简介:Formanyclusteringalgorithms,itisveryimportanttodetermineanappropriatenumberofclusters,whichiscalledclustervalidityproblem.Inthispaper,anewclusteringvalidityassessmentindexisproposedbasedonanovelmethodtoselectthemarginpointbetweentwoclustersforin-ter-clustersimilaritymoreaccurately,andprovidesanimprovedscatterfunctionforintra-clustersimilarity.Simulationresultsshowtheeffectivenessoftheproposedindexonthedatasetsunderconsiderationregardlessofthechoiceofaclusteringalgorithm.
简介:Thepapermakesastudyonsixparts,namely,speaking,writing,sentence,cloze,readingandlisteninginanEnglishtestwithstudentsattwodifferentproficiencylevels.Theirscoredatawerestudiedthroughreliabilityanalysis,validityanalysis,factoranalysisandanalysisofvariance(ANOVA)toconductthevalidationofthisEnglishtest.Thepapertriestoexplorethescoredatainternalconsistency,groupdifferencerecognitiontomaintaintheobjectivityandaccuracyinEnglishtests.
简介:
简介:Background:TheGlobalPhysicalActivityQuestionnaire(GPAQ)hasbeenusedtomeasurephysicalactivity(PA)andsedentarytimeinFrance,butnostudyhasassesseditspsychometricproperties.ThisstudyaimedtocomparethereliabilityaswellascriterionandconcurrentvalidityoftheFrenchversionoftheGPAQwiththeFrenchInternationalPhysicalActivityQuestionnairelongform(IPAQ-LF)anduseofanaccelerometerinageneraladultpopulation.Methods:Weincluded92participants(studentsorstaff)fromtheMedicineCampusattheUniversityofLorraine,Nancy(north-easternFrance).TheFrenchGPAQwascompletedtwice,7daysapart,tostudytest-retestreliability.TheIPAQ-LFwasusedtoassessconcurrentvalidityoftheGPAQ,andparticipantsworeanaccelerometer(ActiGraphGT3X+)for7daystostudycriterionvalidity.ReliabilityaswellasconcurrentandcriterionvalidityoftheGPAQweretestedbytheintraclasscorrelationcoefficient(ICC),Spearmancorrelationcoefficientforquantitativevariables,andKappaandPhicoefficientsforqualitativevariables.BothconcurrentandcriterionvalidityofGPAQwereassessedbyBland-Altmanplots.Results:TheGPAQshowedpoortogoodreliability(ICC=0.37-0.94;Kappa=0.50-0.62)andconcurrentvalidity(Spearmanr=0.41-0.86),butonlypoorcriterionvalidity(Spearmanr=-0.22-0.42).LimitsofagreementfortheGPAQandaccelerometerwerewide,withdifferencesbetween286.5min/weekand601.3min/week.Conclusion:TheFrenchversionoftheGPAQprovideslimitedbutacceptablereliabilityandvalidityforthemeasurementofPAandsedentarytime.ItmaybeusedforassessingPAandsedentarytimeinaFrenchadultpopulation.
简介:TheCET4writingtestiswidelyusedinChina,butwhetheritisamorereliableandvalidtestofwritingforCET4thanothermethodsofevaluationisstillnotclear.Thepurposeofthisstudywastoexaminethereliabilityandthevalidityofamodificationofthistest.Incarryingoutthestudy,atotalof60Band4studentsinZhejiangUniversitytookthemodifiedtest,aCET4testandaFirstCertificatePracticeTest.TheresultsofthisstudydemonstratethatthemodifiedversionisasignificantlymorereliableandvalidwritingtestthanthepresentCET4writingtest.TheseresultssuggestthattheCET4writingtestneedsmodification.
简介:Dataenvelopmentanalysis(DEA)modeliswidelyusedtoevaluatetherelativeefficiencyofproducers.Itisakindofobjectivedecisionmethodwithmultipleindexes.However,thetwobasicmodelsfrequentlyusedatpresent,theC2RmodelandtheC2GS2modelhavelimitationswhenusedalone,resultinginevaluationsthatareoftenunsatisfactory.Inordertosolvethisproblem,amixedDEAmodelisbuiltandisusedtoevaluatethevalidityofthebusinessefficiencyoflistedcompanies.AnexplanationofhowtousethismixedDEAmodelisofferedanditsfeasibilityisverified.
简介:Background:Wearablemonitors(WMs)areusedtoestimatethetimespentinsedentarybehaviors(SBs)andlight-intensityphysicalactivities(LPAs)andtheirassociatedenergycost;however,theaccuracyofWMsinmeasuringbehaviorsonthelowerendoftheintensityspectrumisunclear.Theaimofthisstudywastoassessthevalidityof3WMs(ActiGraphGT3X+;activPAL,andSenseWear2)inestimatingtheintensityofSBandLPAinadultsascomparedwiththecriterionmeasureofoxygenuptake(VO2)measuredbyindirectcalorimetry.Methods:Sixteenparticipants(age:25.38±8.58years)woretheActiGraphGT3X+,activPAL,andSenseWear2devicesduring7sedentary-to-lightactivities.VO2(mL/kg/min)wasestimatedbymeansofaportablegasanalyzer,OxyconMobile(Carefusion,YorbaLinda,CA,USA).Alldataweretransformedintometabolicequivalentsandanalyzedusingmeanpercentageerror,equivalenceplots,Bland-Altmanplots,kappastatistics,andsensitivity/specificity.Results:MeanpercentageerrorwaslowestfortheactivPALforSB(14.9%)andLPA(9.3%)comparedwithotherWMs,whichwere>21.2%.NoneoftheWMsfellwithintheequivalencyrangeof±10%ofthecriterionmeanvalue.Bland-AltmanplotsrevealednarrowerlevelsofagreementwithallWMsforSBthanforLPA.KappastatisticswerelowforallWMs,andsensitivityandspecificityvariedbyWMtype.Conclusion:NoneoftheWMstestedinthisstudywereequivalentwiththecriterionmeasure(VO2)inestimatingsedentary-to-lightactivities;however,theactivPALhadgreateroverallaccuracyinmeasuringSBandLPAthandidtheActiGraphandSenseWear2monitors.