資源描述：
《實(shí)驗(yàn)設(shè)計之基本概念及實(shí)際應(yīng)用》由會員上傳分享，免費(fèi)在線閱讀，更多相關(guān)內(nèi)容在行業(yè)資料-天天文庫。
1、2010/6/152010SummerSchoolforCognitiveNeuroscience,Taiwan實(shí)驗(yàn)設(shè)計之基本概念及實(shí)際應(yīng)用張智宏助理教授中央大學(xué)認(rèn)知神經(jīng)科學(xué)研究所遠(yuǎn)流出版社12010/6/15?Definitionofanexperiment?Typesofexperimentaldesign?Independentgroups?Definition?Threatstointernalvalidityandwaystoovercomethem?Within‐subjects(repeated‐measures)?Definition?Threatstointe
2、rnalvalidityandwaystoovercomethem?WhenwesayAcausesB,basicallywearesayingthat:?AandBco‐vary?ChangesinAprecedeschangesinB?ThereisnootherexplanationforchangesinBexceptforchangesinA22010/6/15Test1:60Test2:9032010/6/15?Definition?Dependentvariable?Independentvariable?Manipulation/randomization/se
3、lection?Behavioralmeasures?Reactiontime,errorrate,d’,spatialerrors?Physiologicalmeasures?Heartrate,bloodpressure,respirationrate,cortisollevel?BOLDsignal,EEG,EPR,MEG42010/6/15?Event?Trial?Block?Condition?Session?Totesthypothesesaboutacausalrelationship?Todecidetheeffectofatreatment52010/6/15
4、?Theextenttowhichtheobservedeffectscanbeattributedtothemanipulationsoftheindependentvariables.IndependentDependentvaaaberiablevaaaberiableConfoundingvariable?Schedulingofobservations?Thechoiceoftreatmentsandcomparisons?Selectionofobservationsandmeasures?Selectionofparticipants?Themannerofass
5、igningunitstotreatments62010/6/15?Independentggproupsdesign?Within‐subjectsdesign?Complexdesign?Subjectassignment?Differentgroupofsubjectsareassignedtodifferentconditionsasdefinedbytheindependentvariable.?Balancingconfoundingfactors?Intelligence?Motivation?Conscientiousness?Gender?Majorinsch
6、ool72010/6/15Group1:60Group2:90?Randomggproupsdesign?Matchedgroupsdesign?Naturalgroupsdesign82010/6/15?Allrespectbeingcomparableamongggproupsbutonlythetreatmentreceiveddiffers?ComparablegroupformedbyrandomassignmentStage1Stage2Stage3RandomlyAssignedTreatment1ObservationGroup1RandomlyAssigned
7、Treatment2ObservationGroup292010/6/15?AhiAchieveddbyanyprocedureththtatassignsunitstoconditionsbasedonlychance?Ex.Tossingcoinsordices?Hasdifferentpurposefromrandomsampling?Intactggproupsareused?Extraneousvariablesarenotcontrolled?Selectivesubjectlo
