraph Sentence Wordmean = 1. a5 a6;
pvar
Visual Cubes Lozenges Paragraph Sentence Wordmean = ev1-ev6,
Spatial = v1, Verbal = v2;
pcov
Spatial Verbal = cv;
path
= 1. a2 a3,
Verbal
---> Paragraph Sentence Wordmean = 1. a5 a6;
pvar
Visual Cubes Lozenges Paragraph Sentence Wordmean = ev1-ev6,
Spatial = v1, Verbal = v2;
pcov
Spatial Verbal = cv;
fitindex on(only) = [chisq df probchi rmsea srmsr aic
run;
86
Thefirstmethodistodefinetwomodelsforthetwogroups.The modelspecifications
underthetwoMODELstatementsmustbeexactlythesame.
Thismethodisintuitive,butalittleclumsybecauseyouneedtospecifyallparameters
withmatchingnamesinthemodels(althoughyoucancutandpastethemodel
specificationstoensureanexactcopy).Youalsoneedtospecifyeachparameterinthe
model,includingthedefaultparameters,whichyoumightsometimesmiss.
86
Method 2: Two Groups Fitted by the Same Model
proc calis;
Groups 1 and 2 are fitted
model 1 / group = 1, 2;
by the same model.
path
= 1. ,
Verbal
---> Paragraph Sentence Wordmean = 1. ;
run;
87
Thesecondmethodisverysimpleandintuitive.Youspecifyonemodelandfitthis
modeltothetwogendergroups.Thisensuresthegroupsarefittedexactlybythesame
model.
Theadvantageofthismethodisthatitissimple,intuitive,andnoparameternamesare
necessaryforconstrainingmodels.Also,youdonotneedtospecifyanyofthedefault
parametersexplicitlyforsettingupconstraints.
Thisisanidealspecificationmethodifthismodelisallwhatyouwanttofit.However,if
youaregoingtofitasequenceofmultiplegroupmodelstothegroups,youmightwant
toconsiderthenextmethod.
87
Method 3: Two Groups Fitted by the Two Models

That Are Constrained Through the REFMODEL
proc calis;
path
= 1. ,
Verbal
pvar
Visual Cubes Lozenges Paragraph Sentence Wordmean
Spatial Verbal;
pcov
Spatial Verbal;
The REFMODEL statement makes reference
to all explicit specifications in Model 1.
refmodel 1;
run;
Note: This method is used for the completely constrained
model and the subsequent models with parameter constraints.
88
ThethirdmethodconstrainsthemodelsbytheREFMODELstatement.Asdiscussed
previously,theREFMODELcopiesalltheexplicit specificationsfromthereference
modeltothetargetmodel.Inthisslide,allpathcoefficients, varianceparameters,and
covarianceparametersarespecifiedinModel1,whichisfittedtoGroup1(Males).
Model2,whichisfittedtoGroup2(Females),makesreferencetoModel1withoutany
modificationsorrespecifications.
Noticethatinordertocompletelyconstrainthetwomodelsforthetwogroups,all
parameters,includingthosecouldhavebeensetbydefaultbyPROCCALIS(e.g.,
specificationsinthePVARstatementandPCOVstatement),mustbespecifiedexplicitly
inModel1.Thisway,Model2willcopyalltheseparameterspecificationsviathe
REFMODELstatementspecification.
UnlikeMethod1,thismethoddoesnotrequiretheuseofparameternamesfor
constraintsacrossmodels.ConstraintsaredoneviatheREFMODEL statement.
AlthoughnotasintuitiveasMethod2,thismethodwouldbeusefulifyouneedto
considerfittingasequenceofmultiplegroupmodels,whichwillbeillustratedlater.
88
Fit Summary of the Completely Constrained

Multiple-Group Model
Fit Summary
Chi-Square
Chi-Square DF
Pr > Chi-Square
Adjusted GFI (AGFI)
RMSEA Estimate
Bozdogan CAIC
26.0154
29
0.6247
0.0968
0.9235
0.0000
52.0154
103.7130
90.7130
1.0000
Not a bad fit.
89
Thecompletelyconstrainedmodelprovideagoodfitofthedata. Themodelfitchi
squareisnotsignificant.TheRMSEAisperfect,althoughtheSRMSRisnotverygood.
TheAGFIandtheCFIarealsogood.TheAIC,theCAIC,andtheSBCarealsoprinted.
Theseindicescannotbeinterpretedbytheirabsolutevalues,butwillbeusefulwhen
youcomparethefitsofdifferentmultiplegroupmodels.Youwillusetheseindicesto
selectthebest multiplegroupmodelforthedatalater.
89
Fitting Less Restrictive Multiple-Group Models

Completely constrained multiple-group model: Error
variances, structural covariances, and loadings are all
constrained
Release the constraints on error variances
Release the constraints on structural covariances
Release the constraints on the loadings Completely
unconstrained
90
WehavefittedthecompletelyconstrainedmultiplegroupmodelbytheREFMODEL
method.Wecanfurtherfitlessconstrainedmultiplegroupmodelbymodifyingour
PROCCALIScode.
Wecanreleasetheconstraintsonerrorvariances.
Thenwecanreleasetheconstraintsonthestructuralcovariances (amonglatent
variables).
Finally,wecanreleasetheconstraintsonthepathcoefficients (orloadings).
Iamgoingtoshowthesestepbystep.
90
Release the Constraints on Error Variances

Males
Females
v1
v1
1.
Spatial
a2
Visual
1.
Cubes
Spatial
a3
a2
Visual
Cubes
a3
Lozenges
cv
Lozenges
cv
1.
Verbal
a5
Paragraph
Sentence
a6
1.
Verbal
Sentence
a6
WordMean
v2
a5
Paragraph
WordMean
v2
Common parameter names for the error variances are removed.

91
Thepathdiagramsforthemultiplegroupmodelthatreleasestheconstraintsonthe
errorvariancesareshownabove.
Exceptfortheerrorvarianceparameters,alltheremainingparametersarelabeled.This
meansthatonlytheerrorvarianceparametersarenotinvariantacrossthemodelsfor
thegroups.
91
Releasing the Constraints on the Error Variances

proc calis;
path
= 1. ,
Verbal
pvar
/* Visual Cubes Lozenges Paragraph Sentence Wordmean */
Spatial Verbal;
pcov
Spatial Verbal;
refmodel 1;
run;
Comment out the error variance specifications in the PVAR statement, and let PROC CALIS
set two distinct sets of default error variances for the two models.
92
IntermsofPROCCALISspecification,thismeansthatthemodelforfemalesmakes
referencetothemodelformaleswithregardtothoseconstrainedparametersonly.
Thiscouldbedoneveryeasilyfrommodifyingthecompletelyconstrainedmultiple
groupmodel.AllyouneedtodoistocommentoutthePVARstatementspecifications
fortheobservedvariables.
WhenModel2makesreferencetoModel1,onlythoseexplicitspecificationswouldbe
constrainedbetweenthetwomodels.Becausetheerrorvariancesarenotspecifiedin
bothmodels(thatis,commentedoutfromthepreviouscode),PROCCALISwould
generatedifferentsetsofdefaulterrorvarianceparametersfor thetwomodels.In
otherwords,theerrorvarianceconstraintsarereleasedinthis PROCCALISspecification.
92
Fit Summary of the Multiple-Group Model with

Constraints on Loadings and Structural Covariances
Fit Summary
Chi-Square
Chi-Square DF
Pr > Chi-Square
Adjusted GFI (AGFI)
RMSEA Estimate
Bozdogan CAIC
22.0334
23
0.5182
0.0903
0.9163
0.0000
60.0334
135.5913
116.5913
1.0000
93
Themodelfitchisquarestillisnotsignificant,indicatingagoodmodelfit.TheRMSEA,
theAGFI,andtheCFIareallgood.However,theSRMSRdoesnotindicateagoodmodel
fit.
93
Release the Constraints on the Structural

Variances and Covariances
Males
1.
Spatial
a2
Females
Visual
1.
Cubes
Spatial
a3
a2
Verbal
a5
Cubes
a3
Lozenges
1.
Visual
Lozenges
Paragraph
Sentence
a6
1.
Verbal
a5
Paragraph
Sentence
a6
WordMean
WordMean
Common parameter names for the structural variances and covariance are removed.
94
Howaboutreleasingtheconstraintsonthestructuralcovariances?
Inthepathdiagram,onlythepathcoefficientsareconstrainednow(byusingthesame
setofparameternames).Thismeansthatonlythepatheffectsareinvariantacrossthe
modelsforthegroups.
94
Releasing the Constraints on the Error Variances

and Structural Covariances
proc calis;
path
= 1. ,
Verbal
/*
pvar
Visual Cubes Lozenges Paragraph Sentence Wordmean
Spatial Verbal;
pcov
Spatial Verbal;
*/
refmodel 1;
run;
Comment out the PVAR and PCOV statements, and let the PROC CALIS set
two distinct sets of default variances and covariances for the two models.
95
Thisnewmultiplegroupmodelcanbespecifyingbycommentedouttheexplicit
specificationsofthestructuralcovariances (variancesandcovariances amonglatent
variables)inModel1.
WhenModel2makesreferencetoModel1,itcopiestheexplicitspecificationsinthe
PATHstatementofModel1.Errorvariances,structuralvariances andcovariances inthe
twomodelsarenowsetbydefaultandareunconstrainedbetweenthetwomodels.
95
Fit Summary of the Multiple-Group Model with

Loading Constraints
Fit Summary
Chi-Square
Chi-Square DF
Pr > Chi-Square
Adjusted GFI (AGFI)
RMSEA Estimate
Bozdogan CAIC
18.2915
20
0.5682
0.0539
0.9179
0.0000
62.2915
149.7796
127.7796
1.0000
96
Themodelfitchisquareisnotsignificant.Now,theSRMSRisacceptable.TheAGFI,the
RMSEA,andtheCFIcontinuetobeverygood.
96
Release the Constraints on the Loadings

Males
1.
Spatial
Females
Visual
1.
Cubes
Spatial
Cubes
Lozenges
1.
Verbal
Lozenges
Paragraph
Sentence
Visual
1.
Verbal
WordMean
Common parameter names for the loadings are removed.
Paragraph
Sentence
WordMean
97
Finally,forthecompletelyunconstrainedmultiplegroupthepathdiagramsforthetwo
groupsarethesame,butnoparameternames(exceptforfixedvaluesof1)areusedto
denoteconstraints.
97
Completely Unconstrained Multiple-Group Model

proc calis;
path
= 1.
Verbal
---> Paragraph Sentence Wordmean = 1.
path
= 1.
Verbal
---> Paragraph Sentence Wordmean = 1.
run;
,
;
,
;
The REFMODEL statement is not used here because the two models are
not constrained with each other.
98
Becausethetwomodelsforthegroupsaretotallyunrelated,you shouldnotneedto
usetheREFMODELstatementanymore.Instead,thetwomodelsare definedexactlyby
thesamePATHstatementspecifications.However,becausenocommonparameter
namesareusedforthepathcoefficients,thetwomodelsarenot constrained.
98
Fit Summary of the Completely Unconstrained

Multiple-Group Model
Fit Summary
Chi-Square
Chi-Square DF
Pr > Chi-Square
Adjusted GFI (AGFI)
RMSEA Estimate
Bozdogan CAIC
16.4795
16
0.4200
0.0449
0.9077
0.0205
68.4795
171.8746
145.8746
0.9984
The unconstrained SEM model for the groups gives you the best fit,
but it is also the least interesting multiple-group model.
99
Allfitindicesindicateverygoodfitofthecompletelyunconstrainedmultiplegroup
model.
99
Chi-Square Difference Tests for the Nested

Multiple-Group Models
Constrained
Completely
Constrained
Constrained Loadings and
Loadings and
Struct. Cov.
Constrained
Loadings
3.892
Structural Covariances
(p=0.32)
Constrained Loadings
Completely Unconstrained
7.724
3.742
(p=0.44)
(p=0.71)
9.536
5.539
1.182
(p=.27)
(p=.41)
(p=.23)
There are no significant differences between the multiple-group models.
100
Whichmodelisthebestforthedata?
Chisquaredifferencetestsprovideastatisticalmethodtoseeifmodelsaresignificantly
differentfromeachother.Thisslidesshowsthechisquaredifferencetestsfor
comparingthefourmultiplegroupmodels.
Asallpvaluesarebiggerthan0.05,itmeansallthesemultiplegroupmodelsarenot
significantlydifferentfromeachother.
100
Comparing Model Fits by Using Various Fit Indices

Constrained
Completely
Constrained
Chi-Square
Loadings and
Constrained
Completely
Loadings
Unconstrained
16.4795
Struct. Cov.
26.0154
22.0334
18.2915
29
23
20
16
Pr > Chi-Square
0.6247
0.5182
0.5682
0.4200
0.0968
0.0903
0.0539
0.0449
Adjusted GFI (AGFI)
0.9235
0.9163
0.9179
0.9077
Chi-Square DF
RMSEA Estimate
Bozdogan CAIC
0.0000
0.0000
0.0000
0.0205
52.0154
60.0334
62.2915
68.4795
103.7130
135.5913
149.7796
171.8746
90.7130
116.5913
127.7796
145.8746
1.0000
1.0000
1.0000
0.9984
Absolute indices: Chi-square, SRMSR (smaller is better)

Parsimonious indices: AGFI (larger is better),
RMSEA, AIC, CAIC, SBC (smaller is better)
Incremental indices: Bentler CFI (larger is better)
101
Wecanalsocomparethefourmodelsbymeansofthefitindexvalues.
Themodelfitchisquarevaluealwaysfavorsthemodelwiththelargestnumberof
parameters.So,accordingtothemodelfitchisquare,thecompletelyunconstrained
modelisthebestmodel.TheSRMSRalsofavorsthecompletelyunconstrainedmodel
simplybecauseitcanbeviewedasamonotonetransformationofthechisquarevalue.
However,youshouldnotselectyourbestmodelbasedontheabsoluteindicessuchas
modelfitchisquarevalueortheSRMSRbecausetheseindicesdonottakemodel
parsimonyintoaccount.Complicatedmodelsmighthaveperfectmodelfitchisquare
andSRMSRvalues(i.e.,0).Butthesecomplexmodelsshouldnotbeselectedasthebest
modelsbecausetheyhaveverylittlescientificvalue.
TheAGFI,theRMSEA,theAIC,theCAIC,andtheSBCalltakesmodelparsimonyinto
account.FortheAGFI,thelargerthebetter.Forotherindices, thesmallerthebetter.All
theseparsimoniousindicespointtothecompletelyconstrainedmodelasthebest
multiplegroupmodelforthedata.
Lastly,theincrementalfitindexBentler CFIfavorsthecompletelyconstrainedmodeltoo.
However,virtuallyallmultiplegroupmodelinthiscomparisonareequallygood
accordingtotheCFI.NoticethatincrementalindicessuchastheCFImeasureshowa
targetmodelmeasuresbetterthanasocalledbaselinemodel.Theydonottakemodel
parsimonyintoaccount.Inaddition,theydependonhowgoodthe baselinemodelis
usedinthecomputingformula.Ifthebaselinemodelisverybad (suchasthecommonly
useduncorrelatedness model),allcompetingmodelswouldhavegoodincrementalfit
onlybecausethebaselinemodelismuchworse.Forthisreason, incrementalfitindices
mightnotserveasgoodcriteriaformodelselection.
101
Analyzing Direct and

Indirect Effects
102
102
Democracy and Industrialization Data: Direct and

Indirect Effects of Industrialization
Press60
1.
Dem60
Gnppc60
Enpc60
1.
Freop60
lam2
lam3
Fair60
lam4
Legis60
Indust
Press65
Indlf60
Dem65
1.
lam2
Freop65
lam3
Fair65
lam4
Legis65
103
AnalyzingdirectandindirecteffectsissomethinguniquetoSEM.
Letuslookatthemodelforthedemocracyandindustrialization data.Onlythe
structuralpartoftheSEMisshowntoillustratetheidea.
103
Industrialization Effects on Democracy in 1960

and 1965
On democracy in 1960
A direct effect:
Indust --->Dem60
Dem60
On democracy in 1965
A direct effect:
Indust --->Dem65
An indirect effect:
Indust --->Dem60 --->Dem65
Indust
Dem65
Total effect = direct effect + indirect effect

104
First,letuslookattheeffectofindustrializationonthedemocracymeasurein1960.
ThedirecteffectofIndust onDem60referstothepathIndust >Dem60.Thiseffect
canbeestimateddirectlyfromanySEMsoftware.
Onthedemocracymeasurein1965,industrializationhasadirect andindirecteffect.
ThedirecteffectreferstothepathIndust >Dem65.Theindirecteffectisindicatedby
thetrackIndust > Dem60>Dem65.
Whenyouaddupthedirectandindirecteffect,itgivesyouthe totaleffect.
InSEM,thedirecteffectsareestimatedasthepathcoefficients.Indirecteffectsand
totaleffectsarefunctionsoftheparameterestimates.Fortunately,PROCCALIScan
computethesefunctionsefficientlyanditcanalsoprovidestandarderrorestimatesfor
theseeffects.
104
Factors Affecting Mental Abilities: A Model Inspired by

Marjoribanks (1974)
S1
S2
S3
P1
1.
P2
P3
1.
M1
M2
M3
1.
Social Status
Parental
Encouragement
Family Size
Achievement
Motivation
Mental Ability
1.
A1
A2
A3
105
Thisslideshowsamoreinterestingexampleaboutanalyzingdirect,indirect,andtotaleffects.
TheexampleisinspiredbyamodelofMarjoribanks (1974).Thecurrentmodelisasimplification
andthedataaregenerated.Theresultsheredonotrepresenttheoriginalstudy,butwould
servewellforillustrationpurposes.
Themainideaofthestudyistomodelthementalabilityofstudents.Thementalabilityisa
latentconstruct,whichissupposedtobedetermined(predicted) byparentalencouragement
andachievementmotivation,bothofwhichareformulatedaslatentconstructinthemodel.
Tworemotecauses(predictors),socialstatusandfamilysize,havedirecteffectsonparental
encouragementandachievementmotivation.However,thesetworemotecausesaffectthe
mentalabilityonlyindirectly.Socialstatusisalsoformulated asalatentvariable,whilefamily
sizeisanobservedvariable.Forallthelatentvariables,observedindicatorsareusedandthey
arerepresentedbysmallrectanglesinthepathdiagram.
Therearesomemotivatingquestionsaboutthispathdiagramregardingthedirectandindirect
effects.Forexample,
1.Eventhoughsocialstatusdoesnotaffectthementalability,it doeshaveanindirecteffecton
thementalabilityviaparentalencouragementandachievementmotivation.Onewouldlikethe
SEMsoftwaretocomputethisthisindirecteffectanditssignificance.
2.Parentalencouragementhasadirectandanindirecteffectsonthementalability.Whatisthe
overalltotaleffectofparentalencouragementonthementalability.Onewouldalsolikethe
SEMsoftwaretocomputealltheseeffectsandtheirsignificance.
105
Factors Affecting Mental Abilities: PROC CALIS Code

proc calis data=mental nobs=115 effpart;
path
/* Structural Model */
SocialStatus --->
The EFFPART option analyzes

the effect partitioning in the
model.
ParentalEncouragement FamilySize
AchievementMotivation,
FamilySize
--->
ParentalEncouragement
--->
AchievementMotivation MentalAbility,
AchievementMotivation
--->
MentalAbility,
/* Measurement Model */
SocialStatus
---> S1 S2 S3
= 1.,
ParentalEncouragement ---> P1 P2 P3
= 1.,
AchievementMotivation ---> A1 A2 A3
= 1.,
MentalAbility
= 1.;
---> M1 M2 M3
fitindex on(only)=[agfi srmsr rmsea bentlercfi] noindextype;

run;
106
Now,thePATHspecificationforthetargetmodelshouldbeeasyforyou.Youcan
specifythemeasurementmodelandthestructuralmodelbythemultiplepathsyntax.
Youcanlookatthepathdiagramandwritedownthepathsinthe PATHstatement.
Noticethateachpathinthepathdiagramrepresentsadirecteffectofonevariableon
anothervariable.
TheonlynewoptionintroducedhereistheEFFPARToptioninthe PROCCALIS
statement.EFFPARTstandsforeffectpartitioning.Inotherwords,itpartitionsthetotal
effectsofanyvariableonanyothervariableintodirectandindirecteffects.PROCCALIS
willcomputetheseeffectsandthestandardizedversionallwithstandarderror
estimatesprovided.
106
Fit Summary
Fit Summary
0.0936
Adjusted GFI (AGFI)
0.7341
RMSEA Estimate
0.1431
0.8087
Not a very good model fit.
107
Themodelfitactuallydoesnotlooktoogoodforthissimulated data.Butthisisnotthe
concernhere.Wewanttostudytheeffectpartitioning,assuming thatwearesatisfied
withthemodelfit.
107
Factors Affecting Mental Abilities: Estimation Results

S1
S2
1.
1.20**
S3
P1
1. .58**
.75**
Social Status
-.29*
Family Size
P2
.25**
P3
1.47**
Parental
Encouragement
.22*
1.57**
1.
A1
1.18**
A2
M2
1.
-1.73*
2.11**
M3
2.07**
Mental Ability
1.32**
Achievement
Motivation
-.13**
M1
Negative effect of
parental
encouragement on the
mental ability!
.60**
A3
* Marginally significant
** Significant
~ Fixed
108
Beforedivingintotheresultsfortheeffectpartitioning,Iwanttolookattheestimates
showninthepathdiagram.Iwanttothrowinonemoremotivationtostudydirectand
indirecteffectsinastructuralequationmodel.
Inthispathdiagram,estimatesareshownwiththeirsignificancemarked.Twoasterisks
afteranestimatemeanstheestimateisstatisticallysignificant.Oneasteriskafteran
estimatemeansthattheestimateismarginallysignificant.
Iwanttofocusontheeffectsofparentalencouragementonmentalability.Thedirect
effectis1.73.Thismeansthatparentalencouragementhasanegativeeffectonmental
ability.Thissoundsalittlestrangeatthefirstglance.Butifwelookatthebiggerpicture
inthepathdiagram,wecanunderstandwhythatisso.Noticethatparental
encouragementhasapositiveeffectonachievementmotivation,whichinturnshasa
positiveeffectonmentalability.Thewholepicturesuggeststhatpurelyparental
encouragementdonotnecessarilyaffectmentalabilityinapositiveway.Sometimes,
themoreencouragementwouldonlyaddmorepressuretotheindividualsmental
performancehencethenegativedirecteffectonmentalabilityobservedinthepath
diagramresult.However,whentheparentalencouragementcanaffectsomethingmore
internaloftheindividualsnamely,theindividualsachievementmotivation,thenitwill
resultinahighermentalabilityscore.Hence,thereisapositiveindirecteffectof
parentalencouragementonthementalability.
Insum,aninterestingquestioninthispathdiagramresultisthatwhatistheoverall
totaleffectofparentalencouragementonmentalability,giventhatithasanegative
directeffectandapositiveindirecteffect?
108
Partitioning of the Effects: A Prerequisite
Stability Coefficient of Reciprocal Causation = 0
Stability Coefficient < 1
Total and Indirect Effects Converge
NOTE: The stability coefficient is 0, which is less

than one. The condition for converged total and
indirect effects is satisfied.
109
Beforeyoucananalyzedirectandindirecteffects,youshouldcheckwhethera
prerequisiteissatisfied.Inordertostudytheeffectpartitioninglegitimately,theso
calledstabilitycoefficientmustbelessthen1.PROCCALISprovidessuchacheck.When
youseethesemessagesfromthePROCCALISoutput,youcouldproceedtoexamine
youreffectpartitioningresults.
109
Partitioning of the Effects: Total Effects

Total Effects
Effect / Std Error / t Value / p Value
FamilySize
Achievement
Motivation
Mental
Ability
Parental
Encouragement
SocialStatus
A1
-0.1287
0.0360
-3.5769
0.000348
1.0000
1.5701
0.5176
3.0337
0.002416
0.6523
0.0942
6.9258
<.0001
A2
-0.1522
0.0420
-3.6274
0.000286
.
.
.
.
.
.
.
.
-0.1699
0.0572
-2.9696
0.002982
1.1821
0.1084
10.9012
<.0001
.
.
.
.
.
.
.
.
1.3196
0.4124
3.1995
0.001377
1.8560
0.6051
3.0672
0.002161
.
.
.
.
.
.
.
.
0.3376
0.4045
0.8346
0.4039
0.7710
0.1036
7.4398
<.0001
.
.
.
.
.
.
.
.
0.4244
0.1280
3.3159
0.000914
0.2516
0.0692
3.6356
0.000277
.
.
.
.
.
.
.
.
MentalAbility
.
.
.
.
.
.
.
.
0
Details
omitted.
110
WiththeEFFPARToption,PROCCALISproducestablesfortotal,direct,andindirect
effectsseparately.Thesetablescouldbelarge.Ijustannotate theseresultshere.Some
resultsarenotshown.
Thistableisabouttheestimatesofthetotaleffects,theirstandarderrors,tvalues,and
significancelevels.
110
Partitioning of the Effects: Direct Effects

Direct Effects
FamilySize
Achievement
Motivation
Mental
Ability
Parental
Encouragement
SocialStatus
A1
1.0000
A2
1.1821
0.1084
10.9012
<.0001
.
.
.
.
.
.
.
.
1.3196
0.4124
3.1995
0.001377
.
.
.
.
.
.
.
.
0
.
.
.
.
.
.
.
.
-1.7343
0.9047
-1.9169
0.0553
.
.
.
.
.
.
.
.
MentalAbility
.
.
.
.
.
.
.
.
0
.
.
.
.
.
.
.
.
Details
omitted.
0.2516
0.0692
3.6356
0.000277
111
Thistableisaboutthedirecteffects,theirstandarderrors,tvalues,andsignificance
levels.
111
Partitioning of the Effects: Indirect Effects

Indirect Effects
FamilySize
Achievement
Motivation
Mental
Ability
Parental
Encouragement
SocialStatus
A1
-0.1287
0.0360
-3.5769
0.000348
1.5701
0.5176
3.0337
0.002416
0.6523
0.0942
6.9258
<.0001
A2
-0.1522
0.0420
-3.6274
0.000286
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
1.8560
0.6051
3.0672
0.002161
.
.
.
.
.
.
.
.
0.7710
0.1036
7.4398
<.0001
.
.
.
.
.
.
.
.
-0.1699
0.0572
-2.9696
0.002982
2.0719
1.0483
1.9763
0.0481
0.4244
0.1280
3.3159
0.000914
.
.
.
.
.
.
.
.
MentalAbility
Details
omitted.
112
Thistableisabouttheindirecteffects,theirstandarderrors,tvalues,andsignificance
levels.
112
Customized Effect Analysis

The EFFPART option displays all logical possible effects
of the variables
Columns: Five variables, each of which serves as a
predictor at least once:
o FamilySize
o AchivementMotivation
o MentalAbility
o ParentalEncouragement
o SocialStatus
Rows: Sixteen variables, each of which serves as an

outcome variable at least once (all variables except for
SocialStatus)
113
Whenyouhavelargetableslikethoseshowninpreviousslides,youarelikelytobe
doingexploratoryanalysiswithoutspecificquestionsinyourmind.Theeffecttables
couldgetverylargeandyoumighthaveadifficulttimetolook fortheparticularresults
thatyouareinterestedin.Forexample,thecolumnsoftheeffecttablesconsistoffive
variables,eachofwhichservesasapredictoratleastonceinthepathdiagram.These
fivevariableshavedirectorindirecteffectsontherowvariables.Therowsconsistof
sixteenvariables,eachofwhichservesasanoutcomevariableatleastonce.Inthe
currentpathdiagram,itincludesallvariablesexceptfortheSocialStatusvariable.
However,ifyouhavespecificresearchquestionsinyourmind,youarerecommendedto
docustomizedeffectanalysisthatissupportedbyPROCCALIS.
113
Customized Effect Analysis on the Structural Model

What are the total,
direct, and indirect
effects of this remote
cause on other latent
constructs?
.25**
Social Status
-.29*
Family Size
.22*
-.13**
What is the total effect

of parental
encouragement on the
mental ability?
Parental
Encouragement
1.57**
-1.73*
Mental Ability
1.32**
Achievement
Motivation
114
Inthebeginning,wealreadyhavethesemotivatingquestions.
1.Forthesocialstatusvariable,Whatarethetotal,direct,andindirecteffectsofthis
remotecauseonotherlatentconstructs,especiallyonthementalability?
2.Whatisthetotaleffectofparentalencouragementonthemental ability?
114
Factors Affecting Mental Abilities: Customized

Effect Analysis
proc calis data=mental nobs=115;
path
/* Structural Model */
SocialStatus -->
FamilySize
--->
ParentalEncouragement FamilySize AchievementMotivation,

--->
--->
MentalAbility,
/* Measurement Model */
SocialStatus
---> S1 S2 S3
= 1.,
= 1.,
= 1.,
MentalAbility
= 1.;
---> M1 M2 M3
The EFFPART
statement defines the
customized effects of
interest.
effpart
SocialStatus
-> ParentalEncouragement AchievementMotivation

MentalAbility,
ParentalEncouragement -> MentalAbility;

run;
115
PROCCALISsupportsthecustomizedeffectanalysis.ThiscanbedonebytheEFFPART
statement,asshowninthePROCCALIScodeinthisslide.
First,youwanttostudytheeffectpartitioningofsocialstatusonthesethreevariables:
parentalencouragement,achievementmotivation,andmentalability.Hence,youusethe
followingcodeintheEFFPARTstatement:
SocialStatus -> ParentalEncouragement AchievementMotivation
MentalAbility,
Second,youwanttostudytheeffectpartitioningofparentalencouragementonmental
ability.Hence,youusethefollowingcodeintheEFFPARTstatement:
ParentalEncouragement -> MentalAbility;
115
Effects of Social Status

Effects of SocialStatus
Total
Direct
Indirect
0.2516
0.2516
Direct effect
only
0.0692
0.0692
3.6356
3.6356
0.000277
0.000277
0.6523
0.2193
0.4330
Direct and indirect

effects
0.0942
0.1147
0.1203
6.9258
1.9125
3.5985
<.0001
0.0558
0.000320
0.4244
0.4244
MentalAbility
Indirect effect
only
0.1280
0.1280
3.3159
3.3159
0.000914
0.000914
116
TheeffectpartitioningresultsfromPROCCALISareshowninthisslideandthenextone.
Theeffectsofsocialstatusonthethreespecificlatentvariablesareshowninthistable.
Ontheparentalencouragement,socialstatushasadirecteffect only,whichispositive
andsignificant.
Ontheachievementmotivation,socialstatushasbothadirectandanindirecteffects.
Bothoftheseeffectsaresignificant.Thetotaleffectisthesumofthedirectandindirect
effect.Thetotaleffectisalsosignificant.
Onthementalability,socialstatushasonlyanindirecteffect,whichisalsosignificant.
116
Effects of Parental Encouragement on Mental Ability
Effects of ParentalEncouragement
MentalAbility
Total
Direct
Indirect
0.3376
-1.7343
2.0719
0.4045
0.9047
1.0483
0.8346
-1.9169
1.9763
0.4039
0.0553
0.0481
Parental encouragement has a positive total effect on

the mental ability, although the total effect is not
significant.
117
Thisslideshowstheeffectpartitioningofparentalencouragementonmentalability.
Thedirecteffectisnegative,asshownpreviouslyinthepathdiagram.Thisdirecteffect
ismarginallysignificant.
Theindirecteffectispositiveandisstatisticalsignificant.Thisisapieceofcomforting
informationparentalencouragementdoesaffectthementalabilitypositively,butonly
throughitseffectonachievementmotivation.
Thetotaleffect,whichisthesumofdirectandindirecteffect,however,isnotsignificant.
ThisexampleshowsthatSEMeffectanalysiscanshowsomeeffect patternsthatsimply
cannotbeanalyzedbylinearregressionanalysisadequately.The SEMeffectanalysis
providessomethingmoredetailedandrefinedregardingthetotalityofthetheory.In
thisregard,thecustomizedeffectanalysissupportedbyPROCCALISisveryuseful.
117
Standardized Effects of Social Status

Standardized Effects of SocialStatus
MentalAbility
Total
Direct
Indirect
0.6675
0.6675
0.0833
0.0833
8.0141
8.0141
<.0001
<.0001
0.6990
0.2350
0.0597
0.1207
0.1153
11.7105
1.9478
4.0231
<.0001
0.0514
<.0001
0.4960
0.4960
0.4640
0.0850
0.0850
5.8374
5.8374
<.0001
<.0001
118
PROCCALISalsoprovidesthestandardizedresultsforeffectanalysis.Standarderrors,t
values,andpvaluesarealsocomputedforthestandardizedeffectestimates.
118
Standardized Effects of Parental Encouragement

on Mental Ability
Standardized Effects of ParentalEncouragement
MentalAbility
Total
Direct
Indirect
0.1487
-0.7639
0.9126
0.1705
0.3096
0.3487
0.8722
-2.4675
2.6171
0.3831
0.0136
0.008869
Parental encouragement has a positive

standardized total effect on the mental ability,
although the standardized total effect is not
significant.
119
Thisslideshowsthestandardizedeffectsofparentalencouragementonmentalability.
Thepatternisquitesimilartotheunstandardizedversion.
119
Testing Specific
Hypotheses
120
120
Testing Hypotheses about the Measurement Model

v1
v2
S1
v3
S2
1.
g2
S3
P1
g3
P2
P3
1.
M1
M2
1.
Social Status
Parental
Encouragement
Family Size
Achievement
Motivation
r2
M3
r3
Mental Ability
1.
A1
A2
A3
121
Testingspecifichypothesesisaninterestingtopic.Herewelookatsomeexamples.
Inthementalabilitymodel,youhavesomeindicatorvariablesforthelatentvariables.
Twolatentvariablesareselectedtoillustratethetestingofspecifichypotheses.
Forthementalabilityfactor,onemightwanttotestthehypothesisthattheloadings
(pathcoefficients)arethesamefortheM2andM3indicators.Inthepathdiagram,r2
andr3arelabeledasthepathcoefficients.Youwanttotestwhetherr2andr3areequal
withinthemodel.
Forthesocialstatusfactor,younotonlywanttotestthehypothesisthattheloadings
(pathcoefficients)arethesameforthethreeindicators,butyoualsowanttoseeiftheir
correspondingerrorvariancesarethesameinthepopulation.In thepathdiagram,g2,
g3,v1,v2,andv3areparametersofinterest.Youwanttotestsimultaneouslywhether
g2,g3areequalto1andv1,v2,andv3arethesameinthepopulation.
121
Specific Hypotheses
Parallel items for measuring SocialStatus
H1 : g2 = 1
H2: g3 = 1
H3: v1 = v2
H4: v2 = v3
Equality of loadings for MentalAbility items M2 and M3

H5 : r2 = r3
Sum of the loadings for M2 and M3 is two times as

much as the sum of the loadings for S1 and S2
H6: (r2 + r3) / (g2 + g3) = 2
122
Thetestofequalloadingsandequalerrorvariancesforthesocialstatusitemsisatest
ofparallelitems.Thiscouldbestatedmoreformallyasthefollowingfourcomponent
hypothesesH1,H2,H3,andH4,asshownintheslide.Thesefour hypothesesneedtobe
testedsimultaneously.Rejectionofthesimultaneoustestmeanstheitemsarenot
parallel.
Thetestofequalloadingsforthemeasurementindicatorsofthe mentalabilityfactoris
simpler.ItisstatedinH5.RejectionofH5meansthatr2andr3arenotequalinthe
population.
Finally,youcaninventanystrangehypothesisthatcanbeexpressedasacontinuous
functionofthemodelparameters.Forexample,H6statesthattheratioofthesumofr2
andr3tothesumofg2andg3is2.Thishypothesismayormaynotmakesense.Butitis
includedheretodemonstratetheflexibilityofPROCCALIS.
122
PROC CALIS Hypotheses Testing: h() = 0

Parallel items for measuring SocialStatus:
H1 : h1 = g2 1 = 0
H2: h2 = g3 1 = 0
H3 : h3 = v1 v2 = 0
H4: h4 = v2 v3 = 0
Equality of loadings for MentalAbility items M2 and M3

H5 : h5 = r2 r3 = 0
Sum of the loadings for M2 and M3 is two times as

much as the sum of the loadings for S1 and S2
H6: h6 =2(g2 + g3) (r2 + r3) = 0
123
BeforeIshowyouthePROCCALIScode,itisusefultoreformulatethehypothesesinto
theformsthatmatchthePROCCALISinput.
PROCCALIStestshypothesesoftheformh()=0,whereh()isanycontinuousfunction
ofthemodelparameters(forexample,theerrorvariancesandthepathcoefficientsin
themodel).
Thehypothesesinthepreviousslidecouldallberewritteninthisrequiredform,as
showninthisslide.Withtheseforms,youarereadytospecifythosehypothesesin
PROCCALIS.
123
Testing Specific Hypotheses about the Measurement

Model Using PROC CALIS
path
SocialStatus ---> ParentalEncouragement FamilySize AchievementMotivation,
FamilySize
---> AchievementMotivation,
ParentalEncouragement ---> AchievementMotivation MentalAbility,
AchievementMotivation ---> MentalAbility,
SocialStatus
---> S1 S2 S3
= 1. g2 g3,
Specify g2, g3, r2, r3, v1, v2,
= 1.,
and v3 explicitly.
= 1.,
MentalAbility
---> M1 M2 M3
= 1. r2 r3;
Use the SIMTEST statement
pvar S1-S3 = v1-v3;
to test simultaneously
simtest parallel_social_items=[h1 h2 h3 h4];
hypotheses. Use the
testfunc h5_equal_load_m2_m3 h6_proportional_sum;
TESTFUNC statement to test
h1 = g2 - 1;
individual hypotheses.
h2 = g3 - 1;
Use the SAS programming
h3 = v1 - v2;
statements to define the
h4 = v2 - v3;
parametric functions in the
tests.
h5_equal_load_m2_m3 = r2 - r3;
h6_proportional_sum = 2*(g2 + g3) - (r2 + r3);
run;
124
First,youhavetolabelornametheparametersinthecorrectlocationsofthemodel
specification.Forexample,g2andg3arethepathcoefficientsforS2andS3,respectively;andr2
andr3arethepathcoefficientsforM2andM3,respectively.Noticethatyoudidnotnamethese
parametersintheprecedingmodelspecifications.Namingtheseparameterswereoptional
becauseyoudidnotreferencethem.However,becauseyouaregoingtorefertothese
parametersinthehypothesistesting,youmustnameorlabelthemintherespectivelocations
now.Similarly,theerrorvariancesforS1S3arenamedasv1v3,asshowninthePVAR
statement.
ThemaintoolsfortestingspecifichypothesisinPROCCALISare theSIMTESTSandtheTESTFUNC
statements.
TheSIMTESTstatementenablesyoutotestsimultaneoushypothesesliketheparallelhypothesis
withfourcomponenthypotheses.Herewehaveh1,h2,h3,andh4, allofwhicharetreatedjust
asthenamesofthehypothesesthataredefinedlater.
TheTESTFUNCstatementenablesyoutotestindividualhypotheses liketheequalityofloadings
andtheproportionalityhypothesesdescribedpreviously.HereIuselongnamessuchas
h5_equal_load_m2_m3andh6_proportional_sumtoremindmeofthenatureofthetarget
hypotheses.
NowIusethesocalledSASprogrammingstatementstodefinethehypotheses:h1h4,
h5_equal_load_m2_m3,andh6_proportional_sum.TheSASprogrammingstatementsarejust
likecommonmathematicalequations.ThesesixSASprogrammingstatementsdefinethe
parametricfunctionsinthetargethypotheses.PROCCALIStestsallparametricfunctions
equalingzero.
124
Individual Tests of Parametric Functions:

TESTFUNC Results
Tests for Parametric Functions

Parametric
Function
h5_equal_load_m2_m3
h6_proportional_sum
Value
Standard
Error
t Value
p Value
0.04147
-0.27995
0.24290
1.02816
0.17072
-0.27228
0.8644
0.7854
Both individual hypotheses are supported.
125
TheTESTFUNCspecificationproducestheresultsshowninthistable.
YoufailtorejecttheequalityofloadingsforM2andM3becausethepvalueisbigger
than0.05.So,theequalityoftheloadingsissupported.
Youalsofailtorejecttheproportionalsumhypothesis(pvalue=0.79).
125
Tests for Parallel Social Status Items:

SIMTESTS Results
Simultaneous Tests
Simultaneous
Test
Parametric
Function
Function
Value
DF
Chi-Square
p Value
h1
h2
h3
h4
0.19873
-0.25004
-0.24067
-0.64108
4
1
1
1
1
24.23862
3.57013
8.37521
0.17774
1.49312
<.0001
0.0588
0.0038
0.6733
0.2217
parallel_social_items
Overall parallelism hypothesis is not supported for the SocialStatus items, although
the equality of error variances is supported.
126
TheSIMTESTSstatementspecificationproducestheoutputshowninthistable.
Fortheparallelhypothesis,thesimultaneoustestisrejected(p<.0001).Theparallel
itemhypothesisisnotsupported.
PROCCALISalsoprovidesindividualtestsforthecomponenthypotheses.Thiswouldbe
usefulfordoinganadhocanalysistoprobewhatfailsthesimultaneoushypothesis.For
example,bothh1andh2areatleastmarginallysignificant.But h3andh4arenot
significant.Recallthath1andh2areabouttheequalityofthe loadings(pathcoefficients)
whileh3andh4areabouttheequalityoferrorvariances.Thecurrentresultsshowthat
theitemsmighthavethesameerrorvariancesbutnotthesameloadingsinthe
population.
126
Model Modifications
127
127
When the Model Does Not Fit Well

Fit Summary
Chi-Square
Chi-Square DF
Pr > Chi-Square
Adjusted GFI (AGFI)
RMSEA Estimate
196.7455
59
<.0001
0.0936
0.7341
0.1431
0.8087
Large SRMSR and RMSEA

Small AGFI and CFI
Model modification: suggests ways to improve the model
fit
Lagrange multiplier (LM) tests: which parameters you can
add to significantly decrease the model fit chi-square value
128
Thementalabilitymodeldidnotfitwell.TheSRMSRandtheRMSEAarelarge,whilethe
AGFIandtheCFIaresmall.Whenyouencounterabadmodelfit, itwouldjeopardize
yourinterpretationsofthemodelparameters,effectanalysis,hypothesistesting,and
etc.
Modelmodificationisastatisticaltechniquethatsuggestsways toimproveyourmodel
fit.Themostcommonmodelmodificationtechniqueisdonethroughthesocalled
Lagrangemultiplier(LM)tests.Essentially,theLMtestssuggestwhichparametersyou
couldaddtothemodeltosignificantlylowerthemodelfitchisquarevalue.Whenthe
modelfitchisquareislowered,mostotherfitindices(butnotall,especiallythose
parsimoniousindicesthattakemodelcomplexityintoaccount)mightalsoimprove.
128
Using the MODIFICATION Option
proc calis data=mental nobs=115 modification;

path
SocialStatus --->
ParentalEncouragement FamilySize
FamilySize
--->
--->
--->
MentalAbility,
SocialStatus
---> S1 S2 S3
= 1. ,
---> P1 P2 P3
= 1.,
---> A1 A2 A3
= 1.,
MentalAbility
---> M1 M2 M3
= 1.;
run;
129
TheoptionyoucanusetodomodelmodificationinPROCCALISis theMODIFICATION
optioninthePROCCALISstatement.YoucansimplyaddthisoptiontothePROCCALIS
statementwhenyourunyourmodel.ThisexampleshowsthattheLMtestsformodel
modificationisrequestedfortheoriginalmentalabilitymodel, whichdoesnothavea
verygoodmodelfit.
129
LM Tests for Paths

Rank Order of the 10 Largest LM Stat for Path Relations
Parm
To
From
LM Stat
Pr > ChiSq
Change
P2
P1
P1
56.19414
<.0001
-0.73639
P2
56.19349
<.0001
-0.72904
A2
M2
19.17647
<.0001
0.22842
A2
18.57947
<.0001
-2.31463
A2
MentalAbility
17.20340
<.0001
0.95581
A1
17.04464
<.0001
0.27042
A1
15.86099
<.0001
1.88904
FamilySize
A2
14.43548
0.0001
-1.14590
A1
MentalAbility
13.88705
0.0002
-0.75314
A2
P3
12.96818
0.0003
-0.57151
Adding the P2 <--- P1 (or P1 <--- P2) path reduces your model fit chi-square by 56 approximately.
Adding the A2 <--- M2 path reduces your model fit chi-square by 19 approximately.
130
PROCCALISoutputseveraltablesfortheLMtests.Theresultsareshownindifferent
tables,accordingthetypeoftheparameters.ThistableshowstherankingofLM
statisticsforaddingthe(singleheaded)pathsintothementalabilitymodel.Itgivesyou
thetenpathsthatcanimprovethemodelfitchisquarestatisticthemost.
Thetoponeisthep1>p2path.TheLMstatistic56.19meansthatifyouincludethis
pathintothemodel,youcanexpecttoreducethemodelfitchisquarebyabout56.This
isasubstantialimprovementbecauseyoucangetthisbigimprovementbyjustlosing
onedegreeoffreedom.Thesecondoneisthep2>p1path.Essentially,thiswillgive
thesameamountofmodelimprovementasthefirstpath.Thethirdoneisnotthat
dramatic,butstillgiveyouasubstantialimprovement.AddingtheM2>A2path
reducesthemodelfitchisquareby19.
Doyouwanttoaddthesepathsintoyourmodel?Letusdiscussthisafterweexamine
moreresultsabouttheLMtests.
130
LM Tests for Error Variances and Covariances

Rank Order of the 10 Largest LM Stat for Error Variances and Covariances
Error
Error
of
of
LM Stat
Pr > ChiSq
Change
Parm
P2
P1
56.19312
<.0001
-1.96473
A1
12.26622
0.0005
0.46050
A2
12.08031
0.0005
-0.48351
FamilySize
A2
11.22650
0.0008
-1.88205
M2
A2
10.26408
0.0014
1.55895
S2
S1
7.78117
0.0053
1.55314
MentalAbility
A2
7.48800
0.0062
0.78161
A1
6.95709
0.0083
-0.52904
P2
A3
6.54315
0.0105
0.76007
A3
A2
6.21429
0.0127
-0.67173
Adding the error covariance (P2 <---> P1) reduces your model fit chi-square by
56 approximately.
131
ThistableshowstheLMtests(statistics)fortheerrorvariancesandcovariances.Onthe
topofthelististhecovariancebetweentheerrorsofP2andP1.Addingthecovariance
betweentheerrorsofthesetwovariablesreducesthemodelfitchisquarestatisticby
56.ThisisactuallythesameimprovementthatwehaveseenforaddingeithertheP2
>P1or
P1>P2path.Thenextoneinthelisthasamuchlessimprovement. TheLMstatisticis
only12.26.
Forthisparticularlymodel,thesetwotablesareallthatPROCCALISproducesfortheLM
statistics.Thequestionnowiswhichparameterorparametersyouwanttoaddtothe
model.ThiscouldnotbeansweredbyjustlookingattheLMstatistics.Butitmightalso
involvesomejudgmentabouthowreasonabletheaddedparametersare.Dothese
addedparametersrenderyourmodeluninterpretable,orevencontradictorytoyour
theoreticalclaims,despitethefactthattheyimproveyourmodelfitsubstantially?
131
Notes on the LM Statistics

Chi-square reductions are linear approximations
Chi-square reductions are not additive
Modifications suggested might not be substantively
meaningful
132
Beforegivingananswertothecurrentmodelmodificationanalysis,someimportant
generalpointsabouttheLMstatisticsarediscussed.
First,themodelfitchisquarereductionsasindicatedbytheLMteststatisticsareonly
linearapproximations.Thismeansthatifyouactuallyrefitthe modelbyaddingthe
suggestedparameter,theactualchisquarereductionmightbemoreorless.
Second,thechisquarereductionsassuggestedbytheLMteststatisticsarenot additive.
Thatmeansthatyoucannotaddtwoormoreparametersintothemodelandexpectthe
actualreductioninthenewmodelisexactlythesumofthecorrespondingLMstatistics.
Usually,theactualreductionwouldbesmaller(althoughitcouldbelarger).
Lastbutnotleast,modificationsuggestedbytheLMstatisticsmightnotbe
substantivelymeaningful.
Allthesethreepointsareimportantindecidingwhichparameter youwanttoaddtothe
currentmentalabilitymodelforimprovingthemodelfit.
132
Adding Error Covariance between P1 and P2

S1
S2
S3
P1
1.
P2
P3
1.
M1
M2
M3
1.
Social Status
Parental
Encouragement
Family Size
Achievement
Motivation
Mental Ability
1.
A1
A2
A3
133
ConsideringthetopsuggestionsfromtheresultsoftheLMteststatistics,Iwouldadd
thecovariancebetweenP1andP2.Theaddedparameterisshowninthepathdiagram.
Basically,allthetopLMsuggestions theP2>P1andP1>P2paths,andtheerror
covariancebetweenP1andP2arejustdifferentmanifestationsofthesamelackoffit
aboutacovarianceelementintheoriginalmodel.Thatis,thecovariancebetweenP1
andP2wasnotwellexplainedbytheoriginalmodel.Addingeitherofthesewillleadto
abetterfittingofthecovariancebetweenP1andP2.Inaddition,addingeitherofthese
willgiveyouanapproximatemodelfitchisquareimprovementof56.But,youwould
notgetthreetimesofthisamountbyaddingallthesethree.In fact,ifyouweretoadd
allthesethreeparameters,itisverylikelythatyourmodelis notidentified,meaning
thatyouwouldnotgetuniqueestimates.
Amongthetopthreechoices,theerrorcovarianceischosenbecausetheinterpretation
ofaddederrorcovarianceisalittlecleaner. P1andP2aremeasurementindicatorsof
thesamefactor(ParentalEncouragement).Theerrorcovarianceinterpretationisthat
thesetwoindicatorshavesomesortofcorrelationthatisunexplainedbytheircommon
factor.Theaddederrorcovariancerepresentsthecovarianceexplainedbysome
unknownsources.However,ifIweretoaddeithertheP1>P2orP2>P1paths,it
wouldcreatesomeconflictswithpurportedcommonfactorstructureforthetwo
indicatorvariables.
Notethatthecurrentconclusionisbasedonaverygeneralargumentthataimsat
preservingtheoriginalfactorvariablestructure.Itisnotauniversalprinciple.Inpractice,
youhavetoalsoconsiderthesubstantivegroundsoftheaddedparameters.
133
Adding Covariance between the Errors of P1 and P2
proc calis data=mental nobs=115 modification;

path
SocialStatus ---> ParentalEncouragement FamilySize
FamilySize
---> AchievementMotivation MentalAbility,
---> MentalAbility,
SocialStatus
---> S1 S2 S3
= 1. ,
---> P1 P2 P3
= 1.,
---> A1 A2 A3
= 1.,
MentalAbility
---> M1 M2 M3
= 1.;
pcov P1 P2;
run;
134
NowthatIhavedecidedtoaddthecovariancebetweenP1andP2, Irefitthemodelby
addingthePCOVstatementspecificationforthetwovariables,asshownintheSAS
codeinthisslide.IalsousetheMODIFICATIONoptiononemoretimetoseeifthere
couldbeanyfurthersuggestedimprovements.
134
Before and After Adding the Error Covariance

between P1 and P2
Before
After
Fit Summary
Chi-Square
Chi-Square DF
Fit Summary
196.7455
59
Chi-Square
110.6388
Chi-Square DF
58
Pr > Chi-Square
<.0001
Pr > Chi-Square
<.0001
0.0936
0.0661
Adjusted GFI (AGFI)
0.7341
Adjusted GFI (AGFI)
0.8062
RMSEA Estimate
0.1431
RMSEA Estimate
0.0892
0.8087
0.9269
Improve the model fit chi-square a

lot more than 56.
135
BeforeyouaddtheerrorcovariancebetweenP1andP2,yourmodelfitchisquarewas
about197.Afteraddingthecovariance,themodelfitchisquareisabout111.This
improvementisactuallylargerthanwhattheLMstatisticsuggested,whichwas56.
Otherfitindicesalsoimprove.TheSRMSRandtheRMSEAarenowclosetobe
acceptable.TheAGFIandtheCFIareboostedtohigherlevels.
135
A New Set of LM Tests for Paths

Rank Order of the 10 Largest LM Stat for Path Relations
Parm
To
From
LM Stat
Pr > ChiSq
A2
A2
Change
M2
23.93741
<.0001
0.23830
22.22260
<.0001
-2.30398
A2
MentalAbility
21.70998
<.0001
0.90790
A1
19.34281
<.0001
1.96513
A1
MentalAbility
18.10545
<.0001
-0.75752
A1
17.30632
<.0001
0.32504
A1
M2
15.06992
0.0001
-0.17617
A2
FamilySize
15.06786
0.0001
-0.17675
FamilySize
A2
14.29265
0.0002
-0.89658
A1
11.75569
0.0006
-0.35962
Adding the A2 <--- M2 path now reduces your model fit chi-square
by 24 (was 19 before adding the error covariance).
136
ThenewsetofLMtestsforpathssuggeststheadditionoftheM2>A2path.TheLM
statisticisabout24.Ifyoucomparethisresultwiththefirst LMresultsregardingthe
samepath,younoticethattheLMstatisticschangesasthefittedmodelchanges.
Previously,thesamepathhadanLMstatisticof19.Thisillustratesthenonlinearityand
nonadditivityoftheLMstatistics.
136
A New Set of LM Tests for Error Variances

and Covariances
Rank Order of the 10 Largest LM Stat for Error Variances and Covariances
Error
Error
of
of
LM Stat
Pr > ChiSq
Change
Parm
A2
14.11717
0.0002
-0.53966
A1
14.00823
0.0002
0.51242
FamilySize
A2
13.96827
0.0002
-1.98162
M2
A2
11.86015
0.0006
1.65462
A1
11.75570
0.0006
-0.57402
MentalAbility
A2
10.30441
0.0013
0.86542
P1
A1
9.02987
0.0027
0.53357
S2
S1
8.94243
0.0028
1.66876
MentalAbility
FamilySize
8.05365
0.0045
2.49300
MentalAbility
6.94045
0.0084
0.98947
137
ThereisalsoanewsetofLMtestsforaddingerrorcovariances.
Youmightwanttoimproveyourmodelfurtherbyaddingsomeparametersfromthese
twoLMtables,althoughIwillnotattempttodomorehere.
137
Customized LM Tests
Principled modification process
Restrict the set of parameters of interest for
the LM tests
138
ModelmodificationbyusingtheMODIFICATIONoptioniskindofblindsearch
procedurethatyoutrytoimproveyourmodelwithoutanydefinitedirections.As
discussedbefore,theLMteststatisticsmightnotgiveyousuggestionsthatare
substantivelymeaningful.
However,insomeoccasionsyoumightwanttorestrictyourattentiontocertainsetof
potentialpathsorparametersinyourmodel,ratherthanallpossibleparameterspace
searchedbytheMODIFICATIONoption.
Ifyouwanttodosuchaprincipledmodificationprocess,youcanusethecustomizedLM
testssupportedinPROCCALIS.
138
Customized LM Tests by Using the LMTESTS

Statement
path
FamilySize
---> MentalAbility,
SocialStatus
---> S1 S2 S3
= 1. ,
---> P1 P2 P3
= 1.,
---> A1 A2 A3
= 1.,
MentalAbility
---> M1 M2 M3
= 1.;
lmtests corr_err=[coverr] path=[LV->LV LV->MV];

run;
Explore the set of LM tests called
corr_err, which contains all the
potential error covariance
parameters (COVERR) to be freed.
Explore the set of LM tests called path, which

contains all potential latent variable paths (LV->
LV) and measurement paths (LV -> MV) to be
freed.
139
ThecustomizedLMtestsdefinesetsofparametersofinterestso thatyourmodel
modificationprocess(orLMstatisticsoutput)wouldbelimitedtothosesetsof
parameters.PROCCALISprovidestheLMTESTSstatementsyntaxtoachievethe
customizedLMtests.
Thementalabilitymodelisusedagain.ThistimeIdefinetwosetsofparametersof
interest.ThefirstsetofLMtestsiscalledcorr_err (itisjustanameyouassign).This
setofparameterscontainstheparameterregionCOVERR,whichis akeywordthat
denotesallerrorcovariancesinthemodel.ThesecondsetofLM testsiscalledpath
anameyouassign.Thissetofparametersdonotexhaustallpathsinthemodel.It
containstheparameterregionsLV>LVandLV>MV,whicharekeywordsthatdenotes
thelatentvariable(LV)tolatentvariable(LV)pathsandthelatentvariable(LV)to
manifestvariable(MV)paths,respectively.Therefore,thiscustomizedsetpath
excludespathsfromobservedvariablestoobservedvariables,or fromobserved
variablestolatentvariablessothatthefactorstructuresofthemodelcouldnotbe
potentiallydestroyedbyaddingthesepaths.TheLMtestsforthesepathsaresimplynot
includedintheresultsforthepath set.
139
Customized LM Tests for Error Covariances

Rank Order of the 10 Largest LM Stat for Set corr_err
Parm
Type
Var1
Var2
LM Stat
Pr > ChiSq
Change
COVERR
P2
COVERR
P1
56.19312
<.0001
-1.96473
A1
12.26622
0.0005
COVERR
0.46050
A2
12.08031
0.0005
-0.48351
COVERR
FamilySize
A2
11.22650
0.0008
-1.88205
COVERR
M2
A2
10.26408
0.0014
1.55895
COVERR
S2
S1
7.78117
0.0053
1.55314
COVERR
MentalAbility
A2
7.48800
0.0062
0.78161
COVERR
A1
6.95709
0.0083
-0.52904
COVERR
P2
A3
6.54315
0.0105
0.76007
COVERR
A3
A2
6.21429
0.0127
-0.67173
140
ThistableshowsthecustomizedLMtestsoftheCORR_ERR set.Essentially,thistable
isthesameasoneofthestandardtablesproducedwiththeMODIFICATIONoption
becausebothtableshavethesameparameterregionCOVERR.
140
Customized LM Tests for Paths

Rank Order of the 10 Largest LM Stat for Set path
Parm
Type
Var1
Var2
LM Stat
Pr > ChiSq
Change
DV_DV
A2
DV_DV
A2
18.57947
<.0001
-2.31463
MentalAbility
17.20340
<.0001
DV_DV
0.95581
A1
15.86099
<.0001
1.88904
DV_DV
A1
MentalAbility
13.88705
0.0002
-0.75314
DV_DV
S2
MentalAbility
8.97262
0.0027
-0.38910
DV_DV
S3
6.27192
0.0123
0.32436
DV_DV
S2
6.19233
0.0128
-0.40928
DV_DV
MentalAbility
5.57439
0.0182
0.29376
DV_DV
5.33788
0.0209
0.39218
DV_DV
FamilySize
5.33730
0.0209
-1.20671
The measurement path A2 <--- ParentalEncouragement reduces the

model fit chi-square by 19.
141
ThesecondcustomizedsetofLMtestssuggeststhataddingthedependentvariable(DV)
todependentvariable(DV)pathA2< ParentalEncouragementimprovesthemodelfit
themostamongstallpathsinthepath set.Thechisquareimprovementisabout19,
whichisstatisticallysignificant.
141
Adding a Path from Parental Achievement to A2

S1
S2
S3
P1
1.
P2
P3
1.
M1
M2
M3
1.
Social Status
Parental
Encouragement
Family Size
Achievement
Motivation
Mental Ability
1.
A1
A2
A3
142
Addingthefirstpathsuggestedbythesecondsetofcustomizedsetisrepresentedby
thepathdiagramshownabove.ThepathinredshowsthatA2,whichisanindicatorof
AchievementMotivation,isnowalsoanindicatorofParentalEncouragement.Although
thefactorvariablefunctionalrelationshipispreservedinthissuggestedpathdiagram,
A2becomesfactoriallycomplex.ThealsoimpliesthatA2mightnothavebeenagood
(unique)measureofachievementmotivation.
142
Adding the ParentalEncouragement ---> A2 path

path
FamilySize
---> MentalAbility,
SocialStatus
---> S1 S2 S3
= 1. ,
---> P1 P2 P3 A2
= 1.,
---> A1 A2 A3
= 1.,
MentalAbility
---> M1 M2 M3
= 1.;
run;
143
Nonetheless,youaddthisnewpathforA2,asshownintheabove PROCCALIScode.All
youneedtodoistoaddA2asoneoftheobservedindicatorsof the
ParentalEncouragementfactor.
143
Before and After Adding the

ParentalEncouragement ---> A2 Path
Before
After
Fit Summary
Chi-Square
Chi-Square DF
Pr > Chi-Square
Fit Summary
196.7455
59
Chi-Square
Chi-Square DF
168.2618
58
<.0001
Pr > Chi-Square
<.0001
0.0936
0.0896
Adjusted GFI (AGFI)
0.7341
Adjusted GFI (AGFI)
0.7675
RMSEA Estimate
0.1431
RMSEA Estimate
0.1291
0.8087
0.8468
The model fit improves quite a bit.
144
Thesetwotablescomparethefitindicesbeforeandafteradding the
ParentalEncouragement>A2path.Themodelfitchisquareactuallydropsmorethan
19,whichwassuggestedbytheLMstatisticintheprecedingresults.Allotherfitindices
improvequiteabittoo.
Finally,acautionaboutallmodelmodification:youshouldvalidateyournewly
establishedmodelbynewdata.Thereasonisthatthemodelmodificationprocessis
subjecttothecapitalizationonchance.Usingaprincipledmodificationprocessbythe
customizedLMtestsmightnotavoidthechanceproblemcompletely.Confirmation
fromnewdataisalwaysrecommended.
144
More About PROC CALIS .

Many other different modeling languages: COSAN,
FACTOR, LINEQS, MSTRUCT, and RAM All
support multiple-group analysis and mean
structures
Other estimation methods (default ML): GLS,
WLS (ADF), ULS, DWLS, and FIML
Standardized solutions with standard error
estimates
Analysis of missing patterns (SAS/STAT 9.3)
145
Inthisworkshop,ImostlyusethePATHmodelinglanguagetofit SEM.Ialsobriefly
mentionedtheLISMODasaninterfacefortheLISRELmodel.There areactuallyquitea
fewmoremodelinglanguageinPROCCALIS:COSAN,FACTOR,LINEQS, MSTRUCT,and
RAM.Alloftheselanguagessupportmultiplegroupanalysisandmeanstructure
analysis.
IhavealsousedthedefaultML(maximumlikelihood)estimationmethodinthis
workshop,butPROCCALISsupportsmanyotherestimationmethodsaswell:GLS
(generalizeleastsquares),WLS(weightedleastsquares),ULS(unweightedleastsquares),
DWLS(diagonallyweightedleastsquares),andFIML(fullinformationmaximum
likelihood).
Ihaveonlyusedunstandardizedresultsinmostexamples,butPROCCALISalsoprovide
standardizedsolutionswithstandarderrorestimates.
Finally,analysisofmissingpatternswillbeavailablewiththe FIMLmethodinSAS/STAT
9.3.IhopetoaddmorefunctionalitiestoPROCCALISinthefuture.
145
Glossary
Manifest Observed variables (measured variables) in the data set.
Latent Unobserved variables.
Endogenous Dependent /mediating variables; at least one single-headed arrow points to it;
used as an outcome variable in an equation; can also be a predictor variable in other
equations.
Exogenous Independent variables; no single-headed arrows point to it; never used as an
outcome variable in the model; used only as a predictor in the model.
Factor A latent (unmeasured) variable that is treated as a hypothetical construct
(systematic source) in the model.
Error An exogenous term for uncertainty (unsystematic source) associated with an
endogenous manifest variable (or any endogenous variable, in a more general definition).
Disturbance An exogenous term for uncertainty (unsystematic source) associated with an
endogenous latent variable.
Path diagram representation
Rectangles: Observed / manifest variables.
Ovals / circles : Latent variables (factors, errors, and disturbances). Errors and
disturbances are not necessarily put into ovals/circles.
146
146
Glossary
Single-headed arrows: Directed paths, direct effects, path coefficients; specified in the
PATH statement.
Double-headed arrows that point to individual variables: Variance parameters of
exogenous variables or error variance parameters of endogenous variables; specified in
the PVAR statement.
Double-headed arrows that point to two distinct variables: Covariance parameters
between exogenous variables or error covariance parameters between endogenous
variables; specified in the PCOV statement.
Fit assessment
model fit chi-square statistic: Nonsignificance means that the theoretical model is
supported; not a very practical index because it almost always rejects all approximating
models that are practically useful.
AGFI (adjusted goodness-of-fit index) and Bentlers CFI (comparative fit index): Two
popular fit indices that indicate good model fit when their values are above 0.9.
SRMR (standardized root mean square residual) and RMSEA (root mean squared error
approximation): Two popular fit indices that indicate good model fit when their values
are below 0.05.
AIC, CAIC, and SBC: Information criteria for comparing competing models. The smaller
the better.
147
147
References
Arbuckle, J. (2008), AMOS 17.0 Users Guide, Crawfordville, FL: Amos Development
Corporation.
Bentler, P. M. (1985), Theory and Implementation of EQS: A Structural Equations
Program, Manual for Program Version 2.0, Los Angeles: BMDP Statistical Software.
Bentler, P. M. (1995), EQS, Structural Equations Program Manual, Program Version 5.0,
Encino, CA: Multivariate Software.
Bollen, K. A. (1989), Structural Equations with Latent Variables, New York: John Wiley
& Sons.
Fuller, W. A. (1987), Measurement Error Models, New York: John Wiley and Sons.
Holzinger, K. J. and Swineford, F. A. (1939), A Study in Factor Analysis: The Stability
of a Bi-Factor Solution, Supplementary Educational Monographs, No. 48. Chicago:
University of Chicago,Dept. of Education.
Jreskog, K. G. (1973), A General Method for Estimating a Linear Structural Equation
System, in A. S. Goldberger and O. D. Duncan, eds., Structural Equation Models in
the Social Sciences, New York: Academic Press.
148
148
References
Keesling, J. W. (1972), Maximum Likelihood Approaches to Causal Analysis, Ph.D. thesis,
University of Chicago.
Marjoribanks, K., ed. (1974), Environments for Learning, London: National Foundation
for Educational Research Publications.
McDonald, R. P. (1978), A Simple Comprehensive Model for the Analysis of Covariance
Structures, British Journal of Mathematical and Statistical Psychology, 31, 5972.
McDonald, R. P. (1980), A Simple Comprehensive Model for the Analysis of Covariance
Structures: Some Remarks on Applications, British Journal of Mathematical and
Statistical Psychology, 33, 161183.
Nowak, T. P., Hoffman, D. L., and Yung, Y. F. (2000), Measuring the Customer
Experience in Online Environments: A Structural Modeling Approach, Marketing
Science, 19(1), 2242.
Wiley, D. E. (1973), The Identification Problem for Structural Equation Models with
Unmeasured Variables, in A. S. Goldberger and O. D. Duncan, eds., Structural
Equation Models in the Social Sciences, New York: Academic Press.
149
149

Sem in Sas Jsm2010 - Yung

Hochgeladen von

Dokumentinformationen

Originaltitel

Copyright

Verfügbare Formate

Dieses Dokument teilen

Dokument teilen oder einbetten

Freigabeoptionen

Stufen Sie dieses Dokument als nützlich ein?

Sind diese Inhalte unangemessen?

Copyright:

Verfügbare Formate

Sem in Sas Jsm2010 - Yung

Hochgeladen von

Copyright:

Verfügbare Formate

Introduction to Structural

Equation Modeling Using

SAS/STAT 9.22 or later

Copyright 2010, SAS Institute Inc. All rights reserved.

Causal Model, Prediction, and Path Diagram

This is the essential representation.

Copyright 2010, SAS Institute Inc. All rights reserved.

Structural Equation Modeling versus

Direct and indirect effects

Copyright 2010, SAS Institute Inc. All rights reserved.

Other Names for Structural Equation Modeling (SEM)

Copyright 2010, SAS Institute Inc. All rights reserved.

Copyright 2010, SAS Institute Inc. All rights reserved.

A Very Brief History of PROC CALIS

TCALIS (SAS 9.2, 2008): experimental version

Copyright 2010, SAS Institute Inc. All rights reserved.

Structure of the Workshop

A brief description of the process of SEM

The PATH modeling language in PROC CALIS

Specifying models and interpreting results

LISMOD a language tailored to LISREL users

Second Part: Advanced Modeling

Analyzing direct and indirect effects

Testing specific hypotheses

Copyright 2010, SAS Institute Inc. All rights reserved.

Emphases of the Workshop

Introducing the structural equation methodology

Copyright 2010, SAS Institute Inc. All rights reserved.

Illustrating the Process of

Copyright 2010, SAS Institute Inc. All rights reserved.

A Structural Equation Model of Web-Surfing

Copyright 2010, SAS Institute Inc. All rights reserved.

Key Features of SEM

Copyright 2010, SAS Institute Inc. All rights reserved.

Basics: A Simple Regression

Copyright 2010, SAS Institute Inc. All rights reserved.

A Simple Linear Regression Model

Assumption: Variables are centered.

Copyright 2010, SAS Institute Inc. All rights reserved.

Measures of the Number of Hen Pheasants

Copyright 2010, SAS Institute Inc. All rights reserved.

Regression Analysis by PROC REG

Copyright 2010, SAS Institute Inc. All rights reserved.

Results Obtained from PROC REG

Copyright 2010, SAS Institute Inc. All rights reserved.

Regression Equation, Path Diagram, and the PATH

Copyright 2010, SAS Institute Inc. All rights reserved.

Regression Model Specified by PROC CALIS

proc calis data=hens;

Copyright 2010, SAS Institute Inc. All rights reserved.

Results from PROC CALIS for the Pheasant Data

The same estimate

Copyright 2010, SAS Institute Inc. All rights reserved.

Specification with Parameter Names

Without an explicit error term

proc calis data=hens;

Use the PVAR statement to specify

Copyright 2010, SAS Institute Inc. All rights reserved.

Hen Pheasants Results with Parameter

Parameter name specified

Parameter names specified

Copyright 2010, SAS Institute Inc. All rights reserved.

Keys to the PATH Modeling Language