Language-Testing-and-uation.ppt

LanguageTestingandEvaluation IntroductiontoLanguageTesting Testingisanindispensablepartofeveryteachingandlearningexperience Rightfromthecommencementofformaleducation weareconfrontedwithamyriadoftests fromsmallquizzes mid termandfinalexaminations torigidmatriculationexaminations nottomentionmanyothertestsafterourgraduation onthegradesofwhichhangdecisionsconcerningourcareer Asamatteroffact wearehauntedbyteststhroughoutourlifetime Whatisa test Apsychologicaloreducationaltestisaproceduredesignedtoelicitcertainbehaviorfromwhichonecanmakeinferenceaboutcertaincharacteristicsofanindividual Carroll 考試是由一定數(shù)量的題目組成 用于測(cè)定人的某一心理品質(zhì) 或?qū)W習(xí)掌握程度的工具 教師出試題實(shí)際上是創(chuàng)設(shè)一個(gè)語言情景 學(xué)生的答題過程則是他的語言能力和知識(shí)掌握程度的一種表現(xiàn) 考試的目的就是要測(cè)定學(xué)生在某一方面的能力或?qū)W習(xí)情況 舒運(yùn)祥2001 AssessmentToputitsimply assessmentmeanstodiscoverwhatthelearnersknowandcandoatacertainstageofthelearningprocess Withdifferentassessmentpurposesinmind wemayadoptdifferentassessmentmethods Aclosestudyontheassessmentpurposeswillmakeitclearthatallthepeopleinvolvedineducationhavesomereasonstoconsiderassessmentnecessary Theyareadministrators teachers parentsandstudents Administrators Administratorsprovidemoneyandpersonnelforeducation Theyneedtoknowwhethertheprogrammestheyhaveplannedareworkingwell Theonlywaytodothisistodiscoverhowwellthepupilsaredoingwiththeircourses Ifassessmentresultsaredifferentfromwhattheyhaveexpected theyneedtoreplantheprogrammessothatbetteraccomplishmentcanbeachievedlater Teachers Teachersputtheadministrators plansintopractice Inthecontinuumoflearning teachersneedtoknowwhathasbeendoneandwhatneedstobedonenext whatthepupilsalreadyknoworcandoandwhattheydonotknoworcannotdoyet Iftheprogramiswellplanned itprettymuchdependsontheteachers performancewhethertheprogrameventuallyleadstosuccess Parents Nobodyismoreanxiousthanparentstoknowhowtheirchildrenaredoinginschool Unabletowatchtheirkidsintheclass parentsvaluethefeedbackabouttheirchildren sperformancefromtheteachersandtheschool Students Finally studentsneedtoknowwhatthey veaccomplished beawareofwhattheyneedtoworkontext andbuilduptheirconfidenceandsatisfactionfromwhattheyhaveachieved Reasonsfortest assessmentuse Toknowpeople slanguageability Todiscoverhowsuccessfulstudentshavebeeninachievingtheobjectiveofacourseofstudy Toprovidefeedbackforlearners Toevaluatetheeffectivenessofteaching Todiagnosestudents strengthsandweaknesses Toplacestudentsinanappropriateprogram 語言能力 興趣 態(tài)度等 考試 觀察 面談 數(shù)值 價(jià)值 判斷 測(cè)量 標(biāo)準(zhǔn) 量化 評(píng)價(jià)過程 考試 測(cè)量和評(píng)價(jià)三者之間既有聯(lián)系又有區(qū)別 評(píng)價(jià)過程實(shí)際上包含了測(cè)量 如果測(cè)量以考試為手段 那么評(píng)價(jià)中也包含了考試 外語教學(xué)中為了了解學(xué)生的語言能力 興趣 知識(shí)掌握得情況 就要對(duì)其進(jìn)行評(píng)價(jià) 但能力 興趣等心理品質(zhì)是一個(gè)抽象概念 為了有利于評(píng)價(jià) 我們通過具體做題目 考試 等方式 對(duì)抽象的概念進(jìn)行定性 定量的描述 最后還需借助一定的標(biāo)準(zhǔn)解釋考試成績 以確定學(xué)生的語言水平 這樣一個(gè)過程就是評(píng)價(jià) Test Measurement evaluation Assessment 外語測(cè)試 語言學(xué) 心理學(xué)與教學(xué)法 理論語言學(xué) 應(yīng)用語言學(xué) 心理語言學(xué) 社會(huì)語言學(xué) 學(xué)習(xí)心理學(xué) 外語教學(xué)法 測(cè)量學(xué) 教育統(tǒng)計(jì)學(xué) 教育測(cè)量學(xué) 計(jì)算機(jī)科學(xué) 外語測(cè)試與有關(guān)學(xué)科 TeachingandTestingTestingandteachingarecloselyinterrelated Languagetestingbothservesandisservedbyresearchinlanguagelearningandlanguageteaching Theeffectoftestingonteachingandlearningisknownasbackwash Backwashcanbeharmfulorbeneficial Well madetestsofEnglishcanbringaboutbeneficialbackwashwhileill designedoneswillleadtomisguidedteachingandlearning Goodtestscanhelpcreatepositiveattitudetowardslearning Goodtestscanhelpfosterpositiveclassroomatmosphereandhelpthemmasterthelanguage TeachingObjectivesTeachingContentTeachingMethodsTesting Notalltests however aresufficienttomeetourrequirement i e toprovideanaccuratemeasurementofthecandidates languageproficiencyfortheirspecificuses Moreoftenthannot acandidate sperformanceonalanguagetestdoesnotmatchhisperformanceinrealuse Ifblindfaithisplacedinthetestresultregardlessofthevalidityandreliabilityofthetestforaspecificsituation itwillgiverisetosevereproblems Whenthetestscoreisregardedastheonlyyardstickofaperson slanguageproficiency studentswilltendtostudytesttakingskillsratherthanthelanguageitself Theybecomefiendsattestinsteadofmastersofthelanguage Thenweareboundtofallintotheexamination centeredteaching whichshouldbeavoided Clearly thisisundesirable Teachingcontents TeachingMethods Testing KindsofTestsandTestingTestscanbecategorizedfollowingdifferentcriteria suchastheintendeduseofthetest theapproachtotesting testingcontent thewaytointerpretthescoreandscoringmethods First testsareusedtoobtaininformation Givingatestislikeaskingaquestion itisarequestforinformation Ifyouasktherightsortofquestion yougettherightsortofanswer Ifyouaskasillyquestion yougetasillyanswer Inthiswaytheycanbecategorizedintoproficiencytests achievementtests diagnostictests placementtestsandaptitudetests Proficiencytests 水平測(cè)試 Proficiencytestsaredesignedtomeasurepeople soverallmasteryofEnglishortomeasurehowsuitablecandidateswillbeforperformingacertaintaskorfollowaspecificcourse TOEFL EPT NMETetc Forexample doesthestudentknowenoughEnglishtofollowacertainuniversitylecturegiveninthemediumofEnglish DoesthestudentknowenoughEnglishinordertofunctionefficientlyinaparticulartypeofemployment Thecontentofaproficiencytestisnotbasedonthecontentorobjectivesoflanguagecourseswhichpeopletakingthetestmayhavefollowed Itisinnowayrelatedtoanysyllabusorteachingprogram indeed manyproficiencytestsareintendedforstudentsfromseveraldifferentschools countriesandevenlanguagebackgrounds Achievementtests 成績測(cè)試 Achievementtestsaredirectlyrelatedtolanguagecourses theirpurposebeingtoestablishhowsuccessfulindividualstudents groupsofstudents orthecoursesthemselveshavebeeninachievingobjectives Theyareoftwokinds finalachievementtestsandprogressachievementtests Finalachievementtestsareusuallyformalexaminations givenattheendofacoursestudyorattheendoftheschoolyear Oftenittakestheformofanexternaltestwhichissetbyofficialexaminingboards ministriesofeducation orbymembersofteachinginstitutions Progressachievementtestsareintendedtomeasuretheprogressthatstudentsaremaking Theyareusuallydesignedbyteacherstochecktheextenttowhichthestudentshavemasteredthematerialtaughtintheclassroom Beingthemostwidelyusedtypesoftests progresstestsareusuallyaneffectiveteachingdevicetomotivatelearningandreinforcewhathasbeentaught Diagnostictests 診斷測(cè)試 Diagnostictestsareusedtoidentifystudents strengthsandweaknesses Theyareintendedprimarilytoascertainwhatfurtherteachingisnecessary Inourteaching wemaysometimesbetemptedtoconcentrateonfollowingthesyllabusandignoretheneedsofourstudents However ifwedothis alotofourstudentsmayfail howeverthoroughlywehavecoveredthesyllabus Itisveryimportanttotakeintoaccounttheneedsofourstudentsateverystageinourteaching Justasitisnecessaryfordoctorstodiagnoseanillnessinordertocuretheirpatients soteachersmustdiagnoseproblemsinordertoteacheffectively Veryfewtestsareconstructedforpurelydiagnosticpurposes Usuallyadiagnostictestformspartofanothertypeoftest especiallyaclassroomprogresstest Achievementandproficiencytestsarealsofrequentlyusedfordiagnosticpurposes areasofdifficultyarediagnosedinsuchtestssothatappropriateremedialactioncanbetakenlater Placementtests 摸底測(cè)試 Placementtestsareintendedtoprovideinformationwhichwillhelptoplacestudentsatthestage orinthepart oftheteachingprogrammostappropriatetotheirabilities Typicallytheyareusedtoassignstudentstoclassesatdifferentlevels PlacementtestsareusuallygeneralandconcentrateontestingawideandrepresentativerangeofabilityinEnglishinordertolookforwardtothelanguagedemandswhichwillbemadeonstudentsduringthecourse AptitudeTests 學(xué)能測(cè)試 AptitudeTestisdesignedtopredictastudent slikelihoodofsuccessandeaseinlearningaforeignlanguage Languagelearningaptitudedoesnotrefertowhetherornotanindividualcanorcannotlearnaforeignlanguage itisassumedthatvirtuallyeveryonecanlearnaforeignlanguagegivenadequateopportunity Languagelearningaptitudereferstothe predictionofhowwell relativetootherindividuals anindividualcanlearnaforeignlanguageinagivenamountoftimeandundergivenconditions 例如 本子在盒子上 Pasyhojtzov 盒子在桌子上 Pasydeupthoj 椅子在桌子下 Pasydeupfgrip 筆在盒子里 Pasyhojcmak 書在椅子下 Pasygripftjav 1 書在桌子上 A Pasydeupttjav B PasydeupftjavC Pasygripfdeupt D Pasytjavtgrip 2 筆在本子下 A Pasyhojtmak B Pasyzovtmak C Pasyzovfmak D Pasyhojfmak 3 本子在盒子里 A Pasyhojczov B Pasyhojctjav C Pasyhojtzov D Pasyhojttjav HistoryofTestingModernlanguagetestinghasahistoryofabout100years Broadlyspeaking thedevelopmentoftestingcanbedividedintothreestages thepre scientificstage thescientificstageandthecommunicativestage ThePre scientificStage Thenineteenthcenturyandtheearlydecadesofthetwentiethcentury basically intuitive orsubjective dependingonthepersonalinclinationoflanguageteachers Studentsevaluatedinavarietyofways notonlyonskillsinusingthelanguagebutalsoonfactsaboutEnglish Languageknowledgeoftentakingupheavyweighting studentshadtolabelpartsofasentenceandmemorizenumerouslistsoflanguagepatterns Abundantwritinginvariousformsincludingtranslation essay dictation precis andopen endedanswersbasedonreadingcomprehension Westillseesomeofthesetestingformsinuseandtheyturnouttobequitesoundforadvancedstudents TheScientificStage Subjectivetestsinthepre scientificstageattackedforlackofreliability astrongneedforteststhatwouldbeobjectiveandfairer Testingspecialistswithlinguistictrainingenteredthescene Languagetheoriessuggestedthatlanguagecouldbebrokenintosub skillsandcomponentsandcouldbeevaluated scientifically bitbybit Objectivetestsweredevisedthatmeasuredperformanceorrecognitionofseparatesounds specificgrammaticalfeatures orvocabularyitems Subjectivewrittentestsbegantobereplacedbyobjectivetests Answersbeingstandardized scoredevenbyuntrainedpeople Specialistsstartedtoevaluatetestsstatistically lookingattheeffectivenessofeachquestionandmeasuringtheexamination sreliabilityandvalidity TheCommunicativeStage Teststodaymainlyconcernedwithevaluatingrealcommunication abroaderviewoflanguageability Notonlymeasurelanguageskillsbutalsocommunicativecompetence Thebestexams combinevarioussub skills Tomeasuremorethanisolatedlanguageskills howwellapersoncouldfunctioninhissecondlanguage Tousethebestfeaturesofsubjectiveandobjectivetesting Theclozetestreflectsthiscompromise Fromastoryoressay wordsareremovedatregularintervals Afterconsideringthecontext theexamineehastofillinthemissingwords Thus thetaskisholistic thatis grammarandvocabularyandoverallmeaningaretestedsimultaneously Butthescoringisquiteobjective Inresponsetothesebroaderviewsofcommunicativelanguageabilityandcommunicativelanguageuse mucheffortisbeingdirectedtowarddevelopingtests thatnotonlymeasureawiderangeoflanguageabilities includinggrammatical discourse sociolinguistic andstrategiccompetencies butthatarealso authentic inthattheyrequiretesttakerstointeractwithandprocessboththeexplicitlinguisticinformationandtheimplicitillocutionaryorfunctionalmeaningofthetestmaterial CharacteristicsofDiscretepointtests 分離式測(cè)試 Integrativetests 綜合性測(cè)試 andCommunicativetests 交際性測(cè)試 StructuralisttestingapproachCorrespondingtothetraditionalviewonthenatureoflanguageandlanguagelearning conventionaltestsadoptthestructuralistapproachtotesting Whatistobetestedisrevealedbyastructuralcontrastiveanalysisbetweenthetargetlanguageandthelearner smothertongue Morrow Basedonthecontrastiveanalysescarriedoutatallthelevelsofstructure grammatical syntactic phonological testitemsareconstructed Thesameapproachisadoptedastohowtotest Languageisbrokendownintoseparateelementsanddiscreteitemsareconstructed eachofwhichideallyrevealsthetesttaker smasteryofonelevelofthelanguageintermsofoneofthefourskills listening speaking readingandwriting Norm referencedConventionaltestsrelyonquantitativeassessment Followingthebehavioristviewoflanguagelearningthroughhabitformation testsposequestionstoelicitresponseswhichshowwhetherornotcorrecthabitshavebeenestablished Correctanswersarerewardedandnegativeonespenalizedinsomeway Passingatestinvolvesmakingaspecifiedproportionofcorrectresponses Consequently scoresofconventionalteststendtorevealsimplythequantityofitemsansweredcorrectly Nothingcanbeinferredaboutthequalityofthetesttaker slanguageperformance 分離式測(cè)試的理論基礎(chǔ) 1 結(jié)構(gòu)主義語言學(xué) 教學(xué)內(nèi)容可被分解成語音 詞法和句法 詞匯等方面 語言能力可分解為聽說讀寫等方面 2 行為主義學(xué)習(xí)論 復(fù)雜的學(xué)習(xí)行為可分解為最基本的單元 刺激 反應(yīng) S R 的聯(lián)結(jié) 新行為主義代表斯金納提出 刺激 反應(yīng) 強(qiáng)化 3 聽說教學(xué)法 聽說領(lǐng)先 反復(fù)實(shí)踐形成習(xí)慣 以句型為中心 排斥或限制母語 對(duì)比語言結(jié)構(gòu) 確定教學(xué)難點(diǎn) 及時(shí)糾正錯(cuò)誤 培養(yǎng)正確的語言習(xí)慣和廣泛利用現(xiàn)代化教學(xué)手段 分離式測(cè)試的特點(diǎn)和問題 1 個(gè)人的語言技能發(fā)展是不平衡的 語言測(cè)試可以從聽說讀寫四個(gè)方面分別進(jìn)行 2 語言是有結(jié)構(gòu)并按層次構(gòu)成的 測(cè)試可按層次由低到高進(jìn)行 如對(duì)聽能的測(cè)試可分為 1 音素 元音 輔音 語調(diào) 重音等 2 詞匯 3 句子 4 小段對(duì)話或短文等 3 語言情景是無限的 語言成分是有限的 測(cè)試成分更有效 4 母語和外語在相互轉(zhuǎn)換時(shí) 其形式 結(jié)構(gòu)和意義等方面由一定的不同之處 也是測(cè)試重點(diǎn) 5 試題采用只有一個(gè)正確答案的題型 以保證客觀性與科學(xué)性 6 評(píng)分采用客觀打分 任何人閱卷結(jié)果都一致 分離式測(cè)試以行為主義學(xué)習(xí)理論 結(jié)構(gòu)主義語言學(xué)和聽說教學(xué)法的理論為依據(jù)分別對(duì)考生的語音 詞匯 語法等語言知識(shí)和聽說過讀寫等方面的語言技能進(jìn)行測(cè)試 題量大 每題的測(cè)試目標(biāo)明確 有助于教師了解學(xué)生在學(xué)習(xí)中的長處和不足 測(cè)試數(shù)據(jù)能量化分析 打開科學(xué)測(cè)試的一頁 它的最大弱點(diǎn)是重語言形式而輕語言應(yīng)用 把現(xiàn)實(shí)中的語言分解成無上下文的單句 其測(cè)試結(jié)果很難反映學(xué)生實(shí)際應(yīng)用語言能力 正是由于這些弱點(diǎn) 使綜合測(cè)試和交際性測(cè)試應(yīng)運(yùn)而生 He sgoingtocallusback A won theB didn theC doesn theD isn the Didn tyougooutyesterday A Yes Ididn tB Yes IdidC No IdidD No Iwentout IntegrativetestingIntegrativetesting bycontrast requiresthecandidatetocombinemanylanguageelementsinthecompletionofatask Thismightinvolvewritingacomposition makingnoteswhilelisteningtoalecture takingadictation orcompletingaclozepassage 綜合性測(cè)試的理論基礎(chǔ) 1 語用期待語法假設(shè) 美國語言學(xué)家奧勒 JohnW Oller jr 認(rèn)為交際過程中 語境對(duì)于確定語音或詞義有著至關(guān)重要的作用 正常的語言使用過程中存在兩種語境 語言自身語境 linguisticcontext 和語言外語境 extralinguisticcontext 言語和體勢(shì)語構(gòu)成語言自身語境 語言知識(shí) 語言外的現(xiàn)實(shí)世界構(gòu)成語言外語境 語境中語言加工能力為 語用期待語法 規(guī)則 奧勒認(rèn)為一個(gè)人的語言能力是由這樣一個(gè)期待語法生成系統(tǒng) expectancygeneratingsystem 構(gòu)成的 完型填空和聽寫是最有效的語用測(cè)試形式 e g 1 Hellen willyoubeatthepartytonight Yes butIhavesomuchhomeworktodothatIreally A can tB mustn tC won tD shouldn t2 Wasitthere youwereawaytoanswerthephone There snodoubtaboutit A thatB whichC whileD where context A Whatareyoulookingfor B Oh mygod Ican tfindmywalletanywhere IputitjustherebeforeIwentintoanswerthephone A Wasittherewhileyouawaytoanswerthephone B Thereisnodoubtaboutit Lookatthefollowingdialogue 1 A Canyoutellmethetime B Well themilkmanhasjustcome 2 A Thereisthetelephone B Iaminthebath A OK 3 A Whatarethepolicedoing B I vejustarrived 4 A CanIbuyyouadrink B Actually I dratherhavethemoney 5 A IthinkIcouldmakeyouveryhappy B Why Areyouleaving 2 格式塔心理學(xué) 強(qiáng)調(diào)感知的整體性 認(rèn)為整體不等于部分之和 整體是先于部分而存在 并制約著部分的性質(zhì)和意義 綜合性測(cè)試特點(diǎn) 綜合性測(cè)試也稱語用測(cè)試 pragmatictest 其測(cè)試方法正好與分離式相反 綜合性測(cè)試將人的語言能力視為一個(gè)整體 每次對(duì)考生的多種能力的多個(gè)方面同時(shí)進(jìn)行測(cè)試 這類考試始終在語境中測(cè)試學(xué)生使用語言的能力 因此結(jié)果更接近于考生實(shí)際使用語言能力 以完型和聽寫為主要形式的綜合性測(cè)試仍屬間接測(cè)試 無法直接測(cè)出考生的語言表達(dá)能力 需與其它測(cè)試結(jié)合使用 才能取得滿意效果 ThecommunicativetestingThecommunicativeapproachtolanguagetestingissometimeslinkedtotheintegrativeapproach However althoughbothapproachesemphasizetheimportanceofthemeaningofutteranceratherthantheirformandstructure thereareneverthelessfundamentaldifferencesbetweenthetwoapproaches Communicativetestsareconcernedprimarily ifnottotally withhowlanguageisusedincommunication Successisjudgedintermsoftheeffectivenessofthecommunicationratherthanformallinguisticaccuracy Inpractice however sometestsofacommunicativenatureincludethetestingofusageandalsoassessabilitytohandletheformalpatternsofthetargetlanguage Indeed fewsupportersofthecommunicativeapproachwouldarguethatcommunicativecompetencecouldeverbeachievedwithoutaconsiderablemasteryofthegrammarofalanguage 交際性測(cè)試的理論依據(jù)是社會(huì)語言學(xué)和交際法 也稱意念 功能法 以社會(huì)語言學(xué)為理論基礎(chǔ)的交際教學(xué)法提出以語言功能項(xiàng)目為綱 以培養(yǎng)交際能力為本的教學(xué)法 海姆斯 D H hymes 認(rèn)為 交際能力應(yīng)體現(xiàn)在四個(gè)方面即 形式正確 合理可行 內(nèi)容得體 實(shí)際使用 語言使用時(shí)應(yīng)遵循 得體法則 ruleofappropriateness 因此交際能力強(qiáng)的人的語言應(yīng)該語法正確 符合人的感知特點(diǎn) 內(nèi)容得體 語言表達(dá)的是人們生活中真實(shí)的文化行為 分離式和綜合測(cè)試無法測(cè)定真實(shí)語言交際能力 交際性測(cè)試的特點(diǎn) 是行為測(cè)試 在真實(shí)或接近于真實(shí)情景中 通過完成任務(wù)來測(cè)試語言掌握程度 因此又稱 基于任務(wù)的 task basedtests 考試 如 考生聽一段錄音后寫一份電傳回電 將聽與寫結(jié)合起來 雖然有其優(yōu)點(diǎn) 但考試材料 語境和任務(wù)的真實(shí)性仍然有爭(zhēng)議 同時(shí)作為主觀測(cè)試 評(píng)分的客觀性問題阻礙它在大規(guī)?？荚囍械膽?yīng)用 但目前 有人將多項(xiàng)選擇題引入到交際測(cè)試 如 Youwereapplyingtoauniversityandneededaletterofrecommendation Youwenttoaprofessor whowasalsoyourfriend andsaid A I dappreciateitifyoucouldwritealetterofrecommendationforme B Iwanttoaskyoutowritealetterofrecommendationforme C Iwonderifyoucouldwritealetterrecommendationme D Hey givemerecommendationletter 正確 得2分 語法正確 不夠得體 得1分 較得體 語言有誤 得1分 語言有誤 內(nèi)容不得體 不得分 ContrastofCommunicativeandTraditionalApproacheswithTesting 1 Whatislanguage C asystemforcommunicatingmeaning T asystemofknowledge2 Whatislanguagelearning C learntocommunicatebyusingT learngrammarrulesandstructures3 pedagogicconcerns C communicativecompetence meaning contentT linguisticcompetence form 4 Pedagogictechniques C analytic top down roughly tunedinput interactionT synthetic bottom up finely tunedinput transmission5 Affective C Learner centered initiator performer problem solver T Teacher centered controller knowledge imparter assessor 6 testing C Integrative Holistic Criterion referenced Production Communicativecompetence Function UseRawdata SurfacestructureT Analytical Discretepoint Norm referenced Reception Linguisticcompetence Form Usage Idealization Deepstructure Norm referencedtestingNorm referenced NR testsaredesignedtoenablethetestusertomake normative interpretationsoftestresults Thatis testresultsareinterpretedwithreferencetotheperformanceofagivengroup ornorm Soastudent sperformanceiscomparedwiththeperformancesoftheotherstudentsinthesamegroup Forexample isthestudentinthetoporinthebottompartofthegrouporinthemiddleofthegroup Howmanystudentsinthegrouparebetterthanthestudentandhowmanyareworse Theperformancecharacteristics ornorms mosttypicallyusedasreferencepointsaretheaveragescoreofthegroup Criterion referencedtestingCriterion referenced CR testsaredesignedtoenablethetestusertointerpretatestscorewithreferencetoacriterionlevelofabilityordomainofcontent Anexamplewouldbethecaseinwhichstudentsareevaluatedintermsoftheirrelativedegreeofmasteryofcoursecontent ratherthanwithrespecttotheirrelativerankingintheclass Thus allstudentswhomasterthecoursecontentmightreceivean A irrespectiveofhowmanystudentsachievethisgrade Inotherwords weuseCRteststofindoutwhetherastudentcanperformaparticulartaskornot Forexample canthestudentwritealetteraskingforinformationaboutsomething 常模參照考試 Norm referencedtests 主要用于甄別 分類 選拔等 例如編班考試 外語競(jìng)賽 中考和高考等 標(biāo)準(zhǔn)參照考試 criterion referencedtests 常用于各類證書考試 如高中會(huì)考 期末考試 大學(xué)四 六級(jí)等 QualitiesofTestsToevaluateatest wemustinthefirstplacebeclearofthebasicqualitiesthatmakeatestuseful BachmanandPalmer Testusefulnessincludessixtestqualities reliability validity authenticity interactiveness impact andpracticality Reliability 信度 Reliabilityisoftendefinedasconsistencyofmeasurement Areliablescoreisfreefromerrorsofmeasurementandconsistentacrossdifferentcharacteristicsofthetestingsituation Forexample ifthesametestweretobeadministeredtothesamegroupofindividualsontwodifferentoccasions intwodifferentsettings its