n Italy 1996.docx

n Italy 1996.docx

《n Italy 1996.docx》由会员分享，可在线阅读，更多相关《n Italy 1996.docx（18页珍藏版）》请在冰豆网上搜索。

nItaly1996

EvolutionofCensusStatisticsonEnterprisesinItaly1996-2006:

fromtheTraditionalCensustoaRegisterofLocalUnits.

MonicaConsalvi,LuigiCostanzo,DanilaFilipponi（Istat）

1.Introduction

Intenyears,from1996to2006,Istathascompletelyreformedtheproductionofcensusstatisticsonenterprisesofindustryandservices.In1993,theEuropeanCommissionrequiredthememberStatestorealisebusinessregistersbasedonadministrativedata,tobeusedfortheyearlyproductionofharmonisedofficialstatisticsonthewholepopulationofnon-agriculturalenterprises,whereastheeconomicCensusesarenormallytakeneverytenyears.

In1996IstatstartedtheprojectoftheItalianBusinessRegister（BR）,namedStatisticalRegisterofActiveEnterprises（ASIA）.ASIAhasbeendevelopedthroughthestatisticalintegrationofdifferentadministrativesources,coveringtheentirepopulationofenterprisesofindustryandservices,otherminorarchivesavailable（coveringparticularsectors）,andstructuralbusinessstatisticscurrentlyproducedbyIstat.Inordertoassessthereliabilityofthemethodologyappliedandtotestthedataquality,aspecial“mid-term”Censuswastakenin1998,whoseresultssubstantiallyconfirmedthevalidityofASIAasaproductionprocessaswellasintermsofdataproduced.

So,theCensusof2001（CIS）couldtakeadvantageofthesupportofASIA,thatmadepossibleimportantinnovationsinthesurveytechnique.Moreover,thecomparisonbetweentheBRandtheCensusmadepossibletomeasurethecoverageofbothsourceswithoutperformingapost-enumerationsurvey,andeventoidentifythemissingunitsandtointegratethemintheCensusdisseminationfile.

ThenextstepwasfillingthegapbetweenCensusandtheBRwiththeproductionofterritorialdataonenterprisesthroughtheimplementationinsideASIAofaregisteroflocalunits（ASIA-UL）.Tobuildandupdatethisnewfeaturehasbeenorganisedayearlysurveyonthelocalunitsoflargeenterprises（IULGI）.

Thispaperprovidesanoverviewoftheevolutionprocessthatledfromthetraditionalenumerationofeconomicactivitiestoanintegratedsystemofstatisticalproduction,thatcanbedefinedasacontinuousCensussinceitprovideseveryyearstatisticalinformationontheterritorialdistributionofeconomicactivitiesandtheemployment,sofaravailablethroughtheCensuseverydecade.

2.ThedevelopmentoftheItalianBusinessRegister（ASIA）

2.1Theexperimentalphase

InItaly,asignificantknowhowontheuseofadministrativearchivesforstatisticalpurposesstartedtodevelopsincetheendofthe80s,whenseveralexperimentalstudies,insideandouttheIstat,exploredthetechnicalfeasibilitytosetupastatisticalbusinessregister.

In1994,eventocomplywiththerequirementsoftheEU,IstatopenedacomplexprojectinordertoimplementanItalianBR,whosefirststepwastheproductionofafeasibilitystudy.Theworkgroupinchargedefineditsagendaasfollows:

a.Definitionofametadataframework;

b.Studyofthemainavailableadministrativearchives（definitionofunitsandcharacters,classificationsused,coverage,maintenanceandupdatingprocedures）;

c.Developmentofa“metadatatranslator”,i.e.asetofrulestoconverttheadministrativedataintostatisticalinformation,byidentifyingthestatistically-relevantunitsamongthelegally-relevantones;

d.Setupofarobustmethodologytoestimate/validatethecharactersoftheidentifiedstatisticalunits.

TheacronymASIA（inItalian,ArchivioStatisticodelleImpreseAttive）wasadoptedandthedevelopmentoftheprojectwasoutlinedinthreemajorsteps.

Thefirstphase,startedin1995,consistedincreatingaprototypeoftheBRforthreeItalianprovinces,inordertotestthemethodologicalsolutionstobeadoptedfortheintegrationofadministrativearchives.Forthispurpose,differentlinkageproceduresanddifferentmethodologiesfortheimputationofmissingdatawereexperimented,andasetofrulesforcheckingtheattributesofstatisticalunitswasimplemented.

Thesecondphaseoftheprojectconsistedinextendingtheexperimenttotheentirenationalterritory.AfirstreleaseofASIA,asaresultofthelogicalandphysicalintegrationofadministrativeandstatisticalsources,wasissuedin1996（t）withreferencetotheyeart-2（1994）.TheconstructionprocessoftheBRwasthencompletedin1997,byperformingaqualitycontrolofthepreviousreleases:

thatyearmicrodatafromASIA,referringto1995,weredisseminatedforthefirsttime.

Finally,inthethirdphase,carriedoutin1998-1999,theBRwasvalidatedthroughafieldsurvey,withaspecialIntermediateCensus.

2.2The1996IntermediateCensus

TheIntermediateCensushadadoubleaim.Asusual,itwasasurveytosupplyterritorialinformationontheeconomicstructureoftheCountrybut,atthesametime,itwasalsoageneralcheckoftheinformationabouttheactiveunitsrecordedintheBR.ThedirectsurveyregardedallthemediumandlargeenterprisesofindustryandservicesrecordedintheBR,and–amongthesmallerones–onlythosewithdiscordancesbetweendifferentadministrativesources（mainlyinthenumberofemployeesorintheactivitystatus）.Alltheothersmallerunitsweresimplycheckedthroughadeskreview.Thequestionnairesweresentbymailand,incaseofmissingresponse,theunitswerecontactedbyphoneordirectlyonfieldbyinterviewers.

Thequestionnaireswerepartlypersonalisedwithpre-printedinformationdrawnfromtheBR.Theenterpriseswereonlyaskedtoconfirmortocorrecttheinformationincaseofvariationsand/orerrors,ifany.Alsothedateofthepossiblevariationwastobereported.

Comparedtothetraditionalcensus,thisorganisationimpliedsomeadvantages:

a）Ahighercoveragerate（about95%）,thankstotheuseofadministrativesources;

b）Lessercostsandlesserburdenfortherespondents,thankstotheinnovationofthesurveytechnique（theunitssurveyedwereonlya15%oftheentireuniverse,andtheyhadtoanswerlessquestions）;

c）Ashorteningofthedataprocessingand,then,abettertimelinessinthedatadissemination（thefinaloutputwasreleasedattheendofOctober1998,justaboutoneyearafterthesurvey’sbeginning）.

ToevaluatetheoutcomeoftheIntermediateCensuswascarriedoutaspecificsurvey,showingtheoverallreliabilityoftheBR（seetab.1）.However,theCensushighlightedtwomainerrorsintheBR:

a）Anover-coverageerror,i.e.theinclusionofunitsrecognizedasnotbelongingtotheobservationfield.Thiswasgenerallyduetoerrorsincodingtheeconomicactivity（especiallyforself-employersandenterpriseswithoutemployees,astheyarenotcoveredbyalltheavailablesources.

b）Anunder-coverageofsomeeconomicsector,suchasconstruction,transportationandtradingintermediation.

Moreover,table1reportsthediscordanceratesforthemaincharactersofthestatisticalunits,betweentheBRandtheCensussurvey（takingintoaccountthatpartoftheerrorisjustduetothetimereferencelag）.

Tab.1-ConcordanceanddiscordancebymaincharactersofstatisticalunitsbetweentheBRandthe1996IntermediateCensus

Characters

Concordance

Discordance

Absolutevalues

Percentagevalues

Absolutevalues

Percentagevalues

Total

Ofwhichduetotimelag

ActivitystatusatDecember31,1995

340,808

93.6

23,471

6.4

n.a.

Economicactivitycode

322,844

88.6

41,434

11.4

1.9

Enterprisename

346,865

95.2

17,413

4.8

2.8

Address

288,278

79.1

76,000

20.9

6.7

Juridicalstatus

350,453

96.2

13,825

3.8

1.9

Numberofemployees

317,905

94.0

20,351

6.0

n.a.

2.3Datadisseminationandthere-engineeringoftheBRInformationSystem

InDecember1998theIntermediateCensusdataweredisseminatedviaInternet,throughaDataWarehouse.Forthefirsttimeuserswereallowedtocreatebythemselvestablescomparingthe1996datawiththosefromthepreviousCensuses（upto1971）.Theadvantagesofthistoolofdatadisseminationwereahigherinformationdetail,thecustomisableelaborations,adatabasethatcouldbequeriedviaInternetinrealtime,andfreeaccess/downloadatanylevel.TheDataWarehouserepresentedthepivotoftheCensus’disseminationplan.

Thisdisseminationapproachrequiredare-engineeringoftheBRInformationSystemthroughtherealisationofarelationaldatabase.Theinformationsystemcontainsthehistoricalinformationofthestatisticalunitssince1996i.e.thevaluesoftheirmaincharactersovertheyears.Therelationaldatabasewasrealisedin1999,withthelogicalandconceptualstudyofthedatabaseandthephysicalrealisationofitsfirstfunctions,navigation,visualisation,updating.Inthedatabasesomemetadatawerealsoincluded:

theprocedureusedforcharacterattribution（imputationmodel,estimation,directlyfromsurvey,etc.）;theproductionprocess（survey,integrationofadministrativeregisters）;thesourceofthedata;whetherchangesoccurringintheperiodarevariationsoradjustments;reliabilityofdata（withreferencetothegeneratingprocessandsources）.

Afterthefirstset-upoftheregister,in1999beganthedevelopmentofamultipleupdatingprocedure.Sincerecordedunitsdonothavethesamestatisticalweight,theupdatingprocedureofASIAwasdifferentiatedbysizeclasses.Simplifying,wecansaythatthesmallunits（upto9employees）,correspondingapproximatelytothe95%oftherecordedunits,areyearlyupdatedbytheintegrationprocessofadministrativesources;thecharactersofthemediumsizedunits（10to249employees）areupdateddirectlyfromstatisticalsources（SBS/STS,thatcollecttheneededdatathroughanadditionalform）andthe