英语测试理论与实践

1宁波大学硕士研究生2008/2009学年第2学期期末考试答题纸考试科目：英语测试理论与实践课程编号：07G06329A姓名：卢晓娜学号：G08B05020130阅卷教师：李奕成绩：EnglishAssessment:TheoryandPractice1.Definethefollowingtermsinyourownwords1)TestingTestingwhichisoftenusedforpedagogicalpurposes,isaproceduredesignedtoelicitcertainbehaviorfromwhichonecanmakeinferencesaboutcertaincharacteristicsofanindividual.2)AssessmentAssessmentcollectsinformationofstudents’learningforvariouspurposesandusingvarioustechniques.3)EvaluationEvaluationappliesthesystematicwaystogatherinformationforthepurposeofmakingdecisions.4)MeasurementMeasurementistheprocessofquantifyingthecharacteristicsofpersonsaccordingtoexplicitproceduresandrules.5)ReliabilityReliabilityisaqualityoftestscores,andaperfectlyreliablescorewouldbeonewhichisfreefromerrorsofmeasurement.6)ValidityValidityisanimportantqualityoftestinterpretation,whichconcernstheextenttowhichtheinferencesordecisionswemakeonthebasisoftestscoresaremeaningful,appropriateanduseful.7)Norm-referencedtestsNorm-referencedtestsaredesignedtoenablethetestusertomaketheinterpretationsoftestresultswithreferencetotheperformanceofagivengroupornorm.8）criterion-referencedtestsCriterion-referencedtestsaredesignedtoenablethetestusertointerpretatestscorewithreferencetoacriterionlevelofabilityordomainofcontent.9）formativetestFormativetestisthetestteachersusetocheckontheprogressoftheirstudents,toseehowfartheyhavemasteredwhattheyshouldhavelearnedandusethisinformationtomodifytheirfutureteachingplans,suchasinformaltestsorquizzes.210)SummativetestSummativetestisusedattheendoftheterm,semesteroryearinordertomeasurewhathasbeenachievedbothbygroupsandbyindividuals.2．Discussionquestions(Answereachquestioninnolessthan200words.)1)Howcanweimprovethetestreliability?Inordertodevelopthetestreliability,wehavetominimizetheeffectsofmeasurementerrorsandtomaximizetheeffectsofthelanguageabilitieswewanttomeasure.Therearetwokindsofreliability:testreliabilityandscorerreliability.Andthreecategoriesaffecttestandscorereliability:1)testmethodfacets;2)attributesofthetesttakerthatarenotconsideredpartofthelanguageabilitieswewanttomeasure;3)randomfactorsthatarelargelyunpredictableandtemporary.Theresultoftheeffectsofallthesefactorsisthatwheneverindividualstakealanguagetest,theyarenotalllikelytoperformequallywell,andsotheirscoresvary.Becausedifferentfactorsaffectdifferentindividual,thetestersshouldusedifferentmethodsoftestingaccordingtothedifferentindividuals.Andweshouldusedifferentapproachedtoestimatingthereliabilityofthetests.Wecanalsoapplythetest-retestandsplit-halftesttoinvestigatethereliabilityinordertofurtherimproveit.2)Howcanweimprovethetestvalidity?Reliabilityisanecessaryconditionforvalidity.Aperfectlyvalidmeasuremustalsobereliable.Foratestscoretobevalid,itmustbereliable.Sotoincreasethetestvalidity,wemustfirstincreasethereliability.Therearefourkindsoftestvalidity:1)contentvalidity;2)criterion-relatedvalidity;3）constructvalidity;4)facevalidity.Inordertoimprovetestvalidity,wemustmakesurethetheoreticalframeworkusedtogeneratetheempiricalpredictionsiscorrect;second,themethodorprocedureusedtotestisappropriate;third,therelationshipamongmultipleindicatorsdesignedshouldrepresentagiventheoreticalconceptandtheoreticallyrelevantexternalvariablesshouldbesimilarintermsofdirection,strength,andconsistency.Forexample,twoindicators,bothofwhicharedesignedtomeasuresocialstatue,shouldhavesimilarcorrelationswithpoliticalinterest.3)IscommunicativelanguagetestingapplicableinChinesecontext?Whyandwhynot?CommunicativelanguagetestingisapplicableinChinesecontext.Communicativelanguagetestingisintendedtoprovidethetesterwithinformationaboutthetestee’sabilitytoperforminthetargetlanguageincertaincontext-specifictasks.Thetestsmeasurehowthetesteesareabletouselanguageinreallifesituations.Ashasbeensaid,thecommunicativelanguagetestingisjustapromisedlandinchinaasthereisnoreal-lifesituationinChinesecontext.Soweloseboththereliabilityandvalidity.However,theaimofthistestistomeasurethetestees’communicativeskills,sowecansetupareal-lifelikesituationandsamplesituationsinperformanceteststotestthestudents’communicative3competence.Insuchaway,thistestcouldbematchedtoabilities.SoinChinesecontext,westillcancreateareal-lifesituationtoletstudentstofinishthecommunicativetasks.4）Howcanwepromotethepositivewashbackoflanguagetesting?Therearenegativeandpositivewashbackoflanguagetesting.Topromotethepositivewashback,weshouldinhibittheformerandencouragethelater.Thenegativewashbackincludesthatteachersmayignoresubjectsandactivitiesthatdidnotcontributedirectlytopassingtheexam,andstudentsmaybepracticingexamtechniquesratherthanlanguagelearningactivities.Accordingtotheseproblems,thetestsshouldbedesignedtoreflectthelearningprinciplesandthecourseobjectivestowhichtheyaresupposedlyrelated.Testprogramshouldn’ttaketoomuchoftimeorelseitwillreducethetimeavailableforinstructionandnarrowcurricularofferings.What’smore,testshouldbeproperlydesignedtomotivatethestudentsandteacher,andtogivethemcorrectfeedbacks.5)Howwouldyouapplyperformanceassessmentinyourlanguageteachingclassroom?Inapplying