linking toefl ibt scores to ielts scores

Linking TOEFL iBT
™
Scores to
IELTS
®
Scores –
A Research Report
ETS
December 17, 2010

LinkingTOEFLiBT™ScorestoIELTS®Scores–AResearchReport


1



Abstract

TOEFL®testscoreshavebeenwidelyacceptedasevidenceofnonnativeEnglishspeakers’proficiencyinthe
useofEnglish.TheTOEFLtestmeasurestesttakers’abilitytocommunicateinEnglish‐mediumcollegesand
universities.

TheTOEFLiBT™testmeasuresacandidate’sabilitytocombinelis tening,reading,speakingandwritingskills
withauthenticandinnovativeintegratedtasks,makingitthemostcomprehensiveEnglish‐language 
proficiencytest.Acceptedbyover7,500scoreusersinmorethan130countries,theTOEFLtestisthemost
widelyrecognizedEnglish‐languagetestintheworld.

Recently,ETSconductedscorecomparisonresearchbetweentheTOEFLtestand
analternativetest,IELTS
(InternationalEnglishLanguageTestingSystem).Fortheresearch,ETSobtainedasampleof1,153students

whohadbothIELTSandTOEFLscores.Usingequipercentilelinking,thecorresponding TOEFLscorethat
wouldpassthesamepercentageoftesttakersforeachIELTSscoreswasobtained.Thescore
comparison
resultsforeachsection(Listening,Speaking,Reading,andWriting)andthetotaltestshowedthatmostofthe
studentsscoredinthemiddletomid‐highscorerangesonbothtests.Limitationsoftheresearchand
suggestionsforfutureresearcharealsodiscussed.


LinkingTOEFLiBT™ScorestoIELTS®Scores–AResearchReport


2



LinkingTOEFLiBT™ScorestoIELTSScores

Introduction

TOEFL®testscoreshavebeenwidelyacceptedbyeducationalandotherinstitutionsthroughouttheworldas
evidenceofnonnativeEnglishspeakers’proficiencyin Englishlanguageuse,andtheTOEFLtesthasreceived
verypositivefeedbackfromscoreusersandinstructorsintheEnglishasaSecond Language(ESL)community.

Thereare
manyoccasionswhenscoresfromdifferenttestsorevenfromdifferentversionsofthesametest
needtobelinkedfordecisionmakingorotherpurposes.Forexample,thescoresontheSAT®IandACT®
tests,twowellknowncollegeadmissionstestsintheUnitedStates,havebeenlinked
toeachothertohelp
universityadmissionsofficesusethescorestoevaluateapplicants.Aresearchstudywascarriedoutbyboth
ETSandACTresearchstafftodevelopascorecompar isontableforthispurpose(Pommerich,Hanson,Harris,
&Sconing,2000.)Anotherexampleisthat,whentheTOEFLiBTtest,
wasdevelopedtobetakenviathe

Internetandtoreplacethecomputer‐adaptivetest(cBT)version,astudywasconductedtolinkTOEFLiBT
scorestoTOEFLcBTscores(Wang,Eign or,&Enright,2007.)

Recently,ETSconductedscorecomparisonresearchbetweentheTOEFLtestandanalternativetest,IELTS
(InternationalEnglishLanguageTestingSystem).AsabestpracticerequiredbytheGuidelinesforPracticeby
theInternationalLanguageTestingAssociation(ILTA,2007),theStandardsforEducationalandPsychological
Testing(AERA,APA,NCME,1999),andtheETSstandards(ETS,2002,p.45,),appropriatepsychometric
proceduresshouldalwaysbeusedtolink
scoresfromtwodifferentassessmentsifthescoresaretobe
compared.Tocomplywiththisstandard,ETSpsychometricstaffconductedaresearchstudytoexploreand
establishempiricalrelationshipsbetweenthescoresonthetwotests.Thisresearchwasdesignedandcarried 
outtoanswerthefollowingtworesearch
questions:

1. WhatTOEFLiBTsectionscoresarecomparabletoIELTSsectionscores?

2. WhatTOEFLiBTtotalscoresarecomparabletoIELTStotalscores?


LinkingTOEFLiBT™ScorestoIELTS®Scores–AResearchReport


3
FindingthecorrespondingTOEFLiBTscoresforeachoftheIELTSscoreswasaccomplishedbywhatisknown
asscorelinkingorscorecomparison(calledscorecomparisonhereafterinthisreport)inthefieldof
educationalmeasurement.

Inthisreport,thesamplesandtheanalyseswillbedescribedintheMethod
section.ThesectionforResults
willpresentscorecomparisonsforthefourskillsectionsandthetotalscores.Thisisfollowedbythe

Discussionsectionthatconsiderstheinterpretationoftheresultsandtheirimplicationsintermsofthe
generalizationofthefindingsfromthisstudy,evaluatesthelimitationsof
theresearch,andmakes
suggestionsforfutureresearch.

LinkingTOEFLiBT™ScorestoIELTS®Scores–AResearchReport


4
Method

DataCollection 

Inordertoacquireempiricalevidenceforlinkingthescoresonthetwotests,scoreswereobtainedfrom
studentswhohadtakenbothtests(Pommerich,2007).Thedata collectioneffortforthisresearchstartedin
2008andcontinuedinto2 009.Thefirstbatchofdata,obtainedin2008,
contained217studentswho
submittedtoETScopiesof theirTOEFLandIELTSscorereports.Inordertoincr easethesamplesizeandthe
stabilityoftheresults,ETSthencontactedalargenumberoftesttakersviae‐mailandencouragedthe
participantstosubmittheirlatestscoresonboth
teststoETS. Ove r1,000stude ntsrespondedandreported
theirscoresonthetwoteststoETS.Thetwobatchesofdatawerethencombinedforthestudy.

Beforeanyanalysesstarted,thefirsttaskwastocleanthedata,whichinvolvedremovingunusablerecords.
Recordsthathadmissingscores
ononeormoresectionsorthathadinvalidscoreswereremovedfrom
subsequentanalyses.Forexample,onereportedIELTSscoreof4.7wasidentifiedasaninvalidscoreandwas
eliminatedfromtheanalysisdataset
1
.Afterthisdatacleaningprocedure,1,1 53recordswereretainedfor

theanalysesdescribedinthenextsection.Closeto1,000recordsincludedtheparticipants’countryof
testinginformation,includingabout70countriesfromaroundtheworld.ThelargestgroupwasfromChina
includingHongKong(41%),followedbyUSA(6% ),
andJapan(5%).Theotherswereallbelow4%.

DataAnalysis

Tounderstandthestatisticalcharacteristicsofthescoresonthetwotestsfromthefinalsampleof1,153
studentsandtoestablish therelationshipsbetweenthescoresonthetwotests,thefollowingtypesof
analysiswereconducted.The
findingsfromtheanalysesarepresentedintheResultssection.

1.Descriptivestatisticswerecomputedtoevaluatesuchinformationastheaveragescoresoneachtestat
boththesectionandthetotalscorelevel,andhowmuchvariation(standarddeviation)wasobservedinthe
scores.Thecorrelationbetweenthescores
onthetwotestswasalsocomputedtoseetowhatextentthe
scoresonthetwotestswererelatedtoeachother.


1
IELTSscoreshave9bandsinone‐halfincrements(e.g.,3,3.5,4,4.5 )

LinkingTOEFLiBT™ScorestoIELTS®Scores–AResearchReport


5
2.Linkingscoresisbasicallyatransformationfromascoreononescaletoascoreontheotherscale
(Holland,2007)andcanbeaccomplished inavarietyofwaysdependingonthenatureofthedataandthe
purposeofthelinking(Kolen&Brennan,1995;Holland&Dorans,
2006).Tofindthecorrespondence

betweentheTOEFLiBTandIELTStestsforeachsectionandforthetotalscore,ETSpsychometricstaff
conductedextensiveanalysesbetweenMarchandAugust2009usingavarietyofstatisticalmethods
includingregression‐basedprediction,conditionalprobability‐basedscorematching,andequipercentile
linking.In
reviewingtheresultsandinconsiderationofthepurposeofthisresearch,wedeemeditmost
appropriatetousetheequipercentilelinkingmethod.Thisapproachhasbeenusedextensivelyinavarietyof
settings(e.g.thelinkingoftheSATtestwiththeACTtest) andallowedustoidentifythe
corresponding
TOEFLiBTscoreswhichwouldrestrict/passasimilarpercentageofthestudentsinthesampleaswouldthe
IELTSscores.Giventheneedforbothteststoscreenoutapplicantsinacomparablemanner,webelievethis 
tobethemostdefensibleandstraightforwardapproach.

Results

DescriptiveStatisticsoftheSample

TOEFLiBTscoresrangefrom0to30foreachsectionand0to120forthetotaltest;thetotaltestscoreisthe
sumofthefoursections.IELTSreportsbothsectionandtotalscoresona9‐bandscaleinone‐half(0.5)
band
increments;thetotaltestscoreistheaverageofthefoursections.Thisgeneralinformationaboutthe
differentscorescalesofthetwotestscanhelpputthescoresoftheresearchsampleinperspective.For
example,becauseofthecompactIELTSscorescales,IELTSscaleswillshowmuch
smallerscorevariation
(standarddeviationvaluesinTable1)thantheTOEFLiBTscales.Also,forsomeIELTSscoresinTables3to7,
thecorrespondingTOEFLiBTscoreswillshowscorerangesduetothelargerTOEFLscales.

Table1givesthesamplesize,themeanscores,andthescore
variations(standarddeviation,scorerange)of
thetwotests.OnIELTS,theReadingandListeningmeanscoreswereboth6.8,andtheSpeakingandWriting
meanswere6.3and6.1,respectively.Thetotaltestmeanscorewas6.6.OntheTOEFLiBTscale,theReading

andListeningmeanscoreswere
bothcloseto21andtheSpeakingandWritingmeanscoreswere20and
21.6,respectively.Thetotalmeanscorewas83.6.

Toevaluatetheoverallperformanceofthestudentsinthisresearchsample,the2009testperformance
informationonthetwotestswasobtainedfromtherespectivewebsitesand
ispresentedinthecolum nfor
2009populationmeaninTable1.

LinkingTOEFLiBT™ScorestoIELTS®Scores–AResearchReport


6

Generallyspeaking,thestudentsinthissample,asagroup,scoredhigheronbothteststhantheirrespectiv e
populationsdidin2009.Inotherwords,thissampleappearedtobemoreablethanthegeneralpopulations
ofIELTSandTOEFLiBT testtakers,respectively.Thiscouldbetheoutcome of
selfselectionofthe
participantsinthatstudentshavinghighertestscoresmighthavebeenmorewillingtoreporttheirscores
thanthosehavinglowerscores.Asaresult,therewereveryfewlowscoresinthedatabutrelativelymore
scoresatthemiddletomid‐high levels. Ourlevel
ofconfidenceinthe resultsforthelowscorelevelswould
thenberelativelylowerthanthatforthemiddletomid‐highlevels.

LinkingTOEFLiBT™ScorestoIELTS®Scores–AResearchReport


7
Table1.MeansandStand a rdDeviations
Test N Mean

Standard
Deviation
Score
Range
 2009
Population
Mean*
IELTSTotal 1,153 6.6 0.8 3.5‐9  5.8
IELTSListening 1,153 6.8 1.1 2.5‐9  6
IELTSSpeaking 1,153 6.3 0.9 1.5‐9  5.7
IELTSReading 1,153 6.8 1.1 2.5‐9  5.8
IELTSWriting 1,153 6.1 0.8 2‐9  5.4
  
TOEFLTotal 1,153 83.6 20.3 23‐119  79
TOEFLListening 1,153 20.9 7.6 1‐30  19.4
TOEFLSpeaking 1,153 20.0 3.8 5‐30  19.7
TOEFLReading 1,153 21.2 7.8 1‐30  19.9
TOEFLWriting 1,153 21.6 4.6 5‐30  20.5
*The2009populationmeanscoresbasedontheinformationprovidedattheTOEFLandIELTSwebsites.

Table2liststhecorrelationcoefficients betweenthescoresfromthesamemeasuresaswellasthescores
fromthetotaltests.Acorrelationcoefficientindicatestheextenttowhichthescoresonthetwotestsare
relatedtoeachother.Acorrelationcoefficientgoesfrom0fornorelation
atallto1foraperfectrelation.
Thecorrelationbetweentotalscoreswillalwaysbehigherthanthatbetweensectionscoresbecausetotal
scorescontainallthesectionscoreinformationandaremorereliable.Asexpected,thetotaltestscore
correlationis0.73,higherthananysectionscorecorrelation,which
variesfrom0.44forWritingto0.68for
Reading.Generallyspeaking,forthesco res tobelinked,moderatelyhighcorrelations(aro und0.7)wouldbe
verydesirable,butmoderatecorrelations(around 0.5to0.6)arenotatypical.






LinkingTOEFLiBT™ScorestoIELTS®Scores–AResearchReport


8
Table2.CorrelationsbetweenIELTSandTOEFLiBTSectionandTotalScores
Score Correlation
IELTSListening&TOEFLListening 0.63
IELTSSpeaking&TOEFLSpeaking 0.57
IELTSReading&TOEFLReading 0.68
IELTSWriting&TOEFLWriting 0.44
IELTSTotal&TOEFLTotal 0.73

Thenextsectionsprovidetheequipercentilelinking resultsforeachofthesectionsandthetotalscore.The
resultsarepresentedintabularform.Whilewewereabletoobtaincorrespondingscoresforthetwotestsat
eachoftheIELTSscorebands,thenumb er ofcasesperscoreband
varied,withgenerallyveryfewcasesin
theverylowestandhighestscorebands.Ingeneral,assamplesizesincrease,resultsbecomemorestable
andreliable.Consequently,inthisstudy,wehavemoreconfidenceintheresultswheretherearerelatively
largercasecounts(i.e.,5%ofthesample)perscore
bandandwerecommendcautionininterpretingthe
resultswherethecasecountsarerelativelysmall.Wehavenoted,byshadingtheareaoftherangeofscores
onthetables,wheresamplesizesperscorebandaregreaterthan5%ofthetota l.

ListeningSectionScoreComparisonResults


Table
3presentstheequipercentilelinkingmethodbasedscorecomparisonresultsonthetwoListening
sections.TheshadedareainTable3indicatesthatmoststudentshadscoresfrom5.5to8.5onIEL TSand7to
29onTOEFLiBTtest,withveryfewstudents(lessthan5%orso)at
theotherscorelevels.Aswasmentioned
earlier,duetothelargerscaleofTOEFLscoresthanIELTSscores,someIELTSListeningscoresarefoundto
correspondtoarangeofTOEFLscoresinsteadofasingleTOEFLscore.Forexample,anIELTSListeningscore
of5.5wouldcorrespondto
TOEFLiBTListeningscoresof7to11.Insummary,thetabulatedresultsindicate
thatanIELTSListeningscoreof6.0wouldcorrespondtoTOEFLListeningscoresof12to19;anIELTSListening
scoreof6.5wouldcorrespondtoTOEF LListeningscoresof20to23;anIELTSListeningscore
of7.0would
correspondtoTOEFLListeningscoresof24to26;andsoforth.Inallcases,theIELTS scoresandtheir
correspondingTOEFLscores(orscoreintervals)wouldpasssimilarpercentagesoftesttakers.


Table3.ListeningSectionScoreComparisons

LinkingTOEFLiBT™ScorestoIELTS®Scores–AResearchReport


9
IELTSScore
TOEFL
Score
9 30
8.5 29
8 28
7.5 27
7 24‐26

6.5 20‐23
6 12‐19
5.5 7‐11
5 4‐6
4.5 3
<=4 0‐2

SpeakingSectionScoreComparisonResults

Table4presentsthescorecomparisonresultsfortheSpeakingsections.Theinformationcanbeinterpreted
inthesamewayasdescribedaboveforListening.AsshownbytheshadedareainTable4,mostscoreswere
withintherangeof5.0to8.0on
theIELTStestand14to27ontheTOEFLiBTtest.Again,anumberofIELTS
scorescorrespondtoTOEFL scorerangesinsteadofsinglescoresduetoscaledifferencesbetweenthetwo
tests.Forexample,thetableshowsthatanIELTSscoreof5.0wouldcorrespondtoTOEFL
Speakingscoresof
14to15;anIELTSSpeakingscoreof5.5wouldcorrespondtoTOEFLSpeakingscoresof16to17;anIELTS
Speakingscoreof6.0wouldcorrespondtoTOEFLSpeakingscoresof18to19;anIELTSSpeakingscoreof6.5
wouldcorrespondtoTOEFLSpeakingscoreof
20to22;andsoforth.Inallcases,theIELTSscoresandtheir
correspondingTOEFLscores(orscoreintervals)wouldpasssimilarpercentagesoftesttakers.






LinkingTOEFLiBT™ScorestoIELTS®Scores–AResearchReport

10
Table4.SpeakingSectionScoreComparisons
IELTSScore TOEFLScore
9.0 30
8.5 28‐29
8.0 26‐27
7.5 24‐25
7.0 23
6.5 20‐22
6.0 18‐19
5.5 16‐17
5.0 14‐15
4.5 12‐13
<=4 0‐11

ReadingSectionScoreComparisonResults

Table5presentsthescorecomparisonresultsfortheReadingsections.Theinformationcanbeinterpretedin
thesamewayasdescribedaboveforListeningandSpeaking.AsshownbytheshadedareainTable5,most
scoreswerewithintherangeof5to8.5
ontheIELTStestand4to29ontheTOEFLiBTtest.Thetableshows
thatanIELTSReadingscoreof5.0wouldcorrespondtoTOEFLReadingscoresof4to7;anIELTSReading
scoreof5.5wouldcorrespondtoTOEF LReadingscoresof8to12;an
IELTSReadingscoreof6.0would
correspondtoTOEFLReadingscoresof13to18;anIELTSReadingscoreof6.5wouldcorrespondtoTOEFL
Readingscoresof19to23;anIELTSReadingscoreof7.0wouldcorrespondtoTOEFLReadingscoresof24to
26;andsoforth.Note
thataTOEFLReadingscoreof29wouldcorrespondtobothIELTSReadingscoresof8
and8.5.ThisoccurredbecausewewantedallIELTSbandscorepointstoberepresentedinthetable.Since
thehighestTOEFLiBTscore(30)hastomaptothehighestIELTSbandscore

(9)andpercentiledata suggest
thatanTOEFLiBTscoreof29mapstoanIELTSbandscoreof8,thenithastobethecasethatanTOEFLiBT
scoreof29alsomapsto8.5,i.e.,thereisnoTOEFLiBTscorepointbetween29and30to
maptotheIELTS
bandscoreof8.5.Inallcases,theIELTSscoresandtheircorrespondingTOEFLscores(orscoreintervals)
wouldpasssimilarpercentagesoftesttakers.


LinkingTOEFLiBT™ScorestoIELTS®Scores–AResearchReport


11
Table5.ReadingSectionScoreComparisons
IELTSScore TOEFLScore
9 30
8.5 29
8 29
7.5 27‐28
7 24‐26
6.5 19‐23
6 13‐18
5.5 8‐12
5 4‐7
4.5 3
<=4 0‐2

WritingSectionScoreComparisonResults
Table6presentsthescorecomparisonresultsfortheWritingsections.Theinformationcanbeinterpretedin
thesamewayasdescribedaboveforListening,ReadingandSpeaking.AsshownbytheshadedareainTable
6,mostscoreswerewithintherangeof5

to7onIELTSand 14to28ontheTOEFLiBTtest.Thetableshows 
thatanIELTSWritingscoreof5.0wouldcorrespondtoTOEFLWritingscoresof14to17;anIELTSWriting
score5.5wouldcorrespondtoTOEFLWritingscoresof18to20;anIELTSWriting
scoreof6.0would
correspondtoTOEFLWritingscoresof21to23;anIELTSWritingscoreof6.5wouldcorrespondtoTOEFL
Writingscoresof24to26;andsoforth.Inallcases,theIELTSscoresandtheircorrespondingTOEFLscores
(orscoreintervals)wouldpasssimilarpercentages
oftesttakers.






LinkingTOEFLiBT™ScorestoIELTS®Scores–AResearchReport


12
Table6.WritingSectionScoreComparisons
IELTSScore TOEFLScore
9.0 30
8.5 30
8.0 30
7.5 29
7.0 27‐28
6.5 24‐26
6.0 21‐23
5.5 18‐20
5.0 14‐17
4.5 12‐13

<=4.0 0‐11

LinkingTOEFLiBT™ScorestoIELTS®Scores–AResearchReport


13
TotalScoreComparisonResults

Table7presentsthecomparisonresultsforthetotalscoresonthetwotests.Thecomparisonoftotalscores
canbeinterpretedinthesamewayasdescribedaboveforthesectionscores.Asshownbytheshadedarea
Table7,mostscoreswerewithintherange
of5.5and7.5ontheIELTStestand46and109onTOEFLiBTtest.
Lessthan5%oftesttakersateachscoreorscoreintervalreceivedscoresoutsideofthisrange.Becauseof
themuchlargerTOEFLtotalscorescale(0to120)thanIELTS(0to
9),theTOEFLscorerangecorrespondingto
anIELTSscoretendstobeverybroad.Forexample,anIELTSscoreof5.5wouldcorrespondtoTOEFLscores
of46to59;anIELTSscoreof6.0wouldcorrespondtoTOEFLscoresof60to78;anIELTSscoreof6.5would

correspondtoTOEFLscoresof79to93;anIELTSscoreof7.0wouldcorrespondtoTOEFLscoresof94to101;
andsoforth.Inallcases,theIELTSscoresandtheircorrespondingTOEFLscores(orscoreintervals)would
passsimilarpercentagesoftesttakers.

Table7.TotalScore
Comparisons
IELTSScore TOEFLScore
9 118‐120
8.5 115‐117
8 110‐114
7.5 102‐109
7 94‐101

6.5 79‐93
6 60‐78
5.5 46‐59
5 35‐45
4.5 32‐34
0‐4 0‐31




LinkingTOEFLiBT™ScorestoIELTS®Scores–AResearchReport


14
Discussion

Tobetterunderstandthe scorecomparis onresultsonboththesectionandthetotaltestlevels,itis
importanttoevaluatetheresults inthecontextofthechara cterist icsofthetests,theresearchsample
employed,theintendeduseoftheresults,andthelimitationsofthestudy.Itisalso
worthwhiletothink
aboutwhatfurtherresearchcanbeconductedinthefuturetoenhancethequalityofthescorecomparisons.

EvaluationoftheScoreComparisonResultsandLimitations

Therearetwoconsiderations regardingtheinterpretationoftheresults.Thefirstisthecharacteristicsofthe
twotests.Itis
recognizedthattheTOEFLandIELTStests,whiletheyhavethesamenumberofsections
measuringsimilarskillsandunderlyingconstructs,werebuiltfromdifferentframeworksanddifferentte st
blueprints.Forexample,theTOEFLiBTtestisthelatestversionoftheTOEFLtestandisbuiltonthe
frameworkof

communicativecompetencybasedoncurrentsecondlanguageacquisitiontheo ryandpractice.
TheTOEFLiBTtestcontainsnumerousinnovativetasktypesthatcanonlybepresentedonacomputer‐based
testandtapintoawiderarrayofEnglishproficiencyskillsthanapaper‐basedtest.Asaresult,TOEFLiBT
scoresdonotmeanexactlythesamethingasIELTSscores,andmoderatecorrelations
2
betweenthescores,
asseeninTable2,canbeexpected.Aspreviouslynoted,forthescorestobelinked,moderatelyhigh
correlations(around0.7)wouldbeverydesirable,butmoderatecorrelations(around0.5to0.6)arenot
atypical.

Thesecondconsiderationisthesample.Asin manyotherstudies
ofthisnature,aself‐selectedgroupof
participantsvo lunteeredtoreporttheirIELTSandTOEFLscorestoETS,whichwouldlimitgeneralizingthe
resultstosamplesthatdiffersignificantlyfromthisresearchsample.Inparticular,thenumbersof
participantswithrela tivelylowandveryhighscores weresmall.Futuresamples
thatincludemorecaseswith
lowerandhigherscoreswouldberecommendedifanunderstan dingof themappingofscoresatthesescore
levelsisdesired.Whilethepresentcasecountsaresufficienttoevaluatethemappinginthemiddle ofthe
scorescales,cautionshouldbeexercisedwhencomparing
thelowerandhigherscores.Asdescribed
previously,thetablesincludeshadingtoindicatethosescorelevelswherewehaverelativelymore
confidenceinthescorecomparability.



2
Correlationcanbeusedasanindexthatisusedinevaluatingthepotentialforlinkingthescoresfromdifferenttests
thatareintendedtomeasuresimilarconstructsbutareconstructedaccordingtodifferentframeworks.

LinkingTOEFLiBT™ScorestoIELTS®Scores–AResearchReport

15
FutureResearch

Thescorecomparisonresultsfromthisresearchweregroundedintheempiricalinformationthatwas
obtainedfromthestudentswhotookbothIELTSandTOEFLiBTtests,andwerebasedonanappropriate
psychometricprocedurethatusedscoresfromthetwoteststoestablishtherelationshipsbetween thetwo

setsofscores.Asinanyotherscorecomparisonresearch,thecurrentresearchandthefindingsforthescore
comparisonbetweentheIELTSandTOEFLiBTtestswillbenefitfrommoreresearchinthefuture,especial ly 
fromfutureresearchstudiesbasedonmorecontrolledsamplesintermsofsamplesize,
sampleselection,
datacollection,andstudentbackgroundinformation suchasnationality,nativelanguage,yearsoflearning
English,etc.Withalargerandbettercontrolledsample,thedatacollectionqualitywillbeimprovedanda
sufficientnumberofstudentsacrossallscoreswillbeprovidedtoenhancethescorecomparisonresults.


LinkingTOEFLiBT™ScorestoIELTS®Scores–AResearchReport


16
Reference
AmericanEducationalResearchAssociation,AmericanPsychologicalAssociation,&NationalCouncilon
MeasurementinEducation.(1999).Standardsforeducationalandpsychologicaltesting.Washington,DC:
AmericanEducationalResearchAssociation.
ETS.(2002).ETSStandardsforqualityandfairness.Princeton,NJ:ETS.
Holland,P.W.(2007).Aframeworkandhistoryforscorelinking.In
N.J.Dorans,M.Pommerich,&P.W.
Holland(Eds.),Linkingandaligningscoresandscales(pp.5‐30).NewYork,NY:Springer.

Holland,P.W.,&N.J.(2006).Linkingandequating.InR.L.Brennan(Ed.),Educationalmeasurement(4the
ed.,pp.187‐220).Westport,CT:PraegerPublishers.
InternationalLanguage
TestingAssociation.(June2007).ILTAGuidelineforPractice.RetrievedMay12,2010
from />Kolen,M.J.,&Brennan,R.L.(1995).Testequatingmethodsandpractices.NewYork,NY:Springer.
Pommerich,M.,Hanson,B.A.,Harris,D.J.,&Sconing,J.A.(2000).Issuesincreatingandreporting
concordanceresultsbasedonequipercentilemethods(ACTResearchRep.No.2000‐1).IowaCity,IA:ACT.
Pommerich,M.(2007).Concordance:TheGood,theBad,andtheUgly.InN.J.Dorans,M.Pommerich,&P.
W.Holland(Eds.),Linkingandaligningscoresandscales(pp.199‐216).NewYork,NY:Springer.


Wang,L.,Eignor,D.R.,&Enright,M.K.(2008).AFinalAnalysis.InC.A.Chapelle,M.K.Enright,&J.Jamieson
(Eds.),BuildingaValidityArgumentforanewTestofEnglishasaForeign Language(pp.259‐318).NewYork,
NY:Routledge.




Copyright © 2010 by Educational Testing Service. All rights reserved. The ETS logo, LISTENING. LEARNING. LEADING. and TOEFL are registered
trademarks of Educational Testing Service (ETS). TOEFL iBT is a trademark of ETS. IELTS is a registered trademark jointly owned by the British Council, IDP
Education PTY Ltd, and University of Cambridge Local Examinations Syndicate.
