©2010-2026 宁波自信网络信息技术有限公司 版权所有
客服电话：0574-28810668 投诉电话：18658249818
浙ICP备2021020529号-1 | 浙B2-20240490
关注我们 ：
1、Chapter3FundamentalsofStatisticalStudiesDefinitionofStatisticsStatisticsisthestudyofvariation,interactionsand,operationaldefinitions.Statisticsisthestudyofdatatoprovideabasisforactiononapopulationorprocess.TypesofStatisticalStudiesEnumerative studies arestatisticalinvestigationsthatleadtoactiononast
2、aticpopulation(thatis,agroupofitems,people,etc.thatexistinagiventimeperiodand/oratagivenlocation).Analytic studies arestatisticalinvestigationsthatleadtoactiononadynamicprocess.Sincethisisatextonqualityimprovement,theprimaryfocuswillbeonanalyticstudies.EnumerativeStudiesBasicConceptsApopulation(orun
3、iverse)isthetotalityofunits,items,orpeopleofinterestthatexistinagiventimeperiodand/orgivenlocationAdministration.Aframeisalistoftheitemsinapopulation.Thegapisthedifferencebetweentheframeandthepopulation.Asample isaportionoftheframeunderinvestigation,andisselectedsothatinformationcanbedrawnfromitabou
4、ttheframe.BasicConceptsNonrandom samples areselectedonthebasisofconvenience(convenience sample),theopinionofanexpert(judgment sample),oraquotatoensureproportionalrepresentationofcertainclassesofitems,units,orpeopleinthesample(quota sample).Allnonrandomsampleshavethesameshortcomingtheyresubjecttoanun
5、knowndegreeofbiasintheirresults.Thisbiasiscausedbytheabsenceofaframe.Nonrandomsamplesshouldbeusedonlywhenbetterinformationistoocostlytoobtain.Random samples areselectedsothateveryelementintheframehasaknownprobabilityofselection.Typesofrandomsamplesincludesimple,stratified,andcluster.Allrandomsamples
6、allowgeneralizedstatementstobemadeabouttheframefromthesample.Arandomsampleisselectedbyoperationallydefiningaprocedurethatutilizesrandomnumbersintheselectionofthesampleditemsfromtheframetoeliminatebiasandholduncertaintywithinknownlimits.Sevenstepsareinvolvedinselectingasimplerandomsample:Step 1.Count
7、thenumberofelementsintheframe,N.Step 2.Numbertheelementsintheframefrom1throughN.IfNis25,thentheelementsintheframeshouldbenumberedfrom01through25.Allelementsmustreceiveanidentificationnumberwiththesamenumberofdigits.Step 3.Selectapageinatableofrandomnumbers.Forexample,selectingapageandstartingpointyi
8、elds.193040092878311398676069611339043462052856735990548709422934276212493869294093Step 4.Ontheselectedpageofrandomnumbers,selectacolumnofnumbers,selectastartingpointinthatcolumn,anduseasmanydigitsastherearedigitsinN(twodigitsinthecaseofN=25).Forexample,beginningwiththetopofourcolumn,selecting,andus
9、ingthefirsttwodigitsofeachnumberinthatcolumn,thefirstrandomnumberis19.Sample NumberIdentification NumberItems119S209I304D405E513M612LStep 5.Determinethenecessarysamplesize.Step 6.Fromthechosencolumnontheselectedpage,selectthefirstsixtwo-digitnumbersbetween01and25,inclusive.Ifanumberisencounteredthat
10、issmallerthan01(e.g.,00)orlargerthan25(e.g.,31),ignorethenumberandcontinuedownthecolumn.Ifanacceptablenumberappearsmorethanonce,ignoreeveryrepetitionandcontinuemovingdownthecolumnuntilsixuniquenumbersbetween01and25havebeenselected.Ifthebottomofthepageisreachedbeforesixuniquerandomnumbersareobtained,
11、gotothetopofthepageandmovedownthenexttwo-digitcolumn.Step 7.Finally,analyzetheinformationasabasisforaction.Twoimportantpointstorememberare:1.differentmethodsofmeasurementwillalsoyielddifferentresults.2.differentsamplesofsizesixwillyielddifferentresults,andRandomsamples,however,donthavebias,andthesam
12、plingerrorcanbeheldtoknownlimitsbyincreasingthesamplesize.ConductinganEnumerativeStudyThefollowing13stepspresentaguideforconductinganenumerativestudy.Thestepsarethesamewhetherthestudyisbasedonacompletecountofthepopulationorasample.Step 1.Specifythereason(s)youwanttoconductthestudy(forexample,toestim
13、atetheaveragenumberofsickdaysperemployeeintheXYZCompanyin2002).Ifthisaverageisgreaterthan8.0days,thenanewhealthcareplanwillbeinstituted.Ifitislessthanorequalto8.0days,thecurrentplanwillbemaintained.Step 2.Specifythepopulationtobestudied.Inourexample,thepopulationwouldbeallfull-timeemployeesintheXYZC
14、ompanyin2002.Anemployeeisconsideredfull-timein2001ifhehadfull-timestatusdesignationatanytimeduringtheyear.Step 3.Constructtheframe(alistofallfull-timeemployees).Everyonewhowillusethestudysresultsasabasisforactionmustagreethattheframerepresentsthepopulationuponwhichtheywanttotakeaction.Step 4.Perform
15、secondaryresearch(suchastheexaminationofpre-publisheddata)todeterminehowmuchinformationisalreadyavailableabouttheproblemunderinvestigation.Forexample,checktheHumanResourcesDepartmentsrecords.Step 5.Determinethetypeofstudytobeconducted(forexample,mailsurvey,personalinterviews,analysisofunits).Inthise
16、xample,wewouldanalyzeemployeeabsenteecardsfor2002.Step 6.Makeitpossibleforrespondentstogiveclear,understandableinformation,and/orfortheresearchertoelicitclear,understandableinformation.Forexample,themethodforanalyzingabsenteecardsshouldbeclearandstraightforward.Considertheproblemofnon-response.Inthi
17、sexample,besurenoabsenteecardsaremissing,andmakesurethedatagatherersknowhowtointerprettheabsenteecards.Step 7.Establishthesamplingplantobeused,determinetheamountofallowableerrorintheresults,andcalculatethecostofthesamplingplan.Atthisstage,Steps1and2mayneedrevisionsduetocostconsiderations.Forexample,
18、wemaydecidetodrawasimplerandomsampleofemployeeabsenteecardsusingrandomnumbers,atacostof$1percard,assuminganallowableerrorofonequarterofadayintheestimate.Step7requiresarandomsample.Theresultofanonrandomsampleinanenumerativestudyisworthnomorethanthereputationofthepersonwhosignsthereport.Step 8.Establi
19、shprocedurestodealwiththenon-responsesanddifferencesbetweeninterviews,testers,inspectors,andsoon.Forexample,beabletoassessdifferencesincollecteddataduetodifferencesinthedatagatherersabilities.Step 9.Prepareunambiguousinstructionsforthedatagatherersthatcoverallphasesofdatacollection.Step 10.Establish
20、plansfordatahandling,includingformatoftables,headings,andnumberofclasses.Step 11.Pretestthedata-gatheringinstrumentanddata-gatheringinstructions.Step 12.Conductthestudyandthetabulations.Itiscriticalthatthestudybecarriedoutaccordingtoplans.Fromthegathereddata,calculatethesamplingerrorsofinterest.This
21、isimportantsothatthestudysuserscanunderstandthedegreeofuncertaintypresentinthestudyresults.Inourexample,wewouldcalculatethestandarderrorfortheaveragenumberofdaysabsentperemployee.Step 13.Interpretandpublishtheresultssothatdecisionmakerscantakeappropriateaction.Forexample,iftheaveragenumberofdaysabse
22、ntperemployeeisgreaterthaneight,thenestablishthenewhealthcareprogram.AnalyticStudiesBasicConceptsAnalyticstudiesarestatisticalinvestigationsthatleadtoactionsthatwillimproveorinnovateaprocessovertime.Theconceptofapopulationdoesnotexistforananalyticstudybecausefutureprocessoutput,whichdoesnotyetexist,
23、cannotbepartofthepopulation.Aframecannotexistwithoutapopulation.Lackofapopulationandframemakesitimpossibletodrawarandomsampletostudyaprocesssbehavior.Modelscanbeusedtostudythesecause-and-effectsystems.Thesemodelsincludeflowchartsofaprocess,simulationsoftheprocess,andcause-and-effectdiagrams.Inananal
24、yticstudy(unlikeanenumerativestudy),thereisnounderlyingstatisticaltheorythatenablesquantificationofuncertainty.Expertopinionisinvaluableinunderstandingthemagnitudeoftheuncertaintycausedbychangestoaprocess.Generally,thisuncertaintyisbestexplainedbyanexpertwhoisinvolvedwiththeprocessunderstudy.Improve
25、mentorinnovationofaprocessisaccomplishedusingthePlan-Do-Study-Act(PDSA)cycle.ThePDSAcycleisusedtonarrowthedifferencebetweenprocessperformanceandcustomer(eitherinternalorexternal)needsandwants.TheaimofthePLANstageistodetermineanimprovementorinnovationtoaprocessthatwillnarrowthedifferencebetweenproces
26、sperformanceandcustomerneedsandwants.(i.e.oldflowcharttonewflowchart)TheDOstageofthePDSAcyclerequiresthatplannedexperimentsbeconductedtodeterminehowbesttoimplementthePLANestablishedinthepreviousstage.Experimentsshouldbeconductedonasmallscale,usingalaboratory,anofficeorplantsite,or,withhisconsent,eve
27、nacustomerslocation.TheaimoftheSTUDYstageistodetermineifthePLANhasbeeneffectiveindecreasingthedifferencebetweenprocessperformanceandcustomerneedsandwants.IntheACTstage,thePLANshouldbeintegratedintotheprocess.ErrorsinAnalyticStudiesAtype one error occurswhenactionistakenonaprocesswhenitshouldhavebeen
28、leftalone.Atype two erroroccurswhenwefailtotakeactiononaprocesswhenactionisappropriate.DesignofAnalyticStudiesWecanincreaseourdegreeofbeliefinapredictionfromananalyticstudybyconsidering:BuildingofknowledgeinasequentialfashionTestingoverawiderangeofconditionsSelectionofunitsforthestudyBuilding of Kno
29、wledge in a Sequential FashionExperimentsperformedintheiterationsofthePDSAcycle,incombinationwiththeoryabouttheprocessfromsubjectmatterexperts,mayincreasethedegreeofbeliefinpredictionsaboutthefuturebehavioroftheprocessunderstudy.Thisdegreeofbeliefisincreasedassequentialpredictionsabouttheprocesssfut
30、urebehaviorcomeclosertotheactualperformanceoftheprocess.Testing over a Wide Range of ConditionsThedegreeofbeliefinthepredictivevalueoftheknowledgegainedfromananalyticstudyisincreasediftheanalyticstudyyieldsthesameresultsoverawiderangeofconditions.Onlyanexpertinthesubjectmatterunderstudycananswerques
31、tionssuchashowwidearangeofconditionsisadequatetohaveadegreeofbeliefsufficienttomakeaprediction,orhowclosetoactualconditionsmusttheexperimentalconditionsbetohaveadegreeofbeliefhighenoughtomakepredictions.Selection of the Units for the StudySincethereisnoframeinananalyticstudy,therecanbenorandomsample
32、andquantifiabledegreeofuncertaintyinapredictionaboutthefuture.Judgmentsamplesareusedtoconductanalyticstudies.AnalysisofDatafromAnalyticStudiesA Stable ProcessAstableprocessprovidesaforumforasubjectmatterexperttoconductexperimentstogainknowledgetopredictthefuturebehavioroftheprocess.Graphical Analysi
33、sThegeneralapproachforconductinganalyticstudiesreliesongraphicaltechniques,suchascontrolcharts,whichutilizebothstatisticalknowledgeandsubjectmatterknowledgetolearnabouttheprocesstopredictitsfuturebehavior.DistinguishingBetweenEnumerativeandAnalyticStudiesIfa100percentsampleoftheframeanswersthequestionunderinvestigation,thestudyisenumerative;ifnot,thestudyisanalytic.
©2010-2026 宁波自信网络信息技术有限公司 版权所有
客服电话：0574-28810668 投诉电话：18658249818
浙ICP备2021020529号-1 | 浙B2-20240490
关注我们 ：
