7/25/2019 tau equivalent.pdf
1/65
ComparingClassicalTestTheorywithCFA
and
HowToUseTestScoresin
SecondaryAnalyses
LatentTraitMeasurementand
StructuralEquationModels
Lecture#8
March6,2013
PSYC948:Lecture#8
7/25/2019 tau equivalent.pdf
2/65
TodaysClass
ComparingclassicaltesttheorytoCFA Theuseandmisuseofsumscores
ReliabilityforsumscoresunderCFA
HowtouseCFAtotestassumptionsinCTT
WhattodowhenSEMisntanoption Secondaryanalyses
PSYC948:Lecture#8 2
7/25/2019 tau equivalent.pdf
3/65
DataforTodaysClass
Datawerecollectedfromtwosources: 144experiencedgamblers
Manyfromanactualcasino
1192collegestudentsfromarectangularmidwestern state
Manynever
gambled
before
Today,wewillcombinebothsamplesandtreatthemas
homogenous onesampleof1346subjects
Laterwewilltestthisassumption measurementinvariance(calleddifferentialitemfunctioninginitemresponsetheoryliterature)
Wewillbuildascaleofgamblingtendenciesusingthefirst24items
oftheGRI Focusedonlongtermgamblingtendencies
PSYC948:Lecture#8 3
7/25/2019 tau equivalent.pdf
4/65
PathologicalGambling:DSMDefinition
Tobediagnosedasapathologicalgambler,anindividualmustmeet5of10definedcriteria:
PSYC948:Lecture#8
1. Ispreoccupiedwithgambling
2. Needstogamblewithincreasingamountsofmoneyinorderto
achievethe
desired
excitement
3. Hasrepeatedunsuccessfuleffortstocontrol,cutback,orstopgambling
4. Isrestlessorirritablewhenattemptingtocutdownorstopgambling
5. Gamblesasawayofescapingfromproblemsorrelievingadysphoricmood
6. Afterlosingmoneygambling,oftenreturnsanotherdaytogeteven
7. Liestofamilymembers,therapist,orotherstoconcealtheextentofinvolvementwithgambling
8.
Hascommitted
illegal
acts
such
as
forgery,fraud,theft,orembezzlementtofinancegambling
9. Hasjeopardizedorlostasignificantrelationship,job,educational,orcareeropportunitybecauseof
gambling10. Reliesonotherstoprovidemoney
torelieveadesperatefinancialsituationcausedbygambling
4
7/25/2019 tau equivalent.pdf
5/65
Final12ItemsontheScale
Item Criterion Question
GRI1 3 Iwouldliketocutbackonmygambling.
GRI3 6
IfIlostalotofmoneygamblingoneday,Iwouldbemorelikelytowanttoplay
againthefollowingday.
GRI5 2
Ifinditnecessarytogamblewithlargeramountsofmoney(thanwhenIfirst
gambled)forgamblingtobeexciting.
GRI6 8 Ihave gonetogreatlengthstoobtainmoneyforgambling.
GRI9 4 IfeelrestlesswhenItrytocutdownorstopgambling.
GRI10 1 ItbothersmewhenIhavenomoneytogamble.
GRI11 5 Igambletotakemymindoffmyworries.
GRI13 3 Ifinditdifficulttostopgambling.
GRI14 2 IamdrawnmorebythethrillofgamblingthanbythemoneyIcouldwin.
GRI15 7 Iamprivateaboutmygamblingexperiences.
GRI21 1 Itishardtogetmymindoffgambling.
GRI23 5 Igambletoimprovemymood.
PSYC948:Lecture#7 5
7/25/2019 tau equivalent.pdf
6/65
The12itemanalysisgavethismodelfitinformation:
GRI12ItemAnalysis
PSYC948:Lecture#7 6
Themodelindicatedthemodeldidnotfitbetterthanthesaturatedmodel butthis
statisticcanbeoverlysensitive
ThemodelRMSEAindicatedgoodmodelfit
(wantthistobe.95)
TheSRMRindicatedthefitwell(wantthisto
be
7/25/2019 tau equivalent.pdf
7/65
CLASSICALTESTTHEORY
PSYC948:Lecture#8 7
7/25/2019 tau equivalent.pdf
8/65
ClassicalTestTheory(CTT)
WhatyouhavelearnedaboutmeasurementsofarlikelyfallsunderthecategoryofCTT:
Writingitemsandbuildingscales
Itemanalysis
Scoreinterpretation
Evaluatingreliabilityandconstructvalidity
Bigpicture:WewillviewCTTasmodelwitharestrictivesetof
assumptionswithinamoregeneralfamilyoflatenttrait
measurementmodels
ConfirmatoryFactorAnalysisisameasurementmodel
PSYC948:Lecture#8 8
7/25/2019 tau equivalent.pdf
9/65
DifferencesAmongMeasurementModels
Whatisthenameof
the
latent
traitmeasuredbyatest?
ClassicalTestTheory(CTT) = TrueScore(T)
ConfirmatoryFactorAnalysis(CFA) = FactorScore(F)
ItemResponseTheory(IRT) = Theta()
Fundamentaldifferenceinapproach:
CTT unitofanalysisistheWHOLETEST(itemsumormean)
Sum=latenttrait,andthesumdoesntcarehowitwascreated
Onlyusingthesumrequiresrestrictiveassumptionsabouttheitems CFA,IRT,andbeyond unitofanalysisistheITEM
Modelofhowitemresponserelatestoanestimatedlatenttrait
Differentmodelsfordifferingitemresponseformats
Providesaframeworkfortestingadequacyofmeasurementmodels
PSYC948:Lecture#8 9
7/25/2019 tau equivalent.pdf
10/65
ClassicalTestTheory(CTT)
InCTT,theTESTistheunitofanalysis: TruescoreT:
Bestestimateoflatenttrait:Meanoverinfinitereplications
Errore:
Expectedvalue(mean)of0,expectedtobeuncorrelatedwithT
esaresupposedtowashoutoverrepeatedobservations
SotheexpectedvalueofTisYtotal Intermsofobservedvarianceofthetestscores:
Observedvariance=truevariance+errorvariance
GoalistoquantifyreliabilityReliability=truevariance/(truevariance+errorvariance)
Because
the
CTT
model
does
not
include
individual
items,
itemsmustbeassumedexchangeable(andmoreitemsisbetter)
PSYC948:Lecture#8
Ytotal
TrueScore
error
?
?
10
7/25/2019 tau equivalent.pdf
11/65
ClassicalTestTheory,continued
CTTunitofanalysisistheWHOLETEST(sumofitems) Wanttoascertainhowmuchofobservedtestscorevariance
isduetotruescorevarianceversuserrorvariance
Quantifyerrorvarianceinvariousways
ErrorisaunitaryconstructinCTT(anderrorisbad) Goalisthentoreduceerrorvarianceasmuchaspossible
Standardizationoftestingconditions(makeconfoundsconstants)
Aggregation=moreitemsarebetter(errorsshouldcancelout)
Itemsareexchangeable;propertiesarenottakenintoaccount
Followedbygeneralizabilitytheorytodecomposeerror
e.g.,ratervariance,personvariance,timevariance
PSYC948:Lecture#8 11
7/25/2019 tau equivalent.pdf
12/65
AdvantagesofCFAoverCTT
Morereasonableassumptionsaboutitems CTTassumestauequivalentitems
Tau equivalentitems:equalfactorloadings
CFAallowsatestofwhethereachitemrelatestothefactor,aswellaswhether
differentfactorloadingsacrossitemsareneeded
Wouldindicatesomeitemsarebetterthanothers
Comparabilityacrosssamples,groups,andtime CTT:Noseparationofobserveditemresponsesfromtruescore
Sumacrossitems=truescore;itempropertiesarefor thatsampleonly
CFA:Latenttraitisestimatedseparatelyfromitemresponses
Separatespersontraitsfromspecificitemsgiven
Separatesitempropertiesfromspecificpersonsinsample
Advantagesapplytoanylatenttraitmodel
PSYC948:Lecture#8 12
7/25/2019 tau equivalent.pdf
13/65
ReliabilityMeasuredbyAlpha
Forquantitativeitems(itemswithascale althoughusedoncategoricalitems),thisisCronbachs Alpha
OrGuttmanCronbach alpha(Guttman 1945>Cronbach 1951)
Anotherreducedformofalphaforbinaryitems:KR20
Alphaisdescribedinmultipleways:
Isthemeanofallpossiblesplithalfcorrelations
Isexpectedcorrelationwithhypotheticalalternativeformofthe
samelength
Islowerboundestimateofreliabilityunderassumptionthatallitemsaretau
equivalent(moreaboutthatlater)
Asanindexofinternalconsistency
Althoughnothingabouttheindexindicatesconsistency!
PSYC948:Lecture#8 13
7/25/2019 tau equivalent.pdf
14/65
WhereAlphaComesFrom
Thesumof
the
item
variancesisgivenby:
Var(I1)+Var(I2)+Var(I3).+Var(Ik)(justtheitemvariances)
ThevarianceofthesumoftheitemsisgivenbythesumofALLthe
itemvariancesandcovariances:
Var(I1+I2+I3)=Var(I1)+Var(I2)+Var(I3)
+2Cov(I1,I2)+2Cov(I1,I3)+2Cov(I2,I3)
Wheredoesthe2comefrom?
Covariancematrixissymmetric
Sumthewholethingtogettothe
varianceofthesumoftheitems
PSYC948:Lecture#8
I1 I2 I3
I1
12
12
13
I2
21
22
23
I3 31 32 3
2
14
7/25/2019 tau equivalent.pdf
15/65
GuttmanCronbach AlphaforReliability
Numeratorreducestojustthecovarianceamongitems
Sum
of
the
item
variances
Var(X)+Var(Y)=Var(X)+Var(Y)justtheitemvariances
Variance
of
total
Y
(the
sum
of
the
items)
Var(X+Y)=Var(X)+Var(Y)+2Cov(X,Y) PLUScovariances
So,iftheitemsarerelatedtoeachother,thevarianceofthetotalYitemsum
shouldbebiggerthanthesumoftheitemvariances
Howmuchbiggerdependsonhowmuchcovarianceamongtheitems theprimaryindexofrelationship
PSYC948:Lecture#8
Covariance
Version:
k=#items
15
7/25/2019 tau equivalent.pdf
16/65
AssessingReliabilityinOur12ItemGamblingScale
TogettheGuttmanChronbach Alphaofour12itemscale,weneedthecovariancematrix
ThiscanbefoundbytheSAMPSTAToptionunderthe
OUTPUTstatement
Sumofitemvariances=11.834
Sumofitemcovariances=21.575
VarianceofTotalY=11.834+2*21.575=54.984
Alphareliability: ..
. .852
PSYC948:Lecture#8 16
7/25/2019 tau equivalent.pdf
17/65
Reliabilityforwhat?
Thealphareliabilityisthereliabilityfor: Thetotaltestscore
Undertheassumptionthattheitemsaretauequivalent
Tauequivalentmeanseachitemcontributesequally
Inafewslides,wewillseehowthistranslatestoCFA
Whatalphaisnot: Anindexofmodelfit(unidimensionality)
PSYC948:Lecture#8 17
7/25/2019 tau equivalent.pdf
18/65
MeasurementLanguage:DontSayThese
Often,peoplerefertoitemsastappingsomelatenttrait Ithinkthismakestheprocesslesstransparent itemsmeasurethetrait
Whenalphaisused,youcansometimeshearpeoplesaysomething
abouthow
well
the
items
hang
together
Thisiscertainlynottrue
PSYC948:Lecture#8 18
7/25/2019 tau equivalent.pdf
19/65
HowtoGetAlphaUP
PSYC948:Lecture#8 19
7/25/2019 tau equivalent.pdf
20/65
AlphaasReliabilityWhatcouldgowrong?
Alphadoesnotindexdimensionality itdoesnotindextheextenttowhichitemsmeasurethesameconstruct
Thevariabilityacrosstheinteritemcorrelationsmatters,too!
Weuseitembasedmodels(CFA)toexaminedimensionality
PSYC948:Lecture#8 20
7/25/2019 tau equivalent.pdf
21/65
CaseInPoint:All24Items
Lastclassweshowedthatthe24itemsoftheGRIdidnotfitaonefactormodel whatwouldhappenifweneglectedtocheckmodel
fitandusedthetotalscoreasourestimateofgamblingtendency?
Thereliabilityestimate fromthecovariancematrixofalltheitems
(thesaturatedmodelH1)was .861 Wewouldhaveconcludedwehadagoodscaleforgambling
But,fromCFAlastweek,wefoundthatonefactordidntdescribeall
theitems Anysubsequentanalysiswillhavethemisfitbiastheresults
PSYC948:Lecture#8 21
7/25/2019 tau equivalent.pdf
22/65
TestingCTTAssumptionsinCFA
Alphaisreliabilityassumingtwothings: Allfactorloadings(discriminations)areequal,orthattheitemsare
truescoreequivalentortauequivalent
Localindependence(dimensionalitynowtestedwithinfactormodels)
Wecantesttheassumptionoftauequivalencetoovianestedmodel
comparisonsinwhichtheloadingsareconstrainedtobeequal
doesmodelfitgetworse?
Ifso,dontusealpha usemodelbasedreliability(omega)instead.Omegaassumesunidimensionality,butnottauequivalence
Researchhasshownalphacanbeanoverestimateoranunderestimatedependingon
particulardatacharacteristics
TheassumptionofParallelitemsisthentestablebyconstraining
itemerrorvariancestobeequal,too doesmodelfitgetworse? Parallelitemswillhardlyeverholdinrealdata
Notethatiftauequivalencedoesnthold,thenneitherdoesparallel
PSYC948:Lecture#8 22
7/25/2019 tau equivalent.pdf
23/65
AnotherBlastfromthePast:ParallelItems
AnotherCTTmodelthatexiststhatofparallelitems Allitemshavethesamecovarianceandvariance
Goesonestepfurtherthantauequivalence(equalcovariancesbutunequalvariances)
Undertheparallelitemsmodel,thealphareliabilityforthetotaltestscoreiscalledthe
SpearmanBrownreliability Usedtoprophesythenumberofitemsneededtoincreasereliabilitytoadesiredlevel
SpearmanBrownProphesyFormula
ReliabilityNEW=ratio*relold/[(ratio1)*relold+1]Ratio=ratioofnew#itemstoold#items
Forexample:
Oldreliability=.40Ratio=5timesasmanyitems(had10,whatifwehad50)
Newreliability=.77
PSYC948:Lecture#8 23
7/25/2019 tau equivalent.pdf
24/65
Reliabilityvs.ValidityParadox
GiventheassumptionsofCTT,itcanbeshownthatthecorrelationbetweenatestandanoutsidecriterioncannotexceedthereliabilityofthetest(seeLord&
Novick1968)
Reliabilityof.81?Noobservedcorrelationspossible>.9,
becausethatsallthetruevariancetheretoberelatable!
Inpractice,thismaybefalsebecauseitassumesthattheerrorsareuncorrelatedwith
thecriterion(andtheycouldbe)
Selectingitemswiththestrongestdiscriminations(orthestrongestinter
correlations)canhelptopurifyorhomogenizeatest,butpotentiallyattheexpenseofconstructvalidity
Canendupwithabloatedspecific
Itemsthatareleastinterrelatedmaybemostusefulinkeepingtheconstructwell
definedandthusrelatabletootherthings
PSYC948:Lecture#8 24
7/25/2019 tau equivalent.pdf
25/65
UsingCTTReliabilityCoefficients:BacktotheScoreEstimates
ReliabilitycoefficientsareusefulfordescribingthebehaviorofthetestintheoverallsampleVar(Y)=Var(T)+Var(e)
Butreliabilityisameanstoanendininterpretingascoreforagivenindividual
weuseittogettheerrorvariance
Var(T)=Var(Y)*reliability;soVar(e)=Var(Y) Var(T)
95%CIforindividualscore=Y1.96*SD(e)
Givesanindicationofhowprecisethetruescoreestimateisonthemetricofthe
originalvariable
Example:
Y
=
100,
Var(e)
=
9
95%
CI
94
to
106Y=100,Var(e)=25 95%CI90to110
Notethisassumesasymmetricdistribution,andthuswillgooutofboundsofthescale
forextremescores
NotethisassumestheSD(e)ortheSEforeachpersonisthesame
Cuemind
blowing
GRE
example
PSYC948:Lecture#8 25
7/25/2019 tau equivalent.pdf
26/65
95%ConfidenceIntervals:Quantitative
SEMrangesfrom9to55
PSYC948:Lecture#8 26
7/25/2019 tau equivalent.pdf
27/65
REVISITINGCTTFROMA
CFAPERSPECTIVE
PSYC948:Lecture#8 27
7/25/2019 tau equivalent.pdf
28/65
ClassicalTestTheoryfromaCFAPerspective
InCTTtheunitofanalysisisthetestscore:,
InCFAtheunitofanalysisistheitem:
TomapCFAontoCTT,wemustputthesetogether:
,
PSYC948:Lecture#8 28
7/25/2019 tau equivalent.pdf
29/65
FurtherUnpackingoftheTotalScoreForumla
BecauseCFAisanitembasedmodel,wecanthensubstituteeachitemsmodelintothesum:
,
MappingthisontotruescoreanderrorfromCTT:
and
PSYC948:Lecture#8 29
7/25/2019 tau equivalent.pdf
30/65
FamiliarTerms
Thetauequivalentmodelassumes: Allitemsmeasurethefactorthesame: Eachitemhasitsownuniquevariance: 0,
Theparallelitemsmodelsassumes: Allitemsmeasurethefactorthesame: Allitemshavethesameuniquevariance: 0,
Assuch,eachofthesemodelscanbetestedbyusingtheCFAapproach eacharenestedwithinthefullCFAmodel
PSYC948:Lecture#8 30
7/25/2019 tau equivalent.pdf
31/65
TauEquivalence:ModelImpliedCovarianceMatrix
TheCFAmodelimpliesaveryspecificformforthecovariancematrixoftheobserveditems:
Thevarianceofanitem
was:
Thecovarianceofapairofitemsandwas:
UnderTauEquivalence,allloadingsarethesame,meaning: Theitemvariancescanbedifferent(becauseof) Allitemcovariancesarethesame()
Thisiscalledthecompoundsymmetryheterogeneousmodel Wecanactuallyachievethesamemodelwithoutthefactor
PSYC948:Lecture#8 31
7/25/2019 tau equivalent.pdf
32/65
TauEquivalenceModelfor12ItemGamblingScale
ThefollowingtwopiecesofMplussyntaxresultinthesameequivalentmodel: TauEquivalenceasaFactorModel:
TauEquivalenceasaCompundSymmetryHeterogeneousVariancesModel:
PSYC948:Lecture#8 32
7/25/2019 tau equivalent.pdf
33/65
ModelImpliedCovarianceMatrix
Allcovariancesequal/allvariancesdifferent
PSYC948:Lecture#8 33
7/25/2019 tau equivalent.pdf
34/65
TestingforTauEquivalence
TheTauEquivalencemodel(assumedwhenyousumitems)canbetestedagainstthefullCFAmodel
Themodelsarenested,sowecanusealikelihoodratiotest
LoglikelihoodfromCFAmodel: 18,988.425;SCF=2.4309 36parameters(12itemintercepts,11factorloadings,1factorvariance,12uniquevariances)
LoglikelihoodfromTEmodel: 19,051.350;SCF=2.5172 25parameters(12itemintercepts,1factorloading,12uniquevariances)
MLRLikelihoodratiotest: 56.315, .001
Therefore,werejectthetauequivalentmodelinfavoroftheCFAmodel this
meansthe
simple
sum
of
the
items
isnot
sufficient
WeshouldusetheCFAmodelfactorscoreinsteadofasumscore
PSYC948:Lecture#8 34
7/25/2019 tau equivalent.pdf
35/65
ParallelItems:ModelImpliedCovarianceMatrix
TheCFAmodelimpliesaveryspecificformforthecovariancematrixoftheobserveditems:
Thevarianceofanitemwas: Thecovarianceofapairofitemsandwas: UnderParallelItems,allloadingsanduniquevariancesarethesame:
Allitemvariancesarethesame( ) Allitemcovariancesarethesame()
Thisiscalledthecompoundsymmetrymodel Wecanactuallyachievethesamemodelwithoutthefactor
Becauseparallelitemsarenestedwithintauequivalentitems,wedonothaveto
testthismodelasweknowitwillnotfitwhencomparedtotheCFAmodel
PSYC948:Lecture#8 35
7/25/2019 tau equivalent.pdf
36/65
TestScoreReliabilityUndertheCFAModel
CoefficientalphagavereliabilityforthetotaltestscoreundertheTauEquivalentItemsModel
WerejectedthatmodelinfavoroftheCFAmodel
Therefore,coefficientalphawillnotbecorrectforourtotaltestscore(ifwewereto
stillsum
up
the
items)
ThenotionsoftestscorereliabilityundertheCFAmodelnowinvolve
thefactorloadings Butstillcomebacktoclassicalnotionofreliabilitybeingtheproportionofvariancedue
totruescore:
PSYC948:Lecture#8 36
7/25/2019 tau equivalent.pdf
37/65
DerivingReliabilityForSumScores UndertheCFAModel
ToshowwheretotalscorereliabilityundertheCFAmodelcomesfrom,recallourCFAmodelforthetotalscore:
,
MappingthisontotruescoreanderrorfromCTT:
and
WenowmustderivethevarianceforTandE
PSYC948:Lecture#8 37
S V i U d h C A d l
7/25/2019 tau equivalent.pdf
38/65
TrueScoreVarianceUndertheCFAModel
Thevarianceforthetruescore:
PSYC948:Lecture#8 38
E V i U d th CFA M d l
7/25/2019 tau equivalent.pdf
39/65
ErrorVarianceUndertheCFAModel
BecausetheCFAmodelallowsfortheestimationoferrorcovariances(althoughyoushouldntdothat),theerror
varianceundertheCFAmodelbecomes:
Whenerrorcovariancesarenotestimated,thelasttermis
zero,leaving PSYC948:Lecture#8 39
R li bilit f T t l S U d CFA
7/25/2019 tau equivalent.pdf
40/65
ReliabilityforTotalScoreUnderCFA
ThereliabilityofthetotalscorefromCFA,isthen:
2, ThisreliabilitycoefficientiscalledcoefficientOmega() Ifthetauequivalentmodeldoesnotholdisthereliabilityofa
totaltestscore(sumscore) TypicallyishigherthanAlpha
Ifunidimensionalmodelholds,coefficientswillbeclose
PSYC948:Lecture#8 40
Calculating Omega for Our Test
7/25/2019 tau equivalent.pdf
41/65
CalculatingOmegaforOurTest
Wecan
use
Mplus
to
calculate
Omega
for
our
test:
PSYC948:Lecture#8
Here,Omegais.855
41
Omega Under Tau Equivalent Items
7/25/2019 tau equivalent.pdf
42/65
OmegaUnderTauEquivalentItems
Omegaequal
to
Alpha
when
you
use
the
tau
equivalent
items
model
OmegaistheSpearmanBrownreliabilityunderparallelitems
PSYC948:Lecture#8
Here,Omegais.852
whichisequaltothe
Alphawecalculatedusing
thecovariancematrix
42
Recapping: CTT using CFA
7/25/2019 tau equivalent.pdf
43/65
Recapping:CTTusingCFA
Classicaltesttheory andmorespecifically,totaltestscores,isthedominantwaytoassesssubjects
ThisistrueevenunderCFA
Thekeyistobesuretocheckifaonefactormodelfitsthedatabeforeusinganytypeofreliabilitycoefficient
Ifnot,donotuseatestscore
Iftheonefactormodelfits thenasinglescorecanrepresentthetest
Thenextworryisaboutrepresentingtheerrorinthetestscore
(relatedtoreliability) Ifreliabilityishigh(?Howhigh,standardof.8),thenusingthetestscoreina
subsequentanalysisisacceptedpractice
PSYC948:Lecture#8 43
Secondary Analyses with Factor Scores
7/25/2019 tau equivalent.pdf
44/65
SecondaryAnalyseswithFactorScores
Ifyouwanttouseresultsfromasurveyinanewanalysis Best: UseSEM errorinfactorscoresisalreadypartitionedvariance
Similarlygood: Useplausiblevalues(repeateddrawsfromposterior
distributionofeachpersonsfactorscore) essentiallywhatSEMdoes butwithfactorscoresthatvarywithinaperson
CanbedoneinMplus notdescribed
SlightlyLessGood: UseSEMwithsingleindicatorfactorsusingsumscores
Thefocusofthenextsection
Makeerrorvariance=(1reliability)*Variance(Sumscore);factorloading=1
Okay(butwidespread):forscalesthatareunidimensional(andverifiedin
CFA),usesumscoresAssumesunidimensionalityandhighreliability
NotCool: Usefactorscoresonly
PSYC948:Lecture#8 44
What about Using Factor Scores?
7/25/2019 tau equivalent.pdf
45/65
WhataboutUsingFactorScores?
AlthoughCFAfactorscoreshavefewerproblemsthanEFAfactorscores(becausethereisnorotationinCFA),theystillhaveissues:
Theywillbeshrunken(i.e.,pushedtowardsthemean,suchthattheobservedvariance
ofthefactorscoreswillbelessthantheoriginalfactorvariance)
Cangetestimatesoffactordeterminacy howcorrelatedestimatedfactorscores
arewithtruefactorscores(basicallyhowmucherrorisintroducedbyestimatingthe
factorscoresasobservedvariables)
Theyarejustestimatesof
central
tendencyfromadistributionforeachperson,not
knownvalues andusingestimatesasknownvaluesinanothermodelmakesthe
relationshipswithinthatmodellookmoreprecisethantheyare(likeSE=0)
YouCANNOTcreatefactorscoresbyusingtheloadingsassuch:
F=11y1+21y2+21y3 ThisisaCOMPONENTmodel,notaFACTORmodel.
PSYC948:Lecture#8 45
7/25/2019 tau equivalent.pdf
46/65
SINGLEINDICATORMODELS
PSYC948:Lecture#8 46
Single Indicator Models
7/25/2019 tau equivalent.pdf
47/65
SingleIndicatorModels
SingleindicatormodelsareCFAlikemodelswhereafactorismeasuredbya
singleindicator:
Shownhereforthegamblingfactor
PSYC948:Lecture#8
Gambling
Tendencies
Sumof10
GRIItems
,
1
1
47
IdentificationinSingleIndicatorModels
7/25/2019 tau equivalent.pdf
48/65
g
Howisthis
possible?
Isnt
asingle
indicator
factor
model
unidentified?
Wefixthefactorvariance,factorloading,anduniquevariance
Factorvariancerepresentsreliableportion
Singleindicatormodelparameters:
factorvariance; factorloading;itemuniquevariance(assumefactormeanfixedtozeroanditeminterceptissettoitmean)
Ourconstraintsare:
1 (theportionofYthatisreliable) 1 (theportionofYthatisleftover)
PSYC948:Lecture#8 48
AssumptionsinaSingleIndicatorAnalysis
7/25/2019 tau equivalent.pdf
49/65
p g y
Touseasingleindicatoryoumustassume: Theindicatorisunidimensional(onlyonefactor)
ThisistestableinCFA(butifyouhaveasmallsampleishardtodo)
Ifnotpossibletotest,youmustassumeyouhaveonefactor Thisisanassumptionthatthetestis*as*dimensionalinyoursample/population
Thereliabilityoftheindicatorisknown
AlsoobtainablefromCFA
Ifnotpossibletoobtain,thenyoumustuseapreviouslyreportedreliability
coefficient Thisisanassumptionthatthetestis*as*reliableinyoursample/population
PSYC948:Lecture#8 49
SingleIndicatorExampleAnalysis
7/25/2019 tau equivalent.pdf
50/65
g p y
Todemonstrateasingleindicatorexampleanalysis,wewillusethe12itemGRItopredicttheSOGSscore
SOGS=SouthOakGamblingScreen(wecollectedthis)
Note:weassumethishasreliabilityof1.0
The12itemGRIisthesingleindicatorofthegamblingfactor
Step#1:determinethatthesingleindicatorisunidimensional The1factorCFAmodelfitthe12itemGRI
Step#2:getthesingleindicatorreliability FromtheCFAanalysiswefoundthatthereliabilityofthe12itemGRIwas.855
Step#3:estimatethevarianceofthe12itemGRItotalscore WecandothisinMplus foundthevariancetobe55.050
PSYC948:Lecture#8 50
SingleIndicatorAnalysis
7/25/2019 tau equivalent.pdf
51/65
NowthatwehaveourreliabilityoftheGRIandthevarianceoftheGRI,wecanputthese
intothesingleindicatormodel:
PSYC948:Lecture#8 51
SingleIndicatorModelResults:
7/25/2019 tau equivalent.pdf
52/65
Note:47.067+7.982=55.05(thevarianceoftheGRI)
PSYC948:Lecture#8
Gambling
Tendencies
Sumof10
GRIItems
,
1
47.067
(3.459)
7.982
SOGSScore
.167(.011)
2.440(.213)
52
SingleIndicatorModelInterpretation
7/25/2019 tau equivalent.pdf
53/65
Thestandardizedregressionslopeforthegamblingfactor,predictingtheSOGSwas.591
asgamblingwentup1SD,theSOGSscorewentup.591SD
Correlationbetween
gamblingandSOGS
Thegambling
factor
accountedfor35.0%ofthe
varianceintheSOGSscore
PSYC948:Lecture#8 53
ComparisonwithNonSingleIndicator
7/25/2019 tau equivalent.pdf
54/65
Withoutusingthesingleindicator:
PSYC948:Lecture#8 54
7/25/2019 tau equivalent.pdf
55/65
ITEMPARCELING
PSYC948:Lecture#8 55
ItemParceling
7/25/2019 tau equivalent.pdf
56/65
Frequently,sumscoresareusedinSEMunderadifferentlabel:asitemparcels
Evidentlyparcelsoundsmorepolitestuffthatdidntfit
ItemparcelsaresumsofsetsofitemsthatareinsertedintoaSEMwithoutanyfurtherinspection
Frequently,itemparcelswillhidebadfitofmodel Blindparceling=cheating
Asparcelsaresums andtodaysclassisaboutusingsums,wecan
nowdiscuss
parcels
under
CTT
with
CFA
PSYC948:Lecture#8 56
ApplyingOurUnderstandingofTotal/SumScorestoParcels
7/25/2019 tau equivalent.pdf
57/65
Aswe
have
seen
today,
atotal
score
isastatistical
model
Tauequivalentitems
Aswithanystatisticalmodel,ifthemodeldoesnotfit(adequately
representthedata),misleadingresultsoccur
Parcelingitemsmakesanimplicitassumptionabouttheirstructure
thattheytooaretauequivalent Ifthatassumptionisnotvalid,resultscannotbebelieved
Mostusesofparcelingmakenoattempttodetermineifthetau
equivalenceassumption
iscorrect
PSYC948:Lecture#8 57
Revisitingour24ItemGRI
7/25/2019 tau equivalent.pdf
58/65
Asyou
will
recall,
our
24
item
GRI
did
not
fit
the
one
factor
model
Todemonstrateparceling,wewilltakethe12misfittingitemsand
createaparcel(sumtheirscorestogether)
WewillthenaddtheparceltoaCFAmodelwiththeother12items
PSYC948:Lecture#8 58
ThemodelRMSEAindicatedthemodeldidnotfitwell(wantthistobe.95)
TheSRMRindicatedthemodeldidnotfitwell
(wantthistobe
7/25/2019 tau equivalent.pdf
59/65
Thesyntax:
Themodelfitstatistics(adequatemodelfit):
Note:12itemGRIhadRMSEAof.045
PSYC948:Lecture#8 59
HowParcelingHidesPoorModelFit
7/25/2019 tau equivalent.pdf
60/65
Theitem
parcel
hides
poor
model
fit
by
using
anumbers
game
toitsadvantage
Modelwith24itemshad300elementsinsaturatedcovariance
matrix(but48parametersforthatmatrix)
Modelwith12itemsplusparcel(12items)had91elementsin
saturatedcovariance
matrix
(and
26
parameters)
Therelativeratioofparameterstosaturatedcovariancesmakesthe
parcelhidethefitissues Especiallywhentheremainderoftheitemsfitwellalready
PSYC948:Lecture#8 60
ParcelingDoneRight
7/25/2019 tau equivalent.pdf
61/65
Toaddaparcelyoumustfirstexaminethefitofaonefactormodeltotheitems
oftheparcel:
Themodelfitsuggestsaonefactormodeldoesntfit
PSYC948:Lecture#8 61
WhyParcelingisCheatingandWhyYouShouldntDoIt
7/25/2019 tau equivalent.pdf
62/65
Ifyou
didnt
check
the
parcel
before
adding
itto
the
12
item
GRI
modelyouwouldconclude1factormodelfitthedatawell Ifaonefactormodelfits,thenwhatcomesnextistypicallytheuseofitssumscore
Thesumscorefromthe12good+1bad(parcel)modelisjustthesumscorefromthe24itemGRI whichdidntfitaonefactormodel
Thecaveat:the12+1modelsumscorehadanomegareliabilityof.516!
Mostofthelackofreliabilitycomesfromtheestimateduniquevarianceoftheparcel
Youcannotmakeagoodfactorbycheatingwithaparcel!
PSYC948:Lecture#8 62
7/25/2019 tau equivalent.pdf
63/65
CONCLUDINGREMARKS
PSYC948:Lecture#8 63
WrappingUp
7/25/2019 tau equivalent.pdf
64/65
Todaywas
spent
on
comparing
classical
test
theory
(synonymous
withsumscores)toCFA
UnderstandinghowCTTandCFAarerelatedisimportant ManypeoplebelievethatsumscoresareAOK
Theyonlyareiftheyfita1factormodelandhaveahighreliability
Manypeopledontthinkparcelinginvolvessumscores
Thelabelmustbetheproblem
Singleindicatormodelscanbeagoodwaytousesumscoresif: The1factormodelfits
Thereisahighdegreeofreliability
WewillreturntothisoncewediscussSEMmorethoroughly
PSYC948:Lecture#8 64
ComingUp
7/25/2019 tau equivalent.pdf
65/65
Nextweekslecture:multidimensionalCFAmodels Morethanonefactor
Reliabilityforatotaltestscorenolongerapplies(eachfactoris
wherereliabilityisimportant)
Timepermitting:anintroductiontoExploratoryFactorAnalysis
FollowedbyacomparisonofCFAandEFA
AndwhyyoualsoshouldntbedoingEFA ButcouldexplorethedatabetterusingCFAtechniques
PSYC948:Lecture#8 65