![教育統(tǒng)計(jì)學(xué)課件_第1頁](http://file4.renrendoc.com/view/99f99abf01e7f3b4d9dc88e0de504e03/99f99abf01e7f3b4d9dc88e0de504e031.gif)
![教育統(tǒng)計(jì)學(xué)課件_第2頁](http://file4.renrendoc.com/view/99f99abf01e7f3b4d9dc88e0de504e03/99f99abf01e7f3b4d9dc88e0de504e032.gif)
![教育統(tǒng)計(jì)學(xué)課件_第3頁](http://file4.renrendoc.com/view/99f99abf01e7f3b4d9dc88e0de504e03/99f99abf01e7f3b4d9dc88e0de504e033.gif)
![教育統(tǒng)計(jì)學(xué)課件_第4頁](http://file4.renrendoc.com/view/99f99abf01e7f3b4d9dc88e0de504e03/99f99abf01e7f3b4d9dc88e0de504e034.gif)
![教育統(tǒng)計(jì)學(xué)課件_第5頁](http://file4.renrendoc.com/view/99f99abf01e7f3b4d9dc88e0de504e03/99f99abf01e7f3b4d9dc88e0de504e035.gif)
版權(quán)說明:本文檔由用戶提供并上傳,收益歸屬內(nèi)容提供方,若內(nèi)容存在侵權(quán),請(qǐng)進(jìn)行舉報(bào)或認(rèn)領(lǐng)
文檔簡(jiǎn)介
Session6ComparingTwoGroups1/36BetweengroupsSamplingdistributionofmeansNormaldistribution–individualscoresSamplingdistributionofmeans–classmeansCharacteristics:For30ormoresamples,itisnormallydistributedItsmeanisequaltothemeanofthepopulationItsstandarddeviation,calledstandarderrorofmeans,isequaltothestandarddeviationofthepopulationbythesquarerootofthesamplesize.Degreeoffreedom2/36Betweentwogroups–t-testSamplingdistributionofdifferencesbetweentwomeanstofindanindividualscoreinanormaldistribution,weusez-scoretoplaceasamplemeaninadistributionofmeans,weuseat-testtocomparethedifferencebetweentwomeans3/36Betweentwogroups–t-test4/36Researchhypothesis?Thereisnoeffectofgrouponlisteningcomprehension(i.e.,thereisnodifferenceinthemeansoftheexperimentalandcontrolgroups)Significancelevel?.051-or2-tailed?2-tailedDesignDependentvariable(s)?ListeningcomprehensionMeasurement?Scores(interval)Independentvariable(s)?GroupMeasurement?Nominal(experimentalvs.control)Independentorrepeated-measures?IndependentStatisticalprocedure?t-testBetweentwogroups–t-test5/36Betweentwogroups–t-testdf=n1+n2–2=36Criticalvaluefordf=36at.05is2.042t=.845,df=36,p=n.s.(*)6/36Thereareonlytwogroupsofoneindependentvariabletocompare.Youcan’tcross-comparegroups.Eachobservationisassignedtooneandonlyonegroup.Thedataaretrulycontinuous(intervalorstronglycontinuousordinalscores).Themeanandstandarddeviationarethemostappropriatemeasurestodescribethedata.Thedistributionintherespectivepopulationsfromwhichthesamplesweredrawnisnormal,andvariancesareequivalent.Assumptionsunderlyingt-test7/36Strengthofassociationeta2()ByrejectingtheH0,wecanshowthatthereisaneffectofthegroupsoftheindependentvariableonthedependentvariable.Thetwogroupsdifferintheirperformanceonthedependentvariable.Whenthesamplestatisticissignificant,oneroughwayofdetermininghowmuchoftheoverallvariabilityinthedatacanbeaccountedforbytheindependentvariableistodetermineitsstrengthofassociation.Theof.55isaverystrongassociation.Ittellsusthat55%ofthevariabilityinthissamplecanbeaccountedforbyindependentvariable.(45%ofthevariabilitycannotbeaccountedforbytheindependentvariable.Thisvarianceisyettobeexplained.)8/36MindworkKey9/36NonparametriccomparisonoftwogroupsMannWhitneyUWhenthereare‘outliers’inthesample,thedataarenotnormallydistributedandthemedianratherthanthemeanisthebestmeasureofcentraltendencyNon-intervaldata,ranks10/36NonparametriccomparisonoftwogroupsMannWhitneyUScoresforthetwogroupsarecombinedandrankedSumtheranksforeachgroupUsethefollowingformulaetocomputetheteststatisticUUisthesmallerofU1andU211/36Thefirsttablegivesthecriticalvaluesforsignificanceatthep≤0.05levelinatwo-tailed/non-directionaltest,andforthep≤0.025levelinaone-tailed/directionaltest.Thesecondtablegivesthecriticalvaluesforthep≤0.01levelinatwo-tailed/non-directionaltest,andforthep≤0.005levelinaone-tailed/directionaltest.Forsignificance,thecalculatedvalueofUmustbesmallerthanorequalto
thecriticalvalue.N1andN2
arethenumberofobservationsinthesmallerandlargergroup,respectively.12/36NonparametriccomparisonoftwogroupsIfthereareapproximately20ormoresubjectsineachgroup,thedistributionofUisrelativelynormalandthefollowingformulaeisusedwhereN=N1+N213/36McEncry,BakerandWilson(1994)describeaquestionnairewhichwascirculatedtostudentswhousedtheCyberTutorcorpus-basedgrammarteachingcomputerprogramandtostudentswhostudiedgrammarinthetraditionalclassroommanner.Thequestionshadtheformatof‘Howdifficult/interesting/usefuldidyoufindthetask?',andwereansweredonafive-pointLikertscale,
rangingfrom‘verydifficult'(1point)to‘veryeasy'(5points).Aggregatescoreswerefoundforeachsubjectinbothgroups.Thecombinedscoresarerankedandaveragescoresaregiventotiedranks.
Example14/36Computer-taughtRank Classroom-taughtRankT1=36,T2=53U1=10;
U2=34ExampleSPSS15/36StrengthofAssociation:eta2Assumewehad18Ssclassifiedasexceptionallanguagelearnersand12aspoorlanguagelearners.Weaskedteacherstonominatethebestlearnersandthelearnerswhohadthemostdifficulty.Wewanttofindoutwhethergoodvs.poorlearnersshowdifferencesinauditoryshort-termmemory.Aftertestingeachperson,weusetheirscorestorank-orderallthelearnersandperformaUtestwhichgivesusauvalueof3.8,whichshowsgoodlearnersdidsignificantlybetterthanpoorlearnersonthetestofauditoryshort-termmemory.Nowhowmuchofthevariabilityintheshort-termmemoryscoresfromalltheSscanbeassociatedwithbeinga‘good’vs.‘poor’learners?49.8%ofthevarianceintheranksofauditoryshort-termmemorymaybeattributedtolearnerclassificationtype.Oneindependentvariablehasexplainedmuchofthevariabilityinshort-termmemory,butthereisstillvariancetobeexplained.16/36ApplicationExercises–Assignment4Alinguiststudyinghumanmemorywouldliketoexaminetheprocessofforgettingwords.Onegroupofsubjectsisrequiredtomemorizealistofwordsintheeveningjustbeforegoingtobed.Theirrecallistested10hourslaterinthemorning.Subjectsinthesecondgroupmemorizethesamelistofwordsinthemorning,andthentheirmemoriesaretested10hourslaterafterbeingawakeallday.Thelinguisthypothesizesthattherewillbelessforgettingduringsleepthanduringabusyday.Theyrecallscoresfortwosamplesofcollegestudentsareasfollows:17/36Sketchafrequencydistributionpolygon/lineforthetwogroups.Justbylookingatthesetwodistributions,wouldyoupredictasignificantdifferencebetweenthetwotreatmentconditions?Statethenullhypothesisandtheresearchhypothesis.Istheresearchhypothesisrejectedoraccepted?Why?
18/3619/36RelatedsamplesArepeated-measurestudyisoneinwhichasinglesampleofindividualsismeasuredmorethanonceonthesamedependentvariable.Thesamesubjectsareusedinallofthetreatmentconditions.Inamatched-subjectsstudy,eachindividualinonesampleismatchedwithasubjectintheothersample.Thematchingisdonesothatthetwoindividualsareequivalentwithrespecttoaspecificvariablethattheresearcherwouldliketocontrol.Example20/36df=npairs–121/36RelatedsamplesStrengthofassociationeta222/36SignTestOrdinaldataAteacherwonderedwhetherinformationfromconversa-tionalanalysiscouldbeusedtoteachtheseculturalconventionsinanESLclass.Thestudyinvolvedtheconversationalopenings,closings,andturn-takingsignalsusedinphoneconversations.Theteacheraskedallstudentstocallherduringthefirstweekofclass.Eachperson'sphoneskills(theopenings,preclosings,closings,andturn-taking)werescoredona1-10pointscale.Duringtheclassaspecialphoneconversationunittheteacherhaddevelopedwaspresentedandpracticed.Attheendofthecourse,theteacheragainhadstudentscallherandagainrankedthemona10-pointscale.SPSSRelatedsamples–nonparametriccomparisons23/36Relatedsamples–nonparametriccomparisons24/36Relatedsamples–nonparametriccomparisonsTallythe+and–symbolsunder“Change”.DiscardSswhodonnotshowchange.Herewehave3–changes,5+changes,and2nochanges.AssignRtothesmallertotalofchanges.HereR=3.TurntotheSigntesttabletodiscoverwhetherwecanrejectthenullhypothesis.25/3626/36WilcoxonSigned-RanksTestIntervaldataWetabulatethenumbersoferrorsmadebyagroupof10subjectsintranslatingtwopassagesofEnglish,ofequallength,intoChinese.Wewishtotest,atthe5percentlevel,whetherthereisanysignificantdifferencebetweenthetwosetsofscores.Sincewearenotpredictingthedirectionofanysuchdifference,anon-directionaltestwillbeappropriate.Thescoresobservedareasfollows:SPSSRelatedsamples–nonparametriccomparisons27/36Relatedsamples–nonparametriccomparisonsCalculatethedifferences.Rankthem,givingameanrankinthecaseofties.Thepairwithzerodifferenceisdroppedfromtheanalysis.Sumthepositiveandnegativeranks.TakethesmallerrankastheW.Checkwhetherthenullhypothesisisrejected.28/3629/36Relatedsamples–nonparametriccomparisonsWhenthenumberexceeds25SignTestWilcoxonSigned-rankstest30/36Relatedsamples–nonparametric
溫馨提示
- 1. 本站所有資源如無特殊說明,都需要本地電腦安裝OFFICE2007和PDF閱讀器。圖紙軟件為CAD,CAXA,PROE,UG,SolidWorks等.壓縮文件請(qǐng)下載最新的WinRAR軟件解壓。
- 2. 本站的文檔不包含任何第三方提供的附件圖紙等,如果需要附件,請(qǐng)聯(lián)系上傳者。文件的所有權(quán)益歸上傳用戶所有。
- 3. 本站RAR壓縮包中若帶圖紙,網(wǎng)頁內(nèi)容里面會(huì)有圖紙預(yù)覽,若沒有圖紙預(yù)覽就沒有圖紙。
- 4. 未經(jīng)權(quán)益所有人同意不得將文件中的內(nèi)容挪作商業(yè)或盈利用途。
- 5. 人人文庫網(wǎng)僅提供信息存儲(chǔ)空間,僅對(duì)用戶上傳內(nèi)容的表現(xiàn)方式做保護(hù)處理,對(duì)用戶上傳分享的文檔內(nèi)容本身不做任何修改或編輯,并不能對(duì)任何下載內(nèi)容負(fù)責(zé)。
- 6. 下載文件中如有侵權(quán)或不適當(dāng)內(nèi)容,請(qǐng)與我們聯(lián)系,我們立即糾正。
- 7. 本站不保證下載資源的準(zhǔn)確性、安全性和完整性, 同時(shí)也不承擔(dān)用戶因使用這些下載資源對(duì)自己和他人造成任何形式的傷害或損失。
最新文檔
- 現(xiàn)代辦公室空間中的綠色植物應(yīng)用
- 現(xiàn)代制造園區(qū)的投資風(fēng)險(xiǎn)評(píng)估與管理
- 現(xiàn)代企業(yè)經(jīng)營(yíng)中的稅務(wù)籌劃與風(fēng)險(xiǎn)管理
- 國(guó)慶節(jié)主題客堂活動(dòng)方案
- 2024年春九年級(jí)化學(xué)下冊(cè) 第10單元 酸和堿 實(shí)驗(yàn)活動(dòng)6 酸、堿的化學(xué)性質(zhì)說課稿 (新版)新人教版
- Unit7 第2課時(shí)(說課稿)Story time三年級(jí)英語上冊(cè)同步高效課堂系列(譯林版三起·2024秋)
- 2《紅燭》《致云雀》聯(lián)讀說課稿 2024-2025學(xué)年統(tǒng)編版高中語文必修上冊(cè)
- 《4 做陽光少年》(說課稿)-2023-2024學(xué)年五年級(jí)上冊(cè)綜合實(shí)踐活動(dòng)皖教版
- 2025水運(yùn)工程施工監(jiān)理合同(試行)
- 2025企業(yè)聘用臨時(shí)工合同
- DBJT 13-460-2024 既有多層住宅建筑增設(shè)電梯工程技術(shù)標(biāo)準(zhǔn)
- 中國(guó)證監(jiān)會(huì)證券市場(chǎng)交易結(jié)算資金監(jiān)控系統(tǒng)證券公司接口規(guī)范
- 2025屆天津市部分學(xué)校高三年級(jí)八校聯(lián)考英語試題含解析
- 微項(xiàng)目 探討如何利用工業(yè)廢氣中的二氧化碳合成甲醇-2025年高考化學(xué)選擇性必修第一冊(cè)(魯科版)
- 廣東省廣州市黃埔區(qū)2024-2025學(xué)年八年級(jí)物理上學(xué)期教學(xué)質(zhì)量監(jiān)測(cè)試題
- 水產(chǎn)品冷凍加工原料處理與加工技術(shù)考核試卷
- 全新保密協(xié)議模板公安下載(2024版)
- 財(cái)務(wù)管理學(xué)(第10版)課件 第1章 總論
- GB/T 4008-2024錳硅合金
- 《鼻咽癌的診治》課件
- 2024年天津市中考英語試題卷(含答案)
評(píng)論
0/150
提交評(píng)論