How to Design and Evaluate Research in Education.docx

How to Design and Evaluate Research in Education.docx

《How to Design and Evaluate Research in Education.docx》由会员分享，可在线阅读，更多相关《How to Design and Evaluate Research in Education.docx（53页珍藏版）》请在冰豆网上搜索。

HowtoDesignandEvaluateResearchinEducation

HowtoDesignandEvaluateResearchinEducation

By

JackR.FraenkelandNormanE.Wallen

Chapter1

TheNatureofResearch

⏹Waysofknowing

ØSensoryexperience（incomplete/undependable）

ØAgreementwithothers（commonknowledgewrong）

ØExperts’opinion（theycanbemistaken）

ØLogic/reasoningthingsout（canbebasedonfalsepremises）

⏹Whyresearchisofvalue

ØScientificresearch（usingscientificmethod）ismoretrustworthythanexpert/colleagueopinion,intuition,etc.

Chapter1-continued

TheNatureofResearch

⏹ScientificMethod（testingideasinthepublicarena）

ØPutguesses（hypotheses）totestsandseehowtheyholdup

ØAllaspectsofinvestigationsarepublicanddescribedindetailsoanyonewhoquestionsresultscanrepeatstudyforthemselves

ØReplicationisakeycomponentofscientificmethod

Chapter1-continued

TheNatureofResearch

⏹ScientificMethod（requiresfreedomofthoughtandpublicproceduresthatcanbereplicated）

⏹Identifytheproblemorquestion

⏹Clarifytheproblem

⏹Determineinformationneededandhowtoobtainit

⏹Organizetheinformationobtained

⏹Interprettheresults

⏹Allconclusionsaretentativeandsubjecttochangeasnewevidenceisuncovered（don’tPROVEthings）

Chapter1-continued

TheNatureofResearch

⏹TypesofResearch

ØExperimental（mostconclusiveofmethods）

→Researchertriesdifferenttreatments（independentvariable）toseetheireffects（dependentvariable）

→Insimpleexperimentscompare2methodsandtrytocontrolallextraneousvariablesthatmightaffectoutcome

→Needcontroloverassignmenttotreatmentandcontrolgroups（tomakesuretheyareequivalent）

→Sometimesusesinglesubjectresearch（intensivestudyofsingleindividualorgroupovertime）

Chapter1-continued

TheNatureofResearch

（TypesofResearchcontinued）

ØCorrelationalResearch

→Looksatexistingrelationshipsbetween2ormorevariablestomakebetterpredictions

ØCausalComparativeResearch

→Intendedtoestablishcauseandeffectbutcannotassignsubjectstotrtmt/control

→Limitedinterpretations（couldbecommoncauseforbothcauseandeffect…stresscausessmokingandcancer）

→Usedforidentifyingpossiblecauses;similartocorrelation

Chapter1-continued

TheNatureofResearch

（TypesofResearchcontinued）

ØSurveyResearch

→Determine/describecharacteristicsofagroup

→Descriptivesurveyinwritingorbyinterview

→Provideslotsofinformationfromlargesamples

→Threemainproblems:

clarityofquestions,honestyofrespondents,returnrates

ØEthnographicresearch（qualitative）

→IndepthresearchtoanswerWHYquestions

→Someishistorical（biography,phenomenology,casestudy,groundedtheory）

Chapter1-continued

TheNatureofResearch

（TypesofResearchcontinued）

ØHistoricalResearch

→Studypast,oftenusingexistingdocuments,toreconstructwhathappened

→Establishingtruthofdocumentsisessential

ØActionResearch（differsfromabovetypes）

→Notconcernedwithgeneralizationstoothersettings

→Focusoninformationtochangeconditionsinaparticularsituation（mayusealltheabovemethods）

⏹Eachofthesemethodsisvaluableforadifferentpurpose

Chapter1-continued

TheNatureofResearch

⏹GeneralResearchTypes

ØDescriptive（describestateofaffairsusingsurveys,ethnography,etc.）

ØAssociational（goesbeyonddescriptiontoseehowthingsarerelatedsocanbetterunderstandphenomenausingcorrel/causal-comparative

ØIntervention（tryinterveningtoseeeffectsusingexperiments）

Chapter1-continued

TheNatureofResearch

Quantitativev.Qualitative

Quantitative（numbers）

ØFacts/feelingsseparate

ØWorldissinglereality

ØResearcherremoved

ØEstablishedresearchdesign

ØExperimentprototype

ØGeneralizationemphasized

Chapter1-continued

TheNatureofResearch

⏹Meta-Analysis

ØLocateallthestudiesonatopicandsynthesizeresultsusingstatisticaltechniques（averagetheresults）

⏹CriticalAnalysisofResearch（somesayallresearchisflawed）

ØQuestionofreality（areonlyindividualperceptionsofit）

ØQuestionofcommunication（wordsaresubjective）

ØQuestionofvalues（noobjectivityonlysocialconstructs）

ØQuestionofunstatedassumptions（researchersdon’tclarifyassumptionsthatguidethem）

ØQuestionofsocietalconsequences（researchservespoliticalpurposesthatareconservativeoroppressive;preservestatusquo）

Chapter1-continued

TheNatureofResearch

OverviewoftheResearchProcess（Fig.1.4）

⏹Introductionchapter

⏹Problemstatementthatincludessomebackgroundinfoandjustificationforstudy

⏹Exploratoryquestionorhypothesis（relationshipamongvariablesclearlydefined）;goeslastinCh.

⏹Definitions（inoperationalterms）

⏹Reviewofrelatedliterature（otherstudiesofthetopicreadandsummarizedtoshedlightonwhatisalreadyknown）

Chapter1-continued

TheNatureofResearch

OverviewoftheResearchProcess（Fig.1.4）

⏹Methodschapter

⏹Subjects（sample,population,methodtoselectsample）

⏹Instruments（tests/measuresdescribedindetailandwithrationalefortheiruse）

⏹Procedures（what,when,where,how,andwithwhom）;

✓Giveschedule/dates,describematerialsused,designofstudy,andpossiblebiases/threatstovalidity

4.Dataanalysis（howdatawillbeanalyzedtoanswerresearchquestionsortesthypothesis）

Chapter2

TheResearchProblem

⏹StatementoftheProblem（identifyaproblem/areaofconcerntoinvestigate）

ØMustbefeasible,clear,significant,ethical

⏹ResearchQuestions（serveasfocusofinvestigation,seep.28list）

ØSomeinfomustbecollectedthatanswersthem（mustberesearchable）

ØCannotresearch“should”questions

ØSeediagram,p.29

Chapter2-Continued

TheResearchProblem

⏹RQshouldbefeasible（canbeinvestigatedwithavailableresources）

⏹RQshouldbeclear（specificallydefinetermsused…operationalneeded,butgiveboth）

ØConstitutivedefinitions（dictionarymeaning）

ØOperationaldefinitions（specificactions/stepstomeasureterm;IQ=timetosolvepuzzle,where<20sec.ishigh;20-40ismed.;40+islow）

⏹RQshouldbesignificant（worthinvestigating;howdoesitcontributetofieldandwhocanuseinfo）

⏹RQsofteninvestigaterelationships（twocharacteristics/qualitiestiedtogether）

Chapter3

VariablesandHypotheses

⏹Importanttostudyrelationships

ØSometimesjustwanttodescribe（useRQ）

ØUsuallywanttolookforpatterns/connections

→Hypothesispredictstheexistenceofarelationship

⏹Variables（anythingthatcanvaryinmeasure;oppositeofconstant）

⏹Variablesmustbeclearlydefined

⏹Ofteninvestigaterelationshipbetweenvariables

Chapter3-Continued

VariablesandHypotheses

⏹VariableClassifications（Fig.3.4,p.42）

ØQuantitative（variablesmeasuredasamatterofdegree,usingrealnumbers;i.e.age,numberkids）

ØCategorical（novariation…eitherinacategoryornot;i.e.gender,haircolor）

ØIndependent:

thecause（akathemanipulated,treatmentorexperimentalvariable）

ØDependent:

theeffect（akaoutcomevariable）

ØExtraneous:

uncontrolledIVs（seeFig.3.2,p.46）

ØAllextraneousvariablesmustbeaccountedforinanexperiment

Chapter3-Continued

VariablesandHypotheses

⏹Hypotheses–predictionsaboutpossibleoutcomeofastudy;sometimesseveralhypothesesfromoneRQ（Fig3.3）

ØRQ:

WillathleteshaveahigherGPAthatnonathletes?

ØH:

AthleteswillhavehigherGPAsthatnonathletes

⏹AdvantagestostatingahypothesisaswellasRQ

ØClarifies/focusesresearchtomakepredictionbasedonpreviousresearch/theory

ØMultiplesupportingteststoconfirmhypothesisstrengthensit

⏹Disadvantages

ØCanleadtobiasinmethods（consciousorun）totrytosupporthypothesis

ØSometimesmissotherimportantinfoduetofocusonhypothesis（peerreview/replicationisacheckonthis）

Chapter3-Continued

VariablesandHypotheses

⏹Somehypothesismoreimportantthanothers

⏹Directionalv.nondirectional

ØDirectionalsayswhichgroupwillscorehigher/dobetter

ØNondirectionaljustindicatestherewillbeadifference,butnotwhowillscorehigher/dobetter

ØDirectionalmorerisky,sobecareful/tentativeinusingdirectionalones

Chapter4

EthicsandResearch

⏹Examplesofunethicalpractices

ØRequiringparticipationfrompowerless（students）

ØUsingminorswithoutparentalpermission

ØDeletingdatathatdon’tagreew/hypothesis

ØInvadingprivacyofsubjects

ØPhysicallyorpsychologicallyharmingsubjects

⏹APAstatementofethicalprinciplesinresearch

ØEachstudentmustsignoneandhaveitsignedbyworkplacesupervisor

Chapter4-Continued

EthicsandResearch

⏹Protectingparticipantsfromharmrequiresinformedconsent

ØSubjectsmustknowthepurposeofthestudy,possiblebenefits/harm;participationisvoluntaryandtheycanw/drawwithoutpenaltyanytime（Fig.4.3,p.59）

⏹Researchersshouldask:

Couldsubjectsbeharmed?

Isthereanotherwaytogettheinfo?

Istheinfovaluableenoughtojustifystudy?

⏹Researchersmustensureconfidentialityofdata（limitaccess;nonamesifpossible;tellsubjectsconfidentialoranonymous）

⏹Deceivingsubjectsissometimesnecessary（Milgramstudy）,askifresultsjustifyethicallapse

ØWhendeceptionusedsubjectstheyshouldbeokaywithitafter（andtheycanrefuseuseoftheirdata）

Chapter4-Continued

EthicsandResearch

⏹Researchwithchildren

ØParentalconsentrequired（signedpermissionfromparents

ØAPAEthicsinResearchFormaddressesthisalso

⏹RegulationofResearch（NationalResearchActof1974）

ØIffederalfundingreceivedmusthaveanIRBtocheck:

riskstosubjects,informedconsentguidelinesmet,debr