Brief introduction and evaluation questions of TOE

BriefintroductionandevaluationquestionsofTOEFLiBTPurpose:TheTOEFLInternet-basedtest(iBT),introducedinlate2005,measureshowwellthetesteesread,listen,speakandwriteinEnglishandusetheseskillstogetherintheuniversityclassroom,foradmissionofnon-nativespeakersofEnglishtohighereducationinstitutionsinNorthAmerica.Popularity:MoreinstitutionsacceptTOEFLtestscoresthananyothertestscoresintheworld—morethan7,000colleges,universitiesandlicensingagenciesinmorethan130countries,tobeexact.And,withTOEFLiBToffered30-40timesayearatmorethan4,500authorizedtestcentersworldwide,takingitisconvenientandeasy.Registration:online,byphone,bymailandinpersonatTOEFLResourceCenterswithlocalcurrency.Scores:ToearnascoreontheTOEFLtest,testeesmustfinishatleast1taskineachofthereading,listening,speakingandwritingsections.Onewillreceiveascoreonascaleof0—30foreachsection,atotalscoreoutof120fortheentiretest,feedbackonone’sperformanceandadescriptionoftasksthattesttakerscandowithinthescorerangeforeachlevel.Thereisnopassingorfailingscore;scorerequirementsareestablishedbyindividualinstitutionsoragencies.TOEFLiBTscoresarevalidfor2yearsafterthetestdate.Timelength:4hours.Reading:included3sets,eachofwhichcontained12–14itemsassociatedwithareadingpassageofabout700wordsinlength,designedtomeasurebasiccomprehension,inferencingandreadingtolearn.Thereare3typesofitems:traditionalsingle-answerfour-optionmultiplechoice;single-answerfour-choice“insertasentencewhereitbestfits”;completethesummarybyselectingthethreechoicesexpressingimportantideasinthepassage.Listening:consistedofsixsets,twobasedonconversationsandfourbasedonacademiclectures.Eachconversationwasabout3minuteslongandwasfollowedby5multiple-choiceitems,whileeachlecturewas3–5minuteslongandwasfollowedby6multiple-choiceitems.Itemtypesincludetraditionalsingle-answerfour-optionmultiple-choice;multiple-choicewithmorethanonecorrectanswer,sequencingitemsaccordingtotheorderofeventsorstepsinaprocess.Speaking:Thespeakingsectionconsistedof6tasks.2wereindependentspeakingtasks,whichrequiredexamineestoexpressopinionsonfamiliartopics.Theother4wereintegratedtasks,wheretesteesspeakbasedonsomethingheardorread.Foreachtask,examineesweregiven15–30secondstoprepare,and45–60secondstorespond.Thissectionmeasurestheabilitytospeakabouteverydayfamiliartopics,andtosummarize,evaluateandcompareinformationinaspontaneousmanner.Eachexaminee’sresponsetoeachtaskwasscoredonadiscretescaleof0–4bytrainedraters.Thefinalscoreonthetaskwasthescoregivenbythesingleraterortheaveragescoresgivenbytworaters.Writing:includedtwotasks.Onewasanindependentwritingtaskthatrequiredexamineestosupportanopiniononacommontopic.Theotherwasanintegratedwritingtaskthatrequiredexamineestorespondtoaquestiononwhattheyhadreadandheard.Eachexaminee’sresponsetoeachtaskwasscoredona0–5scalebytwotrainedraters.Thefinalscoreontheitemwastheaverageofthescoresofthetworaters.Ifthetworaters’scoreswerediscrepantbymorethanonepoint,athirdscorebyanotherraterwasobtainedandusedtodeterminethefinalscore.EvaluationoftheconstructvalidityofTOEFLiBTTOEFLiBTisbasedonatheoryofcommunicativecompetence,whichincludescomplexabilitiesneededforaparticularrangeofgoalsinrelevantcontexts.Onthewhole,TOEFLiBThasachievedthis,withaclearerfocusontheacademicenvironment.Myevaluationwillfocusonthefollowingquestions:1)Isthelanguageabilityconstructforthetestrelevanttothepurposeofthetest?2)Towhatextentdoesthetesttaskreflecttheconstructdefinition?3)Towhatextentdothescoringproceduresreflecttheconstructdefinition?4)Willthescoresobtainedfromthetesthelpustomakethedesiredinterpretationsabouttesttakers’languageability?Theabovefourareconcernedwiththeappropriatenessofthetaskcharacteristicswithrespecttotheconstructdefinition.5)Whatcharacteristicsofthetestsettingarelikelytocausedifferenttesttakerstoperformdifferently?Thefifthoneisconcernedwithwhetheralltesteeshaveanequalopportunitytoperformattheirbest.