Experimental-Design-(实验设计)

017Experimentaldesign2目录contents1234Definition16785Significance9ComponentsTypesCompareConceptualframeworkApplicationSummaryCasediscussionDefinitionofexperimentaldesign1Definitionofexperimentaldesign1Experimentaldesignisadetailedexperimentdesignwhichisdonebeforetheexperiment,itchangesoneormoreprocessvariablestoobservetheresponsevariablechanges,itisaprocedureforobtaintheresearchdata,andgainyieldvalidandobjectiveconclusion(Winer,1971).Experimentaldesignisaresearchdesignthateliminatesallfactorsthatinfluenceoutcomeexceptforthecausebeingstudied(independentvariable),allotherfactorsarecontrolledbyrandomization,investigator-controlledmanipulationoftheindependentvariable,andcontrolofthestudysituationbytheinvestigator,includingtheuseofcontrolgroups(Medicaldictionary).2SignificanceofexperimentaldesignSignificanceofexperimentaldesign2Discussion:(1min)ThesignificanceofexperimentaldesignYoucandivideintofourgroupsfreely,andchooseonetotworepresentative(s)toanwer.Significanceofexperimentaldesign22.1Thedesignofanyexperimentisofutmostimportancebecauseithasthepowertobethemostrigidtypeofresearch(mostprecise/strict).Experimentaldesignisthemostrigorousofallresearchdesigns.2.2Beusedtoexploringtheeffectivenessofaninterventionortreatment,howeffectivetheinterventionis?Inotherwords,theexperimentaldesigngivesevidencethatwhetheronevariablecancausechangetoanothervariableornot.2.3Thescientificexperimentdesigncanuselesstesttimestoachievetheexpectedtesttarget(gettwicetheresultwithhalftheeffort).22.4Experimentaldesignisthemainstepinconductingaresearch,acleardefinitionindetailsoftheexperimentmakesthedesiredstatisticalanalysespossible,andalmostalwaysimprovestheusefulnessoftheresults(Barker,2016).2.5Experimentaldesignistheprocessofplanningastudytomeetspecifiedobjectives,itisveryimportanttoensurethattherighttypeofdataandasufficientsamplesizeandpowerareavailabletoanswertheresearchquestionsofinterestasclearlyandefficientlyaspossible(Barker,2016).Significanceofexperimentaldesign3Componentsofexperimentaldesign3.3Studygroups(Participants)Componentsofexperimentaldesign33.2Constants3.1Variables3.4Intervention/Treatment3.5Outcome/Results3.6Instrument33.1VariablesIV:Independent/Experimentalvariableisthefactor(s)thatispurposefullychangedormanipulatedbytheexperimenter.DV:Dependent/Responsevariableisthefactor(s)thatresponsetothechangeorinotherwords,respondtotheIV(beusedtomeasuretheoutcome/result).EV:Extraneousvariablesarenotinthestudybutcanaffectthestudyvariablesandtheirrelationship.Componentsofexperimentaldesign33.2ConstantsConstantsarethefactorsthatremainthesameorarepurposefullykeptthesameduringanexperiment.3.3Studygroups(Participants)E:ExperimentalgroupismanipulatedbyreceivingthetreatmentC:ControlgroupisnotmanipulatedComponentsofexperimentaldesign33.4Intervention/TreatmentX/T:Thefactorthatisbeingmanipulatedistypicallyreferredtoasthetreatmentorintervention.3.5Outcomes/ResultsO:PretestandPost-test(canbemorethanonce)3.6InstrumentAdeviceorscaleforrecording,measuring,orcontrolling,especiallysuchadevicefunctioningaspartofacontrolsystem.Itusuallybeusedtomeasuretheoutcomeorresultinexperimentaldesign.ComponentsofexperimentaldesignTypesofexperimentaldesign44Question:(1min)1.Doyoukonwhowmanytypesofexperimentaldesign?2.Whatisthedifference(s)between/amongthem?Thesame,youcandivideintofourgroupsfreely,andchooseonetotworepresentative(3)toanwer.Typsofexperimentaldesign44.1Pre-experimentaldesign4.2Trueexperimentaldesign4.3Quasi-experimentaldesign(Campbell＆Stanley,1963)Typesofexperimentaldesign44.1Pre-experimentaldesignPre-experimentaldesignsaresonamedbecausetheyfollowbasicexperimentalstepsbutfailtoincludeacontrolgroup.Inotherwords,asinglegroupisoftenstudiedbutnocomparisonbetweenanequivalentnon-treatmentgroupismade.Thecharacteristicsofpre-experimentaldesignsare:ManipulationofindependentvariablesLimitedcontrolovertheextraneousvariablesNorandomizationandcontrolgroupTypesofexperimentaldesign44.1Pre-experimentaldesign--TypesTypesofexperimentaldesign44.2Trueexperimentaldesign--CharacteristicRandomselectionofparticipantsandtherandomassignmentoftheparticipantstogroupsinthestudy.Exposureofsomeofthesubjectstotheonetreatment(experimentalgroup);andnoexposureofsomeoftheremainingsubjects(controlgroup)Typesofexperimentaldesign44.2Trueexperimentaldesign--CharacteristicCausalitybetweentheIVandtheDVisexaminedunderhighlycontrolledcondition(Kerlinger＆Lee,2000).Soitisconsideredthemostpowerfulquantitativemethodbecauseofrigorouscontrolofvariablesandcompletecontrolovertheextraneousvariables.(confidentlydeterminedthateffectontheDVisdirectlyduetothemanipulationoftheIV)Typesofexperimentaldesign44.2Trueexperimentaldesign--Types(1)WithoutcomparisongroupPost-testonlycomparisongroupdesign(2)Withcomparisongroup--WithrepeatedmeasuresRepeatedmeasuresdesign--WithoutrepeatedmeasuresPretest/post-testcomparisonorcontrolgroupdesign(*)Randomizedclinicaltrials(*)Randomizedblockdesign(*)Solomonfour-groupdesignFactorialdesignsMultivariatedesignsNesteddesignsTypesofexperimentaldesign(*)(*)(*)