英语教育测量与评价

研究生学位课程试卷院（系、所）外国语学院专业英语考试科目英语教育测量与评价第二学期研究生姓名戎竞雄学号132300176考试成绩导师评语：导师签字年月日说明一、凡学位课程考试试 题 快递公司问题件快递公司问题件货款处理关于圆的周长面积重点题型关于解方程组的题及答案关于南海问题 、试卷必须与本封面一起装订。阅卷导师务必用红笔批卷，并在本封面规定位置打分、写完评语后在二周（论文考试一个月）内交院（系、所）办公室教务员，教务员及时做好成绩登记，在学期结束前或第二学期初将成绩单交研究生处统一整理归档。试题、试卷由院（系、所）办公室保管。二、学位课程考试用纸除计算机专用打字纸、16开小方格稿子纸外，一律使用研究生处统一印制的“学位课程考试纸”。三、该封面请用A4纸双面打印，将此说明打印于封面背面。上海师范大学 标准 excel标准偏差excel标准偏差函数exl标准差函数国标检验抽样标准表免费下载红头文件格式标准下载 试卷2013~2014学年第二学期考试日期2014年8月日科目：英语教育测量与评价学科教学＋课程与教学论专业硕士13 年级 六年级体育公开课教案九年级家长会课件PPT下载六年级家长会PPT课件一年级上册汉语拼音练习题六年级上册道德与法治课件 姓名戎竞雄学号_132300176__项目一二三四五六七八总分分值4555100得分我承诺，遵守《上海师范大学考场规则》，诚信考试。签名：__戎竞雄_____I.Answerindetailthefollowingquestions:(45%,15pointsforeach)1.Suggestthedifferencesbetweenproficiencytestsandachievementtests.Giveexamplesifnecessary.Answer:Aproficiencytestassessesthegeneralknowledgeorskillscommonlyrequiredforentryintoagroupofsimilarinstitutions.OneexampleisTOEFL.Proficiencytestsarenorm-referencedtestsbecauseNRTshaveallthequalitiesdesirableforproficiencydecisions.Whileanachievementtestmustbedesignedwithveryspecificreferencetoaparticularcourse.Andtheachievementtestsareoftendirectlybasedoncourseobjectivesandwillthereforebecriterion-referenced.Suchtestswilltypicallybeadministeredattheendofacoursetodeterminehoweffectivelystudentshavemasteredtheinstructionalobjectives.Achievementtestsmustbenotonlyveryspecificallydesignedtomeasuretheobjectivesofagivencoursebutalsoflexibleenoughtohelpteachersreadilyrespondtowhattheylearnfromthetestaboutthestudents'abilities,thestudents'needs,andthestudents'learningofthecourseobjectives.Oneexampleisthetestsattheendofthecourse.2.Thefollowingaretwodifferentkindsofscoredistribution..Whatinformationdothesetwofiguresconveyus?(DiscussfromthescoredistributionofNRTandthatofCRT)Answer:Thefirstfigureconveysusthatitisanorm-referencedtest,whichisdesignedtomeasuregloballanguageabilities(forinstance,overallEnglishlanguageproficiencyincludinglisteningability,readingcomprehension,andsoon).Eachstudent'sscoreonsuchatestisinterpretedrelativetothescoresofallotherstudentswhotookthetest.Suchcomparisonsareusuallydonewithreferencetotheconceptofthenormaldistribution(familiarlyknownasthebellcurve).ThepurposeofanNRTistospreadstudentsoutalongacontinuumofscoressothatthosewithpoorlanguageabilitiesareatoneendofthenormaldistribution,whilethosewith"high"abilitiesareattheotherend(withthebulkofthestudentsfallingnearthemiddle).Thesecondfigureshowsitisacriterion-referencedtest(CRT),whichisusuallyproducedtomeasurewell-definedandfairlyspecificobjectives.Oftentheseobjectivesarespecifictoaparticularcourseorprogram.Eachstudent'sscoreismeaningfulwithoutreferencetotheotherstudents'scores.Astudent'sscoreonaparticularobjectiveindicatesthepercentoftheknowledgeorskillinthatobjectivethatthestudenthaslearned.Moreover,thedistributionofscoresonaCRTneednotnecessarilybenormal.Ifallthestudentsknow100%ofthematerialonalltheobjectives,thenallthestudentsshouldreceivethesamescore.ThepurposeofaCRTistomeasuretheamountoflearningthatastudenthasaccomplishedoneachobjective.Inmostcases,thestudentswouldknowinadvancewhattypesofquestions,tasks,andcontenttoexpectforeachobjective.3.Whatisreliabilityandvalidity?Whatistherelationshipbetweenreliabilityandvalidity?Toassessacandidate’sorallanguageabilityinanoraltest,theexaminingbodyoftenaskstwoexaminerstoscorethatcandidate’sperformance.Similarly,whenanexaminerisgradingacompositionforacertaintest,i.e.TEM4,thesamecompositioncanbemarkedbythesameexaminerontwooccasions.Explainindetailwhysuchmeasuresshouldbetaken.Answer:Thetestreliabilityisdefinedastheextenttowhichtheresultscanbeconsideredconsistentorstable.Testvalidityisdefinedhereasthedegreetowhichatestmeasureswhatitclaims,orpurports,tobemeasuring.Ifteachersadministeraplacementtesttotheirstudentsononeoccasion,theywouldlikethescorestobeverymuchthesameiftheyweretoadministerthesametestagain.Thedegreetowhichatestisconsistent,orreliablecanbeestimatedbycalculatingareliabilitycoefficient,whichcangoashighas+1.0foraperfectlyreliabletestoraslowas0whentheresultsonthetestaretotallyunreliable.Oncethetestsareadministeredtwiceandthepairsofscoresforeachstudentarelinedup,simplycalculateaPearsonproduct-momentcorrelationcoefficientbetweenthetwosetsofscores.Thecorrelationcoefficientwillprovideaconservativeestimate(thatisunderestimate)ofthereliabilityofthetestovertime.Thisreliabilityestimatecanbeinterpretedasthepercentofreliablevarianceonthetest.Testvalidityisdefinedhereasthedegreetowhichatestmeasureswhatitclaims,orpurports,tobemeasuring.Forexample,ifatestclaimstomeasureproficiencyinGermanlisteningcomprehension,thatisjustwhatitshouldassess.II.Discussion(55%,17pointsfor1and2,21pointsfor3)1.Lookatthefollowingtableandanswerthequestionsthatfollow:1)Calculatethetotalstandardscoresforthetwostudents2)Comparethetotalstandardscoresbetweenthetwostudents,seewhichstudentscoredhigher,andexplainbrieflywhyateacherhadbetterusethetotalstandardscoresinsteadofthetotalrawscores.SubjectMeanSDStudentAStudentBPsychology8168580Writing8598091ListeningComprehension7057685Reading74109366Literature8839095Total424417Answer:1)SubjectMeanSDStudentAStandardscoresStudentBStandardscoresPsychology816850.66780-0.17Writing85980-0.556910.667ListeningComprehension705761.2853Reading7410931.966-0.8Literature883900.667952.33Total4243.884175.032)从上图所计算出的标准分可以看出虽然学生A的总分比学生B的总分高，但学生B的考试得分其中三项的标准分显然比学生A要高，也就是大部分的标准分比学生A要高。从标准分总分来看，学生A的标准分总分为3.88，学生B的标准分总分为5.03，高于学生A。所以学生B考得好些。不同考试是不同质的，把不同考试的分数求和是没有意义的，同时也难以真实反映学生的整体情况。所以正确的做法是先将各门课程的考分转换成标准分，再求和，标准分求和后的总分加以比较，从而判断学生A，B在考试中的优劣，这样才更科学。2.UseSPSS16.0(1)tocalculatethecorrelationcoefficientbetweentwosetsofwritingscoresmarkedbytwoteachersforthesamegroupofstudentsandthenmakethescatterplot.Discussifthereisanycorrelationbetweenthesetwosetsofscores.(2)todothedependentsampleTtesttoseewhetherthereisanysignificantdifferencebetweenthetwoteachers’markingofthesamepaper.Reportanddiscusstheresult.Ifthereexistssignificantdifference,givesuggestionsastohowtosolvethisproblem.(3)tocalculatethecorrelationcoefficientbetweenthreesetsofwritingscoresmarkedbythreeteachersforthesamegroupofstudentsandmakethescatterplotbetweenRater1andRater2;Rater1andRater3,andRater2andRater3.Discussifthereisanycorrelationbetweenthesethreesetsofscores.(4)todoANOVAtoseewhetherthereisanysignificantdifferenceamongthethreeteachers’markingofthesamepaper(useLSDandSNK,anddrawMeansPlot).Reportanddiscusstheresult.（1）相关性A组教师B组教师A组教师Pearson相关性1.321显著性（双侧）.073N3232B组教师Pearson相关性.3211显著性（双侧）.073N3232从以上的相关性显著性值0.073,大于0.05，以及散点图，可以看出两位教师所批改的写作分数无相关。（2）成对样本检验成对差分tdfSig.(双侧)均值标准差均值的标准误差分的95%置信区间下限上限对1A组教师-B组教师-3.2508.3161.470-6.248-.252-2.21131.035从上 表 关于同志近三年现实表现材料材料类招标技术评分表图表与交易pdf视力表打印pdf用图表说话 pdf 成对样本检验中的Sig.(双侧)值0.035，小于0.05，可以看出两位老师所打的分，有明显差异。可以通过算标准分来解决此问题。（3）相关性Rater1Rater2Rater3Rater1Pearson相关性1.632**.599**显著性（双侧）.000.000N555555Rater2Pearson相关性.632**1.652**显著性（双侧）.000.000N555555Rater3Pearson相关性.599**.652**1显著性（双侧）.000.000N555555**.在.01水平（双侧）上显著相关。从以上的相关值和散点图可以看出，Rater1andRater2的相关度不高；Rater1andRater3属于正相关；Rater2andRater3属于较强的正相关。（4）分数组别Nalpha=0.05的子集1Student-Newman-Keulsa35568.9515569.4025571.75显著性.302将显示同类子集中的组均值。a.将使用调和均值样本大小=55.000。从以上表格中alpha的值为0.302，大于0.005，可以看出三个数值没有显著差异，也就是三位老师的打分没有明显差异。3.Ateacherwantstoapplyanewteachingmethodinoneofthetwoclassesheisteachingandtoseewhetherthismethodcanbeeffective.Beforehebeginshisexperiment,heusestheT-TestofSPSStocomparethescoresofthemid-termexaminationbetweenthetwoclassesandhefindsthesignificancevalueis0.547.Then,headoptsthenewmethodinEC(experimentclass)andforCC(comparisonclass),hestillusestheoldmethod.Afterthetwomonthsofexperiment,theteacherusestheT-TestofSPSSagaintocomparethescoresofthefinalexaminationbetweenthetwoclassesandthistime,hefindsthesignificancevalueis0.03.DiscusstheimportanceofthetwosignificancevaluesfromtwoT-TestofSPSS.答：从这位老师对实验班和对照班所做的期中考试的T检验的显著性水平值为0.547，大于0.05，可以看出前侧两个班无差异；两个月后期末考试的T检验的显著性水平值为0.03，小于0.05，可以看出后侧两个班有差异，由此，可以看出这位教师所采用的新方法对学生是有效的。