1、外文翻译原文STUDIESOFTRENDSINTHIRDPARTYLOGISTICSUSAGEWHATCANWECONCLUDETHISARTICLEEVALUATESTHEVARIOUSLONGITUDINALSTUDIESOFTHIRDPARTYLOGISTICS3PLUSAGEANDPROPOSESMETHODSANDAPPROACHESTOIMPROVETHERIGOROFFUTURERESEARCHINTHISAREATHREEWELLESTABLISHEDMETHODOLOGIESFORCOMBININGTHERESULTSOFMULTIPLESTUDIESAREREVIEWEDL
2、ITERATUREREVIEWS,METAANALYSISANDREPEATEDSURVEYANALYSIS,ANDTHENCONSIDEREDINAPPLICATIONTOTHEFINDINGSOFTHEVARIOUS3PLUSAGESTUDIESSINCE1991BEGINNINGWITHANOVERVIEWOFTHERESULTSTHATEMERGEFROMSIMPLYCOMBININGANDCOMPARINGTHEKEYLONGITUDINALDATA,THEARTICLEMOVESONTOAMOREDETAILEDANALYSISOFTHEMEASUREMENTSCALES,DATA
3、COLLECTIONMETHODS,ANDSAMPLESUTILIZEDINTHEDIFFERENT3PLSURVEYSTHEARTICLEENDSWITHCONCRETERECOMMENDATIONSFORREORIENTINGTHEONGOINGSURVEYSTOINCREASETHEIRRIGORANDVALIDITY,ANDTHUSINCREASETHEIRPREDICTIVEVALUETOBOTHACADEMICSANDLOGISTICSMANAGERSTHIRDPARTYLOGISTICS3PLHASBEENTHESUBJECTOFACADEMICINVESTIGATIONANDA
4、NALYSISINTHEUNITEDSTATESSINCETHELATE1980SANDEARLY1990SINPARTICULAR,THEUSEOF3PLHASBEENASSESSEDBOTHINSINGLESTUDIESBARDIANDTRACEY1991MALTZ1993RABINOVICHELAL1999SHEFFI1990SINKANDLANGLEY1997ANDINANNUALLYREPEATEDSURVEYSARMSTRONG20032004LANGLEY19961997LANGLEYETAL19981999LANGLEYETAL2000LANGLEYETAL20012002LA
5、NGLEYETAL2003LANGLEYETAL2004LIEBANDPELUSO1999LIEBETAL2000LIEBANDKENDRICK2002LIEBANDBENT/20032KMLIEB2000LIEBANDRANDALL1996THEPURPOSEOFTHISARTICLEISTOEVALUATETHEFINDINGSOFTHESESTUDIESANDTODETERMINEWHATTHIRDPARTYUSAGETRENDSHAVEBEENIDENTIFIEDOVERLIMEBASEDONTHERESULTSOFTHISEXERCISE,WEPROVIDEPROCEDURALINS
6、IGHTSUNDERLYINGCURRENTPROJECTIONSOFGROWTHINTHISIMPORTANTMARKETWESEPARATEOURLITERATUREREVIEWINTOTWOSUBSECTIONSTHEFIRSTSUBSECTIONREVIEWSTHREEWELLESTABLISHEDMETHODOLOGIESFORCOMBININGTHERESULTSOFMULTIPLESTUDIES,ANDEXAMINESTHENECESSARYPRECONDITIONSFORSUCHANALYSESTOBEEFFECTIVELYUTILIZEDBYFARTHEMOSTCOMMONA
7、PPROACHTOCOMBININGSTUDIESISTHEFAMILIARLITERATUREREVIEWASECOND,MOREFORMALOPTIONFORINTEGRATINGRESULTSFROMMULTIPLESTUDIESISMETAANALYSISASDEFINEDBYWOLF1986,“METAANALYSISREFERSTOTHEANALYSISOFANALYSESTHESTATISTICALANALYSISOFALARGECOLLECTIONOFANALYSISRESULTSFROMINDIVIDUALSTUDIESFORTHEPURPOSEOFINTEGRATINGTH
8、EFINDINGS“HOWEVER,RECOGNIZINGTHATMETAANALYSISWITHLONGITUDINALDATAISRELATIVELYRARE,WEALSOCONSIDERTHEAPPLICABILITYOFMETHODSDEVELOPEDTOANALYZEREPEATEDSURVEYSFIREBAUGH1997,WHICHEXPLICITLYCONCERNSITSELFWITHPOOLINGDATACOLLECTEDOVERTIMETHESECONDSUBSECTIONOFTHELITERATUREREVIEWLAYSOUTTHEFINDINGSOFTHEVARIOUS3
9、PLUSAGESTUDIESSINCE1991,WITHSPECIALATTENTIONTOTWOANNUALSURVEYS,REFERREDTOHEREASTHE“LIEBSERIES“LIEBANDPELUSO1999LIEB,MILLER,ANDPETERS2000LIEBANDKENDRICK2002LIEBANDBENTZ2003,2004LIEB2000LIEBETAL1993LIEBANDRANDALL1996ANDTHE“LANGLEYSERIES“LANGLEY19961997LANGLEY,NEWTON,ANDTYNDALL19981999LANGLEYNEWTON,AND
10、ALLEN2000LANGLEY,ALLEN,ANDTYNDALL2001,2002LANGLEY,ALLEN,ANDCOLOMBO2003LANGLEYALLEN,ANDDALE2004，ASWELLASASUMMARYOFTHEREPORTEDFINDINGSFROMATHIRDPRIVATESURVEYSERIESARMSTRONG2003,2004INPARTICULAR,WECOMPAREANDCONTRASTTHELIEBANDLANGLEYSERIESANDTHEPOTENTIALAPPLICABILITYOFAMETAANALYTICAPPROACHINASSESSINGTHE
11、IRCOMBINEDRESULTSTHEMETHODOLOGYSECTIONCOVERSTHEAPPLICATIONOFTHEAFOREMENTIONEDCOMBINATORIALTECHNIQUESTOTHELIEBANDLANGLEYSERIESWEBEGINWITHANOVERVIEWOFTHERESULTSTHATEMERGEFROMSIMPLYCOMBININGANDCOMPARINGTHEKEYLONGITUDINALDATAFOUNDWITHINTHESETWOSERIES,ASWELLASACOMPARISONWITHTHEDATACOLLECTEDBYARMSTRONG200
12、32004THESERESULTSLEADTOAMOREDETAILEDANALYSISOFTHEMEASUREMENTSCALES,DATACOLLECTIONMETHODS,ANDSAMPLESUSEDINEACHSETOFSURVEYSINTHESUMMARYANDIMPLICATIONSSECTION,WERELUCTANTLYCONCLUDETHATTHEAVAILABLESURVEYS,ALTHOUGHUNQUESTIONABLYUSEFULTOPRACTITIONERSWHONEEDA“SENSE“OFTHEMARKETPLACE,DONOTPROVIDEACOMPLETEBAS
13、ISFORRIGOROUSLYESTIMATINGTRENDSIN3PLUSAGETHEREFORE,WEENDTHEARTICLEWITHCONCRETERECOMMENDATIONSFORREORIENTINGTHEONGOINGSURVEYSTOINCREASETHEIRRIGORANDVALIDITY,ANDTHUSINCREASETHEIRPREDICTIVEVALUETOBOTHACADEMICSANDLOGISTICSMANAGERSLITERATUREREVIEW,METAANALYSIS,ANDREPEATEDSURVEYSACCUMULATINGKNOWLEDGEACROS
14、SSTUDIESTOCONFRONTTHEORYWITHDATAISESSENTIALTOPROGRESS“FARLEYETAL1995LITERATUREREVIEWMOSTSTUDIESINTHESOCIALANDBEHAVIORALSCIENCESDONOTRESULTINDEFINITIVEANSWERSTOTHERESEARCHQUESTIONADDRESSEDWOLF1986MULTIPLESTUDIESMAYUSEDIFFERENTDEFINITIONS,METHODOLOGIES,SAMPLES,ANDMEASUREMENTINSTRUMENTSTOINVESTIGATETHE
15、SAMEPHENOMENABUTMANAGERSNEEDINSIGHTINTOGENERALTRENDSANDTHEVALUESOFRELEVANTPARAMETERSSOTHATTHEYCANPLANFORFUTURECONTINGENCIESANDSETREALISTICGOALSFORTHEIROPERATIONSSIMILARLY,ACADEMICSACTINGASSOCIALSCIENTISTSWANTTOIMPROVEUPONTHELESSONSLEARNEDFROMEARLIERWORK“LIGHTANDPILLEMER1984THUS,VIRTUALLYALLRESEARCHB
16、EGINSWITHALITERATUREREVIEWTHATSUMMARIZESPREVIOUSRESULTSANDINDICATESTHECONTRIBUTIONOFTHEWORKBEINGREPORTEDTHEINTENTISTOSYNTHESIZETHECURRENTSTATEOFKNOWLEDGEFORTHEISSUEUNDERINVESTIGATION,USUALLYINANARRATIVEFORMATRESEARCHERSHAVEIDENTIFIEDANUMBEROFSHORTCOMINGSINTHISAPPROACHTOCHARACTERIZINGTHE“STATEOFTHEAR
17、T“FOLLOWINGARESOMESPECIFICCHARACTERISTICSOFLITERATUREREVIEWS1SELECTIVEINTHEIRCHOICEOFSTUDIESTOBEREVIEWED,BASEDONTHEAUTHORSJUDGMENTASTOTHEQUALITYOFTHESTUDY2SUBJECTIVEINHOWTHEYWEIGHTTHERESULTSOFVARIOUSSTUDIES,SINCETHEREARENOFORMALRULESFORCOMBININGSTUDIES3NOTWELLSUITEDTOCOMBININGALARGENUMBEROFSTUDIES,A
18、GAINBECAUSEOFTHELACKOFSYSTEMATICMETHODSFORUNDERSTANDINGRELATIONSHIPSAMONGTHESTUDIESDIFFERENCESINDEFINITIONS,METHODOLOGIES,ANDSAMPLESAREUSUALLYNOTADDRESSEDWITHANALYTICALTOOLSLIGHTANDPILLEMER1984,WOLF1986INTEGRATIVELITERATUREREVIEWSINTHETHIRDPARTYLITERATUREDIFFERBASEDONTHEIRPURPOSEMOSTREVIEWSAREPERFOR
19、MEDINCONJUNCTIONWITHPROJECTSADDRESSINGSPECIFICISSUESFOREXAMPLEMURPHYANDPOIST1998DERIVETESTABLEPROPOSITIONSFORUSAGEBASEDONINTERPRETATIONOFPREVIOUSWORKBYSINKETAL1996LIEBANDRANDALL1996ANDSTANKANDMALTZ1996BOYSONETAL1999CONSIDERWORKBYBARDIANDTRACEY1991LIEBANDRANDALL1996ANDSINKANDLANGLEY1997ANDCONCLUDETHA
20、T“NONEOFTHEPREVIOUSACADEMICWORKHASRESEARCHEDWAYSTOINTERNALLYPLAN,ORGANIZE,OPERATE,ANDCONTROLTHIRDPARTYRELATIONSHIPS“THESEPROJECTSPECIFICREVIEWSAREDESIGNEDTOIDENTIFYAPPROPRIATETOPICSFORRESEARCH,BUTARENOTSUITABLEFORIDENTIFYINGTRENDSINTHEVALUESOFKEYQUANTITIESSUCHASACTUAL3PLUSAGETHEEMPHASISISONPOSITIONI
21、NGNEWWORK,RATHERTHANSYNTHESIZINGPREVIOUSEFFORTSTODRAWCONCLUSIONSTHESEREVIEWSHAVESOMEOFTHESHORTCOMINGSIDENTIFIEDABOVE,INTHATTHEIRSELECTIONISNECESSARILYPARTIALANDSUBJECTIVE,ASISTHEWEIGHTINGOFTHEVARIOUSSTUDIESBUTRESEARCHERSINTHIRDPARTYUSAGEHAVEALSOPRODUCEDASECONDKINDOFREVIEWLIEBANDRANDALL1996DRAWCONCLU
22、SIONSBASEDONSUCCESSIVESTUDIESCOMPLETEDIN1991,1994AND1995THESECONCLUSIONSCONTINUELOBEUPDATEDBASEDONREPEATEDSURVEYSBYLIEBANDHISCOLLABORATORS,THELIEBSERIESREFERREDTOEARLIERSIMILARLY,LANGLEYANDVARIOUSCOLLABORATORSCOMMENTONTRENDSINTHEIRSERIESOFREPEATEDSURVEYS,EARLIERIDENTIFIEDASTHELANGLEYSERIESBUTFORMALT
23、RENDANALYSISANDPROJECTIONISNOTANOBJECTIVEOFEITHERAUTHORINSTEAD,THEYCONCENTRATEONFMDINGSFROMTHEMOSTRECENTWORKANDUSEPREVIOUSSURVEYSPRIMARILYFORCONTEXTSOMEEXAMPLESINCLUDETHEFOLLOWINGFROM1996TO2001THEPERCENTAGEOF3PLUSERSREMAINEDRELATIVELYCONSTANTBETWEEN68PERCENTAND73PERCENTHOWEVERTHISPERCENTAGEINCREASED
24、TO78PERCENTIN2002AND2003,ANDTO79PERCENTIN2004LANGLEYETAL2004“SIXTYTLVE69PERCENTOFTHE94RESPONDENTSINDICATEDTHATTHEIRCOMPANIESUSETHIRDPARTYLOGISTICSSERVICES,THISUSAGERATEISNEARLYIDENTICALTOTHATREPORTEDINI997“LIEBANDRANDALL1998,ASREPORTEDINLIEB2000BOTHLIEBANDLANGLEYASKRESPONDENTSABOUTFUTUREOUTSOURCINGP
25、LANS,BUTNEITHERATEMPTSTOCOMBINERESULTSFORPREDICTIONFROMPREVIOUSSURVEYSTHROUGHFORMALRULES,NORDOTHEYTYPICALLYINCLUDEMORETHANONEORTWOSTUDIES,EXCEPTINDESCRIPTIVEFIGURESORTABLESLANGLEYALLEN,ANDDALE2004LIEBANDRANDALL1996ASNOTEDABOVELIGHTUNDPILLEMER1984ANDWOLF1986HAVERAISEDCONCERNSABOUTTHEUTILITYOFINTEGRAT
26、IVEREVIEWS,WHICHLACKFORMALRULESFORWEIGHTINGANDCOMBININGIHEVARIOUSSTUDIESTHESECONCERNSSEEMTOAPPLYTOTHESTUDIESANDCOMMENTSOFBOTHLIEBCIALANDLANGLEYETALTHEWORKINTHISARTICLEISANATTEMPTTOADDRESSTHESECONCERNSANDAPPLYFORMALRULESTOTHETASKOFDRAWINGCONCLUSIONSFROMTHESEMULTIPLESTUDIESTHEREISONEOTHERWELLKNOWNSERI
27、ESOFSTUDIESOFTHETHIRDPARTYLOGISTICSMARKET,THATOFARMSTRONGANDASSOCIATES20032004INTERESTINGLY,PUBLISHEDARMSTRONGDATAUSUALLYCHARACTERIZEMARKETSIZEINDOLLARSSPENTONTHIRDPARTYSERVICES,RATHERTHANPERCENTUSAGEHOWEVERARMSTRONGSREPORTSAREPROPRIETARYANDITISNOTCLEARWHAT,IFANYFORMALRULESAREUSEDINMARKETPROJECTIONS
28、THEREFORE,THEARMSTRONGSERIESWILLNOTBEUSEDINDRAWINGSTATISTICALINFERENCESINTHISSTUDYASDETAILEDBELOW,THESERIESWILLBEUSEDONLYASAREFERENCEBENCHMARKFORDESCRIPTIVESTATISTICSINCLUDEDINFIGURE1ANDTABLE1METAANALYSISMETAANALYSISISASETOFPROCEDURESTOQUANTITATIVELYCOMBINERESULTSOFMULTIPLERESEARCHSTUDIESFARLEY,LEHM
29、ANN,ANDSAWYER1995METAANALYSISWASFORMULATEDASAWAYTOADDRESSRECOGNIZEDPROBLEMSINSTANDARDLITERATUREREVIEWS,INCLUDINGSELECTIVEINCLUSIONOFSTUDIESBASEDONTHEREVIEWERSOPINIONS,EMPHASIZINGONESTUDYOVERANOTHERBASEDONSUBJECTIVECRITERIA,ANDLACKOFCONSISTENCYINTHEINTERPRETATIONOFSTUDYRESULTSANDSPECIFICCHARACTERISTI
30、CSWOLF1986INMETAANALYSISANINDIVIDUALDATAPOINTREPRESENTSTHERESULTOFANENTIRESTUDY,RATHERTHANTHEOUTCOMEASSOCIATEDWITHASINGLEPARTICIPANTINASTUDYTHESTUDYLEVELDATAPOINTSARETHENSUBJECTEDTOSTATISTICALANALYSISOURINTERESTISINESTIMATINGTHECHANGEOVERTIMEINTHELEVELOFTHIRDPARTYLOGISTICSACTIVITYINTHEUNITEDSTATESSI
31、NCETHEREAREANUMBEROFDIFFERENTSTUDIES,SOMEDONEONLYONCEEGMURPHYANDPOIST1998ANDOTHERSTHATAREREPEATEDTHELIEBSERIESANDLANGLEYSERIES,ITSEEMSPLAUSIBLETOCONSIDERAMETAANALYTICAPPROACHTOESTIMATINGTHETHIRDPARTYLOGISTICSGROWTHRATEUSINGDATAFROMTHEMAXIMUMNUMBEROFSTUDIESBUTASUCCESSFULMETAANALYSISHASTODEALWITHFOURM
32、AJOROBSTACLESWOLF19861DIFFERENCESINDEFINITIONS,MEASUREMENTTECHNIQUES,ANDSAMPLEMAKEUPMUSTBEMINIMIZEDANDACCOUNTEDFOROTHERWISE,THEMETAANALYSISWILLFAILBECAUSEITLUMPSTOGETHER“APPLESANDORANGES“2ITISNECESSARYTOEVALUATETHE“QUALITY“OFEACHSTUDYSDESIGN,SOTHAT“POOR“STUDIESARENOTWEIGHTEDEQUALLYWITH“GOOD“STUDIES3
33、ITISIMPORTANTTOSEEKOUTANYSTUDIESTHATMAYHAVENONSIGNIFICANTFINDINGSINOURCASE,NOEVIDENCEOFCHANGEINTHIRDPARTYUSAGEMANYMETAANALYSESLIMITTHEMSELVESTOPUBLISHEDRESEARCHBUTEXPERIENCEHASSHOWNTHATSIGNIFICANTORSTRIKINGFINDINGSAREMORELIKELYTOBEPUBLISHED,SOUSINGONLYPUBLISHEDWORKISLIKELYTOLEADTOBIASEDESTIMATES4THE
34、USEOFMULTIPLERESULTSFROMONESTUDY,EGESTIMATESOFUSAGEFORSEVERALLOGISTICSACTIVITIES,MUSTBEEXPLICITLYDEALTWITHINANYAGGREGATEDSTATISTICALTESTSOURTREATMENTOFTHIRDPARTYLOGISTICSSTUDIESWILLENDEAVORTOADDRESSALLOFTHESEISSUESINSOMEFASHION,SUBJECTTOTHELIMITATIONSOFTHEUNDERLYINGSTUDIESTHEMSELVESCONCLUSIONTHIRDPA
35、RTYLOGISTICSWASIDENTIFIEDASASEPARATEINDUSTRYANDSERVICEONLYINTHELATE1980S,ANDTHEREHAVEBEENNUMEROUSSTUDIESOFTHEINDUSTRYASITHASGROWNTHERESEARCHPRESENTEDHEREISANATTEMPTTOFORMALLYCOMBINETHERESULTSOFSOMEOFTHESESTUDIESTOBETTERESTIMATETHEGROWTHRATEOFTHIRDPARTYLOGISTICSINGENERALANDOFWAREHOUSEANDTRANSPORTATIO
36、NOUTSOURCINGASWELLWEWEREFORTUNATETOACCESSTWOSEPARATESETSOFSTUDIESOFTHESEPHENOMENA,LEDBYROBERTLIEBANDJOHNLANGLEYJRRESPECTIVELYWEESTIMATEDTHATFROM1996TO2004MARKETPENETRATION,MEASUREDINPERCENTCOMPANIESOUTSOURCING,HASBEENINCREASINGFROM58PERCENTANNUALLYFORLOGISTICSOUTSOURCINGINGENERALTRANSPORTATIONOUTSOU
37、RCINGSEEMSTOHAVEEXPERIENCEDAPPROXIMATELYTHESAMERATEOFINCREASEDPENETRATION,WHILETHIRDPARTYWAREHOUSINGAPPEARSTOHAVEGROWNSOMEWHATFASTERBUTTHEFUNCTIONALINCREASESHAVENOTBEENUNIFORMTHEDATAWEHAVESUGGESTTHATFUNCTIONALOUTSOURCINGINTHELARGECOMPANYSEGMENTDURINGTHISPERIODDOESNOTSHOWCLEARCUTGROWTHWEADOPTEDAMETAA
38、NALYSISPERSPECTIVETOARRIVEATTHESEESTIMATES,INTHESENSETHATWEESTIMATEDEFFECTSIZESACROSSMULTIPLESTUDIESINRETROSPECT,ALTHOUGHMEASURINGGROWTHRATESREQUIRESMULTIPLESTUDIES,THEREMAYBEMORESTRAIGHTFORWARDAPPLICATIONSINLOGISTICSFORMETAANALYSISPROCEDURESFOREXAMPLE,MANYSTUDIESHAVETRIEDTORELATECUSTOMERSERVICETOCU
39、STOMERSATISFACTIONUSINGMETAANALYSISTOCOMBINETHESESTUDIESCOULDRESULTINMORERELIABLE,CONSISTENTESTIMATESOFTHEIMPORTANCEOFSERVICETOOVERALLSATISFACTIONMANYSTUDIESHAVELOOKEDATTHEORDEROFIMPORTANCEOFVARIOUSCOMPONENTSOFCUSTOMERSERVICEMETAANALYSISOFFERSANORGANIZEDMETHODOFCOMBININGTHERESULTSOFTHESESTUDIESANDDR
40、AWINGSTRONGCONCLUSIONSASTHELOGISTICSFIELDMATURESANDTHENUMBEROFSTUDIESPROLIFERATES,INFORMALREVIEWSOFMULTIPLESTUDIESBECOMEMORECHALLENGINGITBECOMESHARDERANDHARDERTORECONCILE“DISPARATEDEFINITIONS,VARIABLES,PROCEDURES,METHODS,SAMPLES,ANDSOON“WOLF1986YET“ACCUMULATINGKNOWLEDGEACROSSSTUDIESTOCONFRONTTHEORYW
41、ITHDATAISESSENTIALTOPROGRESS“FARLEYLEHMANN,ANDSAWYER1995INLOGISTICSASINOTHERFIELDSWEDONOTCLAIMTHATMETAANALYSISISANEASYWAYTOANSWERFUNDAMENTALLOGISTICSQUESTIONSABOUTTHEPRESENCEANDSIZEOFMARKETS,THEINFLUENCEOFKEYPERFORMANCEINDICATORS,ORTHEIMPORTANCEOFINDIVIDUALCUSTOMERSERVICEELEMENTSBUTWEBELIEVETHATMETA
42、ANALYSISMAYPROVIDEAFRAMEWORKFORMOREINDEPTHSTUDIESANDAPPRECIATIONOFINDIVIDUALSURVEYRESULTSANDADISCIPLINEDWAYTOCOMBINETHESERESULTSTOIMPROVEOURUNDERSTANDINGOFIMPORTANTLOGISTICSPHENOMENAWELOOKFORWARDTOWORKINGWITHOURCOLLEAGUESTOSEEIFTHISTECHNIQUEANDITSCOUSIN,REPEATEDSURVEYANALYSIS,CANBEASUSEFULTOOURDISCI
43、PLINEASITALREADYHASPROVENTOBEINOTHERSOCIALANDBEHAVIORALSCIENCESSOURCEBRYANASHENBAUM,ARNOLDAALTZ,ELLIOTRABINOVICHSTUDIESOFTRENDSINTHIRDPARTYLOGISTICSUSAGEWHATCANWECONCLUDEJSUMMER2006,VOL48ISSUE4,P84108,25P,3DIAGRAMS,1CHART译文研究第三方物流的趋势我们的结论摘要本文评价了第三方物流（3PL）纵向研究的多方面的使用和建议的方法以及提高这个领域内未来严谨研究的思路。三种已经得到确认的
44、研究方法结合了多种学习和评估的结论（文学评论,荟萃分析和重复调查分析），然后考虑种群中的应用研究发现各种第三方物流从1991年以来就开始被使用。开始于一个从简单的结合和比较关键的纵向资料的大致结果，文献移到更为详细的分析测量。规模、数据采集方法和样品用于不同的第三方物流调查。文章结尾具体建议,为调整正在进行的调查来增加他们的严格与有效性，因此增加双方学者和物流经理的预测值。在美国80年代末和90年代初,第三方物流一直在美国被学术调查和分析特别是，第三方物流的使用已被两个单项研究评估（BARDIANDTRACEY1991MALTZ1993RABINOVICHELAL1999SHEFFI1990S
45、INKANDLANGLEY1997）,并在每年多次调查（ARMSTRONG20032004LANGLEY19961997LANGLEYETAL19981999LANGLEYETAL2000LANGLEYETAL20012002LANGLEYETAL2003LANGLEYETAL2004LIEBANDPELUSO1999LIEBETAL2000LIEBANDKENDRICK2002LIEBANDBENT/20032KMLIEB2000LIEBANDRANDALL1996）这篇文章来说，目的是评估这些研究的结果，久而久之确定哪些第三方使用趋势已经被鉴定。基于上述工作的结果，我们对目前的增长提供相关
46、的预测见解程序在这个重要的市场我们将我们的资料回顾分成两个小部分。第一部分的回顾是有三个有效的方法去将多个研究结果相结合，并为了这种分析研究了有效的必要条件成为有效的利用。到目前为止，最常用的相结合研究是熟悉的文献回顾。第二，更多的正式选择对于多个研究的集成结果是荟萃分析。正如沃尔夫定义（1986年），“荟萃分析是指分析报告的解析。大收集个别研究分析结果的统计分析是为了集成调查结果。”然而，确认到到带有纵向数据的荟萃分析是比较少见的，我们也考虑去发展分析反复调查的方法的适用性（格伦菲尔鲍1997年），其中明确担心的问题集中在长时间下的公用数据收集。第二部分的回顾，展示出自1991年以来的各种使
47、用第三方物流的研究结果,特别注意两个年度的调查，在这里指“LIEB系列”（LIEBANDPELUSO1999LIEB,MILLER,ANDPETERS2000LIEBANDKENDRICK2002LIEBANDBENTZ2003,2004LIEB2000LIEBETAL1993LIEBANDRANDALL1996）和“LANGLEY系列”LANGLEY19961997LANGLEY,NEWTON,ANDTYNDALL19981999LANGLEYNEWTON,ANDALLEN2000LANGLEY,ALLEN,ANDTYNDALL2001,2002LANGLEY,ALLEN,ANDCOLOMB
48、O2003LANGLEYALLEN,ANDDALE2004）。我们还有一个调查报告的摘要是来自第三个私人调查系列（ARMSTRONG2003,2004）。特别是，我们比较和对比了LIEB和LANGLEY系列和评估了他们的综合结果的荟萃分析的开发潜力。这节的方法涵盖了上述LIEB系列和LANGLEY系列的组合技术应用。我们用一个结果的总结开始，这个总结在简单的结合和比较了这两个系列里的关键纵向数据后出现，以及和ARMSTRONG收集到的数据进行一个比较（2003，2004年）。这些结果导致了对衡量尺度，数据收集方法和在每个组使用的调查样本的更详细分析。在总结和影响部分，我们勉强地得出这样的结论提
49、供调查，但无疑有助于需要一个“感官”市场的开发者，不为第三方物流的使用趋势估提供一个完整的基础。因此，我们结束与重新调整正在进行调查，以增加他们的严密性和有效性，从而提高他们的预测值都学者和物流管理的具体建议的文章。文献回顾在社会和行为科学的大多数研究结果中没有问题的解决方法（沃尔夫1986年）和明确答案。多项研究可能使用不同的定义，方法，样本，以及和调查、测量仪器同样的现象。然而，经理们需要将总的趋势转化成为相关参数值，使他们可以规划未来的突发事成为他们的业务洞察现实的目标。同样地，作为社会科学家的学者需要从行事“提高对早期作品”（光明与PILLEMER1984年）中吸取教训。因此，几乎所有的研究开始于一个文献回顾，总结了以前的结果，并指出工作的贡献然后上报。其目的是为了合成受调查问题的知识现状，通常是一种叙述的格式。研究人员已经确定了一些“艺术评论”的缺点。以下是一些具体的文学评论的特点1在他们对研究选择选择性进行审查时，对提交的判断为基础，加以质量的研究2主观上他们权衡各种研究的结果，因为有研究相结合而没有正式的规则。3不适合由于系统方法的研究之间的关系的认识缺乏大量的研究。不同的是，方法的差异和通常不同样品处理与分析工具。基于他们的目标，在第三方文献差异综合运用了文献资料。大多数评论是结合项目进行解决的具体问题。例如。MURPHY和POIST（1
