Eindhoven University of Technology

MASTER

The preferred characteristics of coworking spacesthe relation between user characteristics and preferred coworking space characteristics : anattribute based stated choice experiment

van de Koevering, J.G.C.

Award date:2017

Link to publication

DisclaimerThis document contains a student thesis (bachelor's or master's), as authored by a student at Eindhoven University of Technology. Studenttheses are made available in the TU/e repository upon obtaining the required degree. The grade received is not published on the documentas presented in the repository. The required complexity or quality of research of student theses may vary by program, and the requiredminimum study period may vary in duration.

General rightsCopyright and moral rights for the publications made accessible in the public portal are retained by the authors and/or other copyright ownersand it is a condition of accessing publications that users recognise and abide by the legal requirements associated with these rights.

• Users may download and print one copy of any publication from the public portal for the purpose of private study or research. • You may not further distribute the material or use it for any profit-making activity or commercial gain

Thepreferredcharacteristicsofcoworkingspaces

Therelationbetweenusercharacteristicsandpreferredcoworkingspacecharacteristics:anattributebasedstatedchoiceexperiment.

J.G.C(Jasper)vandeKoeveringMaster’sthesisEindhoven,2017

1

Thepreferredcharacteristicsofcoworkingspaces

Therelationbetweenusercharacteristicsandpreferredcoworkingspacecharacteristics:anattributebasedstatedchoiceexperiment.

ColophonMaster’sthesisfortherequirementoftheMasterofScience(MSc)degreeatEindhovenUniversityofTechnologyFacultyoftheBuiltEnvironmentDepartmentofArchitecture,BuildingandPlanningChairofRealEstateManagementandDevelopmentStudentJ.G.C.(Jasper)vandeKoeveringStudentnumber:0829742GraduationcommitteeChairmangraduationcommittee:Prof.Dr.T.A.(Theo)ArentzeFirstgraduationmentor:Dr.Ir.H.A.J.A.(Rianne)Appel-MeulenbroekSecondgraduationmentor:Ir.M.(Minou)Weijs-PerréeEindhoven,February2017Coverphoto:http://www.inkorsivo.com

2

PrefaceThisreportpresentsmymaster’sthesisforthecompletionofthemastertrackRealEstateManagement&DevelopmentofthedepartmentArchitectureBuildingandPlanning(ABP)atEindhovenUniversityofTechnology(TU/e).Duringthepastyear,Ihavestudiedthepreferredcharacteristicsofcoworkingspacesanddeterminedwhetherthepreferencesdifferonusercharacteristics.Thefulfilmentofmyresearchwouldnothavebeenpossiblewithouttheadvice,supportandcooperationofothers.Iamgratefulandwishtothankeveryonewhohelpedmeduringtheprocess,especiallythefollowingpersons.ThisstudywasnotpossiblewithoutthesupervisionofRianneAppel-Meulenbroek,MinouWeijs-PerréeandTheoArentze.Theyinspiredmewiththeirconstructivefeedbackandcriticalview.Besidesmysupervisors,IalsowouldliketothankPetervanderWaerdenandMandyvandeSande–vanKasteren,whohelpedmewiththeresearchdesignofthisstudy.Secondly, I want to thank ‘Het Academisch Fonds Vastgoedkunde’ which providedme a scholarship to perform thisresearch independently. Furthermore, I want to thank the participating coworking spaces which made this researchpossible.Finally,abigshout-outtomyparentsforgivingmetheopportunityoffollowingthismasterattheEindhovenUniversityofTechnologyandtheirunconditionalsupport,helpandguidance.J.G.C.(Jasper)vandeKoeveringTilburg,February2017

3

ExecutivesummaryIntroductionTrendsinthemarkethavecreatedanewparadigminofficespaceusage(Calder&Courtney,1992).Manyindustriesandsectorsareshiftingtowardsanincreasinglycomplex,unpredictableanddynamicknowledgeenvironment(Saurinetal.,2008).Inamarketthatrequiresconstantchanges,flexibilityinthedurationofthelease,theuseofservicesandthechoiceof space isdesired (Gibson&Lizieri,1999). Inaddition to flexibility, trends in the labormarket suchas the increasingnumberofself-employedworkers,growthintheuseoftechnologiesandadecreasingamountofworkspaceallocatedtoindividuals, has influence the use of office space (Green, 2014; De Vries & Van de Besselaar, 2013; Barber, Laing, &Simeone,2005).Therearedifferentspaceneeds(surfaces,layout,spacesetc.),contract(foraday,amonth,ayearetc.)andservices (catering, ICT,administrationetc.) (Laterveer,2011;Ketting,2014).Thechangingdemandofofficespaceusagehaveledtothegrowthofanewtypeofproperty;themulti-tenantoffice(Peltier,1992).Thedefinitionofamulti-tenantofficeisabuildinginwhichofficespaceandpossiblyanumberofsharedfacilitiesand/orservicesareofferedtomultipleorganization(Weijs-Perrée,Appel-Meulenbroek,DeVries,&Romme,2016).Fourtypesofmulti-tenantofficescanbedistinguishedaccordingtoWeijs-Perréeetal.(2016),namely:theregularbusinesscenter,theservicedoffice,theincubatorandthecoworkingspace.Eachofthesetypesofmulti-tenantofficesfocusonadifferentpartmarketregardingthetypeoftenant,atmosphere,servicesspacesandtheoverarchinggoal.Theemergenceofcoworkingspaces was a reaction to the absence of an appropriate workplace for self-employed workers, freelancers andentrepreneurswhobecametiredofthedistractioninlocalcoffeebarsorthelonelinessofworkingfromhome(Moriset,2014;Fuzietal.,2014).Fromthepreviousdecade,agrowingtrendregardingcoworkinghasbecomevisible.Thispopularityistheresultof‘’ashiftingattitudetowardswork’’(Sykes,2014,p.142).AccordingtorecentstudyofDeskmag(2015),therewere7,800coworkingspacesworldwidein2015incomparisonwith310in2009.Thenumberofcoworkerswas510,000in2015 incomparisontothe43,000 in2011.Thedefinitionofacoworkingspace is,asreportedbySpinuzzi (2012),aworkspaceforthecommunity,wherepeoplecansitdownandcollaboratewitheachother.Itisalow-costworkspaceandeasilyaccessiblewithsharedknowledgefromdifferentbusinessbackgroundthatcancontributetotheownbusiness.Theconceptofcoworkingreceivedalotofattentionfromvariousmediasbutisalmostignoredbytheacademicliterature(Moriset,2013;Merkel,2014).Sincecoworkingisarelativelynewconceptofmulti-tenantoffice,littleresearchhasbeendone in this field of user characteristics and coworking space characteristics. With the aid of this study, the usercharacteristicsofcoworkersandtheirpreferredcoworkingspacecharacteristics in theNetherlandsbecomeclear.Thetypical characteristics of coworking spaces are an under-exposed subject as well and will be discussed in this study.Eventually,theaimofthisstudyistousergroupsofcoworkingspacesbasedontheirpreferencesandanalyzewhethertheseusergroupsdifferonusercharacteristics.Thefollowingresearchquestionisformulated:Candifferentusergroupsbeidentifiedbasedontheirpreferencesforcoworkingspacecharacteristicsanddotheydiffer

onusercharacteristics?Tobeabletoanswerthisresearchquestion,thefollowingsub-questionsmustbeanswered:

Whatarethetypicalcharacteristicsofcoworkingspaces? Whichtypesofuserscanbeidentifiedincoworkingspacesandwithwhichcharacteristicscantheybedescribed? Whataretheusercharacteristicsofcoworkersingeneral? Whataretheuserpreferencesforcoworkingspacecharacteristics? Candifferentusergroupsbeidentifiedbasedontheirpreferences? Whatarethedifferencesbetweentheseusergroupsbasedonusercharacteristics?

Whentheuserpreferencesforcoworkingspacecharacteristicsbecomesclear,coworkingspacedevelopersandprovidersshouldbemorecapableofofferingamoresuitableworkenvironmentforcoworkers.Thecoworkingspacecanbemoreeffectively developed because providers and developers can respond to the needs and requirements of the user.Moreover, thisstudy isalsorelevant forusersandfutureusersofcoworkingspaces.Multiplecoworkingspaces intheNetherlandswill be analyzed in this study and corresponding characteristics comeacross.Users and futureusers candeterminewhichcoworkingspacecharacteristicssuitstheminordertodecidewheretheywanttowork.Thishelpsthecoworkers to searchmore specificwhat fits their personal needs. This study provides also an overview of the users’motivationstoworkinacoworkingspace.Inthismanner,futureusersknowwhattheintentionsareofcurrentusersinordertoseeifitcorrespondstotheirownintentions.

4

LiteratureReviewBradNeuberg,theinitiatorofthecoworkingmovement,startedthefirstcoworkingspacein2005inSanFransisco.Thecoworking spacewasmeant to be a physical space inwhich independent andmobileworkers could come together.Neubergstartedthisspacebecausehewantedtoworkbyhimself,butinthemiddleofacommunity.Coworkingspacesare of different sizes and have various characteristics; however, as unique as they are, they all have commoncharacteristics.KwiatkowskiandBuczynski(2011)developedfivecore-valueswhichreflectthecoworkingphilosophyandcanbeconsideredasthecradleofthecoworkingphenomenon.Thefollowingcore-valuescanbedistinguished:

• Collaboration:thewillingnesstocooperatewithotherstocreatesharedvalue;• Openness:freesharingofideas,informationandpeople;• Community:agroupwithasharedpurposeandsharedthoughtsaboutcoworking;• Accessibility:coworkingspacesareaccessible,bothsociallyandeconomically,foreverytypeofworker;and• Sustainability:resourcesareusedtogether,whichbringseconomicalandecologicalbenefits.

Togetherwithvariouscharacteristicsofmulti-tenantoffices(location,officeexterioranddivision,officedécorandfacilitiesand services),whicharederived fromprevious reserach (Leesman,2013; Liebregts,2013;VanSusante,2014;Hartog,2015), the typical characteristics of coworking spaces are developed. The following coworking space characteristicsdistinguishacoworkingspacefromanothertypeofmulti-tenantoffice:location,officeexterioranddivision,officedécor,facilitiesandservices,communityandsustainabilityandaccessibility.Itisofimportancetoemphasizethatthepresentedcharacteristics are only the typical characteristics of coworking spaces; therefore, characteristics that truly define acoworkingspace.Allthecharacteristicsoccurinatypicalcharacter/capacity.Coworkingisoftenassociatedwithself-employedworkersandfreelancers;however,itbecameobviousthattheusersofcoworkingspacesareamoreheterogeneousgroup.Accordingtotherelevantliterature,thefollowingusergroupsmakeuseofacoworkingspaceandareusedinthisstudy:self-employedworkers,freelancersandentrepreneurs,employeeofacompany(2-10;11-50;50ormore)andstudents.Eachusergroupcontainsagreatvarietyofuserswithadifferentsetof user characteristics.With the aid of geographic characteristics (e.g. country of the coworking space), demographiccharacteristics(e.g.gender,age,nationalityandeducation)andpsychographiccharacteristics(motivationstoworkinacoworkingspace)thecoworkerswillbedescribed.Eachcharacteristic,forboththeindependentanddependentcharacteristic,consistsofmultiplevariables.Thefollowingconceptualmodel shows the relation between the user characteristics and the user preferences for coworking spacecharacteristics:

Figure3.1:ConceptualmodelResearchDesignToclarifythemethodologyofthisresearch,themethodsaresplitintotwoforms,namely:methodsfordatacollectionandmethods for data analysis. To collect the data for the independent factor user characteristics of coworkers, surveyquestionsareaskedtodeterminethegeographic,demographicandpsychographiccharacteristics.Tocollectthedataforthe dependent factor user preferences, an attribute based stated choice method is used. In this type of method, arespondent(coworker)isplacedinaparticularframeworktocomparealternativesofcoworkingspacesthataredescribed

5

bymultipleattributes(Adamowicz,Boxall,Williams,&Louviere,1998).Therespondenthastochoosetheirmostpreferredcoworkingspace/alternative. Intotal,66coworkingspaces intheNetherlandswereapproached if theywerewillingtocooperateinthestudy.Thesecoworkingspacesareselectedfrommultiplecoworkingwebsites.Ane-mailwassendtotheoperatorsofcoworkingspacesthroughouttheNetherlandswithashortintroduction/explanationoftheresearchandtherelevancefortheoperators.Twenty-fivecoworkingspaceswerewillingtocooperateinthestudy.Dataiscollectedwiththe use of an online questionnaire. In total, 219 coworkers have indicatedwhat their preferences are for the typicalcoworkingspacecharacteristics.Eachcharacteristicistranslatedintoanattributewithmultipleattributelevels.Themostimportanttypicalcharacteristicsareselectedinorderthatthealternativesrepresentthebesthypotheticalsituationofacoworkingspace.Theattributelevelsindicateaproperrangeofvariationforeachattribute(Adamowiczetal.,1998).Thefollowingattributesandattributelevels(betweenthebrackets)areused:

• Location:accessibilityofthelocation(bycarandpublictransport;bycar;bypublictransport);• Officeexterioranddivision:layoutofthespace(openlayout;half-openlayout;closedlayout);• Office exterior and division: diversity in supply spaces (basic coworking space; standard coworking space;

premiumcoworkingspace);• Officedécor:atmosphereandinterioraesthetics(industrial;modern;homey);• Facilitiesandservices:receptionandhospitality(noreceptionandnohost;receptionbutnohost;receptionand

activehost);• Communityandsustainability:events(none;sometimes;often);• Accessibility: diversity of tenants (no diversity of tenants; moderate diversity of tenants; strong diversity of

tenants);and• Accessibility:typeofleasecontract(nocontract;shortterm(day,weekormonth);longterm(yearorlonger).

Multiplemethodsareusedforthedataanalysis.Theusercharacteristicsofthecoworkersingeneralwillbedescribedbydescriptivestatistics(occurrenceofvariables,mean,standarddeviationandrange)andgraphicsthatshowthedistributionofthevariables.Itbecomesclearwhatthegeographic,demographicandpsychographiccharacteristicsareoftheusersinthesample.Forinstance,whatistheirlevelofeducation,sectoroforganizationandmotivationstoworkinacoworkingspace. To analyze what the user preferences are for coworking space characteristics, amultinomial logit model isestimated.With the aidof amultinomial logitmodel, an estimation canbemadeofwhichof the characteristics of acoworkingspacearethemostpreferredforcoworkers.Thistypeofmodelestimatestheutilityweightsofparameters.Therespondents in the sample attain a certain level of utility to a characteristic of a coworking space (attributelevel/parameter).Theparameterwiththehighestutilityhasthelargestprobabilitytobechosenandcanbeconsideredasmostpreferred.Theparameterwiththelowestutilityhasthelowestprobabilitytobechosenandcanbeconsideredastheleastpreferred.Tofindsimilargroups(classes)inthedataofpreferredcharacteristicsofcoworkingspaces,alatentclasslogitmodelisestimated.Inthecontextofthisresearch,coworkerswithmatchingpreferencesformalatentclass.Thepreferredcharacteristicsinalatentclasswillbethesamebuttheusercharacteristicswilldiffer.Withtheaidofmultiplechi-squaretestsandindependentsamplesT-tests,thedifferencesonusercharacteristicsbetweentheestimatedclassesaredetermined.ResultsThis study provides insight in the user characteristics of coworkers in the Netherlands. A number of demographiccharacteristics are remarkable. First of all, the coworkers in the Netherlands are highly educated (higher vocationaleducation or a degree on university). The most dominant user group in the sample are the self-employed workers,freelancersandentrepreneurswithashareof53%.Almost70%ofthisusergrouphasaboard/ownerposition.Almosthalfof the coworkers in the sample is active in the sector consultancy (25%) or IT (21%). These three demographiccharacteristicsarequiteinlinewiththeusercharacteristicsofcoworkersonagloballevelaccordingtomultiplestudiesofDeskmag(Deskmag,2013;Deskmag,2016).Themostimportantmotivationtoworkinacoworkingspaceistoseparatework andprivate life. The secondmost importantmotivation is the vibrant and creative atmosphere followedby theaffordableaccommodation.According to the 219 coworkers in the sample, the following attributes are the most preferred characteristics of acoworkingspace,inorderofimportancewithbetweenthebracketsthemostpreferredattributelevel:

1. Typeofleasecontract(noleasecontractorashortleasecontract(day,weekormonth));2. Accessibilityofthelocation(bycarandpublictransport);3. Layoutofthespace(half-openlayout);4. Diversityoftenants(moderateorstrongdiversity);5. Receptionandhospitality(receptionbutnohosttostimulateuserinteraction);

6

6. Events(sometimesaneventinthecoworkingspace);and7. Atmosphereandinterioraesthetics(homey);

Locationisoneofthemostimportantcharacteristicsinrealestateingeneral.However,thisresearchhasshownthatthetypeofleasecontracttakesamoreimportantplaceregardingthistypeofmulti-tenantoffice.Thestudymadeitclearthatattentionshouldbepaidtothestructureoftheleasecontract.Coworkersattachvaluetonocontractorashortcontract(e.g.arentalperiodofaday,aweekoramonth).Alotofattentionispaidtotheatmosphereandinterioraestheticsincoworking spaces while little utility value is assigned to this attribute. Themultinomial logit model did not estimatesignificantutilityvaluesfortheattributediversityinsupplyspaceswhichcanbeconsideredasnotapreferredcharacteristicofacoworkingspace.Withtheaidofalatentclasslogitmodeltwolatentclassesareestimatedconsistingofcoworkerswithcorrespondingpreferredcoworkingspacecharacteristics.Class1consistsof82coworkerswhichhasnopreferencesaccordingthecoworkingspacecharacteristics.Thisclassisperhapsstillveryheterogeneousorselectedtheattributelevelscompleterandomly.Class2consistsof137coworkerswithaslightlypreferencesformultipleattributelevels.Variouschi-squaretestsand independentsamplesT-testsareperformedtoanalyzethedifferencesbetweentheestimatedclasses(class1andclass2)basedonusercharacteristics.Onlyasignificantdifferenceisfoundbetweenthetwoestimatedclassesregardingtheratiovariablehoursworkinginthecoworkingspace.Thecoworkersinclass2worksignificantlymorehoursperweekinthecoworkingspacethanthecoworkersinclass1.Thismeansthatcoworkerswhospendmoretimeinthecoworkingspaceprefersometimesaneventinthecoworkingspace,apremiumcoworkingspace,accessibilitybypublictransport,astrongdiversityoftenantsandnorentalcontract.Apparently,thosecoworkingspacecharacteristicsbecomesmoreimportantwhenmoretimeisspentinthecoworkingspace.ConclusionandRecommendationsComparedtothestudiesofDeskmag(Deskmag,2013;Deskmag,2014;Deskmag,2016),whichareperformedonagloballevel, the coworkers in the Netherlands find it important to work on an external (remote), relatively affordable andenergetic/creative location and attach less value to the present community in the coworking space. The threemostmentioned motivations concern in particular the characteristics of the coworking space, and to a lesser extent thecommunitypresentinthecoworkingspace.Locationisoftenseenasthemostimportantcharacteristicsofrealestate.Thisstudyshowedthatthestructureoftheleaseismoreimportant.Theliteratureandtheresultsofthisstudyindicatethatthelowthresholdandflexibilityofthecoworkingspaces(rentprice,leaseperiod,multiplelocations)areattractivetocoworkersandcontributepositivelytothepopularityofthistypeofmulti-tenantoffice.Manycoworkingspaceshaveaspecificappearanceandgreatattentionispaidtotheinteriordesignwhichisoneoftheimportantfactorsofthecreativeandenergeticatmosphereinacoworkingspace.However,resultsshowedthatlittleutilityvalueisassignedtothisattributeandendedinlastplaceofthepreferredcharacteristics.Lessattentionneedstobepaidtotheappearanceofthespaceandmoreattentiontothestructureoftheleasecontract(periodandprice).It isofimportanceforoperatorstoofferamoderateorstrongdiversityoftenants,areception and to organize sometimes an event in the coworking space. For the developers of coworking spaces it isimportanttocreateahalf-openlayout.Coworkerswhospendmoretimeinthecoworkingspacehaveacertainpreferenceforsometimesaneventinthecoworkingspace,apremiumcoworkingspace,accessibilitybypublictransport,astrongdiversity of tenants and no rental contract. Providers and developers of coworking spaces have to focus on thosecharacteristicsinordertoattractandsatisfythecoworkerswhichspendrelativelymoretimeinthecoworkingspace.Usersandfutureuserscan,withtheaidofthisstudy,determinewhichcoworkingspacecharacteristicssuitstheminordertodecidewheretheywanttowork.Inaddition,usersandfutureuserscandecideifthemotivationstoworkingacoworkingspacecorrespondtotheirpersonalneedsandrequirements.Fewlimitationsofthisresearchcanbedrawnup.Ithasbeennoticedthatthequestionnairewasfarmorestartedthanfinished. For instance, the Dutch questionnaire wasmore than 400 times started of which 219 respondents actuallyfinished it. Perhaps the attribute based stated choicemethod had a deterred effect on the respondents, due to thecomplexityandcomprehensibilityofthechoicesets.Whatcanbelearnedfromthis,isthattheattributelevelsshouldbeunderstandableandclearataglance.Moreover,itmightbebettertoaskquestionswiththeuseofanattributebasedstated choice method further down the questionnaire. Perhaps this type of method has a deterred effect on therespondentsbecauseofthedegreeofdifficulty.Thesampleofthisresearchisrelativelysmall(219).Withalargersamplesize,morerelationsbetweenusercharacteristicsanduserpreferencescouldhavebeenexplored.Finally,perhapsotherusercharacteristicshavetobeappliedinordertofindsignificantrelationsbetweentheestimatedusergroups.Onlythedemographic variablehoursworking in the coworking space is significantlydifferentbetween the twoestimatedusergroups.

7

TableofcontentsColophon....................................................................................................................................................1

Preface.......................................................................................................................................................2Executivesummary....................................................................................................................................3

Tableofcontents........................................................................................................................................7

Listwithtablesandfigures.........................................................................................................................9

Chapter1:Introduction............................................................................................................................121.1Background..............................................................................................................................................................121.2Researchobjectiveandresearchquestions.............................................................................................................131.3Practicalrelevance...................................................................................................................................................141.4Theoreticalrelevance...............................................................................................................................................141.5Researchdesigndraft..............................................................................................................................................141.6Researchapproachandoutline................................................................................................................................15

Chapter2:LiteratureReview....................................................................................................................172.1Themulti-tenantofficemarket................................................................................................................................172.2Theemergingphenomenonofcoworking...............................................................................................................192.3Thefivecore-valuesofcoworking............................................................................................................................212.4Motivationsforcoworking.......................................................................................................................................232.5Characteristicsofcoworkingspaces........................................................................................................................252.6Marketsegmentationandusercharacteristics.......................................................................................................262.7Usergroupsincoworkingspaces.............................................................................................................................282.8Conclusion................................................................................................................................................................29

Chapter3:ResearchDesign......................................................................................................................313.1Résumétheoreticalframeworkandconceptualmodel...........................................................................................313.2Methodology............................................................................................................................................................32

3.2.1Methodsfordatacollection.............................................................................................................................323.2.2Methodsfordataanalysis................................................................................................................................33

3.3Operationalizationoftheindependentanddependentvariables...........................................................................353.3.1Independentfactor...........................................................................................................................................353.3.2Dependentfactor.............................................................................................................................................37

3.4Questionnairedistributionanddescriptionofthepopulation.................................................................................413.5Reliabilityofthestudy..............................................................................................................................................413.6Validityofthestudy.................................................................................................................................................423.7Conclusion................................................................................................................................................................42

Chapter4:DataDescription......................................................................................................................444.1Datacollection.........................................................................................................................................................444.2Datapreparation.....................................................................................................................................................444.3Datadescriptionoftheusercharacteristics.............................................................................................................464.4Conclusion................................................................................................................................................................52

Chapter5:DataAnalysisandResults........................................................................................................545.1Preferredcoworkingspacecharacteristics(multinomiallogitmodel).....................................................................545.2Classesofpreferredcharacteristics(latentclasslogitmodel).................................................................................595.3Conclusion................................................................................................................................................................64

Chapter6:ConclusionsandRecommendations........................................................................................676.1Researchresults.......................................................................................................................................................676.2Limitationsofthestudy............................................................................................................................................696.3Practicalimplications...............................................................................................................................................70

8

Referencelist...........................................................................................................................................73

Appendices..............................................................................................................................................79AppendixA-Valuedaspectsinmulti-tenantoffices......................................................................................................79AppendixB–Codebookofquestionnaire.....................................................................................................................80AppendixC-Questionnaire(English,Dutch,German,Italian).......................................................................................82AppendixD–Partnersindatacollection.......................................................................................................................98AppendixE–Conceptualmodeloftheindependentvariable........................................................................................99AppendixF–Conceptualmodelofthedependentvariable..........................................................................................99AppendixG–OutputNLOGITMultinomialLogitModel..............................................................................................101AppendixH–OutputNLOGITLatentClassLogitModel(2,3and4classesestimated)..............................................102AppendixI–Part-worthutilitiesofthedependentvariable........................................................................................106AppendixJ–Frequencytablesoftheusercharacteristicsofthelatentclasses..........................................................107

9

ListwithtablesandfiguresChapter1:IntroductionFigure1.1:Preliminaryconceptualmodel.........................................................................................................................14Figure1.2:Outlineofthechaptersandcorrespondingresearchquestions......................................................................15

Chapter2:LiteratureReviewTable2.1:Input(mentionedcharacteristicsofcoworkingspaces)....................................................................................24Table2.2:Output(motivationstoattendacoworkingspace)...........................................................................................24Table2.3:Outcome(motivationstoattendacoworkingspace).......................................................................................24Table2.4:Typicalcharacteristicsofcoworkingspaces......................................................................................................26Table2.5:Usergroupsofcoworkersaccordingtotherelevantliterature........................................................................28

Chapter3:ResearchDesignFigure3.1:Conceptualmodel............................................................................................................................................31Figure3.2:Statedpreferencemethods(Adamowiczetal.,1998).....................................................................................32Table3.1:Geographicvariables(coworkingspacerelatedcharacteristics)......................................................................35Table3.2:Demographicvariables(personalcharacteristics).............................................................................................36Table3.3:Demographicvariables(work-relatedcharacteristics)......................................................................................36Table3.4:Demographicvariables(coworkingspacerelatedcharacteristics)...................................................................37Table3.5:Psychographiccharacteristics............................................................................................................................37Figure3.3:StatisticaldesignABSC-method(Sanko,2011).................................................................................................37Table3.6:Experimentaldesign..........................................................................................................................................39Table3.7:Experimentaldesignwithcorrespondingattributelevels.................................................................................40Table3.8:Version,choicesetandalternatives..................................................................................................................40Table3.9:Examplechoiceset(version2,choiceset1,alternative16,10and21)...........................................................41

Chapter4:DataDescriptionTable4.1:Effectcodingattributelevels.............................................................................................................................46Figure4.1:Genderofcoworker.........................................................................................................................................47Table4.2:Frequencytableagegroups..............................................................................................................................47Figure4.2:Distributionvariableage..................................................................................................................................47Figure4.3:Distributionvariableagegroups......................................................................................................................47Figure4.4:Shareofhighlevelofeducation.......................................................................................................................48Table4.3:Frequencytablevariableeducationgroups......................................................................................................48Table4.4:Crosstabpositionandusergroup......................................................................................................................49Table4.5:FrequencytablesectorandcomparisontoDeskmag(2016)............................................................................49Table4.6:Crosstabincomeandusergroup.......................................................................................................................49Figure4.5:Transporttocoworkingspace..........................................................................................................................50Table4.7:Hoursincoworkingspace..................................................................................................................................50Table4.8:ComparisonresultstotheDeskmagstudies.....................................................................................................50Table4.9:Overviewresults................................................................................................................................................50Figure4.6:Stackedbarchartmotivations.........................................................................................................................51

Chapter5:DataAnalysisandResultsTable5.1:Descriptivestatisticsmultinomiallogitmodel..................................................................................................55Table5.2:Part-worthutility,totalutilityandsignificanceoftheparametersofthemultinomiallogitmodel.................56Figure5.1:Totalutilityofattributes..................................................................................................................................56Table5.3:Totalutilityperpresentedalternative...............................................................................................................58Table5.4:Totalutilitypervisitedcoworkingspace(practicalrelevance)..........................................................................58Table5.5:DescriptivestatisticsLatentClassLogitModel..................................................................................................59Table5.6:Part-worthutility,totalutilityandsignificanceoftheparametersofthelatentclasslogitmodel...................60Figure5.2:Totalutilityofattributesclass2.......................................................................................................................60Table5.7:Numberofrespondentsandpreferredattributesperlatentclass...................................................................61

10

Figure5.3:Gender(χ2=0,122;p=0,73)……………………………………………………………………………………………………………………...62Figure5.4:Education(χ2=1,389;p=0,24).......................................................................................................................62Figure5.5:Usergroup(χ2=1,587;p=0,66)……………………………………………………………………………………….........………..……….62Figure5.6:Positionorganization(χ2=2,37;p=0,67)……..................................................................................................62Figure5.7:Sectororganization(χ2=2,439;p=0,66)…………………………………………………………………………..………………………62Figure5.8:Income(χ2=3,807;p=0,28)..........................................................................................................................62Figure5.9:Transporttocoworkingspace(χ2=1,143;p=0,77).......................................................................................62Figure5.10:Motivation1(χ2=6,476;p=0,26)................................................................................................................63Figure5.11:Motivation2(χ2=10,471;p=0,063)............................................................................................................63Figure5.12:Motivation3(χ2=2,732;p=0,75)................................................................................................................63Table5.8:DescriptivestatisticsindependentsamplesT-test............................................................................................64Table5.9:IndependentsamplesT-test..............................................................................................................................64

11

12

Chapter1:IntroductionThischapterwillintroducethesubjectofthismaster’sthesisandexplainhowthisresearchwillbeperformed.First,section1.1detailsthebackgroundofthesubjectofcoworkingandusercharacteristics.Theresearchgapisalsoaddressedinthissection.Then,insection1.2,theresearchobjectiveisdescribed,followedbytheresearchquestionsandthepreliminaryconceptualmodel.Insections1.3and1.4,thepracticalandtheoreticalrelevanceoftheresearcharepresented.Adraftoftheresearchdesignisgiveninsection1.5.Theresearchapproachandtheoutlineofthismaster’sthesisareprovidedinsection1.6.

1.1BackgroundAwell-knownphraseineconomicsisthat“supplyanddemandshouldbeinbalance”.Inotherwords,inahealthymarket,supplywill respond to demand. Currently, there is a structural vacancy in the officemarket. The vacancy rate in theNetherlandsis15.8%ofthetotalavailablespace,whichcorrespondstoalmost8.0millionm2officespace(DTZZadelhoff,2015).There isastructuralvacancybecausethesupplyno longermeetsthedemandsoftenants; there isamismatchbetweensupplyanddemand(DeVries&vandenBesselaar,2013).“Mismatchofdemandandsupplyofofficespaceonthemarket, locationorbuilding levelcausesanoversupplyofofficebuildings”(Remøy,2010,pp.48). Inordertomeetfuturedemand,realestatemarketsmustrespondtomarkettrendstopreventacontinuouslygrowingmismatchbetweenofficespacesupplyanddemand.

Trends in themarket have created a new paradigm in office space usage (Calder & Courtney, 1992).Manyindustriesandsectorsareshiftingtowardsanincreasinglycomplex,unpredictableanddynamicknowledgeenvironment(Saurinetal.,2008).Inamarketthatrequiresconstantchanges,theconceptofcorporateaccommodationmustbynaturebeflexible,inorderfororganizationstobeabletoadapttomarketdevelopmentsandorganizationalchanges(Laterveer,2011).Flexibilityinthedurationofthelease(contract),flexibilityintheuseofservicesandinthechoiceofspace(Gibson&Lizieri,1999).Untilthe1990s, leasesforlongerthan25yearswererelativelynormal;currently, long-termleasesareriskybecauseofuncertainandvolatileeconomiccircumstances(Ketting,2014;Laterveer,2011).Inadditiontoflexibility,othertrendsinthelabormarkethasinfluencestheuseofofficespace,suchastheincreasingnumberofself-employedworkers,growthintheuseoftechnologiesandadecreasingamountofworkspaceallocatedtoindividuals(Green,2014;DeVries&VandeBesselaar,2013;Barber,Laing,&Simeone,2005;).Newwaysofworking,newgenerationsandnewlifestylesbringnewofficespaceneeds(vanMeel&Vos,2001),suchasdifferentspaceneeds(surfaces,layouts,spaces,etc.),contracts(foraday,amonth,ayear,etc.)andservices(catering,ICT,administration,etc.)(Ketting,2014;Laterveer,2011;Calder&Courtney,1992).Thechangingdemandsofofficespaceusagehaveledtothegrowthofanewtypeofproperty,namely:themulti-tenantoffice(Peltier,1992).

AccordingtoWeijs-PerréeandAppel-Meulenbroek(2015),thedefinitionofamulti-tenantofficeisabuildinginwhichofficespaceandpossiblyanumberofsharedfacilitiesand/orservicesareofferedtomultipleorganizations.Thecompetitiveadvantagesofthis,whencomparedtosingle-tenantoffices,arethebenefitsofsharingfacilitiesandservices,creatingcost-savingsandknowledgespillovers(Fuzi,Clifton&Loudon,2014).Multi-tenantofficesherebyaddvaluetoorganizationsandtheirusers(Hartog,2015).Severalstudies(Weijs-Perrée&Appel-Meulenbroek,2015;Ketting,2014;vandenBerg&Strijnenbosch,2009;Calder&Courtney,1992)haveanalyzedthedifferenttypesofmulti-tenantoffices.AccordingtoWeijs-PerréeandAppel-Meulenbroek(2015),adistinctioncanbemadebetweenfourtypesofmulti-tenantoffices:theregularbusinesscenter,theservicedoffice,thecoworkingspaceandtheincubator.Eachofthesetypesfocusonadifferentpartofthemarket,suchasonthetenanttype,atmosphere,services,spacesandtheoverarchinggoal(Weijs-Perrée&Appel-Meulenbroek,2015).Inthepreviousdecade,agrowingtrendregardingcoworkinghasbecomevisible.Thispopularityistheresultof“ashiftingattitudetowardswork”(Sykes,2014,p.142).The2008crisisledpeopleleavingthetraditionalworkplace,astheywereeitherforcedtoorchoseto,andtostartingbusinessesontheirown.The“freelanceeconomy”thenarose(Fuzietal.2014,pp.4).Freelancersandself-employedworkerswerenotabletofindanappropriateworkplace.Someworkersbecametired of the loneliness ofworking at home and the frontiers between private and professional lives became fuzzy ordisappeared(Moriset,2014).Otherworkerswenttolocalcafesorlibrariestoworkandwereexhaustedbythedistractionstheyexperiencedthere(Fuzietal.,2014).Theriseofcoworkingspaceswasareactiontotheabsenceofanappropriateworkplaceforthisgroupofworkers;itcanbeseenasabottom-upsolution(Merkel,2015).Coworkingspacesarecreativeandenergeticplaceswherepeoplecaninteract,share,buildandco-create(Fuzi,2015).AccordingtotheGlobalCoworkingSurveybyDeskmag(2015),anonlinemagazineaboutthecoworkingphenomenon,therewere7,800coworkingspacesworldwidein2015incomparisonwith310in2009,andthenumberofcoworkersin2015was510,000incomparisonwith43,000in2011.Itisclearthatcoworkingspaceshavebecomeaverypopulartypeofmulti-tenantoffice.“Thegrowthof

13

coworkingspaceshasbeen100%annuallyforthelastsevenyears”(Green,2014,pp.53).Inadditiontodistractionandloneliness,therearefurtherreasonstousecoworkingspaces.Forinstance,financialbenefits(savemoneybysharingspaceandservices),economicbenefits(increasesalesoraccesstocapital),environmentalbenefits(sharingrecourses)andawiderangeofpositivesocialeffects(increasesocialcircleandbuildfriendships)(Kwiatkowski&Buzynski,2011).Deskmag(2015)foundthatcoworkersvalueworking inacoworkingspaceduetocasualsmalltalk,sharingknowledge,enjoyingother’scompanyandbrainstormingwithcoworkers.

Coworkingisoftenassociatedwithfreelancersandself-employedworkers(Parinno,2015).However,accordingtomultiplesurveys(Fuzi,2015;Parinno,2015;Merkel,2015;Weijs-Perrée&Appel-Meulenbroek,2015;Gandini,2014;Fuzietal.2014;Sykes,2014;Moriset,2013;Spinuzzi,2012;Leforestier,2009),thesecoworkersareamoreheterogeneousgroup then simply freelancers and self-employed workers. The following user groups can be distinguished from theliterature:self-employedworkers, freelancers,entrepreneurs,small firms, large firms,extendedworkersandstudents.However, many studies on coworking have not examined the coworkers in much detail. In the academic literature,coworkersareassignedtooneoftheabovementionedgroups;however,thecharacteristicsperuserhavenotreceivedmuch attention. Each user has other geographical characteristics (e.g., residential, city or village), demographiccharacteristics(e.g.,age,gender,incomeandlevelofeducation),psychographiccharacteristics(e.g.,interests,activitiesandvalues)andbehavioralcharacteristics(e.g.,brandloyalty,readinesstobuyordegreeofloyalty)(Kotler,1994).Multiplestudies(Deskmag,2015;Gandini,2015;Sykes,2014;Spinuzzi,2012;Leforestier,2009)wroteaboutbasicsubjects,suchaswhatiscoworking,whocoworksandwheredoesithappen.Coworkinghasbeenwellcoveredbyavarietyofmedia,buthasbeenalmostcompletelyignoredintheacademicliterature(Moriset,2013;Merkel,2014).Thereisrelativelylittlecriticalunderstandingofcoworkingand itsassumedeffects,althoughtherearecertainlystudiesthat focusonspecificsubjects. For instance, Parinno (2015) studied whether the physical co-presence of coworkers stimulates knowledgeexchange. Parinno found that an organizational platform where coworkers can meet and ask questions is of moreimportanceintheexchangeofknowledge.Moriset(2013),whichstudiedcoworkingspacesinrelationtothecreativityinthecity,foundthatcoworkingspacescontributesacreativeworkingenvironmentwithinthecity.Fuzi(2015)performedresearchonhowtopromoteentrepreneurshipinspareentrepreneurialenvironmentsandfoundthatcoworkingspaceshavetobecombinedwithotherconcepts,suchasincubators,topromoteentrepreneurship.

Mostoftherelevantcurrentacademicresearchconcernstheoutputandoutcomeofcoworking.Thefocusinthecurrentstudyisontheinputofcoworking.Inputregardstheresourcesusedtocreateaproduct(newworkplaceconcept);outputconcerns theconditions thatarecreated through the input (increasedknowledgesharing)andoutcome is theresultantchangesorbenefits (increased innovationwithin theorganization) (VanderVoordtetal.,2016).Despite theimportanceofcoworkingspaces inthecurrentmarketofmulti-tenantoffices,toolittleattentionhasbeenpaidtothecharacteristics(theinput)ofcoworkingspaces,inparticular,userpreferencesforthecharacteristicsofcoworkingspaces.Each user has different user characteristics and therefore probably different preferences regarding coworking spacecharacteristics.

1.2ResearchobjectiveandresearchquestionsThe research objective of this study was derived from the description of the background (1.1). It is clear that usercharacteristicsandthecorrespondingpreferencesofcoworkingspacecharacteristicsareanunderexaminedsubject.Tocomplement the limited researchon this subject, thepreferencesofusers incoworking spaceswillbeexamined.Theresearchobjectiveisthereforeasfollows:Toidentifytheusercharacteristicsofthedifferentusersofcoworkingspacesanddeterminewhethertheirpreferences

forcoworkingspacecharacteristicsdifferonusercharacteristics.Themainquestionwillbeansweredthroughseveralsub-questionsontwotopics.Thefirsttopicexaminesthesubjectofcoworking and specifically coworking space characteristics. The second topic includes a study on user characteristics.Accordingtotheabovementionedresearchobjective,thefollowingresearchquestionwasformulated:

Candifferentusergroupsbeidentifiedbasedontheirpreferencesforcoworkingspacecharacteristicsanddotheydiffer

onusercharacteristics?

14

Tobeabletoanswerthisresearchquestion,thefollowingsub-questionsmustbeanswered: Whatarethetypicalcharacteristicsofcoworkingspaces? Whichtypesofuserscanbeidentifiedincoworkingspacesandwithwhichcharacteristicscantheybedescribed? Whataretheusercharacteristicsofcoworkersingeneral? Whataretheuserpreferencesforcoworkingspacecharacteristics? Candifferentusergroupsbeidentifiedbasedontheirpreferences? Whatarethedifferencesbetweentheseusergroupsbasedonusercharacteristics?

Figure1.1showsthepreliminaryconceptualmodelexaminedinthisstudy.Theindependentfactor(usercharacteristics)influencesthedependentfactor(userpreferencesforcoworkingspacecharacteristics).

Figure1.1:Preliminaryconceptualmodel

1.3PracticalrelevanceThisstudyprovidesfurtherinsightintotherelationbetweenusercharacteristicsandtheirpreferencesforcoworkingspacecharacteristics.Thisstudyisrelevantforproviders,developersandusersofcoworkingspaces. Byunderstandingthepreferencesofthetenants/coworkers,coworkingspacescanbedevelopedmoreeffectivelybecauseprovidersanddeveloperscanrespondtotheneedsoftheuser.Theworkenvironmentcanbeadjustedtotheuserpreferencesresultingfromtheresearch.Thisinvestigationisalsorelevantforprovidersanddeveloperbecausethedifferentusercharacteristicsofcoworkersarerevealed.Inthismanner,itbecomesclearwhoactuallyusethesetypesofmulti-tenant offices. Eventually it becomes clear if the preferred characteristics of coworking spaces differ on usercharacteristics.Anexampleofthisisthatself-employedworkersmighthaveapreferenceaccordinganopenlayoutwithanindustrialatmosphere,whileemployeesofafirmattachvaluetoaclosedlayoutwithamodernatmosphere.Inthisway,providersanddeveloperswillbeabletofocusonaspecificgroupwithcertainpreferences.

Secondly,thisresearchisofinteresttocurrentandfutureusersofcoworkingspaces.Itgivesinsightinthepresentvariantsandthecharacteristicsofcoworkingspaces.MultiplecoworkingspacesintheNetherlandswillbeanalyzedandcorrespondingcharacteristicscomeacross.Thisstudyprovidesalsoanoverviewoftheusers’motivationstowork inacoworkingspace.Inthismanner,futureusersknowwhattheintentionsareofcurrentusersinordertoseeifcorrespondstotheirownintentions.

1.4TheoreticalrelevanceThisstudyoffersimportanttheoreticalinsightsintocoworkingandcoworkingspaces.Firstofall,theconceptofcoworkinghasreceivedalargeamountofattentionfromvariousmedia,buthasbeenalmostcompletelyignoredbytheacademicliterature(Merkel,2014;Moriset,2013).Bymeansofthisinvestigation,scientificknowledgeintheareaofcoworkingwillbeexpandedbymeansofdetailingusercharacteristicsanduserpreferences.Todate,verylittleattentionhasbeenpaidto thecharacteristicsofcoworkingspaces.Sincecoworking isa relativelynewtypeofmulti-tenantoffice,nopreviousresearchhasbeenperformedintotheinput(characteristics)ofcoworkingspaces.Whilethisisthecasewithsingle-tenantofficesandother typesofmulti-tenantoffices.Furthermore,coworking ismostlyassociatedwith freelancersandself-employedworkers (Parinno,2015), this researchalsoexamines if theremoreusergroupscanbe identified.Eachusergroupcontainsagreatvarietyofuserswithadifferentsetofusercharacteristics,andnopreviousstudyhasdefinedthesecharacteristics.However,thedistinctivevalueofthisinvestigationcomesthroughansweringwhetherusergroupscanbeidentifiedbasedontheirpreferencesregardingthecharacteristicsofcoworkingspacesandiftheseusergroupsdifferonusercharacteristics.

1.5ResearchdesigndraftThechapterregardingcoworkingandusercharacteristicswillbecomposedbasedonaliteraturereview.Thequantitativeanalysiswillrevealwhichusergroupsmakeuseofcoworkingspacesandwhattheirpreferencesareforcoworkingspacecharacteristics.Toachievethemostrepresentativepicture,coworkersfromdifferentlocationsanddifferentcoworkingproviderswillbesurveyedbytheuseofanonlinequestionnaire.ThecoworkingspacesintheNetherlandswillbeselectedfrommultiplewebsites.Apartofthetotalpopulationofcoworkerswillformthesampleofthisresearch,theso-called

15

sample-basedresearch(Baarda&DeGoede,2006).Theresearchunitinthisstudyis‘theuserofacoworkingspace’.Inthecoworkingspaces,thecoworkers(respondents)willbeselected.Everyonewhoispresentinacoworkingspaceisseenas a user of a coworking space and is coveredby the researchunit.Datawill be collected through theuseof surveyquestionsandanattributebasedstatedchoicemethod.Thedatawillbeanalyzedbyseveraldescriptiveanalysismethods,amultinomiallogitmodel,alatentclasslogitmodel,chi-squaretestsandindependentsamplesT-tests.

Thedataforthe‘usercharacteristics’factorwillbecollectedthroughsurveyquestions.Questionswillbeaskedaboutgeographic,demographicandpsychographiccharacteristicstoidentifytheuser.Behavioralcharacteristicswillnotbe included inthequestionnairebecause it ismainly focusedonactualcustomerbehaviortowardsaproduct,suchasbrandloyalty,readinesstobuyanddegreeofloyaltythanontheuser.Thedataforthe‘userpreferencesofcoworkingspacecharacteristics’factorwillbecollectedthroughtheuseofastatedpreferencemethod.Thismethodattemptstousediscretechoices(AoverB;BoverAandC)inordertoinferpositionsoftheitems(A,BandC)(Louviere,Hensher&Swait,2000).TherespondentsmustmakediscretechoicesconcerningthedifferentcoworkingspacesA,BandC, inordertodecidewhichcharacteristicstheyprefer.

Toanalyzetheusercharacteristicsofcoworkers,multipledescriptiveanalysismethodswillbeused.Todefinetheuserpreferences forcoworkingspacescharacteristics,amultinomial logitmodelwillbeestimated.Differentgroupsofcoworkerswillbeidentifiedbasedonpreferredcoworkingspacecharacteristics,byestimatingalatentclasslogitmodel.Inthismodel,anumberofclasseswillbeestimatedbasedonmatchinguserpreferences.Theclasseswillbecharacterizedbyusingchi-squaretestsandindependentsamplesT-tests,whichdeterminesthedifferencebetweentheestimatedusergroups(classes)basedonusercharacteristics.

1.6ResearchapproachandoutlineThissectionintroducesthesubjectofthisstudyandexplainshowtheresearchwillbeperformedthroughoutthisthesis.In the followingchapter,a literaturereview isperformed.Thesubjectofcoworking (dependent factor)will initiallybeaddressed,asthiswillformtheframeworkforthesubjectofusercharacteristics(independentfactor).Inthefirstpartofchapter2, thecharacteristicsofacoworkingspacewillbedescribedand the first sub-question, “Whatare the typicalcharacteristicsofcoworkingspaces?”willbeanswered.Chapter2willdescribewhichusergroupscanbe identified incoworkingspacesandhowtoanalyzetheirusercharacteristics.Thischapterwillalsoprovideananswertothesecondsub-question, “Which types of users can be identified in coworking spaces and with which characteristics can they bedescribed?”Inchapter3,theresearchdesignwillbeclarified.Themethodologyofthedatacollectionanddataanalysiswillbeexplainedinfurtherdetail.Theoperationalizationofthevariables,aswellasthereliabilityandvalidityofthisstudy,willalsobepresentedinchapter3.Thedataaredescribedinchapter4,wheretheindependentvariablesareanalyzed.Inthischapter,thethirdsub-question“Whataretheusercharacteristicsofcoworkersingeneral?”willbeanswered.Thedataanalysisformsthecentraltopicinchapter5;thedataarehereanalyzedusingmultiplemethods.Thesub-questions“Whataretheuserpreferencesforcoworkingspacecharacteristics?”,“Candifferentusergroupsbeidentifiedbasedontheir preferences?’’ and ‘’What are the differences between theseuser groups basedonuser characteristics’’will beansweredinthischapter.Finally, inchapter6,theconclusionsofthisstudywillbedrawn.Theresearchquestion“Candifferentusergroupsbeidentifiedbasedontheirpreferencesforcoworkingspacecharacteristics?”willbeanswered,andmultiple recommendations will be given. Figure 1.2 shows the structure of this thesis, showing the chapters andcorrespondingsub-questions.

Figure1.2:Outlineofthechaptersandcorrespondingresearchquestions

C1• Introduction

C2• LiteratureReview• Sub-question1 and 2

C3• Research Design

C4• Data Description• Sub-question 3

C5• DataAnalysisandResults• Sub-question 4,5and6

C6

• Conclusions andReccomendations• Sub-questions1- 6• Researchquestion

17

Chapter2:LiteratureReviewTheresearchquestionsaddressedbyliteraturereviewareasfollows:“Whatarethetypicalcharacteristicsofcoworkingspaces?”and “Which types of user can be identified in coworking spaces andwithwhich characteristics can they bedescribed?”

Tocreateaclearviewofwhatcoworkingspacesare,themarketofmulti-tenantofficeswillfirstbeintroducedinsection2.1.Whatisthedefinitionofamulti-tenantoffice?Howdidtheyarise?Andwhichtypescanbedistinguished?Insection2.2,themaintopicofthisinvestigationwillbeintroduced.Thebasicprinciplesofcoworkingwillbediscussed,suchas,whatiscoworking?Howdiditemerge?Howcanitbedefined?Section2.3takesaccountofthefivecore-valuesofcoworkingonwhichthecoworkingphenomenonisbased.Insection2.4,themotivationsofcoworkingarehighlighted.Thetypicalcharacteristicsofcoworkingspacesarepresentedinsection2.5.

Coworkingisoftenassociatedwithfreelancersandself-employedworkers(Parinno,2015).However,accordingtomultiplesurveys,thenumberofusergroupsislarger.Eachusergroupcontainsagreatvarietyofuserswithdifferentsets of user characteristics. In this chapter, also a literature review of user characteristics will be carried out. Anintroductioninmarketsegmentationandusercharacteristicsisgiveninsection2.6.Thedifferentusergroupsfoundincoworkingspacesareidentifiedinsection2.7.Finally,insection2.8,theconclusionsoftheliteraturereviewaregivenandbothsub-questionsaddressedinthischapterwillbeanswered.

2.1Themulti-tenantofficemarket“Amulti-tenantoffice isabuilding inwhichspacesandsomecommon facilitiesand/orservicesareoffered tomultipleorganizations”(Weijs-Perrée&Appel-Meulenbroek,2015,p.23).Theemergenceofmulti-tenantofficeswasinitiatedinthelate1960s,inresponsetoanincreasingdemandforofficespacewithsecretarialandtelephonyservices(Peltier,1992).Inthe1980s,manyofficebuildingswerebuiltinresponsetoeasilyavailablefunding.Realestatewasapopularinvestmentproductbythen.Landlordsofferedlucrativedealstoprovidersofmulti-tenantoffices(forexample,HQBusinessCentersandOfficePLUS,Regusoriginated intheearly1990s) intheformofrent-freeperiodsorfree interiordesign.Landlordsofferedincentivestoenticemulti-tenantofficeproviderstolocatethemselvesintheirbuildings,multi-tenantofficeswereseenasaprofitablebusiness.Themarketmultipliedasaresult(Peltier,1992).Inthemid-1980s,fewcenterswerebuiltduetoaturnaroundintheeconomy.Theoversupplyofmulti-tenantofficesdecreased(Peltier,1992).Attheendofthe1980s,therentalmarketformulti-tenantofficescollapsedduetothesavingandloancrisisandtherealestatecrisis,astenantscancelledordidnotrenewtheirleasecontracts(Peltier,1992).Asubstantialgrowthinself-employedworkersandsmallbusinesses,duetoachanginglabormarket,resultedintheearly1990stoanincreaseddemandforofficespaceinmulti-tenantoffices(Calder&Courtney,1992).Aseriesofmergesbetweenprovidersensuredthatanumberofmajorproviders arose in themarketofmulti-tenantoffices (Peltier, 1992). Providers andowners attempted todifferentiatethemselves,asaresultofalargercompetitivemarket,byofferingotherservicesandfacilitiestoattractaspecificgroupoftenants.Forexample,byofferingaservicepackageinprimelocationsthatattractedlargerorganizations.Providersenteredaspecificmarket.AccordingtoKetting(2014)thetwoarchetypalmulti-tenantofficeshavehistoricallybeentheservicedofficeandtheincubator.Bothofthesetypesemergedfromdifferentphilosophies.Thepurposeofanincubatoristohelpstartup companies tomature. The incubator find its origins due to the forced resignations of employees of industrialenterpriseswhichstartedontheirown(Ketting,2014).Thepurposeofaservicedofficeistoofferflexibilityinservices,spaceandleaseduration.Servicedofficespacesoriginatedintakingcareoftenantsbyofferingmultipleservices(Somsen,2002;Ketting,2014).

Theriseofmulti-tenantofficeswasaresultofcompaniesbecomingawareoftherapidlyevolving,changingandvolatilemarkets(Somsen,2002).Workingenvironmentsneedtobeadaptable,sincebusinessesrequireflexibility,suchasflexibilityinthedurationofthelease,flexibilityintheuseofservicesandflexibilityinthechoiceofspace(Gibson&Lizieri,1999).Inadditiontoflexibilityandtheincreasingnumberofself-employedworkers,othertrendsinthelabormarket,suchasgrowthintheuseoftechnologiesandadecreasingamountofworkspaceallocatedtoindividuals,influencedtheuseofofficespace(DeVies&vandeBesselaar,2013;Barber,Laing,&Simeone,2005;Green,2014).Eachorganizationsoonerorlaterneedsaworkspaceandprefersitflexiblemanner.Multi-tenantofficesprovidethatkindofflexiblespace(Calder&Courtney,1992).Furthermore,municipalitieshaveincreasedtheirfocusonfacilitatinglow-costofficespaceinmulti-tenantoffices.Inthisway,businesseshaveeasieraccesstoaccommodation,whichhaspositiveeffectsontheeconomy(Mensen&vanRijt-Veltman,2005).Byoutsourcingrealestatemanagement,organizationsbecameflexibleandabletoanticipatethevolatilemarkets(Somsen,2002).Corporaterealestatewasconsideredasanon-corebusinessbymostoftheorganizations.Organizationshadtofocuson the fields inwhichdistinctive value couldbeachieved, andoutsource theones inwhich theydonothavea

18

competitiveadvantage(Peltier,1992;Laterveer,2011;McAllistar,2002).Furthermore,outsourcingleadstocostsavings(reducingcapital investmentsandoperationalcosts)andspreadsrisk.Providerssawopportunitieshereandstartedto‘bundle’spaceandservices(McAllistar,2002).Inthisway,themulti-tenantofficetypeofservicedofficespacearose.Along rental periodprovided a steady flowof income for the landlord, a leaseperiodof longer than25 yearswas thestandard in the 1970s and 1980s. It appeared to be inflexible to meet the demands of a more volatile economicenvironment(Hamilton,Lim&McCluskey,2005).Long-termcontractsprovidedaweakrelationshipbetweentenantsandlandlords,becausetherewere limited incentivesto improvetheofferedqualityandservices.The landlordtreatedthebuildingpurelyasafinancialproduct.Providersofmulti-tenantoffices,incontrasttosingle-tenantoffices,ensuredthequalityandservicesofthemanagementoftherealestate.Atailoredproduct,adjustedtothedemandsofthetenant,ensuredagoodrelationshipbetweenlandlordandtenant(Ketting,2014;McAllistar,2002).Inamulti-tenantoffice,theoccupiertooknoresponsibilityforthepropertyandtookadvantagesofafully-servicedoffice(GibsonandLizieri,1999).

Thebenefitforthelandlordofamulti-tenantoffice,comparedtoasingle-tenantoffice,isthatwhenatenantfailstopaytheirrent,(e.g.,afterbankruptcy),itisalessseverecircumstancethaninasingle-tenantbuilding.Inamulti-tenantoffice,thisreductioninincomeofleasecanbebufferedbytheotherremainingtenants.Inasingle-tenantoffice,thesoletenantdisappears,whichisverydetrimentaltothelandlord.Typesofmulti-tenantofficesSeveralstudies(Weijs-Perrée,Appel-Meulenbroek,DeVries,&Romme,2016;Ketting,2014;VandenBerg&Stijnenbosch,2009;Calder&Courtney,1992)haveanalyzed themulti-tenantofficemarket.Eachstudyassignedadefinition to theconceptofmulti-tenantofficesandidentifiedthedifferenttypesofmulti-tenantoffices,eachinadifferentmanner.Weijs-Perrée et. al (2016) analyzed the types of multi-tenant offices, which were identified in the previous literature, andassessed whether the existing types varied in certain characteristics (objectives, business models, leases, tenants,facilities/services,spaces,buildingtypeandcapacity).TheclassificationanddefinitionbyWeijs-Perréeet.al(2016)werebasedonpreviousrelevantresearch.Thedifferenttypesofmulti-tenantofficesareasfollows:

• Regularbusinesscenter;• Servicedoffice;• Incubator;and• Coworkingspace.

RegularbusinesscenterAccordingtoWeijs-PerréeandAppel-Meulenbroek(2015),thespecificpurposeoftheregularbusinesscenteristoprovideofficespacewithalowservicelevel.CalderandCourtney(1992)werethefirsttostudythemarketofmulti-tenantoffices,atthattimearelativelynewpropertyindustry.“Businesscentersincludealltypesofcommercialpropertywhichoffersmallunitsonflexiblelettingterms”(Calder&Courtney,1992,p.2).Regularbusinesscentersareprimarilyaimedatprovidingoffice spaceand in addition services suchas cleaning,maintenanceand security (Weijs-Perrée&Appel-Meulenbroek,2015). Compared to the other types, the regular business center has no overarching theme. Thismakes the regularbusinesscentertheplainesttypeofmulti-tenantoffice,withthefocusonprovidingofficespace.AccordingtoWeijs-PerréeandAppel-Meulenbroek(2015)theregularbusinesscenterhasnospecifictargetgroupoftenantsandoffer leasesforlongertimeperiods(e.g.twotofiveyears).ServicedofficeServiced offices offer flexible office space, business services, catering, facility management and different forms oftechnology(Weijs-Perrée&Appel-Meulenbroek,2015;McAllistar,2002).Thetenantsreceiveafullyworkingandequippedworkenvironmentwithseveralservicestocarryontheirbusiness(Price&Spicer,2002).Thereisnotraditionalrelationshipbetweenthelandlordandtheoccupier;therelationshipratherfocusesontheservicethanontheleaseofspace(Price&Spicer,2002).Flexibleaccommodation isdesirable inahighlyvolatilemarket.Flexibility is themost importantassetofservicedofficespace,sinceorganizationsmustbemoreagileintheirproductionofgoodsandservices.Providersdeveloparealestateproductthatrespondstothisdemandforflexibility(McAllistar,2002).AccordingtoSomsen(2002),flexibilityinservicedofficespaceisseeninthreemainaspects.First,theflexibilityinprovidedservices.Intheservicedoffice,thereisastandardsetofservicessuchascleaning,receptionservices,telephoneansweringandutilities(Gibson&Lizieri,1999).Therearealsoadditional services,whicharebasedon the ‘paywhatyouuse’principle. Forexample,printingormailservices,forwhichextracostsarecharged.Thesecondaspectofflexibilityistheflexibilityofspace.Organizationsrentspacedependingontheirdemand.Intimeswhentheorganizationmustdeliverfewerservicesorproducts,theamountofofficespacecanbescaleddown.Thefinalaspectisthedurationofthelease.Thisisbasedonthe‘plugandplay’principle,whichreferstothesimpleoccupationofofficespace.Businessescanrentspaceonadaily,weekly,monthlyoryearlybasis.Trendsinthemarkethavechangedleaseperiods(Ketting,2014).Longtermcontractsarenotinlinewiththeflexibilityneedsoforganizations.Theusersofservicedofficesaresmallandmedium-sizedorganizationsandteamswhousethis

19

spaceasanextensionoftheircoreoffices(McAllistar,2002).Thegroupofusersislarger,accordingtoLaterveer(2011)and Price and Spicer (2002). Self-employed workers, startups, virtual workers, staff overflow and new productdevelopment teams also use these spaces. Serviced office space is often the core office for smaller and startingorganizations. Serviced office space is for larger organizations often an additional office to their core portfoliowhereemployeescanhavemeetingswithexternalpartiesorforinstanceacreativesparringsession(Gibson&Lizieri,1999).IncubatorAsreportedbyLesáková(2012),thegoalofanincubatoristostimulatethedevelopmentofnewbusinesseswithinthelocaleconomy.Anothercommontermforanincubatorisabrewingplace.TheincubatorisanAnglo-Saxoninventionthatfocusesondevelopingstartupsundersupervision,byofferingadviceservices,tostimulatethelocaleconomy(Lesáková,2012). The twomain benefits for the city are new jobs and the additional revenue generated by the activities. Bothelements can revitalize local economies (Weijs-Perrée et. al, 2016; Lesáková, 2012). Moreover, incubators stimulateknowledgeexchangeintheregion.Businessincubatorshelpstartupsbyofferingmarketingassistance,accesstocapital,networking activities, advisory boards andmuchmore. Incubators attempt to fill the gap that startups are unable tobecauseofalackofnetwork,knowledgeandcapital.Whenstartupsarereadyto‘flyout,’thestartupleavestheincubationprogramandnewonesareallowedin(Fuzi,2015;Lesáková,2012;Berg&Strijnenbosch,2009).Thematuredstartupalsoleavetheincubator(themulti-tenantoffice).Thetargetgroupofincubatorsarethekindofstartupsthathaveaninnovativeideabutaremissingaconcretebusinessplantobringtheproducttomarket.Startupscanjoinanincubatorprogramthatwill ensure that the product can be brought to market (Hartog, 2015). Most of the incubators are public initiatives(Lesákova,2012).CoworkingspaceAcoworkingspaceisatypeofmulti-tenantofficethatfocusesoncreatingacommunityamongthetenants insteadofproviding office space (Weijs-Perrée et. al, 2016; Hillman, 2011). According to Moriset (2014), the coworking officeemergedforworkerswhocouldnotfindanappropriateworkplace.Usersbecametiredofthedistractionandlonelinessofworkingathomeoratthelocalcafe(Fuzietal.,2014).Themostcommonusergroupsincoworkingspacesareself-employedworkers,entrepreneursandfreelancers.Coworkingspacesareenergeticandcreativeofficeswherepeoplecaninteract,share,buildandco-create(Fuzi,2015).Coworkingemergedin2005inSanFrancisco,andduringtheaftermathofthefinancialcrisisof2008,theamountofcoworkingspaceincreasedworldwide(Spinuzzi,2012;Merkel,2015).Theongoingriseofcoworkingspacesreflectsthepopularityofthistypeofmulti-tenantofficeandisthereforeaninterestingresearchtopic(Sykes,2014).Inthenextsection,theconceptofcoworkingandthecoworkingspacewillbeexplainedinfurtherdetail.

2.2TheemergingphenomenonofcoworkingAs reported by Bizzarri (2014), the economy is moving from an industrial economy to a knowledge economy, withdigitalizationandInformationTechnologyaskeywords.Fuzi,CliftonandLoudon(2014)wentastepfurtherbystatingthatwearemovingawayfromaknowledgeeconomyandaremovingtowardsacreativeeconomy.Inthecreativeeconomy,innovation,technologyandcreativeideasareimportantvalues.“Creativityisincreasinglyseenasanew‘HolyGrail’intheeconomic world” (Fuzi et al., 2014, pp. 1). Coworking spaces are seen as workspaces that encourage and stimulatecollaborationandserendipitousencounterswithcoworkers,boostingcreativityandinnovation(Fuzietal.,2014).Sykes(2014) mentioned that coworking signposts a shifting attitude towards work focused on collaboration and fun. Thisparadigmshiftregardingworkhasbeencausedbytheredefinitionofwork.Thenewmodelofworkcentersonthecontextofcollaborativeworkandasharedeconomy(Gandini,2014;Merkel,2015).Throughoutthisstudy,theterm‘coworker’willbeusedtorefertotheuserofcoworkingspaces.‘Coworking’willrefertothetask/activity/workcarriedoutinthecoworkingspacebythecoworker.Therearetwowaystowritecoworker,withorwithoutahyphen.Coworkerisoftenmisspelledintheacademicliteratureandinvariousonlineblogs.Coworkerspelledwithout thehyphen refers to theworkerwhoworks individually in a shared community-drivenworkplace. Co-workerspelledwithahyphenreferstotheworkerwithwhomoneworksonaparticularproject(Gandini,2014).Itisalsoimportantthatcoworkingandcoworkingspacesnotbeconfused.Therefore,thesetwoconceptswillnowbeexplained.It isdifficulttodefinecoworkingusingspatialandorganizationalcharacteristics(Parinno,2015).AccordingtoDeskmag(2012), theonlinemagazineaboutcoworking,onedefinitionofcoworkingsounds illusive.Coworking isaclusteringofmultiplewords.AccordingtoMerkel(2015),the‘co’in‘coworking’referstoanaggregationofthewordscommunication,communityandcollaboration. Inordertocreateaclearviewofwhatthisentails, it is importanttonotethevarietyofdefinitionsusedbythefollowingauthorswhendiscussingcoworking:

20

• ClaySpinuzzi(2012,pp.399):“Workingalonetogether”;• JanetMerkel(2015,pp.121):“Anewformofurbansocial infrastructureenablingcontactsandcollaborations

betweenpeople,ideasandconnectingplaces”and“workingalongsideoneanother”(pp.122);• Deskmag(2012,pp.3):“Coworkingisaself-directed,collaborativeandflexibleworkstylethatisbasedonmutual

trustandthesharingofcommonobjectivesandvaluesbetweenmembers”;• Döring(2010,pp.19):“Coworkingisacombinationofworkingindependentlyandinteractingwithothers.Users

ofcoworkingspacescandecidewhere,when,howoftenandhowlongtheywork.”ThedefinitionprovidedbyDeskmag(2012)isthemostaccuratedefinitionofthefour.Deskmagmadeseverallong-termstudiesregardingtheconceptofcoworkingandanalyzedthepreviousliteraturetoestablishthisdefinition.Deskmagisbyfarthemostpopularwebsiteonthistopic,andmostoftheacademicstudiesoncoworkingrefertoit.Coworkingspacesareofdifferentsizesandhavevariouscharacteristics;however,asuniqueastheyare, theyallhavecommoncharacteristics.InanarticlefromDeskmagwrittenbyNinaPohler(2011),itwasarguedthatevery(co)workerhascertainideasabouttheconceptofacoworkingspace.Theimaginationofthisconceptisalittlevagueanddifferentforeach(co)worker.Forthatreason,thereisadiscrepancybetweenthemultipledefinitionsofacoworkingspace:

• BrunoMoriset(2013,pp.1):“Coworkingspaces(CS)areregardedas"serendipityaccelerators,"designedtohostcreativepeopleandentrepreneurswhoendeavortobreakisolationandtofindaconvivialenvironmentthatfavorsmeetingsandcollaboration.”

• Alessandro Gandini (2014, pp. 194): “Coworking spaces are shared workplaces utilized by different sorts ofknowledgeprofessionals,mostlyfreelancers,workinginvariousdegreesofspecializationinthevastdomainoftheknowledgeindustry.”

• Clay Spinuzzi (2012): A coworking space is aworkspace for the community,where people can sit down andcollaboratewitheachother.Itisalow-costworkspaceandeasilyaccessible,containingsharedknowledgefromdifferentbusinessbackgroundsthatcancontributetotheownbusiness.

• Deskmag(2012,pp.3):“Coworkingspacesaresharedworkspacesthatareusedbyvariousindependentindividualsastheirplaceofbusiness.”

ThedefinitionprovidedbySpinuzzi(2012)isthemostaccuratedefinition,becauseimportantcharacteristicsofcoworkingspacesarementioned,includingcommunity,collaborationandaccessibility.Thesecharacteristicsarereflectedinthefivecore-valuesdraftedbyKwiatkowskiandBuczynski(2011),whichwillbepresentedinthenextsection.BradNeuberg, the initiatorof thecoworkingmovement, startedacoworkingspace in2005 inSanFrancisco.The firstcoworkingspacewastheHatFactorySanFranciscoatSpiralMuze,whichwasahealingandrelaxationcenter(Spinuzzi,2012;Merkel,2015).Thiswasmeanttobeaphysicalspacewhereindependentandmobileworkerswouldcometogether(Thecoworkingwiki,2013).BradNeubergstartedthisspacebecausehewantedtoworkbyhimself,butinthemiddleofacommunity.HisthoughtswerethathewantedtoworkinaplacelikeGoogle,butnotforGoogle.Acoworkingspacemustbe a replication of a ‘cool company.’ According to Neuberg, this involves having a shared context.Neuberg said thefollowingthingaboutthisshredcontext:“IgotobarsandIseelotsofpeoplestandingaroundbutnoonetalkstoeachother.WhenIgotoahousepartyandlotsofpeoplearestandingaround,everyonetalkstoeachother,becausethereisasharedcontext.Howtogetthatsharedgroupidentitywithoutbeinginthesamecompany?”(Deskmag,2005).Thisquotedescribes thegenesisofacoworkingspace. In theyearsafter theemergenceofcoworking, thecoworkingmovementspread over the United States of America (USA). During the aftermath of the financial crisis of 2008, the number ofcoworkingspacesquicklyincreased(Merkel,2015).AccordingtoMoriset(2013),thegrowthofcoworkingspacesmayhavebeenfueledbytheburstingofthepropertybubbleinNorthAmericaandEurope,resultinginaneconomicdownturn.InSpain,wheretheburstingofthepropertybubblewasalmosttheworstintheworld,nowhasthesecondhighestnumberofcoworkingspacesinEurope.TheUSAandEuropaarecurrentlythemaincontinentsusingcoworking(Deskmag,2013).Thisincreasewasareactionagainsttheanti-socialenvironmentofothertypesofmulti-tenantoffices(Merkel,2015).Self-employedworkersandfreelancersdonothaveaconnectionwiththeothertenants(smallandlargefirms).Coworkingspacesrespondedtotheneedsoftheseusergroups.Deskmagpresentsanannualsurveyregardingcoworking,inwhichabroadperspectiveisexamined.Theongoingworldwideriseofcoworkingspacesisreflectedinthecurrentstudy.Atthebeginningof2009,justafterthefinancialcrisis,coworkingspacesnumberedapproximately3,010.In2015,thisnumberhadincreasedtoalmost7,800.Theforecastpublishedin2014byDeskmagstatedthatthenumberofcoworkingspacesworldwidewillbeapproximately37,000bytheendof2018.Thenumberofcoworkerswillalsoincreasefrom510,000in2015to2,370,000in2018(Deskmag,2014;Deskmag,2015)

21

2.3Thefivecore-valuesofcoworkingInthissection,thefivecore-valuesofcoworking,asdescribedbyKwiatkowskiandBuczynski(2011),arepresented.Thesecore-values reflect the coworking philosophy and are considered to be the cradle of the coworking phenomenon.AccordingtoKwiatkowskiandBuczynski,thesevaluesarepurposefullyvagueandopentoindividualinterpretation.Inthisway,coworkingspaceprovidersandcoworkinghostscanattractthesortofcoworkerswhoarededicatedtosimilarvalues,whothenbuildasolidfoundationforthecoworkingcommunity.Thefivecore-valuesofcoworkingareasfollows:

• Collaboration:thewillingnesstocooperatewithotherstocreatesharedvalue;• Openness:freesharingofideas,informationandpeople;• Community:agroupwithasharedpurposeandsharedthoughtsaboutcoworking;• Accessibility:coworkingspacesareaccessible,bothsociallyandeconomically,foreverytypeofworker;and• Sustainability:resourcesareusedtogether,whichbringseconomicalandecologicalbenefits

CollaborationandopennessAccordingtoHillman(2011),thecore-valueopennesscanbecombinedwiththecore-valuecollaboration.Opennessisalessoutspokencore-valueascollaboration,buthasalotincommon.Rohlf(2011)statedthatthemoremembersareopento each, themoreopportunities to give and tobenefit fromeachother. Coworking spacesprovide thepossibility forcollaboration with coworkers. According to Spinuzzi (2012), collaboration is a common result of the interaction thatnaturally occurs. Collaboration involves proposals, evaluations, questions and engaged actions (Appel-Meulenbroek,2014).AccordingtoHillman(2011),acoworkingspacecanbetypifiedasa‘highcontact’environment,whereserendipitousmeetingsoftenhappen.Spontaneousinteractionwithpartnersfromdifferentbusinessactivitiesisameaningfulprinciple.Physicalproximitywillnotautomaticallyleadtoameeting,collaborationorpartnership.Therefore,coworkinghostsarecalledupontocreateencounters,interaction,collaborationandmutualtrust(Merkel,2015;Merkel,2015;Moriset,2013).

Thetaskofthecoworkinghostistodevelopaspacethatstimulatesinteractionbetweencoworkers.Throughouttheinterviewswithcoworkinghosts,performedbyMerkel(2015),itcouldbeseenthatthesehostswerewillingtocreateahomeyatmosphereandadiversityofworkoptions.Coworkersneedencouragementtocometoworkeveryday.Merkel(2015)alsostudiedtheroleofthecoworkinghost,usuallytheownersortheoperators,intheircollaboration,interaction,creativity and productivity amongst coworkers. “The host’s activities of curatorial practice can be summarized asassemblingandarranging (people, spaces, objects), creatingand signifyingnewmeanings (collaborations, community,sustainability,opennessandaccessibility),reframing(workdifferently),caring(enablingcommunity)andexhibiting(theworkspaceanditscommunity),allinordertocreatenewwork-relatedandsocialexperiencesinthecity”(Merkel,2015,pp.131).AccordingtoMerkel,therearetwotypesofcoworkinghosts;‘theserviceprovider’andthe‘visionary.’Theserviceprovider focuseson thework aspect regarding theworkingenvironment, the visionary focuseson the ‘co’ aspectsofcoworking,asmentionedintheprevioussection.AccordingtoFuzi(2015),thecoworkinghostplaysanimportantroleincreatingdifferentmodesofengagement thatstimulate interaction,networkingandcollaboration.Collaboration in theformofknowledgeexchangebetweencoworkerscanbestimulatedbyanorganizationalplatformwherecoworkerscanmeetothercoworkersandaskparticularquestions(Parinno,2015).

AccordingtoParinno(2015),thephysicalco-presenceofcoworkersstimulatesknowledgeexchange.However,asmentionedabove,aphysicalpresencealoneisinsufficient.“Thephysicalproximitycannotproduceitsdesiredeffectsifitis not complemented by a certain degree of social and/or professional proximity” (Moriset, 2013, pp.8). Proximity inknowledge exchange is an important factor in coworking (Moriset, 2013). Appel-Meulenbroek (2014) argued that thelayoutofthebuildinghasarelationshipwithknowledgeexchange.Importantfactorsthatensureknowledgesharingarevisibility,placementwithintheroom,centralityandproximity in thebuilding (Appel-Meulenbroek,2014).Thevisibilitymechanism,whichconcernscoworkersonthesamefloor,visibleworkplacesandcompactnessoftheworkplace,isabletoincreasecollaboration.Anopenlayoutstimulatesinteractionbetweencoworkers.Themoreopenthespace,thefewerboundariesforthecommunitytocollaborate(Hillman,2011;Rohlf,2011).

Interactionisnotalwaysinitiatedfromaprofessionalpointofview;thesocialaspectsarejustasimportant.Deijl(2011)statedthathavingotherentrepreneursaroundappearstobemoremotivating.Supportandprofessionalfeedbackonbusinessactivitiesenhancethelearningprocessoftheindividualentrepreneurandinfluencesthehumancapitalaswellasthebusinesspossibilities.Supportisavailableintheformofamentorship,whereamoreknowledgeablepersonhelpsalessexperiencedworker.Inthiscase,thementortakeshis/herprotégéunderhis/herwing(Leforestier,2009).

Workingwithcoworkersinthesamecommunityleadstocollaboration.Forinstance,amarketerrequireshelpwith the front-end development of a new website. A website developer in the coworking space desires to help themarketer,andinreturnthemarketerwillhelpthewebsitedeveloperinthemarketingofhis/hernewcompany.Afreelancejournalistcapturesthisconversationandapproachesthesamedeveloperforanewwebsite.AccordingtoFerguson(2013),thereexistsa‘payitforward’vibepresentincoworkingcommunities;itisarepeatedloopofgivingsomethingandreceivingsomethingelseinreverse.Thishappensinanaturalcontext.Itissustainableascoworkersarenotdependentonresources

22

outsidethecoworkingspace.Theinfrastructure,networkandfacilitiesarepresentwithinthecoworkingspace.Itisatypeofself-sustainingenvironment(Ferguson,2013).

InthecasestudypresentedbyDeijl(2011),empiricalresearchwascarriedoutontheeffectofeconomicgrowth.Thestudyshowedthattheproductivityofthecoworkersincreased,sincetheyworkedinacoworkingspace.‘’Thiscouldbeduetothestructureandexternalcontrolthisworkspaceprovides’’(Deijl,2011,pp.31).Coworkingledtoknowledgediffusionbetweencoworkers,stimulatedinnovationandhadasignificantpositiveeffectontheincomeofthecoworker.CommunityandsustainabilityAccording to Hillman (2011), the core-value sustainability can be combined with the core-value community. In firstattempt, sustainability is associated with the words ‘green’ and ‘ecological’ which reflect the classic thoughts aboutsustainability.Asharedofficeisecologicallysustainablebecauseresourcesareshared.Coworkersallusetheelectricitynetwork,thefurniture,coffeemachineandvariousservicessuchasreceptionandamailservice.Thecostsofrentingspaceandadditionalservicesarerelativelylow,whichissustainablefromanecologicalandorganizationalperspective.Butthereisa lotmoretosustainability. Inacoworkingspace it isallaboutasustaincommunity.AccordingtoFerguson (2013),“Sustainability in a coworking community is about supporting, nourishing, about ‘buoying up’ our fellow coworkers.”Coworkers benefit from the presence of other coworkers. They become part of a larger community of like-minedindividualswithwhomtheycanshare ideasanddobusiness (Spinuzzi,2012),andformcontacts indifferentactivities,differentfieldsanddisciplinesofwork.Incontrast,coworkersmustmastereveryaspectofrunningabusiness,whichisdifficult.Coworkersmustmasteraccountancy,contracts,investmentandsoon.Often,thisisnottheircorebusinessandothercoworkerscanhelpwhichissustainableformanorganizationalperspective(Kwiatkowski&Buczynski,2011).Theresourcesinsidethe‘coworkingsystem’canbeused.Spinuzzi(2012)gaveanaccountoftwotypesofrelationships:thoseoftemporarypartnershipsbetweenbusinesses(goodneighbors)andthoseofcompaniesthatcomplementeachother'sshortcomings(goodpartners).

Themost important theme in coworking is community. A coworking space offers a community consisting ofdifferent kindsofworkers indifferent kindsof knowledge fields. Spinuzzi (2012)namedcoworking spaces communityworkspaces.Thecommunityhasthedistinctivecharactercomparedtoothertypesofmulti-tenantoffices.Forinstance,incubatorsdonotfocusontheinformalprocessofworkingtogetheranddonotstimulatecollaboration, incontrasttocoworkingspaces(Leforstier,2009).Theemphasiswhenworkinginregularbusinesscentersorservicedofficesisnotonthepresentcommunity,butratheronthespaceandservicesprovided(Weijs-Perréeet.al,2016).Thespaceisasupportivevalueintheprocessofcoworking,butiscertainlynotthefocus.Thecommunityoffersemotionalandpracticalsupportforthosewhoarecutofffromtheseactivitiesduetoisolationintheirhomeofficeorcafe.Thepeopleinthecoworkingspacemakethecoworkingspace,thepeoplebuildthecommunitytogether.Hillman(2011)arguedthatthecommunity(WE)isbiggerthanthepersonhimorherself(ME).Themainvalueofthecommunityisthatit isopentoeveryoneandieasilyaccessible.Itconcernssharingexperiences,learningnewthings,experimentinginnewareasandcelebratingeachother'ssuccesses.Coworkersvaluestructureintheirlives,astoomuchautonomycripplestheproductivity(Moriset,2013).Thecommunityhelpscoworkerscreatestructuresanddiscipline.ThisisconfirmedinthestudyofDeijl(2011).

AccessibilityHillman(2011)notedthatacoworkingspaceisquiteremarkable.Eachpersonpresentintheroomistherebecausetheychose to be there. Perhaps the workers of a corporate office do not like their work environment and prefer to besomewhereelse.Coworkers choose for themselveswhereandwhen towork, andare surroundedby coworkerswithsimilarthoughts.Acoworkerdetermineswhetherthecommunityconcernedisinlinewiththeirpersonalneeds(Capdevila,2013).Inotherwords,itisaccessibletoeveryone.Sykes(2014)andDeijl(2011)bothmentionedthattheattractivenessofcoworkingspacesisduetothereflectedflexibilityandmobility.The‘plugandplay’principleisdesirableforstartingorganizations; coworking spaces offer a lowbarrier to immediately commenceworking. The rent prices of coworkingspacesareoftenlowandtheperiodisflexible;forinstance,aday,aweekoramonth(Sykes,2014).Alargenumberofcoworkingspaceprovidershavecoworkingspacesatmultiplelocations.Thisgivestheusertheflexibilitytochoosewheretowork(Merkel,2015;Fuzi,2015;Spinuzzi,2012;Deijl,2011).AccordingtoMerkel(2015),joiningacoworkingspacehasfinancialadvantagesincontrasttorentingaspace.Itisdifficultforastartingorganizationorself-employedworkerstorentofficespaceatthebeginningoftheircareers,astheylacktherequiredcapitalandarenotcredit-qualityratedforalong-termleasecontract(Merkel,2015;Green,2014). Inaddition,rentwithin innercities,wherebusinesseswantto locatethemselves,aredifficulttopay(high)andmaintenancecostsarehigh(Merkel,2015).Bizzarri(2014)statedthateconomicreasonsarethemostimportantreasonsfortherapidemergenceofcoworking.Thelowbarrierofoccupancyduetotherent level isdecisive.Deskmag (2012) found that47%of respondents stated that rentcostswere themost importantreasonforcoworking.Incontrast,Capdevila(2013)arguedthatthemainfactorstoconsiderjoiningwererelatedtothelocation.

23

2.4MotivationsforcoworkingThemostcitedreasonforpeopletojoincoworkingspacesisthatworkingalonenegativelyinfluencestheirlives.“Theyarecutoff fromnetworkingandtrust-buildingopportunities,with limitedaccessto infrastructureandwithoutfirmbarriersbetweentheirpersonalandworklives”(Spinuzzi,2012,pp.402).AccordingtoErler(2010),thevirtualizationofourpersonallives,duetotherisingknowledgeeconomy,hascausedadesireformoresocialization.Theincreasingusageofmobiletechnologyandattitudetowardsworkhasmadeitpossibletoworkanywhereatanytime;wehavebecomedigitalnomads(Merkel,2015;Green,2014;Fuzietal.,2014;Moriset,2013;Spinuzzi,2012).Theincreasingnumberofcoworkingspaceswereareactiontoanappropriateworkplaceforloneeagles;itisabottom-upsolution(Merkel,2015).Workersbecametiredofthelonelinessofworkingathomeorsufferedfromdistractioninlocalcafesorlibraries(Moriset,2014;Fuzietal.,2014; Fuzi, 2015). Green (2014) called this type of worker “the coffee shop entrepreneur”. Although it is likely thatcoworkingspaceshaveemergedfromcafesorothersocialsettings,arecentstudybyDeskmag(2012)indicatedthatonly4% of the respondents previouslyworked from a café. It is perhapsmore an expression. Themajority of coworkers,approximately60%,firstworkedathomebeforejoiningacollaborativespace.Coworkingspacescombinethebestofbothworlds,withelementsofacafe(social,energeticandcreative)andelementsofaworkspace(productiveandfunctional)(Orel, 2015).A formofprofessional isolation isbuilt into the livesof these ‘loneeagles,’ because there is no formofcollaborationandnetworkingwithotherworkers.AccordingtoMoriset(2013),theyareinneedofaso-calledthirdplace.Oldenburg(1989)mentionedthreetypesofplacesforsocialsurroundings,namelythehome(firstplace),theworkplace(secondplace)andthe‘anchors’ofacommunitysuchascafes,clubsorparks(thirdplace).AccordingtoOldenburg,these‘thirdplaces’areimportantforcivilsociety,democracy,civicengagement,andforestablishingfeelingsofasenseofplace.Fuzi(2015)arguedthatthirdplaceswillreplacetheofficeatacertainmoment.Workerswhowanttoescapelonelinessandboredomandwhoareinsearchofserendipityandpotentialinteractionsjoinoneofthesethirdplaces(Moriset,2013).Peoplelosethesenseofcommunitywhentheyworkfromhome;however,theygainindependence.Peoplelosefreedomandautonomywhenworkinginatraditional9-to-5corporatejob;however,theygainafeelingofcommunityandstructure(Fuzietal.,2014).Coworkingspacescanbeseenasthirdplacesregardingbeinganchorsofacommunity.However,thisdependsonthecontractthatdividesasecondplacefromathirdplace.Ifsomeoneisusuallyworkinginacoworkingspace,itthentendstobeaworkplace(secondplace).

Thereareseveralreasonsand/ormotivationsregardingwhyworkers joinacoworkingspace. It is importanttomakeacleardistinctionbetweenthedifferenttypesofmotivations.Multiplevaluedaspectsofcoworkingarementionedintheliterature.“Socialcircles increasesignificantly,businessnetworksgrow,hugejumpsinproductivityareseen,healthandprivate life factors improve, isolationdecreases,andoverone thirdof coworkers surveyedconfirmed that their incomeincreasedsincejoining”(Deskmag,2013,pp.7).Reasonsand/ormotivationstojoinacoworkingspacearevariousinnatureand can be expressed/labelled in three possibleways, namely input, output or outcome (Van deVoordt et al., 2016;Odhiambo,2013).Inthemajorityofthestudies,input,outputandoutcomewereusedinterchangeably,invitingconfusion.Beforecontinuing,itisimportanttomakeacleardistinction.Thisisimportantbecausethegoalofthisstudyregardsthecharacteristicsofcoworkingspaces,whichcanbeconsideredastheinput.Intheenumerationpresentedbelow,theterms‘input,’ ‘throughput,’ ‘output,’ ‘outcome’ and ‘impact’ will be clarified. In particular, the terms ‘input,’ ‘output’ and‘outcome’areofinterest.

• Input:WhichkindofresourcesareusedàPresentedinsection2.5.o Examples:location,desk,chair.

• Throughput:Managementofimplementation

• Output(firstlevelofresults):Whatisproducedthroughtheresourcesintheshortterm;theresults(tangible)àGivenintable

2.2.o Examples:vibrantandcreativeatmosphere,collaboration,networking.

• Outcome(secondlevelofresults):Whatisproducedthroughtheresourcesinthelongterm;theresultantchangesorbenefits

(“Outcomesaretheultimate,possiblyfar-reachingresultsofoutcomes”(Smith,1995,pp.202)).àGivenintable2.3.o Examples:higherincome,higherjobsatisfaction,increasedstimulation.

• Impact(thirdlevelofprojectresults):Longtermorindirecteffectsoftheactivity(difficulttoascertain).

InputàThroughputàOutputàOutcomeàImpact=Addedvalue

24

In the following three tables, thevaluedaspectsof coworking spacecharacteristics (table2.1)and themotivationsofcoworkingareexpressedinoutput(table2.2)andoutcome(table2.3).Themultiplecharacteristics(input)andmotivations(outputandoutcome)arederivedfromtheliterature.Mostofthestudiesaredescriptivestudiesandappointthedifferentformsofinput,outputandoutcomebuttheydonotuseitascertainvariables,likeinthisstudy.Inthisstudytheinputcharacteristicswillbeusedasavariableforthefactoruserpreferences,andtheoutputwillprovidevariablesforthefactorusercharacteristics(motivationstoworkinacoworkingspace).Table2.1:Input(mentionedcharacteristicsofcoworkingspaces)

Table2.2:Output(motivationstoattendacoworkingspace)

Table2.3:Outcome(motivationstoattendacoworkingspace)

Coworkingspacecharacteristics(input) Leforestier(2009)

Deijl(201

1)

Deskm

ag(2

012)

Spinuzzi(2

012)

Capd

evila(2

013)

Deskm

ag(2

013)

Moriset(2

013)

Parinn

o(2013)

Fuzietal.(2014)

Gandini(2

014)

Green

(201

4)

Kojo&Nen

onen

(2014)

Sykes(2014)

Capd

evila(2

015)

Deskm

ag(2

015)

Fuzi(2

015)

Merkel(2015)

Orel(20

15)

Spreitzer(2015)

Deskm

ag(2

016)

Totaal

Flexiblecontract(short,easy-in,easy-outprocedure) x x x x x x x x x x x x 12Collaborativespaces x x x x x x x x x 9Flexiblesharedfacilitiesandservices(fixedandadditional) x x x x x x x x x 9Flexible(shared)workspaces x x x x x x x x 8Socialevents x x x x x x x x 7Accesstotools,resourcesandnetwork x x x x x x x 7Networkingeventsandworkshopps x x x x x x x 7Acoworkinghost x x x x x x 6Quitespaces x x x x x x 6Location x x x x 4Kitchenareas x x x x 4Conferencerooms x x x 3Eventspaces x x x 3Meditationrooms x x x 3Open-layout x x x 3Virtualorganizationalplatform x x x 3Goodaccessibility(car,bike,PT) x x 224houraccess x 1Artshows x 1Informalcouches x 1Security x 1Yogaclasses x 1

Output Leforestier(2009)

Deijl(2

011)

Deskmag(2

012)

Spinuzzi(2012)

Capd

evila(2

013)

Deskmag(2

013)

Morise

t(2013)

Parin

no(2

013)

Fuzietal.(2014)

Gand

ini(2014)

Green(201

4)

Kojo&Nen

onen

(2014)

Sykes(2014)

Capd

evila(2

015)

Deskmag(2

015)

Fuzi(2015)

Merkel(2015)

Orel(20

15)

Spreitzer(2

015)

Deskmag(2

016)

Totaal

Feelingpartofacommunity x x x x x x x x x x x x x x x x x x x 19Sharingideasandknowledgewithcoworkers x x x x x x x x x x x x x x x x x 17Collaborationwithcoworkers x x x x x x x x x x x x x x 14Interactionwithcoworkers x x x x x x x x x x x x 12Professionalsupportfromcoworkers x x x x x x x x x 9Vibrantandcreativeatmosphere x x x x x x x x 8Accesstobroadnetworkofpossiblepartners x x x x x x x 7Affordable(lowcapitalinvestment) x x x x x x x 7Professionalmeetingsandencounterswithcoworkers x x x x x x x 7Emotionalsupportfromcoworkers x x x x x 5Networkingwithcoworkers x x x 3Sharingcontactsofbusinesspartnerswithcoworkers x x 2Easyaccesstocapital x 1

Outcome Leforestier(2009)

Deijl(201

1)

Deskm

ag(2

012)

Spinuzzi(2

012)

Capd

evila(2

013)

Deskm

ag(2

013)

Moriset(2

013)

Parinn

o(2013)

Fuzietal.(2014)

Gandini(2

014)

Kojo&Nen

onen

(2014)

Sykes(2014)

Capd

evila(2

015)

Deskm

ag(2

015)

Fuzi(2

015)

Merkel(2015)

Orel(20

15)

Spreitzer(2015)

Deskm

ag(2

016)

Totaal

Increasedsocialcircle x x x x x x x x x x 10Feelinglessisolated,feelinganorganizationalstructure x x x x x x x x 8Increasedinnovation x x x x x x 6Increasedproductivity(byhigherjobsatisfaction) x x x x x x 6Increasedskillset x x x x x 5Higherincomeinlongterm(byincreasedproductivity) x x x x 4Increasedstimulationofself-reflectionandself-education x x x x 4Gettingabetterwork-lifebalance x x x 3Increasedbusinessnetwork x x x 3Increasedmotivation x x x 3Higherjobsatisfaction(byflexibilityoflabour) x x 2

25

Bymeansofthesetables,theoutputandoutcomeofcoworkingspacesarepresented.Acertaininputwillcauseacertainoutput; a certain output will cause a certain outcome. Concerning the output, the motivations ‘feeling part of acommunity,’ ‘sharing ideas and knowledge with coworkers’ and ‘collaboration with coworkers’ were commonlymentioned. Concerning the outcome, ‘increased social circle’, ‘feeling an organizational structure’, and ‘increasedinnovation’werecommonlymentionedasvaluedaspects.

2.5CharacteristicsofcoworkingspacesSofar,atheoreticalframeworkonthesubjectscoworkingandcoworkingspacesisdeveloped.Thefirstsub-questionis“Whatarethetypicalcharacteristicsofcoworkingspaces?”Theprevioussectionshowedthedifferencebetweeninput,outputandoutcome.Inthisstudy,thecharacteristicsofcoworkingspacesareassociatedwithinput.Inputconcernsthephysical characteristics (e.g., location, building and space) and non-physical characteristics (e.g., community andorganizationalplatform).Inordertoanswerthefirstsub-question,itisofimportancetocreateaclearunderstandingofthe characteristics of coworking spaces. In table 2.1, the valued characteristics of coworking spaces according to theliteraturewerepresented.Thisinformationisofimportancewhencreatinganoverviewofthetypicalcharacteristicsofcoworkingspacesasnooverviewofthesecharacteristicshasthusfarbeenpresented.

Multiplestudies(Rotheetal.,2011;Rovers,2016;Leesman,2013;Liebregts,2013;VanSusante,2014;Hartog,2015)haveexaminedpreferredworkenvironmentcharacteristics.Severalstudiesfocusedonsingle-tenantoffices(Rotheetal.,2011;Leesman,2013;Liebregts,2013;VanSusante,2014)andsomeonmulti-tenantoffices(Hartog,2015;Rovers,2016).However,researchonpreferredworkenvironmentcharacteristicsinmulti-tenantofficesisstilllimited,especiallyincoworkingspaces.Leesman(2015)surveyedovermorethan1,200workplacesandanalyzedtheresponsesofmorethan140,000employees.Thisdatabaseisknownasthe‘Leesmandatabase,’andwasusedinteraliainthestudiesbyLiebregts(2013)andVanSusante (2014). LiebregtsandVanSusantebothapplieda factoranalysis to theLeesmandatabase toidentifycertainfactors.Hartog(2015),whostudiedusersatisfactioninmulti-tenantoffices,usedthefactoranalysisbyVanSusanteandcomposedatablecontainingvaluedaspectsofmulti-tenantoffices(AppendixA).Hartogdividedthephysicalcharacteristicsofmulti-tenantofficesintothefollowing10characteristics:

• Location;• Officeexterioranddivision;• Officedécor;• Facilitiesandservices;• Seclusionrooms;• Officeleisure;• Flexibility;• ICTandequipment• Privacy;and• Officeclimate.

Thelistofcharacteristicsofsingle-tenantofficeswasexpandedtoalistcontainingcharacteristicsofmulti-tenantoffices.Ashortcominginthecurrentstudywasthatnodistinctivevariationwasmadetothevaluedaspects.Input,outputandoutcomeareusedinterchangeably.Inordertodefineanoverviewofthetypicalcharacteristicsofcoworkingspaces,thevaluedcharacteristicsofcoworkingspaces(table2.1)andthelistofcharacteristicsofmulti-tenantoffices(AppendixA)weremerged(table2.4).

Itisofimportancetoemphasizethatthecharacteristicspresentedintable2.4areonlythetypicalcharacteristicsofcoworkingspaces; therefore,characteristics that trulydefineacoworkingspace.For instance, thecharacteristicsof‘security’ and ‘cleaning,’ which are covered by facilities and services, were not taken into account, because thesecharacteristicsdonotdifferentiateacoworkingspacefromanothertypeofmulti-tenantoffice.Inaddition,characteristicslike‘security’and‘cleaning’,havethesamecharacter/capacityineverytypeof(multi-tenant)officeandisthereforenotincludedasatypicalcharacteristicofacoworkingspace.However,multipletypical(multi-tenant)officecharacteristics(like‘coffeeandteavendingmachine’,‘receptionandhelpdesk’and‘collaborativespaces’)areneverthelessincludedintheoverviewoftypicalcoworkingspacecharacteristicsbecausethesecharacteristicsoccurinatypicalcharacter/capacity.Forinstance,theplacewherethecoffeeandteavendingmachineislocatedisacentralspotwherecoworkerscometogetherto collaborate. The reception and helpdesk is more explicitly present in coworking spaces and is, together with thecoworkinghost,abinding factorbetweencoworkers.Thecollaborativeworkspacesalsooccur innormalofficesbut incoworking spaces these are workspaces where separate workers (lone eagles) come together to co-work. Certaincharacteristicshaveanotheridentity.

26

Table2.4:Typicalcharacteristicsofcoworkingspaces

Thecharacteristicslocation,officeexterioranddivision,officedécorandfacilitiesandservicesarederivedfromAppendixA (Hartog, 2015). The characteristics collaboration and openness, community and sustainability and accessibility arederived from the study of Kwiatkowski and Buczynski (2011). The valued aspects (table 2.1) are subdivided over thecorrespondingcharacteristics.Nowthetypicalcoworkingspacecharacteristicsaredetermined,theusergroupsandusercharacteristicswillbeexaminedinthefollowingsection.

2.6MarketsegmentationandusercharacteristicsAccordingtoGoyat(2011),everycompanydesirestofocusonspecificcustomers.Todeterminethegroupsonwhichtofocus,segmentationoftheconsumermarketisnecessary.Marketsegmentationfindsitsorigininmarketing,andisthefirststepinmarketingstrategy.Thesecondphaseistargetmarketingandthethirdphaseispositioningtheproduct(Kotler,1994).AccordingtoLin(2002),Smith(1956)wasthefirsttointroducemarketsegments.Themarketsarecharacterizedbywidelyheterogeneoususercharacteristics,withwidelyspreadcustomers.Bydividingthemarketintosmallergroupswithdistinctneeds,characteristicsorbehaviors,itbecomesviabletoconcentrateonspecificusergroupsratherthanontheentiremarket.Rotheetal.(2011)arguedthatitisincorrecttosaythatemployeeswhoworkinanidenticalwayalsopreferidenticalworkenvironmentcharacteristics.Thechallengeisthereforetounderstandthatdifferentusersareinneedofadifferentworkenvironment(Rotheetal.,2011).Organizationscanadjusttheirmarketingtoparticularusersthroughtheuseofsegmentation inorder toprovideahighervaluethataddresses thespecificneedsofaspecificgroup.“Thepurposeofsegmentation istoconcentratemarketingenergyandforceonsubdividingtogainacompetitiveadvantagewithinthesegment.Itisanalogoustothemilitaryprincipleoftheconcentrationofforcetooverwhelmtheenemy”(Goyat,2011,pp.1).Marketsegmentationalsohelpsbetterunderstandaparticularmarketandtodevelopanappropriateproductforvarioussegments(Guillet&Kucukusta,2014).

There are two differentways of identifying different segments. According to Binge et al. (2008), researchersdistinguishsegmentationaprioriandsegmentationposteriori.Theapriorisegmentationisbasedonthesegmentsofpriorknowledge and theposteriori segmentation on a classification scheme founded onmultiple attributes. Jansen (2009)mentionedthesesametwomethodsofcomingtoaconcretesegmentation,butnamedthemdifferently.Jansendescribedthetop-downapproach,inwhichspecificcriteriaareusedtosegmentthemarket.Fromcoarse(market,country)tofine(characteristicsof theconsumer).Second,thebottom-upapproach, inwhichconsumerdataare lumpedtogetherandanalyzedinordertofindsimilarities.Inthisway,consumerswithsimilarneedstogetherformasegment.Thetop-downmethodcreatesaprioriclassifications;importantinformationmaybeincorrectlyinterpretedusingthismethod(Jansen,2009).Thebottom-upmethodisthemostaccuratewayofcreatingsegments.Inthisthesis,bothmethodsareused.Inthenextsection,apriorisegmentationisusedtoidentifythedifferentusergroupswithincoworkingspaces.Subsequently,thebottom-upmethodisused,bymeansofalatentclasslogitmodel,tocreatesegmentsbasedonsimilaritiesinpreferredcoworkingspacecharacteristics.Asthereisalargeamountofdiversityinthemarkets,itisimportantthatthecustomersareidentifiedfirst.

Location CollaborationandopennessAccessibilityoflocation(mobility) CollaborativespacesTypeoflocation Conferencerooms

EventspacesOfficeexterioranddivision FitnesscentreArchitecture InformalareaswithsofasandcouchesLay-outandsubdivisionofthebuidling KitchenareasLightninginthespace Quitespaces

Officedécor CommunityandsustainabilityAppearanceaesthetics CustomizedsocialnetworkAppearanceofthespace(branding) EcologicalsustainableArtandphotography Networkevents/training/workshopsDiversityofworkplaces Organizationalsustainable

VirtualorganizationalplatformFacilitiesandservices SocialeventsBookingsystemforspacesandworkspots PresentationorreadingsCanteen/restaurantCleaningservices AccessibilityClothingservices DiversityofthetenantsCoffeeandteavendingmachine Leasecontract(price,period,accessibilityofCoworkinghost(stimulatesinteraction)Openinghours(24hoursaccess)Receptionandhelpdesk

27

AccordingtoKotler(1994),marketerscanapplygeographiccharacteristics,demographiccharacteristics,psychographiccharacteristicsandbehavioralcharacteristicstosegmentconsumermarkets.Geographic,demographicandpsychographicarerelatedtotheconsumercharacteristics.Thebehavioralcharacteristicsarerelatedtotheconsumerresponses.Inthismaster’sthesis,segmentationtakesplacebasedontheuserpreferencesofcoworkingspacecharacteristics.Coworkerswithequalpreferenceswillbegroupedtogether.Inordertodescribethesegroups,theusercharacteristicsareofparticularinterest. Therefore, the characteristics which are related to the consumer responses will not be included in furtherresearch.Inordertodeterminetheusercharacteristicsofcoworkers,thefollowingcharacteristicswillbeexamined:

• Geographiccharacteristics;• Demographiccharacteristics;and• Psychographiccharacteristics.

Moscardo et al. (2001) argued that geographic and demographic characteristics are associated with the a priorimethodology. Psychographic characteristics are associatedwith theposteriorimethodology. The characteristicswhichdescribetheconsumercharacteristicsofthecoworkersaredividedintomultiplevariables,whichnowwillbediscussed.Geographiccharacteristics“Geographicsegmentationcallsfordividingthemarketintodifferentgeographicalunitssuchasnations,states,regions,countries, cities, or neighborhoods” (Kotler, 1994, pp. 148). In addition, geographic variables are also important tocompanies,theproductsofwhichrelyontheclimate,suchasseasonalproducts(summerandwintergear).Geographicvariables are necessarywithin this study to determinewhere a specific person lives. Thismethod of segmentation iscommonwhenservingpeopleinaparticulararea.Thefollowinggeographicvariablescanbedistinguishedandarerelevanttothisinvestigation(Kotler,1994;Moscardoetal.,2001;Goyat,2011):

• Locationofresidence;• Locationofcoworkingspace;and

DemographiccharacteristicsDemographic variables are required in order to determine who the customers are. Demographic characteristics arepopulation characteristics.Demographic variables segment amarketbasedon, for example, age, gender and income.Demographicsegmentationisoneofthemostusedformsofsegmentation,becauseitillustratesatypicalmemberofagroup. The following demographic variables can be distinguished and are relevant to this investigation (Kotler, 1994;Moscardoetal.,2001,Goyat,2011;Hartog,2015;Rovers,2016;Budie,2016):

• Gender;• Age;• Nationality;• Educationallevel;• Usergroup(differentusergroupsaccordingtheliteraturearepresentedinthenextsection);• Positioninorganization;• Sectoroforganization;• Levelofincome;• Landofcoworkingspace;• Hoursofworkingperweekwithinthecoworkingspace;and• Transporttocoworkingspace(e.g.,car,publictransport,bicycle).

PsychographiccharacteristicsPsychographic variables concern the lifestyles of consumers; their activities, interests and opinions. “Psychographicdimensionsarethemeasurementsoftheconsumer’smind,whichpinpointhowheorshethinks,feels,reactsandreflects”(Prasad&Arysari,2010,pp.70).Psychographiccharacteristicsarethenot-so-visiblecharacteristics;whycertainusersbuya product. The following psychographic variables can be distinguished and are relevant to this study (Kotler, 1994;Moscardoetal.,2001,Goyat,2011):

• Motivationsforworkinginacoworkingspace(arepresentedintable2.2);• Personality(e.g.introvert,extrovert);and• Benefits(thebenefitsusersareseekinginaparticularproduct).

28

2.7UsergroupsincoworkingspacesThe genesis of multi-tenant offices clearly shows that several types of multi-tenant offices emerged due to thedifferentiationofaspecificusergroup(Ketting,2014).Forinstance,regularbusinesscentersevolvedasareactiontotheorganizationsthatwereinneedofsecretarialandtelephonyservices(Peltier,1992).Subsequently,servicedofficespacesreactedtoorganizationsthatrequireddifferentkindsofservicesandsharedresources(Peltier,1992).Incubatorsevolvedasareactiontostartupsthatwereinneedofassistanceinnon-corefields.Coworkingspacesoriginatedinresponseto‘loneeagles,’whoneededathirdplacetoworkduetoboredomandisolation(Moriset,2013).Adifferenceexistsintheneedsofusersofvarioustypesofmulti-tenantoffices.Eachtyperespondstothespecificneedsandcharacteristicsoftheparticularuser.

Throughout the related literature, it became apparent that the users of coworking spaces are typified as aheterogeneousgroup (Spinuzzi,2012;DeVries&vandeBesselaar,2013).Coworkershaveoverlappingconcerns,anddifferentworkersrequiredifferentkindsofspacesanddifferentkindsofsupport.Contradictionsexistregardingtheneedsforthecharacteristicsofthespace,bothphysicalandnon-physical(Spinuzzi,2012).Inthisstudy,therelationbetweentheusercharacteristicsandtheuserpreferencesisexamined.Thegeographic,demographicandpsychographiccharacteristicsofthecoworkerswilldescribetheusercharacteristicsofthecoworkers.Intherelevantacademicliterature,thecoworkersareassignedtospecificusergroups(seetable2.5)butarenotdescribedinmuchdetaillikethisstudywilldo.Inordertodrawapictureoftheusergroupsincoworking,anoverviewispresentedintable2.5.Table2.5:Usergroupsofcoworkersaccordingtotherelevantliterature

Themostcommonusergroupsaretheself-employedworkers,entrepreneursandfreelancers.Littleattentionhasbeenpaidtotheusercharacteristicsoftheseusergroups.ThestudybyDeijl(2011),whichperformedempiricalresearchamong85coworkers,focusedonhowcoworkingaffectseconomicgrowth;however,itwasquitelimitedintermsofgeographical,demographicalandpsychographicusercharacteristics.Inthemajorityofthestudies,userswerecompartmentalizedintoboxes.User groupswerenominally approached; theybelonged tooneor anotheruser group,butnot to several. Forexample,auserisafreelanceroraself-employedworker,butnotboth.Detailedinformationregardingtheusergroupsincoworkingspacesiscurrentlymissingintheliterature.Mostoftheauthorsaffirmedthatself-employedworkers,freelancersandentrepreneursarethemainusersofcoworkingspaces. The difference between a self-employed worker and a freelancer is largely the connotation. The term ‘self-employedworker’(zzp’erinDutch)coversalargernumberofformsandtypesofbusinessesthanfreelancers.Thegroupof self-employed workers contains professional groups who are unassociated with freelancers, such as independentlawyersandaccountants.Thecreativesectorisoftenattributedtofreelancers.Freelancersmoreoftenworkinsectorssuchasjournalism,communicationanddesign,andlessinsectorssuchasconstruction,transportorhealthcare.Thisissimilartoentrepreneurs,althoughthisdefinitionisalittleclearer.Foranentrepreneur,thisconcernsthebusinessitself.Thismayincludeseveralprojectsinvarioussectors.Anentrepreneurisanentrepreneurialbusinessowner.

Smallandlargefirmsarealsomentionedasusersofcoworkingspaces.Anincreasingnumberoforganizationsincorporatecoworkingintotheirbusinessstrategytodealwithstrategicandmanagementneeds.Forexample,aresearchteam or an individual extended worker can be situated outside their headquarters to becomemore productive andinnovative(Parinno,2015;Capdevila,2013;Spinuzzi,2012).Incontrast,coworkingspacesattractemployeeswhodemandaflexibleworkspaceandworkatflexibletimes(Spreitzeretal.,2015).Furthermore,corporateorganizationsareapplyingcoworking space concepts in their own offices. In this way, organizations create an innovative and productive workenvironmentbetweentheirownwalls.

Deijl(2011)reportedthatstudentsalsomakeuseofcoworkingspaces.Theygotocoworkingspacestostudyandworkontheirschoolassignments.ThisstudyisbasedontheconceptofSeats2Meet,aDutchcoworkingspaceprovider.AtSeats2Meet,norentmustbepaid(economiccapital);rather, leaseoftheworkingspace ischarged insocialcapital

Usergroups Leforestier(2009)

Deijl(2

011)

Deskmag(2

012)

Spinuzzi(2012)

Capd

evila(2

013)

Deskmag(2

013)

Morise

t(2013)

Parin

no(2

013)

Fuzietal.(2014)

Gand

ini(2014)

Green(201

4)

Kojo&Nen

onen

(2014)

Sykes(2014)

Capd

evila(2

015)

Fuzi(2015)

Merkel(2015)

Orel(20

15)

Spreitzer(2

015)

Deskmag(2

016)

Totaal

Freelancers x x x x x x x x x x x x x x x x 16Entrepreneurs x x x x x x x x x x x x x x x 15Self-employedworkers x x x x x x x x x x x x x x 14Smallfirms x x x x x x x x 8Extendedworkers x x x x 4Largefirms x x x 3Students x x 2

29

(knowledge).Usersrentaworkplace,andinexchangeusersmustbeopentothepossibilityofknowledgeexchange.Forthatreason,Seats2Meethaslowbarriersforstudents,andtheycanalsopaywithsocialcapital.Spinuzzi(2012)studiedninecoworkingspacesinAustin,Texas,andinterviewedcoworkersandcoworkingoperatorstoaccuratelydefinecoworking. InSpinuzzi’s study,coworking isanalyzed fromavarietyofperspectives,namelyobjects,actors, outcomes and contradictions in coworking. Throughout the study, it was observed that coworkers are aheterogeneousgroupofworkers(Spinuzzi,2012;DeVries&vandeBesselaar,2013).Incontrasttootherstudies,Spinuzzi(2012)distinguishedtwomaintypesofusergroups.Table2.5showsthatSpinuzziacknowledgedthedifferentusergroups;however,regardingobject,actors,outcomesandcontradictions,thetypesofuserswerereducedtotwogroups,namelygoodneighborsandgoodpartners.

Thegoodneighborsseeparallelworkasanoutcome.Thesecoworkersaresearchingforaprofessionalspaceinwhichtheycanmeetclients.Mostofthetime,thesearesmall-businessownersandconsultants.Thecoworkingspaceisaworkingplacewheretheycanleavetheirbelongingsunattended;theytrusttheircoworkersasneighbors.Itisimportantforthegoodneighbortointeractwiththecoworkerstogatherfeedbackfromdifferentfieldsofwork.Discussingbusinessistheprimaryaimofthisgroup.Likeneighbors,coworkersdonothavetobeconnectedbythesamefieldofworkbutareconnectedbythoughtsofsharingandcollaboration. Thegoodpartnersseecooperativeworkasanoutcome.Thesecoworkersaresearchingforface-to-facemeetingswithcoworkers,generatingcross-talk.Coworkingisworkingwithotherfieldsofbusinessandbuildingtrustthatcouldleadtoapartnership.Mostofthesecoworkersareentrepreneursandfreelancersandprovideservicesforbusinessesinsteadofproductsforcustomers;theydonotseetheirclientsface-to-faceintheoffice.Beingagoodpartnertocoworkersandbeingcooperativeinworkistheprimaryaimofthisgroup.Likepartners,coworkersareconnectedthroughtheirworklivesandsociallives,whichoccurswithinthecoworkingspaces. Accordingtotherelevantliterature,thefollowingsevenusergroupsmakeuseofacoworkingspaceandwillbeusedinthisstudyforthevariableusergroups:

• Entrepreneurs;• Extendedworkers;• Freelancers;• Self-employedworkers;• Smallfirms;• Students;and• Largefirms.

2.8ConclusionInthischapter,thefollowingtwosub-questionsareanswered:“Whatarethetypicalcharacteristicsofcoworkingspaces’’and‘’Whichtypesofuserscanbeidentifiedincoworkingspacesandwithwhichcharacteristicscantheybedescribed?’’.Theclassicalphysicaldesignofacoworkingspaceisanopenfloorplanwithlargetables,inwhichcoworkerscanhaveeyecontact and easily interactwith one another. The social areas in the space, such as the kitchen,meeting rooms andrelaxationspots,turnitintoacollaborativespaceinwhichworkerscanhavesocialandprofessionaldiscussions.Regarding‘typicalcharacteristics,’themaincharacteristicsinvolveddistinguishcoworkingspacesfromothertypesofmulti-tenantoffices.Thedistinctivecharacteristicswereincludedinthisinvestigation;thenon-typicalcharacteristicswereoutsidethescope of this study. The table detailing coworking space characteristics consisted of six characteristics with 33correspondingvariables.Thefollowingtypicalcharacteristicsofcoworkingspacesemergedbasedonthefivecore-valuesandthevariouscharacteristicsofmulti-tenantoffices.

• Location;• Officeexterioranddivision;• Officedécor;• Facilitiesandservices;• Collaborationandopenness;• Communityandsustainability;and• Accessibility.

Coworkingisoftenassociatedwithself-employedworkersandfreelancers;however,itbecameobviousthattheusersofcoworkingspacesareamoreheterogeneousgroup.Accordingtotherelevantliterature,thefollowingusergroupsmakeuseofacoworkingspace:

30

• Entrepreneurs;• Extendedworkers;• Freelancers;• Self-employedworkers;• Smallfirms;• Students;and• Largefirms.

Each user group contains a great variety of users with a different set of user characteristics. With the followingcharacteristics,thejustmentionedusergroupscanbetypified:

• Geographiccharacteristics;• Demographiccharacteristics;and• Psychographiccharacteristics.

Inthischapter,thetheoreticalframeworkisdevelopedwhichisthebasisfortherestofthismaster’sthesis.Inthefollowingchapter,theresearchdesignofthestudyaccordingmethodology,operationalization,reliabilityandvalidityisdescribed.Theconceptualmodelpresentedinfigure1.1canbeexpandedwiththeindependentfactorof‘usercharacteristics’withthreefurthercharacteristics(geographic,demographicandpsychographiccharacteristics)andwiththedependentfactorof‘userpreferencesforcoworkingspacecharacteristics’withsevenfurthercharacteristics(location,officeexterioranddivision,officedécor,facilitiesandservices,collaborationandopenness,community,andsustainabilityandaccessibility).

31

Chapter3:ResearchDesignInthepreviouschapter,aliteraturereviewiscarriedoutonthesubjectscoworkingandusercharacteristics.Frombothsubjects,variablesemergedthatindicateboththecharacteristicsofcoworkingspacesandusersofcoworkingspaces.Theresearchdesign,concerningmethodologyandoperationalizationofthevariablesisfurtherexplainedinthischapter.Firstinsection3.1,arésuméofthetheoreticalframeworkisgiven,whichistranslatedintoamoredetailedconceptualmodel.Thismodelshowstherelationbetweentheindependentanddependentfactorsandcorrespondingvariables.Insection3.2,themethodologyforthedatacollectionandforthedataanalysisisexplainedinmoredetail.Theoperationalizationof the independent and dependent variables is carried out in section 3.3. In section 3.4, the distribution of thequestionnaireisdiscussedaswellasadescriptionofthepopulationofthisresearch.Thereliabilityandvalidityofthestudyisaddressedinrespectivelysection3.5and3.6.Theconclusionoftheresearchdesignisdrawninsection3.7

3.1RésumétheoreticalframeworkandconceptualmodelInthismaster’sthesis,therelationbetweenusercharacteristicsandtheirpreferencesforcoworkingspacecharacteristicsis examined. The preliminary conceptualmodel is presented in chapter one (figure 1.1). In the previous chapter, thecharacteristicsandvariableswithwhichthemodelshouldbeextendedarementioned.Intable2.4,sevencharacteristicswith45correspondingvariablesare identifiedas typicalcharacteristicsofcoworkingspaces.Besidesthedesignof thevariable, the coworking phenomenon is described with reference to the history, definitions and motivations. Theindependentfactor(usercharacteristics)hasalsobeenaddressedinthepreviouschapter.Theusergroupsincoworkingspacesandthecharacteristicsoftheseusers(geographic,demographicandpsychographic)aredescribedinthischapter.Thefinalconceptualmodelisshowninfigure3.1.Thedependentcharacteristiccollaborationandopennessisremovedfromtheconceptualmodelbecausethecorrespondingvariables,whichareaskedinthequestionnaire,betterfitunderthe characteristics office exterior and division. The characteristic collaboration and openness focus on those typicalcharacteristicsthatstimulatecollaborationandopenness,inparticularthelayoutandtypeofspaces(e.g.collaborativespacesandeventspaces).Thesevariablesarealsocoveredbythevariablelayoutandsubdivisionofthebuildingofthecharacteristic office exterior and division. The variable architecture and lightning in the space are covered by thecharacteristicofficeexterioranddivisionbutnotby thecharacteristic collaborationandopenness. In summary,officeexterioranddivisionisamorecompletecharacteristic.

Inordertoprovideaclearimageoftheconceptualmodel,itisimportanttoappointthevariouspartsofit.Theindependent variables (user characteristics) and the dependent variables (user preferences for coworking spacecharacteristics)arecalled ‘factors’,which isacollectionofmultiplecharacteristics.For instance,usercharacteristics iscalledthefactorandthegeographiccharacteristicsarecharacteristicsofthefactorusercharacteristics.Onlythefactorand their corresponding characteristics will be shown in the conceptual model. Each characteristic, for both theindependent anddependent characteristic, consists ofmultiple variables. The variablesof the independent factor areidentifiedinsection2.6.Thevariablesofthedependentfactorareidentifiedinsection2.5.

Figure3.1:Conceptualmodel

32

3.2MethodologyToclarifytheusedmethodologyinthisresearch,themethodsaresplitintotwoforms,namely:methodsfordatacollectionandmethodsfordataanalysis.Methodsforcollectingthedata(surveyquestionsandanattributebasedstatedchoicemethod)willbediscussedfirstinsection3.2.1,followedbythedataanalysismethods(descriptiveanalysis,multinomiallogitmodel,latentclasslogitmodel,chi-squaretestsandindependentsamplesT-tests)insection3.2.2.3.2.1MethodsfordatacollectionTo collect the data for the factor user characteristics, survey questions are asked to determine the geographic,demographicandpsychographiccharacteristicsofauser.CollectingdatawiththeaidofsurveyquestionsisaccordingtoSapsford(1999)areliableandobjectivemethodtocollectinformationaboutpeoples’characteristics.Tocollectdataforthefactorpreferredcharacteristicsofcoworkingspaces,astatedpreferencemethodwillbeused.Thistypeofmethodreferstoanapproach inwhichpreferencedata iscollectedfromsubjects inhypotheticalsituations(Adamowiczetal.,1998). Hypothetical situations are shown based on, for instance, different alternatives, in which an alternative is arepresentationofarealsituation.Aquestionlike‘Whatisimportant?’doesnotdeliveraclearanswer.AccordingtoEeuwijk(2010), when considerations are not necessary tomake a choice (everything is equally important). Respondents findeverythingimportantthat isrelatedtothatparticularchoice. It isthereforenecessarythatrespondentsarefacedwithcertain options to come to a particular choice. A respondent has to make choices according the advantages anddisadvantagesofaparticularchoice.Theextenttowhichtheindividualpreferencescorrespondtothepropertiesofanobjectaredecisiveforthechoice(Eeuwijk,2010).Whenrankings,ratingsandchoicesareofinterestastatedpreferencemethodisrecommended(Adamowiczetal.,1998).

AttributebasedstatedchoiceThere are two types of preference methods: the stated preference method and the revealed preference method(Adamowiczetal.,1998).Instudies,whichusetherevealedpreferencemethod,choicesaremadeinrealworldsituationswhenobservingindividualbehavior.Mostlyusedforcomparingtheinfluenceofpoliciesonconsumerbehavior.Inastatedpreferencemethod,thechoicesarehypothetically(Louviereetal,2010;Sanko,2011).Thistypeofmethodwillbeusedinthisstudybecausehypotheticalsituationsofacoworkingspacewillbepresentedtotherespondentinordertocollectdata.Thecharacteristicsofcertainchoicesareindependentlyvariableofeachother;theyareorthogonal(theircorrelationiszero).Thistypeofresearchmethodisoftenusedsincethemid-1990’sinagricultureandfoodeconomic,environmentaland resource economics and health economics (Louviere et al., 2010). The following figure (figure 3.2) explains thepossibilitiesofthestatedpreferencemethod.

Figure3.2:Statedpreferencemethods(Adamowiczetal.,1998)Inthisthesis,thestatedchoicemethodnamedattributebasedstatedchoiceisused(ABSC-method).Attributebasedstatedchoicemethodsareusedwhenratingsorrankingsarewanted(Louviereetal,2010).Referendumcontingentvaluation,aform of stated choice method, is used for determining, for instance, themonetary value of environmental damages(Adamowiczetal.,1998;Kløjgaardetal.,2012).ByusingtheABSC-method,respondentshavetomakechoicesbetweenalternatives (e.g.coworkingspaceA,B,C)basedonvaryingattribute levels (e.g. location,décorandaccessibility).Thefactoruserpreferences in coworking spaces consists of six characteristics (see figure3.1). Theobjectiveof theABSC-method is to place the respondent in a particular framework to compare choices/alternatives that are described bymultipleattributes.Adamowiczetal.(1998)describedthefollowingadvantagesofanABSC-method:

• Controlofthestimuli(attributesarechosen);• Controlofthedesignmatrix(greaterstatisticalefficiency);and• Morerobustmodels(widerrangeofattributespossibletheninrealworld).

33

3.2.2MethodsfordataanalysisThethirdsub-questionof this researchwillbeansweredbasedondescriptiveanalysis.The fourth, fifthandsixthsub-question of this researchwill be answered based on a data analysis. This analysiswill bemade in chapter 5 but theexplanationof thesedataanalysismethods takesplace in this sub-section.Toanalyze thesub-questions the followingmethodswillbeused:

§ Descriptiveanalysis(chapter4)o Toanalyze theusercharacteristicsof coworkers ingeneral (sub-question3),descriptiveanalyses are

used;§ Multinomiallogitmodel(chapter5)

o To analyze what the user preferences are for coworking space characteristics (sub-question 4), amultinomiallogitmodel(chapter5)willbeestimated;

§ Latentclasslogitmodel,chi-squaretestsandindependentsamplesT-tests(chapter5)o To find different groups (classes) with similar preferences of coworking spaces characteristics (sub-

question5),alatentclasslogitmodel(chapter5)isestimated;ando Todeterminethedifferencesonusercharacteristicsbetweentheestimatedclasses(sub-question6),

multiplechi-squaretestsandindependentsamplesT-testswillbeperformed(chapter5).

DescriptiveanalysisIn the descriptive analysis, the collected data of the samplewill be described by descriptive statistics (occurrence ofvariables,mean, standarddeviationand range) andgraphics that show thedistributionof the variables.Adescriptiveanalysisshallbeappliedonlytheindependentvariables.Nodescriptiveanalysiswillbeappliedtothedependentvariablesbecauseeffectcodingisused(-1;0;1)whichwillbeexplainedinsection4.2.Fortheindependentfactor,itbecomesclearwhatthegeographiccharacteristics,demographiccharacteristicsandpsychographiccharacteristicsareoftheusersinthesample.Forinstance,whatistheirlevelofeducation,sectoroforganizationandmotivationstoworkinacoworkingspace.Thedescriptiveanalysiswilltakeplaceinchapter4.MultinomiallogitmodelChoice behavior exhibits substantial heterogeneity (Fiebig et. al, 2010). Theheterogeneity of a sample, according thepreferencesofparticularattributes,canbeofavaryingidentity(broadrange).Oftenthedependentvariabledrivesthechoiceofamodel.Whenthedependentvariable isbinary(0or1),abinary logitmodelcanbeused. Ifthedependentvariableisacontinuousvariable(forinstancebodyweight),anormallinearregressionmodelcanbeused.Thereasonamultinomial logitmodel is chosen, is that the dependent variable is a nominal variable, which consists of unorderedcategories.Itdoesnotmakesensetousealinearregressioninanominaldatasetbecausethissetconsistsofappointednumbers(1,2,3)tocertainnominaldata.Withtheaidofamultinomiallogitmodel,anestimationcanbemadeofwhichcharacteristics of a coworking space are the most preferred for coworkers. A multinomial logit model assumeshomogeneitybecauseallthedataislumpedtogether(Borgers&Vosters,2010).Totestheterogeneityamongrespondents,a latentclasslogitmodelcanbeestimated.Inthisway,adistinctioncanbemadeinusercharacteristicsrelatedtothepreferredcoworkingspacecharacteristics.Amultinomiallogitmodelestimatesutilityweightsofparameters,whichwillbesetupinsection3.3.2.Therespondents inthesampleattainacertainlevelofutilitytoacharacteristicofacoworkingspace(parameter).Theparameterwiththehighestutilityhasthelargestprobabilitytobechosen,theparameterwiththelowest utility has the lowest probability to be chosen.When adding up the utilities of all the parameters of certainattributes,thetotalutilityvalueofacoworkingspacebecomesclear.Whenpeoplechoosebetweenalternatives,theyprefertheonewiththehighestoverallutilityvalue(Hensher,Rose&Green,2015).Thefollowingequitationdeterminestheutility,byusingtherandomutilitytheory:

Uiq=Viq+Eiq=∑ßnXinq+ϵiqUiq=theoverallutilityofalternativeIforrespondentQViq=thestructuralutilityofalternativeIforrespondentQEiq=therandomutilitycomponentßn=theutilityweightofattributeNXinq=thescoreofalternativeIonattributeNforrespondentQ

34

Withtheaidoftheutilityweights,derivedfromthemultinomiallogitmodel,theprobability(PI)thatalternativeIwillbechosencanbecalculated.Thefollowingequationdeterminesthisprobability:

PI=exp(Vi)/∑jexp(Vj)PI=probabilitythatalternativeIwillbechosenExp(Vi)=stucturalutilityofalternativeI∑jExp(Vj)=sumofstucturalutilityofalternativeJ

Inordertodetermineiftheanalysisgivereliableconclusions,thegoodnessoffitiscalculated.Onlywhenthegoodnessoffitshowthatthemodelsarereliable,conclusioncanbedrawnfromtheresultsintheresearch.AccordingtoKemperman&Timmermans(2008),thegoodnessoffitcanbemeasuredwiththeaidoftwomeasurements:thelog-likelihoodoftheestimatedmodel(LL(ß))andtheMcFadden’srho-square(ρ2).AccordingtoTrain(2002),thelog-likelihoodisoftenusedwithdiscretechoicemodelstomeasurehowwellthemodelsfitthedata.The log-likelihoodfunctionoftheestimatedparameterscanbecalculatedbasedonthefollowingequation:

LL(β)=∑Nq=1∑yqiln(Pqi)

LL(β)=loglikelihoodfunctionatestimatedparameters;N=samplesize;𝑦𝑞𝑖=choiceofpersonqforalternativei.

LL(ß)isthelikelihoodfunctionoftheestimatedparameters,LL(0)isthelikelihoodfunctionwhenalltheparametersareequaltozero.LL(0) isabenchmarkforthefunctionoftheestimatedparameters(VanLaarhoven,2016).Whenall theparametersareequaltozero,thentheestimatedmodelisnobetterthannomodel.Thelowestvaluearho-squarecantakeis0andthehighestis1.Whentherho-squarevalueishigherthan0.2,themodelperformswell.Whentherho-squarevalueisabove0.1itcanbeconsideredasusable(Louviereetal.,2000).Thefollowingequationcomparesthepredictionabilitiesoftheestimatedmodelwithabasemodel(Kerkman,2010):

ρ2=1−(𝛽)/𝐿(0)

LL(β)=loglikelihoodfunctionatestimatedparameters; LL(0)=loglikelihoodfunctionatzeroparametersMc Fadden’s rho-square adjusted takes the number of parameters into account in comparison with the rho-squaremethod.Therho-squareadjustedcanbecalculatedbasedonthefollowingequation:

ρ2adjusted=1−(𝐿𝐿(𝛽)-K/𝐿𝐿(0)) K=numberofestimatedparameters.Latentclasslogitmodel,chi-squaretestsandindependentsamplesT-tests‘’Clusteranalysisistheartoffindinggroupsindata’’(Kaufman&Rousseeuw,2005,pp.1).Bymeansofclustering,structureisappliedtounstructureddata(Kaufman&Rousseeuw,2005).Inthisway,datacanbebetterandmoreeasilyinterpreted(Everitt,Landau&Leese,2001).Inaclusteranalysis,eachobject(respondent)belongstoaclusterandallclusterstogetherformalltheobjects(Everitt,Landau&Leese,2001).Inthisthesis,alatentclasslogitmodelisusedtoestimategroupsinthemultivariatecategoricaldata,theso-calledclasses.Alatentvariableisavariablethatisderivedtroughamathematicalmodel, inthiscasethroughanattributebasedstatedchoicemethod,andisnotdirectlyobservedbydirectmeasuring.WiththeuseofNLOGIT,asoftwareprogramforparameterestimation,dataisestimatedinlatentclasses(between1andthetotalnumberofrespondents inthesample). Inthecontextofthisresearch,coworkerswithmatchingpreferencesaccordingcoworkingspacecharacteristics forma latentclass.Thepreferredcharacteristics ina latentclasswillbethesamebuttheusercharacteristics inaclasswilldiffer.Oftenarespondentdoesnotbelongexactlytooneclass; this isalwaysexpressedintermsofprobability.Therespondentisassignedtotheclasswiththehighestprobability.Everyclassconsistofmultiplerespondentswithdifferentcharacteristics.‘’Oncetheutilityofeachactivityforeachlatentclassandtheprobabilityofbelongingtoalatentclassareknown,theprobabilitythatandindividualmakesacertainchoiceequalstheexpectedvalueacrosssegmentsofthesegmentspecificprobability’’(Kemperman&Timmermans,2008,pp.311).Withthefollowingequation,itcanbedeterminedwhattheprobabilityisofindividualqbelongingtoclassc:

35

𝑒𝑥(𝜃𝑐𝑧′𝑞)=Σ𝑒𝑥𝑝(𝜃𝑐𝑧′𝑞)𝑐/Σ𝑐𝑒𝑥𝑝(𝜃𝑐𝑧′𝑞)𝑃𝑞∈𝑐=theprobabilityofindividualqbelongingtoclassc;𝜃𝑐=thevectorofutilityweightsbelongingtocharacteristicszspecifiedforclassc;𝑧′𝑞=vectorofobservedindividualcharacteristicsofindividualq.

AccordingtoTrain(2002),therho-squarestatisticscanonlybeusedforcomparingnestedmodels.Forcomparingnon-nestedmodelswiththesamedataset, likethemultinomial logitmodelandthe latentclass logitmodel,AICandBICiswidelyused.ThemodelswiththelowestAkaikeInformationCriterion(AIC)andBayesianInformationCriterion(BIC)valuesrepresentthedatabestandcontainsthemostrepresented information.TheAICandtheBICarecalculatedasfollows(Akaike,1973;Schwarz,1978ascitedinNijënstijn,2012):

AIC=-2*LL(ß)–K/NBIC=-2*LL(ß)+K*LN(N)/N

‘’Eachcriterionreasonsfromanothertheoreticalperspective:BICreasonsfromthepointofviewthatoneofthemodelsunderconsiderationistrue.BICtriestoidentifythemodelwiththehighestprobabilityofbeingthetruemodel.AICdoesnotassumeatruemodel,butlooksatthepredictivepowerofthemodelstoselectthemostadequatemodel.AICandBICcangivedifferentconclusionsindifferentconditions’’(Nijënstijn,2012,pp.28).Forclarity,groupsareformedbasedonpreferencesandnotbasedonusercharacteristics.Theformofthegroup(e.g.gender,education,sectororganization)isunknownandshouldbeanalyzedbyperformingmultiplechi-squaretestsandindependentsamplesT-tests.InanindependentsamplesT-test,twogroupsofdataarecompared.Crosstablesbetweenmultipleuser characteristics and their estimated class canbeestablishedwith achi-square test. In theend, it canbedeterminedwhetherpreferencesofcoworkingspacescharacteristicsarelinkedtodifferentusercharacteristics. Achi-squaretestisbasedonthefollowingequation:

χ2=Σ(Oi-Ei)2/ΣEi χ2=chi-square Oi=observedfrequencies Ei=expectedfrequencies

3.3OperationalizationoftheindependentanddependentvariablesData for the independent factor (user characteristics) and dependent factor (user preferences for coworking spacecharacteristics)arecollectedbytheuseofanonlinequestionnaire.Aquestionnaireconsistsofaseriesofquestionstogaininformation about the variables. In order to collect information about the independent factor, operationalization ofgeographic,demographicandpsychographiccharacteristicsisnecessary.Operationalizationistheconversionofvariablesintoameasureableform(Baarda&DeGoede,2006).Thelevelofmeasurement(nominal,ordinal,interval,andratio)andthetypeofitem(howitismeasured;e.g.multiplechoicescaleoropenquestion)willdescribethevariable.3.3.1IndependentfactorInordertocreateapictureoftheusersofcoworkingspaces;geographic,demographicandpsychographiccharacteristicsareused.EachusergroupcontainsagreatvarietyofuserswithadifferentsetofusercharacteristicsandnopreviousstudyhasinvestigatedwhatthesecharacteristicsareinTheNetherlands.Thedifferentusercharacteristics,whicharepresentedinthissub-section,arebasedonexistingliteraturethatismentionedinsection2.6.ThecompletecodebookispresentedinAppendixB.Table3.1:Geographicvariables(coworkingspacerelatedcharacteristics)

Theoperationalizationofthegeographicvariablecountrycoworkingspacehappenedbasedonamultiple-choicequestionwith64options(partisshownintable3.1).Thisquestionnaireisdistributedinseveralcountries.Allthepossiblecountries,wherethequestionnaireisdistributed,areincludedinthislist.Atagivenmoment,thislistisextendedwith12countries.Thesecountriesareshownafter52(option:‘other...’).

Geographicvariable Levelofmeasurement Typeofquestion TypeofitemCountrycoworkingspace Nominal Multiplechoice(64) 1;TheNetherlands,2;Afghanistan,3;America….

36

Table3.2:Demographicvariables(personalcharacteristics)

Inthisstudy,acoworkerwillbetypifiedamongotherthingsbasedonfourpersonalcharacteristicsnamely:gender,age,nationality and education (shown in table 3.2). These four demographic characteristics draw a picture whether therespondentisamaleorafemale,howoldhe/sheis,whathis/hernationalityisandwhathis/herhighesteducationlevelis.The levelofeducationdifferspercountry; thedifferencesarereflected inthequestionnaires (seeAppendixC).Theabove-describeddemographicvariablesareclusteredatonepageinthequestionnaireregarding‘personalcharacteristics’.

Thefollowingfourdemographicvariablesarealsoclusteredatonepageinthequestionnaireunderthetitle‘work-relatedcharacteristics’(shownintable3.3).Table3.3:Demographicvariables(work-relatedcharacteristics)

Coworkingspacesoriginatedasresponseto‘loneeagles’,whichwereinneedofathirdplacetoworkduetoboredomandisolation (Moriset, 2013). The first user group consists of these lone eagles in the form of self-employed workers,freelancers or entrepreneurs. The linebetween theseuser groups is quite thin. ‘’A freelancer is almost always a self-employedworker,butaself-employedworkerisnotalwaysafreelancer’’(DeGroot,VanDijk&Pijlman,N.D.).Thereisdefinitelyadifferencebetweentheseusergroups,butconsideringthatthisdifferenceisrathervague,theself-employedworker,freelancerandentrepreneurarecombinedinoneusergroup.InthestudyofDeskmag(2016),adistinctionismadebetween entrepreneurs and freelancers. They form both a separate user group, which is confusing. According theliterature, small firms and large firms make also use of coworking spaces. The size of these companies is based oncategoriesderivedfromthestudyofHartog(2015),namely:2-10employees(2),11-50employees(3)andmorethan50employees(4).Deijl(2011)andKojo&Nenonen(2014)statedthatstudentsmakealsouseofacoworkingspace,thereforethisusergroupisincludedaswell.Theusergroupofextendedworkersisdeletedbecausethisisnotaconcreteusergroup.

Anotherwork-relatedcharacteristicistheusers’positionintheorganization.Theoperationalizationisbasedonamultiple-choicequestionwithfivepossibleanswers,namely:supportingstaff (1), regularemployee(2),manager(3),board/owner(4)ordoesnotapply(5).‘Doesnotapply’(5)applies,forexample,forafreelancerorastudent.Theseusergroupshavenospecificpositioninanorganization.

The itemsof thedemographiccharacteristicssectororganizationarederived fromprior researchofDeskmag(2016).Deskmagstudiedtheprofessionsofcoworkersin2013-2014and2015-2016.

Thelastdemographiccharacteristicsrelatedtotheworkcharacteristicsisincome.ThesedifferentlevelsofincomearederivedfromthestudyofHartog(2015).Theoption(6)‘Idon’tknow/I’drathernotsay’,isaddedtothislistforthoserespondentsthatdonotwhattosharethatpieceofinformation(e.g.privacy).

Thefollowingtwodemographicvariablesareclusteredatonepageinthequestionnaireunderthetitle‘coworkingspacecharacteristics’. Thevariablecountry coworking spacealsobelongs to thisgroupofquestion,but since this is ageographiccharacteristic,itisalreadymentioned.

Demographicvariables Levelofmeasurement Typeofquestion TypeofitemGender Nominal Multiplechoice(2) 1;Male,2;FemaleAge Ratio OpenquestionNationality Nominal Multiplechoice(64) 1;TheNetherlands,2;Afghanistan,3;America….Education Ordinal Multiplechoice(9) 1;Noeducation/elementaryschool,2;Preparatorysecondary

vocationaleducation,3;Seniorgeneralsecondaryeducation,4;Pre-universityeducation,5;Intermediatevocationaleduction,6;Highervocationaleducation,7;University(bachelor),8;University(master),9;University(PhD)

Demographicvariables Levelofmeasurement Typeofquestion TypeofitemUsergroup Nominal Multiplechoice(5) 1;Self-employedworker,freelancerorentrepreneur,2;Employee

ofacompany(2-10employees),3;Employeeofacompany(11-50

employees),4;Employeeofacompany(morethan50employees),

5;Student

Positioninorganization Nominal Multiplechoice(5) 1;Supportingstaff(deskattendant,receptionistetc.),2;Regular

employee,3;Manager,4;Board/owner,5;Doesnotapply

Sectororganization Nominal Multiplechoice(12) 1;Consultancy,2;Design,3;Commerce,4;IT,5;Art,6;

Management,7;Research,8;Education,9;Projectmanagement,

10;PR,marketing,sales,advertising,communication,11;Writing,

12;Other

Income Ordinal Multiplechoice(6) 1;Lessthan20000ayear,2;20001-30000ayear,3;30001-500001

ayear,4;40001-50000ayear,5;Morethan50000ayear,6;I

don'tknow/I'drathernotsay

37

Table3.4:Demographicvariables(coworkingspacerelatedcharacteristics)

Therespondenthastofillintheaveragehoursofworkingperweekinacoworkingspace,thisisanopenquestion.Thetransportthatismostlyusedfortravelingtothecoworkingspaceisderivedfromamultiple-choicequestion(shownintable3.4).Possibleanswerstothisquestionarebycar(1),bybike(2),byfoot(3)orbypublictransport(4). Thevariablespersonalityandbenefits,mentionedinsection2.6,areclearlynotthefocusofthisstudysotheywillnot be included in the questionnaire. The corresponding question of the psychographic characteristics is about therespondents’motivationtoworkinacoworkingspace(seetable3.5).Therankingquestionofferstenpossiblemotivationswhereintherespondentshavetochoosetheirthreetopmotivationsinorderofimportance(1;mostimportmotivation,3;leastimportant).Themotivationsarebasedontheoutputasshownintable2.2.Thefollowingpsychographicvariableismentionedaswellunderthetitle‘coworkingspacecharacteristics’.Table3.5:Psychographiccharacteristics

Thegeographic,demographicandpsychographiccharacteristicsarenowoperationalizedinrespectivelyone,tenandthreevariables.Inthefollowingsub-section,thedependentfactorwillbeoperationalized.3.3.2DependentfactorSanko(2011)developedaguidelinefortheexperimentaldesignofanABSC-method.Inthestatisticaldesign,attributesandattributes’levelsarecombinedinordertocreatealternativesandchoicegames(see figure 3.3). The attributes can be seen as coworking spacecharacteristics(variables)andtheattributes’levelsasavariationonthatspecificcharacteristic.TheoperationalizationofthedependentvariableisestablishedbasedontheframeworkofAdamowiczetal.(1998). To establish an ABSC-method, the following steps arenecessary:

• Step1:identifyrelevantattributes;• Step2:attributelevelselection;• Step3:experimentaldesigndevelopment;• Step4:questionnairedevelopment;• Step5:samplesizinganddatacollection(happensinchapter4);and• Step6:modelestimation(happensinchapter5).

Step1:IdentifyrelevantattributesGiventhatthedesignistoolargewhenallthetypicalcharacteristicsofcoworkingspaces,asshownintable2.4,areapplied,a selection of the characteristics ismade. A design of nine attributes is deemed feasible (Adamowicz et al., 1998). Ifrespondentshavetomakeachoicebasedonmorethannineattributes, it isconsideredasnotpossiblebecauseittoomuchtooversee.Itisdifficulttosummarizealltheinputcharacteristicsinjustnineattributes.Forthatreason,aselectionhasbeenmadeinthevariables.Intotal,eightattributesarederived.Eachcharacteristicintheconceptualmodel(figure3.1)isrepresented;sixintotalandtwocharacteristicsarementionedtwice.Theattributehastomeettwocriteria’s;itmustbeatypicalcharacteristicofacoworkingspace(derivedfromtable2.4)andithastovaryonthreeattributelevels

Demographicvariables Levelofmeasurement Typeofquestion TypeofitemHoursincoworkingspace Ratio OpenquestionTransportcoworkingspace Nominal Multiplechoice(4) 1;Car,2;Bike,3;Byfoot,4;Publictransport

Psychographicvariables Levelofmeasurement Typeofquestion TypeofitemMotivation1,2and3 Nominal Rankingquestion(10) 1;Affordableaccommodation,2;Thefeelingofbeingpartofa

community,3;Vibrantandcreativeatmosphereinthecoworking

space4;Professionalappearanceforthecompany,5;Professional

supportiveworkenvironment(supportservices),6;The

opportunitytonetworkwithcoworkers(possiblenewprojects),7;

Socialinteractionwithcoworkers,8;Thepossibilityforwork-

relatedconversationswithothercoworkers(knowledgesharing,

knowledgecreation),9;Iwaslookingforaworkplaceoutsidethe

home(separatingworkandprivatelife),10;Flexibility(rental

period,numberofsquaremeters)

Figure3.3:StatisticaldesignABSC-method(Sanko,2011)

38

(step2).Thefollowingeightattributesarearranged(betweenthebracketsthevariableismentionedtowhichtheattributebelongsaccordingtheconceptualmodelinfigure3.1):

• Attribute1:accessibilityofthelocation(location);• Attribute2:atmosphereandinterioraesthetics(officedécor);• Attribute3:layoutofthespace(officeexterioranddivision);• Attribute4:diversityinsupplyspaces(officeexterioranddivision);• Attribute5:receptionandhospitality(facilitiesandservices);• Attribute6:events(communityandsustainability);• Attribute7:diversityoftenants(accessibility);and• Attribute8:typeofleasecontract(accessibility).

Themost important typical characteristics are selected in order that the alternatives represent the best hypotheticalsituationofacoworkingspace.Thefirstattribute,theaccessibilityofthelocation,isanimportantcharacteristicbecausethisoftendeterminesthechoiceofacertaincoworkingspaceaccordingtothestudyofCapdevila(2013).Locationisingeneral themost importantcharacteristic inrealestate.Attributetwo,atmosphereand interioraesthetics, is includedbecausemanycoworkingspaceshaveaspecificappearancethatcouldbeofimportancewhyacoworkingspaceischosen.Thereisgreatattentiontointeriordesignofcoworkingspacesandisconsideredasoneoftheimportantfactorsofthecreativeandenergeticatmosphereinacoworkingspace(Fuzi,2015).Thelayoutofacoworkingspaceistestedinattribute3.Somecoworkerswouldratherworkinanopenareawhereotherspreferaclosedlayout.Attribute4,diversityinsupplyspaces,hasbeentakenintoaccountbecausethisoftenvariesinacoworkingspace.Fromabasiccoworkingspacewithjust a collaborative workspace to a fully equipped coworking space, with for example a fitness center and bar. Thecharacteristic facilitiesandservices isreflected intheattributereceptionandhospitality.AccordingtoFuzi (2015), thecoworkinghostplaysanimportantroleincreatinginteractionandcollaboration.Theextent,towhichauserattachvaluetothisaspect,isexpressedinthatattribute.Inacoworkingspace,thereisoftenacreativeandvibrantatmospherewherevariouskindsofeventsareorganized.Whetherornotauserpreferstheseeventscomesforwardinthesixthattribute.Attribute7(diversityoftenants)andattribute8(typeofleasecontract)arecoveredbytheaccessibilityofthecoworkingspace.Therearecoworkingspaceswithagreatdiversityoftenantsinvarioussectorsbutalsocoworkingspacesthat,forexample,arefocusedonthesectordesign.Accordingtomultiplestudies(Deijl,2011;Spinuzzi,2012;Fuzietal.,2014;Sykes,2014)aflexiblecontractisatypicalcoworkingspacecharacteristic.Aworkplaceoraspacecanberentedinseveralterms.Alltheeightattributesvaryonthreeattributelevels,whichwillbeexplainedinthenextstep.Step2:AttributelevelselectionThe attributes’ levels are specific for one attribute. The levels indicate a proper range of variation for each attribute(Adamowiczetal.,1998).Thefollowingattributelevelsareidentified:Accessibilityofthelocation

• Attributelevel0:bycarandpublictransport;• Attributelevel1:bycar;and• Attributelevel2:bypublictransport.

Atmosphereandinterioraesthetics

• Attributelevel0:industrial;• Attributelevel1:modern;and• Attributelevel2:homey.

Layoutofthespace

• Attributelevel0:openlayout(largeopenspaces);• Attributelevel1:half-openlayout(combinationofopenspacesandconcentrationrooms);and• Attributelevel2:closedlayout(enclosedandseparatespaces).

Diversityinsupplyspaces

• Attributelevel0:basiccoworkingspace(collaborativeworkspace+meetingroomsandkitchenarea);• Attribute level1:standardcoworkingspace(collaborativeworkspace+meetingrooms+kitchenarea+event

spacesandinformalzones);and• Attributelevel2:premiumcoworkingspace(collaborativeworkspace+meetingrooms+kitchenarea+event

spaces+informalzones+fitnesscenterandbar).

39

Receptionandhospitality• Attributelevel0:noreceptionandnohost;• Attributelevel1:receptionbutnohost;and• Attributelevel2:receptionandactivehost(activecoworkinghostthatconnectscoworkerstoeachother).

Events

• Attributelevel0:none;• Attributelevel1:sometimes;and• Attributelevel2:often.

Diversityoftenants

• Attributelevel0:nodiversityoftenants(tenantsinthesamesector);• Attributelevel1:moderatediversityoftenants(tenantsinafewbusinessfields);and• Attributelevel2:strongdiversityoftenants(alotofdifferentbusinessfieldspresentinthecoworkingspace).

Typeofleasecontract

• Attributelevel0:nocontract;• Attributelevel1:shortterm(dayorweekormonth);and• Attributelevel2:longterm(yearorlonger).

Step3:Experimentaldesigndevelopment‘’Adesignisasampleofprofiles(combinationsofdifferentattributes)whichhaveaparticularsetofstatisticalpropertiesthatdeterminestheutilityspecification(s)thatcanbeestimated(i.e.identified)’’(Adamowiczetal.,1998,pp.13).Therearetwotypesofdesigns,whichcombineattributesandattributelevelstocreatealternatives,namely:fullfactorialdesignsandfractionalfactorialdesigns.Inafullfactorialdesign,everypossiblecombinationofattributelevelsisexpressed.Thedifferenteffectsbetweenattributes(main-effects)andthecorrespondingeffectsbetweenattributes(interactioneffects)arebothtakenintoaccount.Thiscreatesmanyalternatives(38=6.561profiles)whichisabigtaskfortherespondent,soafullfactorialdesignisaseriousdrawbackandonlypracticalforsmallproblems(Sanko,2011).Eightattributeswiththreecorrespondinglevelswillgivemanycombinations(notasmallproblemanymore),soafullfactorialdesignisconsiderednotfeasibleasadesign.Inafractionalfactorialdesign,alltheinteractionsbetweenattributesareignored,whichgiveslessscenarios.Onlythemaineffectsaretakenintoaccount.Inafractionalfactorialdesign,theattributesareindependentofeachother.Instatisticalterms,thismeansthattheattributesareorthogonalanddonotcorrelate.Theexperimentaldesign,asshownintable3.6,consistsof27alternatives.These27alternativesarepresentedinninechoicesets(N)withfourchoices/alternativesperchoiceset.Eachalternativeconsistsofeightattributes(characteristics),asmentionedinstep1,andeachattributecanvaryonthreeattributelevels,asmentionedinstep2.Thenumbersinthetablearecorrespondingtotheattributelevelsofstep2.Table3.6:Experimentaldesign

N Attribute1 Attribute2 Attribute3 Attribute4 Attribute5 Attribute6 Attribute7 Attribute81 1 0 0 0 0 0 0 0 02 1 0 0 1 1 1 2 1 23 1 0 0 2 2 2 1 2 14 2 0 1 0 0 1 1 2 25 2 0 1 1 1 2 0 0 16 2 0 1 2 2 0 2 1 07 3 0 2 0 0 2 2 1 18 3 0 2 1 1 0 1 2 09 3 0 2 2 2 1 0 0 210 4 1 0 0 1 1 1 1 111 4 1 0 1 2 2 0 2 012 4 1 0 2 0 0 2 0 213 5 1 1 0 1 2 2 0 014 5 1 1 1 2 0 1 1 215 5 1 1 2 0 1 0 2 116 6 1 2 0 1 0 0 2 217 6 1 2 1 2 1 2 0 118 6 1 2 2 0 2 1 1 019 7 2 0 0 2 2 2 2 220 7 2 0 1 0 0 1 0 121 7 2 0 2 1 1 0 1 022 8 2 1 0 2 0 0 1 123 8 2 1 1 0 1 2 2 024 8 2 1 2 1 2 1 0 225 9 2 2 0 2 1 1 0 026 9 2 2 1 0 2 0 1 227 9 2 2 2 1 0 2 2 1

40

Step4:QuestionnairedevelopmentBoththeindependentvariablesanddependentvariablesareincludedinthequestionnairesystem;thefirst13questionsareabouttheindependentfactor;thelast9questionsareaboutthedependentfactor.

Intable3.7,theexperimentaldesignoftable3.6iscompletedwiththeattributelevelsasmentionedinstep2.Eachrowintable3.5equals’onechoice/alternative,whichisalsoknownas,profiles(27rows;27alternatives;27profiles).Table3.7:Experimentaldesignwithcorrespondingattributelevels

Thequestionnaireconsistsofninechoicesets.Eachchoicesetconsistsofthreerandomlychosenalternatives(alternative1,2 and 3) and one alternative ‘none of these option’which reflects the respondent ratherwouldwork at home (orsomewhereelse)thanatthecoworkingspace.Threeversionsofchoicesetsaredevelopedinorderthatnotalwaysthesamealternativesareplottedagainsteachother; thiswill causeordereffects. Thereare threepossibleordereffects:questionordereffect(ordereffectofthedifferentchoiceoptions),ordereffectwithinthechoiceoptionsandtheattributeordereffect (ordereffectof thealternativesandattributes) (Chrzan,1994).Therespondentgets thechoicesets fromeitherversion1,2or3.Thishappensrandomly inthequestionnairesystem.Thefollowingtable (table3.8)showstheversion,theparticularchoicesetandthealternatives(1,2and3).Thenumbersinthealternativesarecorrespondingtoaprofilefromtable3.7.Forinstance,inversion2,inthe1thchoiceset,alternative1(seetable3.8)iscorrespondingtorow16intable3.7,alternative2iscorrespondingtorow10intable3.7andalterative3iscorrespondingtorow21intable3.7.Anexampleofthischoicesetisalsoshownintable3.9.Intheonlinequestionnaire,allthealternatives(1,2,3and4;noneoftheoptions)areplottednexttoeachotherwhereallthechoices/alternativesarepresentedvertically.Table3.8:Version,choicesetandalternatives

Alternative Attribute1 Attribute2 Attribute3 Attribute4 Attribute5 Attribute6 Attribute7 Attribute81 Bypublictransportandcar Industrial Openlayout Basiccoworkingspace Noreceptionandnohost None Nodiversityoftenants Nocontract2 Bypublictransportandcar Industrial Half-openlayout Standardcoworkingspace Receptionbutnohost Often Moderatediversityoftenants Longterm(yearorlonger)3 Bypublictransportandcar Industrial Closedlayout Premiumcoworkingspace Receptionandactivehost Sometimes Strongdiversityoftenants Shortterm(dayorweekormonth)4 Bypublictransportandcar Modern Openlayout Basiccoworkingspace Receptionbutnohost Sometimes Strongdiversityoftenants Longterm(yearorlonger)5 Bypublictransportandcar Modern Half-openlayout Standardcoworkingspace Receptionandactivehost None Nodiversityoftenants Shortterm(dayorweekormonth)6 Bypublictransportandcar Modern Closedlayout Premiumcoworkingspace Noreceptionandnohost Often Moderatediversityoftenants Nocontract7 Bypublictransportandcar Homey Openlayout Basiccoworkingspace Receptionandactivehost Often Moderatediversityoftenants Shortterm(dayorweekormonth)8 Bypublictransportandcar Homey Half-openlayout Standardcoworkingspace Noreceptionandnohost Sometimes Strongdiversityoftenants Nocontract9 Bypublictransportandcar Homey Closedlayout Premiumcoworkingspace Receptionbutnohost None Nodiversityoftenants Longterm(yearorlonger)10 Bycar Industrial Openlayout Standardcoworkingspace Receptionbutnohost Sometimes Moderatediversityoftenants Shortterm(dayorweekormonth)11 Bycar Industrial Half-openlayout Premiumcoworkingspace Receptionandactivehost None Strongdiversityoftenants Nocontract12 Bycar Industrial Closedlayout Basiccoworkingspace Noreceptionandnohost Often Nodiversityoftenants Longterm(yearorlonger)13 Bycar Modern Openlayout Standardcoworkingspace Receptionandactivehost Often Nodiversityoftenants Nocontract14 Bycar Modern Half-openlayout Premiumcoworkingspace Noreceptionandnohost Sometimes Moderatediversityoftenants Longterm(yearorlonger)15 Bycar Modern Closedlayout Basiccoworkingspace Receptionbutnohost None Strongdiversityoftenants Shortterm(dayorweekormonth)16 Bycar Homey Openlayout Standardcoworkingspace Noreceptionandnohost None Strongdiversityoftenants Longterm(yearorlonger)17 Bycar Homey Half-openlayout Premiumcoworkingspace Receptionbutnohost Often Nodiversityoftenants Shortterm(dayorweekormonth)18 Bycar Homey Closedlayout Basiccoworkingspace Receptionandactivehost Sometimes Moderatediversityoftenants Nocontract19 Bypublictransport Industrial Openlayout Premiumcoworkingspace Receptionandactivehost Often Strongdiversityoftenants Longterm(yearorlonger)20 Bypublictransport Industrial Half-openlayout Basiccoworkingspace Noreceptionandnohost Sometimes Nodiversityoftenants Shortterm(dayorweekormonth)21 Bypublictransport Industrial Closedlayout Standardcoworkingspace Receptionbutnohost None Moderatediversityoftenants Nocontract22 Bypublictransport Modern Openlayout Premiumcoworkingspace Noreceptionandnohost None Moderatediversityoftenants Shortterm(dayorweekormonth)23 Bypublictransport Modern Half-openlayout Basiccoworkingspace Receptionbutnohost Often Strongdiversityoftenants Nocontract24 Bypublictransport Modern Closedlayout Standardcoworkingspace Receptionandactivehost Sometimes Nodiversityoftenants Longterm(yearorlonger)25 Bypublictransport Homey Openlayout Premiumcoworkingspace Receptionbutnohost Sometimes Nodiversityoftenants Nocontract26 Bypublictransport Homey Half-openlayout Basiccoworkingspace Receptionandactivehost None Moderatediversityoftenants Longterm(yearorlonger)27 Bypublictransport Homey Closedlayout Standardcoworkingspace Noreceptionandnohost Often Strongdiversityoftenants Shortterm(dayorweekormonth)

Version Choiceset Alternative1 Alternative2 Alternative31 1 8 21 101 2 6 26 171 3 22 27 11 4 7 25 161 5 12 15 51 6 9 3 231 7 4 18 241 8 2 13 111 9 20 19 142 1 16 10 212 2 8 9 12 3 7 22 182 4 13 3 202 5 17 15 62 6 23 27 122 7 11 14 52 8 19 25 22 9 26 24 43 1 9 10 193 2 21 13 143 3 7 5 203 4 3 23 273 5 6 4 163 6 26 24 173 7 25 8 13 8 22 18 253 9 2 12 11

41

Table3.9:Examplechoiceset(version2,choiceset1,alternative16,10and21)

3.4QuestionnairedistributionanddescriptionofthepopulationThequestionnaire isdevelopedasanonlinequestionnaire.Thequestionnaire is translated intofour languages(Dutch,English,GermanandItalian)andalltherespondentscanchoosetheirpreferredlanguage.ThecompletequestionnairesareenclosedinAppendixC.ThequestionnairewillbedistributedwiththehelpofpartnersfromdifferentuniversitiesandbymultiplevisitstocoworkingspacesintheNetherlands.Severalacademies(PolitechnicodiMilano,Italy;AaltoUniversityHelsinki, Filand; The Barlett School of Architecture Londen, England; Schweizer Hochschulen für AngewandteWissenschaftenZürich,Zwitserland)willhelptospreadtheonlinequestionnaireinforeigncountriesthroughamessagewithanintroductiontextandrelatedlinks(e.g.websiteorviaamessagebythecoworkingoperator).ForthedistributionintheNetherlands,multiplecoworkingspaceswillbevisitedtohandoverthequestionnaire.Anonlinequestionnairewillbesendtothecoworkersthatarewillingtocooperateinthestudytopreferredcharacteristics.Therespondentscanfillinthequestionnairewhenitsuitsthem,consideringtime,placeandinterest.Abroadrangeofcoworkingspaceswillbepartofthisresearch.Forinstance,largecoworkingspacesandsmallcoworkingspaces,coworkingspaceswithonelocationandcoworkingspaceswithmorelocations,coworkingspacesnearbythehighwayandcoworkingspacesnearbythecentralstationetc.Inthisway,avariedsampleofisobtainedisobtained.DescriptionofthepopulationApartofthetotalpopulationofcoworkerswill formthesampleofthisresearch,theso-calledsample-basedresearch(Baarda&DeGoede,2006).Theresearchunitinthisstudyis‘theuserofacoworkingspace’.Everyonewhoispresentinacoworkingspaceisseenasauserofacoworkingspaceandiscoveredbytheresearchunit.Twostudies(Deskmag,2013;Deskmag,2016)drawconclusionsontheusercharacteristicsonagloballevel.Noscientificstudiesaredone,neitheronthe composition of coworkers as on the preferences of coworking in specific countries. Response from the foreigncountriesmustbeofsimilarmagnitudeofthecollecteddataintheNetherlandsinordertomakeacomparison.Ifthisisnotthecase,theforeigncountrieswillbeexcludedfromtheanalysisandonlytheDutchcaseswillbeincluded.Sincethereareonlystudiestocoworkingonagloballevel,thesamplewillbecomparedtothesestudies(Deskmag,2013;Deskmag,2016)andtostudiesonthenationalworkforce.

3.5ReliabilityofthestudyReliabilityandvalidityformstogetherthecoreofwhatisacceptedasscientificproof(Shuttleworth,2008).Ifascientificstudy is reliable, researchersmustbeable to repeat thestudyandgenerate thesameresults. ‘’Statistically significantresultsarethosethatareinterpretednotlikelytohaveoccurredpurelybychanceandtherebyhaveotherunderlyingcausesfortheiroccurrence’’(Kalla,2009,pp.1).Inordertorepeatthestudy,theresultsofthestudymustbesignificantandnotbasedonchance.

In the first part of the questionnaire, respondents have to choose either (nine times) theirmost preferablealternativeofcoworkingspaces,alternative1,2,3or4‘noneoftheseoptions’.Thispartofthequestionnaireissubjectedtoajudgmentoftherespondent,whichcanbeeffectedbytheircurrentmood,timeandplace.Whentherespondentisinanothermood,timeand/orplace,theanswersaccordingtheirpreferredcoworkingspacecharacteristicsmaydiffer.Thisinfluencesthereliabilityofthisstudybutnotmorethanincomparablestudiessothismaybeneglected.Ontheotherhand,thequestionrisesabouthowreliableanattributebasedstatedchoicemethodis.AccordingtoDanieles&Rotaris(2014),thereliabilityofanABSC-methodisrelatedtothediscrepancybetweenrealchoicesandthestatedones.Alltheattributesthatareusedarederivedfromthe literature;thestatedchoices(attributes)areconsistentwiththechoices(attributes) intherealworld.Besidesthat,multipleversionsofthechoicesetsaremadetoavoidordereffects,whichinfluencesthereliabilityofthemethodaswell(Danieles&Rotaris,2014).

The second part of the questionnaire about the personal characteristics, work-related characteristics andcoworkingspacecharacteristicsisnoteffectedbymood,timeandplace,becausethesearefactualanswers.Theanswerscannotdifferpermoment,whichbenefitsthereliabilityoftheresearch.Thestudytopreferredcharacteristicsofcoworkingspacesfordifferentusergroupscanbeconsideredasreliable.

Attribute Alternative1 Alternative2 Alternative3 NoneoftheseoptionsAccessibility Bycar Bycar BypublictransportAtmosphereandinterioraesthetics Homey Industrial IndustrialLayoutofthespace Openlayout Openlayout ClosedlayoutDiversityinsupplyspace Standardcoworkingspace Standardcoworkingspace StandardcoworkingspaceReceptionandhospitality Noreceptionandnohost Receptionbutnohost ReceptionbutnohostEvents None Sometimes NoneDiversityoftenants Strongdiversityoftenants Moderatediversityoftenants ModeratediversityoftenantsTypeofleasecontract Longterm(yearorlonger) Shortterm(dayorweekormonth) Nocontract

YourChoice O X O O

42

3.6ValidityofthestudyAccordingtoShuttleworth(2008),avalidstudymeasuresthedatathatisaimedtobemeasured.Validitycomesintwodifferentforms;externalvalidityandinternalvalidity.Externalvalidityisaboutwhethertheresultsarerepresentativeforthepopulationorgeneralizableforothersituations.Internalvalidityisaboutwhetheritactuallyismeasuredwhatintendedtobemeasured.

Theinternalvalidityregardingthisthesismeansthattheaspectsaremeasuredcorrectly.Wouldtherespondentsinreallifealsomakethesamechoice?Theindependentanddependentvariablesandcorrespondingcharacteristicsandattributesaremeasuredintwodifferentways.Theindependentvariables(geographic,demographicandpsychographiccharacteristics)aremeasured insuchway, that thequestionsarenot influencingthedependentvariable.The internalvalidity can be influenced by so called confounding, where the research group is influenced by the questions of theindependentvariable.ThequestionsabouttheusercharacteristicsofthecoworkerareaskedaftertheABSC-method.Inthis case, the independentvariablesdonot influence thedependentvariables. If thiswas theotherwayaround (firstquestionsabout the independentvariables), respondentsperhapsmadedifferentchoices in thechoicesets,basedonpreviousanswers.Forinstance,thelastquestioninthequestionnaireisaboutmotivationstogotoacoworkingspace.Ifthisquestionwasaskedpriortothechoicesets,maybethisinfluencedthechoiceinthechoiceset.Theintroductionofthequestionnairestatesthatallthequestionsareindependent.Previouschoiceshavenoeffectonthecontinuationofthequestionnaire, which improves the internal validity. The dependent variables (preferred characteristics of coworkingspaces) aremeasured by using an ABSC-method. The research design of this method is based on the framework ofAdamowiczetal.(1998).Tocometothespecificdesign,thequestionnaireisdevelopedbasedonfoursteps(seesection3.3.2).Theinternalvalidityofthismethodisacceptable.

Theexternalvalidityofastudydependsonfouraspects:thecharacteroftheresearch(sampleselectiontype),thesamplesize,theresponserateandthestandarddeviation.Thesefouraspectsinfluencewhetherthesamplecanbegeneralizedtothepopulation.ThecoworkingspacesintheNetherlandsandtheforeigncountriesareselectedbasedonpermissionofthecoworkingspaceoperatortoreachoutthequestionnairetothecoworkers.IntheNetherlands,allthecoworkingspaceshavebeenapproached.Alistofcoworkingspacesiscomposedwiththeaidofwebsitesthatfocusesoncoworking.Intheforeigncountries,thepartnersfromtheuniversitiesselectedmultiplecoworkingspacestospreadthequestionnaire.Theselectionofthecoworkingspacesisaselectiveprocedureaswellastheselectionofthecoworkers,thisdidnotoccurrandomly.Thesamplesizeofthisresearchdependsonthepresenceofcoworkers inthecoworkingspaces.Thespaceswillbevisitedmultipletimestoensurethatthesamplesizeisvaried.AlargesampleandahighresponserateseemsanachievablegoalbecausemultiplecoworkingspacesinTheNetherlandandabroadhaveagreedtocooperatein thedatacollectionof thissurvey.Thesizeof thesampleandtheamountof theresponsewillbereferredto in thefollowingchapter.Thissurveycanbeconsideredasexternalvalid.Theconclusionsofthisstudywillbeinterpretableforothercoworkingspaces.

3.7ConclusionData for this researchwill be collected by using an online-questionnaire. In the online-questionnaire,multiple surveyquestionswillcollectdatafortheindependentfactorusercharacteristics.Withtheaidofanattributebasedstatedchoicemethod,dataiscollectedforthedependentfactoruserpreferences.Inthischapter,allthevariablesareoperationalizedandadescriptionisgivenaboutthedistributionofthequestionnaire.Themethodscanbeconsideredasreliableandvalidonaninternalandexternallevel.Theoperationalizationoftheusercharacteristicsisdividedintogeographicalcharacteristics,demographicalcharacteristicsand psychographic characteristics. The operationalization led to 13 variables in a measurable form (one geographicvariable,10demographicvariablesandonepsychographicvariable).ThedependentvariablehasbeenoperationalizedwithreferencetotheframeworkofAdamowiczet.al.(1998).Eightattributeshavebeendrawnup(accessibilityofthelocation,atmosphereand interioraesthetics, layoutof thespace,diversity in supply spaces, receptionandhospitality,events, diversity of tenants and the type of lease contract) and each attribute varies on three attribute levels,whichindicatesaproperrangeofvariationforeachattribute.Theattributescanbeseenascoworkingspacecharacteristics,theattributeslevelsasavariationonthatspecificcharacteristic.The questionnaire will be spread in multiple coworking spaces in the Netherlands and with the support of severaluniversities (Politechnico diMilano, Italy; AaltoUniversity Helsinki, Filand; The Barlett School of Architecture Londen,England; Schweizer Hochschulen für Angewandte Wissenschaften Zürich, Zwitserland) the questionnaire will bedisseminatedamonguserofcoworkingspacesabroad.

43

Inthischapter,theresearchdesignisdevelopedforthedatacollectionofthismaster’sthesis.Inthefollowingchapter,adescriptionisgivenofthecollecteddata.Thethirdsub-questionabout‘’Whataretheusercharacteristicsofcoworkersingeneral’’willbeanswered.

44

Chapter4:DataDescription Inthefourthchapterofthisresearch,adescriptionofthedataisgiven.Adescriptionofhow,whenandwherethedataisconductedisdiscussedinsection4.1.Moreinformationabouttheresponserateofthequestionnaireisalsoexpressedinthissection.InordertotransformthedataintoasuitableformforthestatisticalanalysisprogramSPSSandparameterestimationprogramNLOGIT5,thedatahastobeprepared.Abriefsummaryofthedatapreparationisexpressedinsection4.2. Insection4.3,anoverviewoftheusercharacteristics incoworkingspaces isgiven.The independentvariablesareanalyzedbymeansofdescriptivestatistics(occurrenceofvariables,mean,standarddeviationandrange)andgraphicsthatshow the distribution of the variables. In this chapter, the third sub-question ‘’What are the user characteristics ofcoworkersingeneral?’’willbeanswered.Finally,theconclusionofchapter4isdrawnupinsection4.4.

4.1DatacollectionThedataforthisresearchiscollectedfrom17thofOctober2016untilthe14thofNovember2016.AtthebeginningofOctober 2016, an e-mail was send to the operators of coworking spaces throughout the Netherlands with a shortintroduction/explanationoftheresearchandtherelevanceforoperators.Inthise-mail,operatorswereaskediftheywouldcooperateinthestudytopreferredcoworkingspacecharacteristics.Duringthedistributionofthequestionnaire,itbecameclearthatpersonalapproachingthepotentialrespondentswasthemosteffectivewayofgatheringdata,eventhoughthequestionnairewassentbye-mail.Duringthecorrespondencewithcoworkingspaceoperatorswasaskedifitwaspossibletovisitthecoworkingspacespersonallytohandoverthequestionnairetothepresentcoworkers.Ifthiswasnotpossible,thecoworkingspaceoperatorcouldsendthequestionnairetotheircoworkersbythemselves.Intotal,66coworkingspacesintheNetherlandswereapproachedofwhich25coworkingspaceswerewillingtocooperateintheresearch.Sixteenofthese25coworkingspaceswerevisitedtospreadthequestionnaireandinninecoworkingspaces,thecoworkingspaceoperatordistributedthequestionnaire.InAppendixD,anoverviewispresentedofalltheapproachedcoworkingspaces.Alltherespondentsreceivedande-mailwithashortintroductiontextandcorrespondinglinkofthequestionnaireinfourlanguages(Dutch,English,ItalianandGerman).Theresponseofthecompletedquestionnairesis219intotal,whichcanbeconsideredasthesampleofthisresearch.Inthefollowingsectioncomesforwardhowthissampleisestablished.Insection4.3,thesamplewillbeclarifiedwithdescriptivestatistics.

4.2DatapreparationDatapreparationistheprocessofcollecting,cleaningandconsolidatingdataintoonefileordatatablefordataanalysis.Thegoalofthedatapreparationiscreatingadatafilethatcanbedirectlyenteredintotheanalysisandestimationsoftware.Inordertoclarifytheprocessinthechapter,itisofimportancetomakeadistinctionbetweendatacollectionmethodsanddataanalysismethods.Therelevantmethodswereearliermentionedinsection3.2.

Tocollectthedataforboththeindependentanddependentvariablesanonlinequestionnaireisused.Togatherdatafortheindependentvariables,surveyquestionsaboutgeographic,demographicandpsychographiccharacteristicsareasked.Togatherdataforthedependentvariables,thedifferentstepsoftheattributebasedstatedchoicemethodaspresentedinsection3.3.2arefollowed.

Toanalyzethedataforboththeusercharacteristicsandpreferredcoworkingspacecharacteristics,twosoftwareprograms are used. By using SPSS, software for statistical analysis, descriptive statistics (frequency tables, descriptivestatics tables and graphics) are made of the independent and dependent variables. By using NLOGIT, software forparameterestimation,amultinomiallogitmodelandalatentclasslogitmodelwillbeestimatedtoexaminerespectivelythepreferencesaccordingthecharacteristicsofacoworkingspaceandtofindgroupsinthemultivariatecategoricaldata.With theuseofmultiplechi-square testand independent samples T-tests, the relationbetween the independent anddependentvariablesisexamined.

With the aid of these methods, respectively the fourth sub-question (‘’What are the user preferences forcoworking space characteristics?’’), the fifth sub-question (‘’Can different user groups be identified based on theirpreferences?’’) and the sixth sub-question (‘’What are the differences between these user groups based on usercharacteristics)willbeanswered.Thistakesplaceinthefollowingchapter.Forboththeanalyticalsoftwareprograms,thedatasethavetobetransformedinordertomakeitsuitableforfurtheranalysis.Firstthedataoftheindependentvariablewillbepreparedandthenthedependentvariable.DatapreparationoftheindependentvariablesofusercharacteristicsForthedatapreparationoftheindependentvariables,dataisdeleted,missingvaluesaredefinedandmultiplevariablesarerecoded.ThedatahastobepreparedforstatisticalanalysisinSPSS.Thefollowingactionsonthedataarecarriedout:

45

• Deletedcountries.Inthefirstattempt,261respondentsfilledinthequestionnaire.CoworkersfromcoworkingspacesinAmerica(1),Belgium(1),England(3),Finland(3),Italy(32),Poland(1)andSwitzerland(1)completedthe questionnaire. The response from America, Belgium, England, Finland, Poland and Switzerland can beconsideredastoolow.Forthisreason,these10caseswillnotbeincludedinfurtheranalysis.The32casesinItalyarealsoexcludedoutofthesample.Thesecasesprovidea‘contaminated’samplebecauseitispossiblethatthesecasesarespecial(forinstance;averyheterogeneousgroup)whichwillbeaveragedifthetwocountriesareaddedtogether. Ifbothcountrieshadthesameorderofmagnitudethanthetwocountrieswere included.Thetotalnumberofrespondents(N)is219andconsistsonlyofcoworkersofDutchcoworkingspaces.

• Missing values. Thedemographic variablehoursworking in coworking spaces and thepsychographic variablemotivations forusinga coworking spaceshave respectively sixand fivemissingvalues.Thesevaluesarenowdefinedasthenumber‘99’sincethisnumberdoesnotappearinthedatafile.Wheninformationaboutthesevariablesisrequiredintheanalysis,thesespecificcaseswillbeexcluded.

• Recodedvariables.Thedemographicvariablesage,education,usergroups,incomeandhoursincoworkingspacearerecodedintodifferentvariables.Recodingintoadifferentvariabletransformsanoriginalvariableintoanewvariableinwhichvaluesaregroupedtogether.Thenextsectionshowsgraduallywhichvariablesarerecodedandwhytheyarerecoded.Therearethreereasonsforrecodingvariables.Thefirstreasonisthatabetterandclearerviewofthevariablecanbecreatedwhencategoriesareapplied.Secondreasonisthatvariablesthathasaskeweddistributioncanapproachanormaldistributionbyrecoding.Thelastreasontorecodevariablesisthatvariablesshouldbeapproximatelyofthesameorderofmagnitude.Whenordinalgroupsaretoosmall,thesegroupshavetomergewithanothergroup.

• Recodedvariables.Thelastquestionintheonlinequestionnaireisaboutthemotivationstoworkinacoworking

space. This is a psychographic variable. The ranking question offers ten possible motivations where therespondentshavetochoosethreemotivations inorderof importance(1;most importantmotivation,3; leastimportant).Inordertoanalyzewhatthemostchosenmotivationis,allthemotivations(motivation1,2and3)are added together in the descriptive analysis. Regarding the data analysis, all the three motivations areconsideredseparately.

• Recodedvalues.Thevariablesectoroftheorganizationincludedtheoption‘other,namely...’inthequestionnaire.When the remaining listed options did not fit with the profession of the coworkers, they could fill in theirprofessionontheblankform.Twenty-threerespondentsusedtheblankformtofillintheirprofession.Thelistofprofession,deductedfromtheresearchofDeskmag(2016),hadabroad interpretationframework.All the23professions are nonetheless assigned to one of the listed professions because it turned out that all the 23professionsfitunderalistedsector.Forinstance,arespondentmentioned‘texteditor’,whichcanincludedinthevalue‘writing(11)’.Anotherrespondentmentioned‘travelagent’whichcanbeincludedinthevalue‘commerce(3)’.

DatapreparationofthedependentvariablesofpreferredcoworkingspacecharacteristicsThedataofthedependentvariableshastobepreparedfortheanalysisinNLOGIT.Thefollowingactiononthedataarecarriedout:

• Recodedpreferences.Inthequestionnaire,therespondentshadtomarktheirpreferredcoworkingspaceoutoffour alternatives (alternative1, alternative2, alternative3andalternative4 ‘noneof theseoption’).Whenarespondent selected, for instance, alternative 3, the recoded preference is 0010. In this way, the selectedalternatives(alternativeswithrecodedpreference1)canbefiltered.

• RecodedvaluesforNLOGIT.Allthevariablesofdependentfactorwerealreadycodedinthequestionnairesystem,

which provided the right quantitative data for the analysis. The eight attributes with the corresponding 24attributelevelshavetoberecodedinordertomakeitsuitabledataforNLOGIT.Thefollowingencryptionsaregiventotheattributelevelsinordertorecode,thisprocessiscalledeffectcoding:

o Attributelevel0:10;o Attributelevel1:01;ando Attributelevel2:-1-1.

46

Forinstance,theattributeeventsconsistofthreeattributelevels.Attributelevel0(none)isrecodedintheeffectcoding1and0.Attributelevel1(sometimes)isrecodedintheeffectcoding0and1.Attributelevel2(often)isrecodedintheeffectcoding-1and-1whichisalsothereferencegroupofeachattribute.Withtheuseofthisscheme,theutilityvaluesknownasparameters(ß)foreachattributelevelcanbeestimated.Ineffectcoding,onlyminus one, zero and one are used to convey all of the necessary information on groupmembership. Also aconstantparameterisestimated,whichisequaltothegrandmeanofalltheobservations.Theparametersareequaltodifferencebetweenthemeanoftheattributeandthegrandmeanofalltheobservations.Thecompleteschemeofeffectcodingoftheattributelevelsisshownintable4.1

Table4.1:Effectcodingattributelevels

4.3DatadescriptionoftheusercharacteristicsSection4.3showsthedatadescriptionoftheusercharacteristics.Insection3.4wasmentionedthattheresponsefromforeigncountrieshavetobeofsimilarmagnitudeofthecollecteddataintheNetherlandsinordertomakeacomparison.Thisisnotthecase,soallthedataoftheforeigncountriesisexcludedforfurtheranalyses.Thissampleconsistsof219coworkers from the Netherlands. Two studies (Deskmag, 2013; Deskmag, 2016) draw conclusions on the followingindependentvariablesonagloballevel:

• Gender(Deskmag,2016);• Usergroups(Deskmag,2013);• Sectoroforganization(Deskmag,2016);• Transporttocoworkingspace(Deskmag,2016);and• Motivationstoworkinacoworkingspace(Deskmag,2016).

TherearenoscientificstudiesonthecompositionofcoworkersintheNetherlands,whichisoneofthereasonswhythisresearchwillberelevant.Togetsomeindication,thedataiscomparedtotheresultsoftheglobalstudiesoncoworkingandtostudiesaboutthegeneralworkforceintheNetherlands.The data description is done on the basis of frequency tables and descriptive figures. The independent factor usercharacteristicsconsistsofthreecharacteristics,withbetweenthebracketsthecorrespondingvariables,namely:

• Geographiccharacteristics(countryofcoworkingspace).Allthecoworkersinthisstudy(219;100%)areworkingfromacoworkingspaceintheNetherlands.Duetothe lowresponseinAmerica(1),Belgium(1),England(3),Finland(3), Italy(32),Poland(1)andSwitzerland(1),thecharacteristicgeographiccharacteristicshasbecomeunnecessary.Forreasonsofcompleteness,thischaracteristicremainslistedintheconceptualmodel.

• Demographiccharacteristics(gender,age,nationality,education,usergroup,positioninorganization,sectoroforganization,income,hoursincoworkingspace,transporttocoworkingspace);

• Psychographiccharacteristics(motivationstoworkinacoworkingspace).ThecompleteconceptualmodeloftheindependentvariableisshowninAppendixE.Thismodelgivesaclearoverviewoftheinvolvedcharacteristicsandcorrespondingvariables.

Attribute Attributelevel Variablename Variablevalue Variablename VariablevalueBycarandpublictransport Aloc 1 Bloc 0Bycar Aloc 0 Bloc 1Bypublictransport Aloc -1 Bloc -1Industrial Asfeer 1 Bsfeer 0Modern Asfeer 0 Bsfeer 1Homey Asfeer -1 Bsfeer -1Openlayout Alayou 1 Blayou 0Halfopenlayout Alayou 0 Blayou 1Closedlayout Alayou -1 Blayou -1Basiccoworkingspace Adiv_r 1 Bdiv_r 0Standardcoworkingspace Adiv_r 0 Bdiv_r 1Premiumcoworkingspace Adiv_r -1 Bdiv_r -1Noreceptionandnohost Arec 1 Brec 0Receptionbutnohost Arec 0 Brec 1Receptionandactivehost Arec -1 Brec -1None Aeven 1 Beven 0Sometimes Aeven 0 Beven 1Often Aeven -1 Beven -1Nodiversityoftenants Adiv_h 1 Bdiv_h 0Moderatediversityoftenants Adiv_h 0 Bdiv_h 1Strongdiversityoftenants Adiv_h -1 Bdiv_h -1Nocontract Atypco 1 Btypco 0Shortterm Atypco 0 Btypco 1Longterm Atypco -1 Btypco -1

Diversityoftenants

Typeofleasecontract

Accessibilityoflocation

Atmosphereandintrioraesthetics

Layoutofthespace

Diversityinsupplyspaces

Receptionandhospitality

Events

47

GeographiccharacteristicsIntotal219Dutchrespondentsfilledinthequestionnaire.Thegoaloftheresearchistoexamineifdifferentusergroupscanbeidentifiedbasedontheirpreferredcoworkingspacecharacteristics.Sincethequestionnaire,infirstattempt,wouldbespreadinmanycountries,theexactcityisnottakenintoaccountbecausetoomanycities(allovertheworld)wouldbementioned.DemographiccharacteristicsTendemographiccharacteristicsareanalyzedinthequestionnairenamely:

• Gender;• Age;• Nationality;• Education;• Usergroup;• Positioninorganization;• Sectoroforganization;• Income;• Transporttoacoworkingspace;• Hoursworkinginacoworkingspace;• Transport.

GenderThesampleof219respondentsconsistedof68%malesand32%females(seefigure4.1). According to The Fact Club (N.D.), 55% of the total work population of theNetherlandsismaleand45%female,whichindicatesthatthedistributionofgenderinthesampleslightlydiffersfromtheworkpopulation.AccordingtoDeskmag(2016),theamountofwomanwhogocoworkingincreasedovertheyears.In2016,38%ofthecoworkerswasafemale,whichisquiteinlinewiththeamount32%ofthisstudy.AgeWitharangeof57years,theyoungestcoworkerinthesamplewas18yearsoldandtheoldestwas75yearsold.Figure4.2showsthattheageincoworkingspacesisnotnormallydistributed.Theaverageageofacoworker(mean)is34.61withadispersionaroundtheaverage(standarddeviation)of11.2years.Thefocusofthedistributionisclearlyunderthemean.Thevariableageisaratiovariablewithnocategories.Thevariableageisrecodedinthevariableagegroupstocreatecategories,whichprovideacleareroverviewofthevariable.Theratiovariableagewillbeusedintheanalysis.Figure4.3showsthedispersionofthisvariable.Table4.2providesafrequencytableofthevariableagegroups.

NationalityThe number of nationalities is quite diverse, although in absolute numbers the Dutch (203) nationality is still bestrepresented.OthercountiesrepresentedinthecoworkingspacesareAfghanistan(1),America(1),Denmark(1),Germany(2), France (2), Hungary (1), Italy (1), Iran (1), Morocco (1), Romania (1), Turkey (2), South Africa (1) and 2 remainnationalitiesthatnotarerepresentedinAppendixB.

Figure4.1:Genderofcoworker

Figure4.2:Distributionvariableage Figure4.3:Distributionvariableagegroups

Frequency Percent CumulativePercent24yearoryounger 35 16,0 16,025-34year 86 39,3 55,335-44year 55 25,1 80,445yearorolder 43 19,6 100,0Total 219 100,0

Table4.2:Frequencytableagegroups

48

EducationThenominalvariableeducationconsistsofeightcategories(initiallyninebutthevariable‘’noeducation(option1)’’hasnotbeenchosenonce).Inordertocreateavariablewiththesameorderofmagnitude,thevariableeducationisrecodedinto the variable education groups. Originally, this variable consists of three group, namely: low level of education(preparatory secondary vocational education), medium level of education (senior general secondary education, pre-university education and intermediate vocational education) and high level of education (senior general secondaryeducation,bacheloratuniversity,masteratuniversityandDoctorofPhilosophy(PhD)atuniversity(CBS,2013)).Thereasontorecodethevariableeducationtoeducationgroupsistocreatelesscategories.Sincethenumberofcoworkerswithalowlevelofeducationisstronglyunderrepresented(1.2%),thislevelisenclosedintheeducationlevelmediumlevelofeducation.Thetwoeducationgroupsarenow:low-mediumlevelofeducationandhighlevelofeducation.Fromthe219respondents, 32 coworkers (13.9%)havea low-medium level of education and187 coworkers (86.1%) ahigh level ofeducation.Afrequencytableofeducationvariablesispresentedintable4.3.Figure4.4showsthedistributionofthehighlevelofeducation(highlightedinyellowintable4.3).Ingeneral,28%ofthetotalpopulationintheNetherlandsishighlyeducated(CBS,2013).Comparedtothehighlyeducatedgroupinthesample,whichcovers86.1%,thereisanextremelysignificantdifference.Anexplanation for thispercentagemaybe thatofficeworkers tend tobehighereducated thanpeoplewhoworkinforinstanceafactory.

UsergroupTherearemultipleusergroupsrepresentedinacoworkingspace.Inthesecondchapterofthisresearch,anoverviewisgivenoftheusergroupsincoworkingspacesaccordingtheliterature.Coworkingisoftenassociatedwithfreelancersandself-employedworkers(Parinno,2015).Thisgroup,ofthebyMoriset(2013)called‘loneeagles’, isthemostdominantgroupinthesamplewithashareof53%.Thenominalvariableusergroupconsistsoffivepossiblecategoriesandisrecodedintothevariableusergroup-groupswithfourcategories:self-employedworkers,freelancersorentrepreneurs,employeeofacompany(2-10employees),employeeofacompany(11ormoreemployees)andstudents.Thereasontorecodethevariableusergroupsistocreatevariableswiththesameorderofmagnitude.Employeesofacompanyalsoappeartobewellrepresentedwithatotalofnear35%.Inthesampleofthisresearch,38coworkersrepresentacompanywith2-10employees,andalso38with11ormoreemployees.Twenty-sixofthetotalrespondents(12%)werestudent,whichisinlinewiththeresearchofDeijl(2011)andKojo&Nenonen(2014)whostatethatthecoworkingspaceisalsoaspottoworkonschoolassignments.Thenumberofusersinausergroupisreflectedincrosstable4.4.

AccordingtotheglobalresearchofDeskmag(2016),44%ofthecoworkersisaself-employedworker,freelanceror entrepreneur, 51% is an employee of a company and 5% is of another user group. The self-employed workers,freelancersandentrepreneursarewith53%morerepresentedintheNetherlandsandthenthe44%globally(Deskmag,2016).IntheNetherlands,theemployeesofacompanyhaveashareofalmost35%,whichalsodifferfromthe44%globally.Asufficientpartinthesamplearethestudentswith12%.StudentsarenotrecognizedasausergroupintheresearchofDeskmag(2016).SomecoworkingspacesintheNetherlands(Seats2Meet)areopenforstudents;itmaybepossiblethatthisisnotthecaseinothercoworkingspacesworldwide.

PositioninorganizationOfthe219coworkersintotalinthissample,sixrespondentsaresupportingstaff,49respondentsareregularemployees,17coworkersaremanager,93respondentsare inpositionofboard/ownerand54respondentsdoesnotapply to thequestion.Itisstrikingthatthepercentageofrespondentswithaboard/ownerpositionis42%(seetable4.4),whichisquitehigh.Intable4.4,acrosstab,whichcrossesthevariablespositionintheorganizationontheverticalaxisandthevariableusergrouponthehorizontalaxis,ispresented.Thecrosstabshowsthat81ofthe117self-employedworkers,freelancersorentrepreneursoccupiesaboard/ownerpositionintheorganization.Thismeansthatalmost70%ofthisusergroupownsanorganization.Thecoworkersthatfilledindoesnotapply,aremostlyself-employedworkers,freelancers,entrepreneurs

Frequency Percent CumulativePercentPreparatorysecondaryvocationaleducation

3 1,2 1,2

Seniorgeneralsecondaryeducation 18 7,6 8,8Pre-universityeducation 5 2,8 11,6Intermediatevocationaleducation 6 2,4 13,9Highervocationaleducation 79 32,3 46,2University(bachelor) 19 12,7 59,0University(master) 81 37,8 96,8University(PhD) 8 3,2 100,0Total 219 100,0

Table4.3:Frequencytablevariableeducationgroups

Figure4.4:Shareofhighlevelofeducation

49

orstudents.Thismakessenseforstudentsbecausetheyhavesimplynopositioninanorganization.Italsomakessensefor theusergroupofself-employedworkers, freelancersandentrepreneursbecausemostof thetime it isaone-manorganizationwithoutspecificpositions.

SectoroforganizationThedemographiccharacteristicsectoroftheorganizationissomewhatdiverse.Thesectorsarederivedoutofpreviousresearch of Deskmag (2016). The sectors consultancy (25%), IT (21%) and PR / marketing / sales / advertising /communication(11%)havethelargestshareinthecoworkingspace.Intable4.5,theresultsofthisresearcharecomparedtotheresultsoftheDeskmagresearch.Thesecondcolumnofthetableshowstheoccurrenceofthevariableinthesample.Theshareofthevariableinthetotaloccurrenceispresentedinthethirdcolumn.TheresultsoftheDeskmagresearcharepresentedinthefourthcolumnandthedifferencesbetweenthetwostudiesinthefifthcolumn.TheresultsofthisresearcharequiteinlinewiththestudyofDeskmag(2016).

IncomeThedistributionofrespondentscouldalsobedescribedbytheirincome.Thevariable‘income’isrecodedinthevariableincomegroups,tocreateanewvariablewhichconsistsof3incomegroupsandacategory‘Idonotknow/Irathernotsay’(seetable4.6).Thereasontorecodethevariableincomeistocreatevariableswiththesameorderofmagnitude.Twenty-sevenpercentofthecoworkersearnslessthen€20.000ayear.Themajorityoftherespondents(36%)earnbetweenthe€20.001and€50.000peryearand14%ofthecoworkersearns€50.000ormoreayear.IntheNetherlands,themodalnetannualincomeis€24.500ayearin2015.The27%ofthecoworkerswhoearnslessthan€20.000ayearconsistsfor55%ofself-employedworkers, freelancersorentrepreneursandfor20%ofstudents (seecrosstab4.6).Therelativelyhighpercentageoflowincomecanbededicatedtobothusergroups.Whatisstrikingisthat,althoughthesurveywasconductedanonymously,still23%oftherespondentsdonotknoworrathernotsharetheirlevelofincome.

Self-employedworker,freelanceror

entrepreneur %

Employeeofacompany(2-10employees) %

Employeeofacompany(11ormoreemployees) % Student % Total Percent

Supportingstaff 2 1% 3 1% 1 0% 0 0% 6 3%Regularemployee 3 1% 20 9% 22 10% 4 2% 49 22%Manager 3 1% 2 1% 12 5% 0 0% 17 8%Board/owner 81 37% 9 4% 3 1% 0 0% 93 42%Doesnotapply 28 13% 4 2% 0 0% 22 10% 54 25%

117 53% 38 17% 38 17% 26 12% 219 100%

Position

Usergroups

Total

Table4.4:Crosstabpositionandusergroup

Frequencysample Percentage Deskmag(2016) DifferenceConsultancy 55 25% 15% 10,0%Design 17 8% 7% 1,0%Commerce 15 6% 3% 3,0%IT 46 21% 27% -6,0%Art 8 4% 3% 1,0%Management 4 2% 5% -3,0%Research 11 5% 4% 1,0%Education 19 9% 5% 4,0%Projectmanagement 11 5% 4% 1,0%PR,marketing,sales,advertising,communication 25 11% 5% 6,0%Writing 8 4% 5% -1,0%Other x x 17% xTotal 251 100,0% 100%

Self-employedworker,freelancerorentrepreneur

Employeeofacompany(2-10employees)

Employeeofacompany(11ormoreemployees) Student Percent

Lessthan20.000ayear 33 11 4 12 60 27%20.001-50.000ayear 47 14 15 2 78 36%Morethan50.000ayear 15 4 12 0 31 14%Idon'tknow/I'drathernotsay 22 9 7 12 50 23%

117 38 38 26 219 100%53% 17% 17% 12% 100%Percent

Usergroup

TotalIncomegroups

Total

Table4.6:Crosstabincomeandusergroup

Table4.5:FrequencytablesectorandcomparisontoDeskmag(2016)

50

TransporttocoworkingspaceFigure4.5showsanoverview,whichtransportthecoworkersusetotraveltothecoworkingspace.The largest sharegoesbycar (51%), followedbybike(22%)andbypublictransport(21%).Just13ofthe219(6%)coworkerswalkto the coworking space. In comparison to the global research of Deskmag(2016),thebiggestpart(51%)goesbycar,22%bybike,18%byfootand8%bypublictransport.Thesampleofthisresearchis,accordingtothesharethatgoesbycarandbybike,quite in linewiththeglobalresults.ThesharethatgoesbypublictransportisbiggerintheNetherlandsthanglobal.Causeforthismay be that everything in the Netherland is relatively easily accessible bypublictransportincomparisontotherestofthecoworkingspacesworldwide.HoursworkingincoworkingspaceTheamountoftimeworkingaweekinthecoworkingspaceisshownintable4.7.Theratiovariableisrecodedintoanewvariablethatreflectsthehoursinaworkday.Whatisstrikingisthat28.8%ofthesampleworks0to8hoursonaverageaweek.Acoworkingspaceisprobablyusedinthiswayasatemporaryworkspace.13.7%ofthesampleworksonaverageafullworkingweekinthecoworkingspace.Theratiovariableofhoursworking inacoworkingspacewillbeusedintheanalyses.Table4.7:Hoursincoworkingspace

Finally,intable4.8,thevariablesgender,usergroup,sectorandwayoftransporttothecoworkingspacearecomparedtotheresultofthestudiesofDeskmag(Deskmag,2013;Deskmag,2016).Thevariablesgender,usergroupandsectoroforganizationarequite in linewitheachother. The variable transport to the coworking space is quitedifferent in TheNetherlandscompared to the restof theworld.Anoverviewof theothervariables (agegroups,nationalities, levelofeducation,positioninorganization,incomeandhoursincoworkingspace)ispresentedintable4.9.

Frequency Percent CumulativePercent0-8hours/week 63 28,8 28,88-16hours/week 33 15,1 43,916-24hours/week 35 16,0 59,924-32hours/week 40 18,3 78,232-40hours/week 30 13,7 91,9Morethan40hours/week 13 5,9 97,8Total 214 97,8 100

Missing 5 2,2219 100,0

Valid

Total

Figure4.5:Transporttocoworkingspace

Table4.8:ComparisonresultstotheDeskmagstudies Table4.9:Overviewresults Percent

24yearoryounger 16%25-34year 39.3%35-44year 24.1%45yearorolder 19.8%

TheNetherlands 93%Othercountries 7%

Low-mediumlevelofeducation 14%Highlevelofeducation 86%

Supportingstaff 3%Regularemployee 22%Manager 8%Board/owner 42%Doesnotapply 25%

Lessthan20.000ayear 27%20.001-50.000ayear 36%Morethan50.000ayear 14%Idon'tknow/I'drathernotsay 23%

0-8hours/week 28.8%8-16hours/week 15.1%16-24hours/week 16.0%24-32hours/week 18.3%32-40hours/week 13.7%Morethan40hours/week 5.9%

Hoursincoworkingspace

Income

AgeGroups

Nationalities

Levelofeducation

Positioninorganization

Percent Deskmag

Male 65% 62%Female 35% 38%

Self-employedworker,freelancerorentrepreneur

53% 44%Employeeofacompany 35% 51%Student 12% xOther x 50%

Consultancy 25% 15%Design 8% 7%Commerce 6% 3%IT 21% 27%Art 4% 3%Management 2% 5%Research 5% 4%Education 9% 5%Projectmanagement 5% 4%PR,marketing,sales,advertising,communication

11% 5%Writing 4% 5%Other x 17%

Car 26% 51%Bike 47% 22%Byfoot 6% 19%Publictransport 19% 8%

Transporttocoworkingspace

Gender

Usergroup

Sectoroforganization

51

PsychographiccharacteristicsThelastusercharacteristicsofthefactor‘usercharacteristics’aremotivationstouseacoworkingspace(psychographiccharacteristics).AccordingtoAhmad(2003),benefitssoughtinaparticularproductisawaytosegmentusersaccordingpsychographicdimensions.Thelastquestioninthequestionnaireisaboutrespondents’motivationtoworkinacoworkingspace.Thisrankingquestionoffers10possiblemotivations.Inorderofimportance(1;mostimportmotivation,3;leastimportant),thecoworkerhadtoselecttheirmotivation.Toshowthedifferencesbetweenthethreemotivations,astackedbarchartispresentedinfigure4.6.Inthisstackedbarchart,allthethreechoiceoptionsareaddedtogetherwhichshowstheshareperrankofthetotalmotivation.Asanexample,motivation1:75respondentschosemotivations1astheirfirstmotivation,39respondentschoseitastheirsecondmotivationand28respondentsastheirthirdmotivation.

1. Iwaslookingforaworkplaceoutsidethehome(separatingworkandprivatelife).2. Vibrantandcreativeatmosphereinthecoworkingspace;3. Affordableaccommodation;4. Socialinteractionwithcoworkers;5. Theopportunitytonetworkwithcoworkers(possiblenewprojects);6. Thepossibilitytowork-relatedconversationswithothercoworkers(knowledgesharing,knowledgecreation);7. Thefeelingofbeingpartofacommunity;8. Flexibility(rentalperiod,numberofsquaremeters);9. Professionalappearanceforthecompany;and10. Professionalsupportiveworkenvironment(supportservices);

Motivation1:’Iwaslookingforaworkplaceoutsidethehome(separatingworkandprivatelife)’isoverallthemostchosenmotivationtogotoacoworkingspace.Seventy-fiverespondents(34.2%)chosethismotivationastheirmainmotivation.AccordingtoMoriset(2013),thefrontiersbetweenprofessionallivesandprivatelivesdisappearwhenworkingfromhome.Theywanttobreaklonelinessandescapefromboredom.Abetterwork/lifebalancecanbeachievedwhenworkingfromsomewhereelse.Fuzietal. (2014)mentionedthatworkersdecidedtogotoacoworkingspacebecausetheybecameexhaustedofthedistractionathome.Themainmotivationaccordingtherespondentsofthequestionnaireisinlinewiththemotivationsintheliteratureoncoworking.

Coworking spaces are creative, social andenergetic places (Fuzi, 2015;Orel, 2015). The secondmost chosenmotivationoverallisthemotivationnumber2:‘vibrantandcreativeatmosphereinthecoworkingspace’withashareof18.7%.Besidesthat,thismotivationisthemostchosensecondmotivationanalsothemostchosenthirdmotivation.

Affordableaccommodationiswith11.7%thethirdmost importantmotivationoveralltocometoacoworkingspace.AccordingtoSykes(2014)aretherentpricesareoftenlowincoworkingspaces.Whatmighthaveinfluencedtherespondentswhopickthismotivationistheproviderofthecoworkingspace.Seats2Meetprovidesaworkspacefreeinexchangefor‘socialcapital’inwhichthecoworkersharestheirknowledge.

Thetopthreemotivationsarecloselyfollowedbythemotivations ‘social interactionwithcoworkers’and‘theopportunitytonetworkwithcoworkers’.Bothmotivationsprovideinformationabouttheinteractionbetweencoworkers.Theliterature(Leforestier,2009;Deijl,2011;Spinuzzi,2012;Moriset,2013)showsthatthis isanimportantmotivationwhichisreflectedinthesampleofthisresearch.

75

38 2815 10 12 23

6 4 8

39

46

2425

22 1814

14 8 8

28

39

2528

22 21 1116

16 12

0

20

40

60

80

100

120

140

160

1 2 3 4 5 6 7 8 9 10

thirdmotivation

secondmotivation

firstmotivation

Figure4.6:Stackedbarchartmotivations

52

4.4ConclusionInthissection,aconclusionisdrawnofthefourthchapteraboutthedatadescriptionofthisresearch.Thefollowingsub-questionisexaminedinthischapter:‘’Whataretheusercharacteristicsofcoworkersingeneral?’’Twenty-fivecoworkingspaceswerewillingtocooperate inthestudytopreferredcoworkingspacecharacteristics.Theresponseofthecompletedquestionnaires in219intotal,whichareallrespondents(coworkers)thatworks inaDutchcoworkingspace.SincetherearenoscientificstudiesonthecompositionofcoworkersintheNetherlands,itisnotpossibletogeneralize thesampleof this study to thepopulationofall thecoworkers in theNetherlands.For that reason, twostudies(Deskmag,2013;Deskmag,2016)thatfocusesonthecompositionofcoworkersonagloballevel(accordinggender,usergroups,sectoroforganization,transporttocoworkingspaceandmotivationstoworkinacoworkingspace)gaveanindicationofthecompositionworldwide.Inordertogiveanswertothesub-question‘’whataretheusercharacteristicsofcoworkers in general?’’, a cross-sectional view is presented of the geographic, demographic and psychographiccharacteristics.Basedonasampleof219respondentscanbeconcluded,withappropriatecaution,thatthecoworkersintheNetherlands:

Ø Aremostlymen(68%)–similartotheworldpopulation(Deskmag,2013);Ø Areonaverage35yearsold;Ø Arehighlyeducated(86%)(highervocationaleducationoruniversity(bachelor,mastersorPhD));Ø Areself-employedworkers,freelancerorentrepreneurs(53%)–similartotheworldpopulation(Deskmag,2016);Ø Havemostlyaboard/ownerposition(42%)–similartotheworldpopulation(Deskmag,2016);Ø Earnsbetweenthe€20.001and€50.000peryear(36%)Ø Areactiveinthesectorconsultancy(25%)andIT(21%)–similartotheworldpopulation(Deskmag,2016);Ø Aregoingtothecoworkingspacebybike(47%)–notsimilartotheworldpopulation(Deskmag,2016);Ø Areworkinglessthan24hours(60%)perweekinthecoworkingspace;andØ Aregoingtothecoworkingspacebecausetheywanttoseparateworkandprivatelife(21.6%),wanttoworkina

vibrantandcreativeatmosphere(18.7%)andareinsearchofaffordableaccommodation(11.7%).Whatthepreferredcharacteristicsofcoworkingspacesareandwhetherdifferentgroupscanbedefinedinthepreferredcharacteristicswillbeestimatedwiththeaidofmultiplemodelsinthefollowingchapter.

53

54

Chapter5:DataAnalysisandResultsIn this chapter, the data analysis will be carried out and the results will be presented. In the previous chapter, thedescriptivestatisticsoftheusercharacteristicsofcoworkersintheNetherlandsaregiven.Thefourthsub-questionisasfollows:‘’whataretheuserpreferencesforcoworkingspacecharacteristics?’’.Thissub-questionwillbeansweredbasedonamultinomiallogitmodelinsection5.1.Thismethodanalysestheresultsoftheattributebasedstatedchoicemethodofthedependentvariable.Thefifthsub-questionisasfollows: ‘’Candifferentusergroupsbeidentifiedbasedontheirpreferences?’’.Inordertofindcategoriesinthemultivariatedataofthepreferencesofcoworkingspacecharacteristics,alatentclasslogitmodelisestimatedinsection5.2.Withtheaidofthismethod,multipleclasseswithequalpreferenceswillbeestimated.Toanswerthesixthsub-question‘’Whatarethedifferencesbetweentheseusergroupsbasedonusercharacteristics?’’ will be answered with the aid of multiple chi-square tests and independent samples T-tests. Theconclusionofthedataanalysisandtheresultsaredrawnupinsection5.3.

5.1Preferredcoworkingspacecharacteristics(multinomiallogitmodel)Withtheaidofanattributebasedstatedchoicemethod,dataforthemultinomiallogitmodeliscollected.Withtheaidofthat model, the fourth sub-question ‘’what are the user preferences for coworking space characteristics?’’ can beanswered.Thefollowingattributesweresubmittedtotherespondents:

• Accessibilityofthelocation;• Atmosphereandinterioraesthetics;• Layoutofthespace;• Diversityinsupplyspaces;• Receptionandhospitality;• Events;• Diversityoftenants;and• Typeofleasecontract.

ThecompleteconceptualmodelofthedependentvariableisshowninAppendixF.The following descriptive statistics can be presented according to the data collectionmethodology of the dependentvariables:

• 219respondentshadtochoose9timestheirpreferredalternative;• 1.971(219times9)choicesetsarepresentedto219respondentswhichisequaltothenumberofobservations;• Eachchoicesetconsistof4choices/alternatives(alternative1,alternative2,alternative3analternative4‘’none

oftheseoptions);• Eachrespondenthadtoface36choices/alternatives(9times4);• Intotal7.884(2259times4)choices/alternativeswerepresented;• In394ofthe1971(20%)choicesets,therespondentselectedtheoption‘noneoftheseoption’(alternative4),

whichrepresentsthatthecoworkerratherwouldworkathome(orsomewhereelse)thaninacoworkingspaces;• In 1.577 (1.971minus 394) (80%) of the choice sets, the respondent selected one of the other alternatives

(alternative1,alternative2oralternative3).Theoverallpreferencesofcoworkingspacecharacteristics,withoutconsideringusercharacteristics,canbedeterminedwiththeaidofamultinomiallogitmodel.First,thegoodnessoffitanddescriptivestatisticsofthemodelwillbedeterminedandthereaftertheutilityvaluesandsignificanceoftheparameters(coworkingspacecharacteristics).GoodnessoffitInordertodetermineifthemultinomiallogitmodelperformswell,thegoodnessoffithastobecalculated.Theunderlyingequationforthismethodispresentedinsection3.2.2.Theloglikelihoodfunctionoftheestimatedparametersis-2474.425whichcanbefoundinAppendixG(greenhighlighted)andreflectsthefunctionofasetofparameters.Theloglikelihoodfunctionatthezeroparametersis-2732.386whichcanbefoundinAppendixH(yellowhighlighted)andreflectstheloglikelihoodof themodelwith nopredictors. The log likelihood is definedup to an arbitrary additive constant. The loglikelihoodfunctionatthezeroparameters(restrictedloglikelihood)isderivedfromtheoutputofNLOGITofthe latentclasslogitmodel.Thiscanalsobecalculatedbymultiplyingthenumberofobservations(1971:219times9)withLn(0.25).Ln(0.25)standsforfouralternatives,sincetheexperimentaldesigncontainedninetimesfouralternatives.

55

ρ2=1−𝐿𝐿(𝛽)/𝐿𝐿(0)LL(β)=loglikelihoodfunctionatestimatedparameters;

LL(0)=loglikelihoodfunctionatzeroparametersρ2=1−(-2474.425)/(-2732.386)=1–0.9056=0.0944

Insection3.2.2wasmentionedthatwhentheMcFadden’srho-squarevalueisabove0.1,itcanbeconsideredasusable(Louviereet.al.,2000).Themodelwhichestimatedthepreferredcharacteristicsofacoworkingspacedidnotperformwellenough(ρ2=0.0944<0.1),howevertheresultliesveryclosetothelimitof0.1.Thisdoesnotmeanthatthemodelisnotvalidbutthatthereismuchunobservedheterogeneitybetweentherespondentsormanyrandomerrors(faultsandindifferences)inthechoicesmade(Kemperman&Timmermans,2008).Therefore,theresultswillbefurtheranalyzedandinterpretedcarefully.Intable5.1,underlyingstatisticsoftheestimatedmodelarepresented.Table5.1:Descriptivestatisticsmultinomiallogitmodel

UtilityvaluesandsignificanceoftheparametersWiththeaidofthemultinomiallogitmodel,theutilityvaluesofthefirsttwolevelsofallattributesareestimatedaswellastheircorrespondingsignificance.Anattributeconsistsofthreeattributelevels(parameters)inwhichtheutilityvalueofthetotalattributeisreflectedinthetotalattributeutilityandtheutilityvalueoftheattributelevelisreflectedinthepart-worthutility(PWU).Eachthirdparameterofanattribute(ß3,ß6,ß9,ß12,ß15,ß18,ß21,ß24)servesasabasepoint/referenceofthecorrespondingattribute(seesection4.2).Relativetothebasepoint,theutilityvaluesandlevelofsignificanceofthatspecificattributeareestimated.Noutilityvalueandsignificanceareestimatedforthethirdlevelparametersbecauseitisabasepoint.Withathree-levelattribute,twoattributelevelsareestimated(ß1,ß2,ß4,ß5,ß7,ß8,ß10,ß11,ß13,ß14,ß16,ß17,ß19,ß20,ß22,ß23).Theutilityvalueofthethirdattributelevelcanbederivedfromthefirstandsecondattributelevel(–(ß1+ß2)).

Theutilityvaluesof theestimatedmodelarepresented in table5.2.Theconstantparameter in this researchreflectstheutilityvalueofthealternative‘noneoftheseoption’ inwhicharespondentratherwouldworkathomeorsomewhereelsethaninapresentedcoworkingspace.Theutilityvalueoftheconstantß0is0.16intheestimatedmodel.Onaverage,whenthetotalutilityvalueofanalternativeisbelow0.16,theprobabilityacoworkerratherwouldworkathome(orsomewhereelse)thaninapresentedalternativeislarger(seetable5.3).Theparametersß1…B24representthemaineffectsoftheattributelevels(Borgers&Vosters,2010).Thesepart-worthutilitiesarepresentedinthefourthcolumnin table 5.2. A positive part-worth utility (PWU) indicates that a parameter has a positive influence (utility) on thepreferencesofacoworkingspace.Anegativepart-worthutilityindicatesthataparameterhasanegativeinfluence(utility)onthepreferencesofacoworkingspace.

Asignificantparametermeansthattheparameterissignificantlydifferentfromthemeansincetherehasbeenmadeuseofeffectcoding(seetable4.1).Ifdummycodingwasused(zeroandone),itwascomparedtozero.Ifaparameterisnotsignificant,ittellssomethingabouthowitdiffersfromtheaverage.Inthefifthcolumn,thesignificanceofthepath-worthutilitiesispresented.Onlythesignificantparameterscanbeinterpretedasso.‘’Statisticallysignificantresultsarethosethatareinterpretednotlikelytohaveoccurredpurelybychanceandtherebyhaveotherunderlyingcausesfortheiroccurrence’’(Kalla,2009,pp.1).Thefollowingsignificancelevelsoccur:

• Whenacoefficienthas3stars(***),thepart-worthutilityissignificantona1%level;• Whenacoefficienthas2starts(**)itissignificantona5%level;and• Whenacoefficienthasnostars,itisnotsignificant.

Thetotalutilityvalueofacoworkingspaceisthesumofthecorrespondingpart-worthutilities(PWU),whichispresented(perattribute) inthesixthcolumn.Thenegativepart-worthutilitiesaremadepositiveforthiscalculationbecause it isabouttheoverallcontributionofaspecificattributeandnotaboutpositiveornegativeinfluencetotheuser’preferencesinacoworkingspace.Thefollowingtable(table5.2)showseverypart-worthutility(PWU)andtheirlevelofsignificance.AnoverviewoftheoutputofNLOGITwhenestimatingamultinomiallogitmodelispresentedinAppendixG.

Loglikelihoodoftheestimatedparameters -2474,43Loglikelihoodofthezeromodel -2732,39McFadden'sRho-square 0,0944McFadden'sRho-squareadjusted 0,0812Numberofobservation 1971AICcriteriumdevidedbynumberofobservations 2.528Numberofparameters 17Numberofclasses 1

56

Table5.2:Part-worthutility,totalutilityandsignificanceoftheparametersofthemultinomiallogitmodel

Thefollowingenumerationandfigure(5.1)shows,inorderofimportance,thecontributionofattributestoacoworkingspace.Sincenoneoftheparameteroftheattributediversityinsupplyspacesissignificant,thisattributewillnolongerbeincluded.

1. Typeofleasecontract(1.42)o Nocontract:0.51***;o Shortcontract:0.20***;o Longtermcontract:-0.71;

2. Accessibilityoflocation(1.0)o Bypublictransportandcar:0.48***;o Bypublictransport:0.02;o Bycar:-0.50***;

3. Layoutofthespace(0.61)o Half-openlayout:0.28***;o Openlayout:0.03;o Closedlayout:-0.30;

4. Diversityoftenants(0.48)o Moderatediversityoftenants:0.14***;o Strongdiversityoftenants:0.10***;o Nodiversityoftenants:-0.24;

5. Receptionandhospitality(0.34)o Receptionbutnohost:0.13***;o Receptionandactivehost:0.04;o Noreceptionandnohost:-0.17***;

6. Events(0.31)o Sometimes:0.15***;o Often:0.004;o None:-0.15***;

7. Atmosphereandinterioraesthetics(0.27)o Homey:0.13;o Modern:-0.01;o Industrial:-0.14**.

AvisualizationoftheestimatedutilitiesispresentedinAppendixI(recommended).

Attribute Attributelevel β-level PWU Significance TotalAttributeUtilityConstant Constant*** β0 0,16 0,008 0,16

Bycarandpublictransport*** β1 0,48 0,000 -

Bycar*** β2 -0,50 0,000 -

Bypublictransport β3 0,02 - 1,00

Industrial** β4 -0,13 0,010 -

Modern β5 -0,01 0,846 -

Homey β6 0,13 - 0,27

Openlayout β7 0,03 0,565 -Halfopenlayout*** β8 0,28 0,000 -Closedlayout β9 -0,30 - 0,61

Basiccoworkingspace β10 -0,04 0,398 -Standardcoworkingspace β11 0,07 0,148 -Premiumcoworkingspace β12 -0,03 - 0,14

Noreceptionandnohost*** β13 -0,17 0,001 -Receptionbutnohost*** β14 0,13 0,003 -Receptionandactivehost β15 0,04 - 0,34

None*** β16 -0,15 0,001 -Sometimes*** β17 0,15 0,001 -Often β18 0,004 - 0,31

Nodiversityoftenants*** β19 -0,24 0,000 -Moderatediversityoftenants*** β20 0,14 0,001 -Strongdiversityoftenants β21 0,10 - 0,48

Nocontract*** β22 0,51 0,000 -Shortterm*** β23 0,20 0,000 -Longterm β24 -0,71 - 1,42

Totalutility 4,56

Diversityoftenants

Typeofleasecontract

Accessibilityoflocation

Atmosphereandinterioraesthetics

Layoutofthespace

Diversityinsupplyspaces

Receptionandhospitality

Events

0 0,2 0,4 0,6 0,8 1 1,2 1,4 1,6

Atmosphereandinterioraesthetics

Events

Receptionandhospitality

Diversityoftenants

Layoutofthespace

Accessibilityoflocation

Typeofleasecontract

Figure5.1:Totalutilityofattributes

57

Thefourthsub-questionisasfollows:‘’whataretheuserpreferencesforcoworkingspacecharacteristics?’’.Inordertoanswerthissub-question,table5.2andthejustpresentedenumerationwillbeusedasabasis.Thepreferencesarebeingtreatedinorderofdescendingimportance(totalutility),sothemostpreferredcharacteristicsofacoworkingspacewillbepresentedfirst:

1. Accordingthe219respondents, thetypeof leasecontract is themostpreferredcharacteristic inacoworkingspace(totalutility:1.42).Thecoworkerssignificantly(p<0.01)prefernocontractorashortcontract(day,weekormonth)butcertainlynotalongcontract(yearorlonger).Thepart-worthutilityoftheattributelevelnocontractisthehighestofallthepreferredcharacteristicsandthepart-worthutilityoftheattributelevellongtermcontract(yearorlonger)isthelowestofallthepresentedcharacteristics.Themostimportantcharacteristicofacoworkingspaceisthetypeofleasecontractwithastrongpreferencefornocontractorashortleasecontract(day,weekormonth).

2. Theaccessibilityofthelocationcomesinsecondplaceofthemostpreferredattributes(totalutility:1.0).Whenthecoworkingspaceisbothaccessiblybypublictransportandcar,mostutilityvalueisassignedbythecoworker(p<0.01).Relativelylittleutilityvalueisassignedtotheparameterbypublictransport(0.02)whichisremarkablesinceallthecoworkingspaceswerewellservedbypublictransport.Agoodaccessiblecoworkingspacebybothpublictransportandcarisofcoursemoreconvenient.Theparticipatingcoworkingspacesinthesampleweremainlylocatedwithinurbanareaswhicharebetterservedbypublictransportthanbycar.Theleastutilityvalueisassignedtoacoworkingspacewhichisonlyaccessiblebycar.

3. Coworkerssignificantlyprefer(p<0.01)ahalf-openlayoutoveranopenlayoutandaclosedlayout.Inahalf-openlayout,thereisacombinationofopenspacesaconcentrationrooms,whereanopenlayoutmainlyconsistsoflarge open spaces. A closed layoutwith enclosed and separate spaces is a drawback for a coworking space.Coworkingspacesareknownfromtheiropenandeasyaccessibleenvironment,aclosedlayoutisincontradictionwiththat.Thetotalutilityvalueofthelayoutofthespace(0.61)indicatesthatitisofinteresttothecoworkerhowthespace isorganized.Therehastobeasufficientamountofspaceforcollaborativework,but it isalsoimportantthatcoworkerscanconcentratesomewhere,whichisharderinacollaborativeenvironmentwithalotofdistraction.

4. Thediversityoftenantsalsoplaysasignificant(p<0.01)roleinthecoworkingspace(0.48).Acoworkingspaceisacommunity-drivenenvironmentof like-minded individualswhere lotsofbusinessfieldcanmeeteachother.Thisalsocomesforwardinthesampleofthestudy.Whenthetenantsinthecoworkingspacerepresentafewora lotbusiness fields, thishasapositiveeffectonthepreferredcharacteristics (0.14and0.10).Nodiversityoftenantsisnotpreferred(-0.24).

5. Receptionandhospitalityfinishedinfifthpositionoverall(0.34).Coworkersprefer(p<0.01)tohaveareceptionbecausetheattributelevelsreceptionbutnohostandreceptionandactivehostbothscoredapositiveutilityvaluewhile the attribute level no reception and no host scored relatively low. An active host that connectscoworkerstoeachotherisalsoapreferredcharacteristicforsomecoworkersinthesample.

6. Thecoworkerprefers(p<0.01)eventsinthecoworkingspace(0.31).Multiplecoworkersinthesamplestatedthattheyprefertohavesometimesaneventinthecoworkingspace,whileothercoworkersalsostatedthatnovalueisattachedtonoevents.Thereisaclearseparationofcoworkersaccordingthisattribute.Thepart-worthutilityoftheattributeleveloftenismerely0.004,whichindicatesthatnovalueisattachedtomanyeventsinthecoworkingspace.Comparedtothepreviousattributes,itappearsthatthisattributeisoneoftheleastimportantcharacteristicsinacoworkingspace.

7. Whatisstrikingisthattheatmosphereandinterioraestheticsarerelativelyunimportantinacoworkingspace(0.27)whileoftenalotofattentionispaidtotheaestheticsofacoworkingspace.Thestylesmodernandindustrialhavethemselvesalowutilitythatreflectsthatithasanegativeinfluenceonthepreferences(p<0.05).

In table 3.7, the experimental design with corresponding attribute levels was presented. The total utility of analternative/coworkingspacecanbecalculatedbasedonthepart-worthutilitiesofeveryattribute.Thisisthesumofalltheattribute level corresponding part-worth utilities. The total utility per alternative is presented in table 5.3. From thealternativewiththehighesttotalutility(alternative8)tothealternativewiththelowesttotalutility(alternative12).Thistable clearly shows inwhich compositionof attribute levels themostutility canbeachieved. Theutility valuesof theattributediversityinsupplyspacesareignoredinthecalculationofthetotalutilityvaluebecauseitisnotsignificant.

58

Table5.3:Totalutilityperpresentedalternative

Alternative 8 is themost preferred alternative of all the presented alternativeswith a total utility value of 1.48. Thiscoworkingspaceisaccessiblebypublictransportandcar,theatmosphereofinterioraestheticsishomeyandhasahalf-openlayout.Whatisstrikingisthatthisalternativedoesnothaveareceptionandahost,lowutilityvalueisassignedtothisattribute levelaccordingtotable5.2.Thiscoworkingspacealternativeorganizesometimesanevent,hasastrongdiversityoftenantsinmultiplesectorsandnoformofleasecontractneedtobesigned.Insecondplacecomesalternative7withatotalutilityvalueof1.04.Therelativelybigdifferenceintotalutilityismainlyduetoattributeslayoutofthespace(half-openoropen)andtypeofleasecontract(nocontractorshorttermcontract).

Theprobabilitythatacoworkerratherwanttoworkathome(orsomewhereelse)islargerifthetotalutilityofthealternativeislowerthantheconstant(0.16).Thesealternativesaremarkedred.Thehigherthetotalutilityvalue,thehighertheprobabilityacoworkerchoosetheofferedalternativeinsteadofworkingfromhomeorsomewhereelse.Thesealternativesaremarkedgreen.Alternative12reflectstheleastpreferredcoworkingspacecharacteristicsofallpresentedalternativeswithatotalutilityvalueof-2.05.Alltheparameterswiththelowestutilityofaspecificattribute,withexceptionoftheparameteroftenoftheattributeevents,comestogetherinthisalternative.Basically,anythingthatdoesnotreflectatypicalcoworkingspace.

Inparticular,thedistributionofattributelevelsofattribute1andattribute8standoutintable5.3.Regardingattribute1,thereisaconcentrationoftheattributelevelbypublictransportandbypublictransportandcarabovetheconstant and a concentration of the attribute level by car under the constant. Public transport is the essence howcoworkersgettothecoworkingspace.Regardingattribute8,thereisaconcentrationinparticularoftheattributelevelsnocontractandshortcontractabovetheconstantandconcentrationoftheattributelevellongtermcontractundertheconstant.Alongrentalcontractgoesagainsttheprinciplesofacoworkingspace.Thepracticalrelevanceofthisstudyisprovidinginsightintheuserpreferencesforcoworkingspacecharacteristics.Byunderstandingthepreferencesofthecoworkers,coworkingspacescanbemoreeffectivelydevelopedbecauseprovidersanddeveloperscanrespondtotheneedsoftheuser.Theworkenvironmentcanbeadjustedtotheuserpreferencesresultingfromtheresearch.Twenty-fivecoworkingspacescooperatedintheresearchofwhich16coworkingspaceswerevisited(seeAppendixD).Inordertodemonstratethepracticalrelevance,thetotalutilitypercoworkingspaceispresentedintable5.4.Table5.4:Totalutilitypervisitedcoworkingspace(practicalrelevance)

Alternative Attribute1 Attribute2 Attribute3 Attribute4 Attribute5 Attribute6 Attribute7 Attribute8 TotalUtility8 Bypublictransportandcar Homey Half-openlayout Standardcoworkingspace Noreceptionandnohost Sometimes Strongdiversityoftenants Nocontract 1,487 Bypublictransportandcar Homey Openlayout Basiccoworkingspace Receptionandactivehost Often Moderatediversityoftenants Shortterm(dayorweekormonth) 1,0423 Bypublictransport Modern Half-openlayout Basiccoworkingspace Receptionbutnohost Often Strongdiversityoftenants Nocontract 1,0225 Bypublictransport Homey Openlayout Premiumcoworkingspace Receptionbutnohost Sometimes Nodiversityoftenants Nocontract 0,726 Bypublictransportandcar Modern Closedlayout Premiumcoworkingspace Noreceptionandnohost Often Moderatediversityoftenants Nocontract 0,665 Bypublictransportandcar Modern Half-openlayout Standardcoworkingspace Receptionandactivehost None Nodiversityoftenants Shortterm(dayorweekormonth) 0,603 Bypublictransportandcar Industrial Closedlayout Premiumcoworkingspace Receptionandactivehost Sometimes Strongdiversityoftenants Shortterm(dayorweekormonth) 0,551 Bypublictransportandcar Industrial Openlayout Basiccoworkingspace Noreceptionandnohost None Nodiversityoftenants Nocontract 0,3321 Bypublictransport Industrial Closedlayout Standardcoworkingspace Receptionbutnohost None Moderatediversityoftenants Nocontract 0,212 Bypublictransportandcar Industrial Half-openlayout Standardcoworkingspace Receptionbutnohost Often Moderatediversityoftenants Longterm(yearorlonger) 0,2018 Bycar Homey Closedlayout Basiccoworkingspace Receptionandactivehost Sometimes Moderatediversityoftenants Nocontract 0,17Constant x x x x x x x x 0,1611 Bycar Industrial Half-openlayout Premiumcoworkingspace Receptionandactivehost None Strongdiversityoftenants Nocontract 0,1420 Bypublictransport Industrial Half-openlayout Basiccoworkingspace Noreceptionandnohost Sometimes Nodiversityoftenants Shortterm(dayorweekormonth) 0,1122 Bypublictransport Modern Openlayout Premiumcoworkingspace Noreceptionandnohost None Moderatediversityoftenants Shortterm(dayorweekormonth) 0,0610 Bycar Industrial Openlayout Standardcoworkingspace Receptionbutnohost Sometimes Moderatediversityoftenants Shortterm(dayorweekormonth) 0,0317 Bycar Homey Half-openlayout Premiumcoworkingspace Receptionbutnohost Often Nodiversityoftenants Shortterm(dayorweekormonth) 0,0027 Bypublictransport Homey Closedlayout Standardcoworkingspace Noreceptionandnohost Often Strongdiversityoftenants Shortterm(dayorweekormonth) -0,014 Bypublictransportandcar Modern Openlayout Basiccoworkingspace Receptionbutnohost Sometimes Strongdiversityoftenants Longterm(yearorlonger) -0,1313 Bycar Modern Openlayout Standardcoworkingspace Receptionandactivehost Often Nodiversityoftenants Nocontract -0,1726 Bypublictransport Homey Half-openlayout Basiccoworkingspace Receptionandactivehost None Moderatediversityoftenants Longterm(yearorlonger) -0,2615 Bycar Modern Closedlayout Basiccoworkingspace Receptionbutnohost None Strongdiversityoftenants Shortterm(dayorweekormonth) -0,5419 Bypublictransport Industrial Openlayout Premiumcoworkingspace Receptionandactivehost Often Strongdiversityoftenants Longterm(yearorlonger) -0,659 Bypublictransportandcar Homey Closedlayout Premiumcoworkingspace Receptionbutnohost None Nodiversityoftenants Longterm(yearorlonger) -0,6614 Bycar Modern Half-openlayout Premiumcoworkingspace Noreceptionandnohost Sometimes Moderatediversityoftenants Longterm(yearorlonger) -0,8224 Bypublictransport Modern Closedlayout Standardcoworkingspace Receptionandactivehost Sometimes Nodiversityoftenants Longterm(yearorlonger) -1,0616 Bycar Homey Openlayout Standardcoworkingspace Noreceptionandnohost None Strongdiversityoftenants Longterm(yearorlonger) -1,2812 Bycar Industrial Closedlayout Basiccoworkingspace Noreceptionandnohost Often Nodiversityoftenants Longterm(yearorlonger) -2,05

Coworkingspace Location Attribute1 Attribute2 Attribute3 Attribute4 Attribute5 Attribute6 Attribute7 Attribute8 TotalUtilityInterpolisCarrousel Tilburg Bycarandpublictransport Homey Half-openlayout x Receptionbutnohost Sometimes Moderatediversityoftenants Nocontract 1,82AmsterdamConnected Amsterdam Bycarandpublictransport Modern Half-openlayout x Receptionbutnohost Sometimes Moderatediversityoftenants Nocontract 1,68Meet&Discover Amersfoort Bycarandpublictransport Homey Openlayout x Receptionbutnohost Sometimes Moderatediversityoftenants Nocontract 1,57Qlubhouse Tilburg Bycarandpublictransport Modern Half-openlayout x Receptionbutnohost Sometimes Moderatediversityoftenants Shortterm(dayorweekormonth) 1,38Tribes Utrecht Bycarandpublictransport Modern Half-openlayout x Receptionbutnohost Sometimes Moderatediversityoftenants Shortterm(dayorweekormonth) 1,38Seats2Meet Eindhoven Bycarandpublictransport Industrial Openlayout x Receptionbutnohost Sometimes Strongdiversityoftenants Nocontract 1,27Seats2Meet Amersfoort Bypublictransport Modern Half-openlayout x Receptionbutnohost Sometimes Strongdiversityoftenants Nocontract 1,17BouncespaceEindhoven Eindhoven Bycarandpublictransport Modern Half-openlayout x Noreceptionandnohost Sometimes Moderatediversityoftenants Shortterm(dayorweekormonth) 1,08Seats2Meet DenBosch Bypublictransport Modern Openlayout x Receptionbutnohost Sometimes Strongdiversityoftenants Nocontract 0,92MeetBerlage Amsterdam Bypublictransport Industrial Openlayout x Receptionbutnohost Sometimes Moderatediversityoftenants Nocontract 0,85Startdock Amsterdam Bypublictransport Homey Closedlayout x Receptionandactivehost Sometimes Moderatediversityoftenants Shortterm(dayorweekormonth) 0,82Seats2Meet Utrecht Bypublictransport Industrial Openlayout x Receptionbutnohost Sometimes Strongdiversityoftenants Nocontract 0,80WickedGrounds Amsterdam Bypublictransport Industrial Half-openlayout x Receptionbutnohost Sometimes Moderatediversityoftenants Shortterm(dayorweekormonth) 0,79BouncespaceAmsterdam Amsterdam Bypublictransport Modern Half-openlayout x Noreceptionandnohost Sometimes Moderatediversityoftenants Shortterm(dayorweekormonth) 0,61HetNieuweKantoor DenBosch Bycar Modern Half-openlayout x Receptionbutnohost Sometimes Moderatediversityoftenants Shortterm(dayorweekormonth) 0,39ALab Amsterdam Bypublictransport Industrial Closedlayout x Noreceptionandnohost Sometimes Strongdiversityoftenants Shortterm(dayorweekormonth) 0,31Constant x x x x x x x x x 0,16

59

Everyvisitedcoworkingspacehasahighertotalutilityvaluethantheutilityvaluewhichisassignedtoworkingfromhome(orsomewhereelse).Thetopthreecoworkingspacesinthesample,basedontotalutilityvalue,areInterpolisCarrousel(Tilburg),AmsterdamConnected(Amsterdam)andMeet&Discover(Amersfoort).Whatisstriking,isthatthetotalutilityvalueofthesecoworkingspacesallresultsinahighertotalutilityvaluethanthetopalternative(8)oftable5.3.(1.48).Allthethreecoworkingspacesareworkingwithabooking-systemfromSeats2Meetinwhichfreeworkspaceisofferedto‘loneeagles’.HetNieuweKantoor(DenBosch)wasonlygoodaccessiblebycarandthereasonsthatALab(Amsterdam)comparativelyscoreslowisthatishasaclosedlayout.Overall,thevisitedcoworkingspacesscoredreallyhighsincethelowestutilityvalueis0,31(seetable5.3).Allthevisitedcoworkingspaceslieveryclosetothemostpreferredcharacteristicsresultingfromtheestimation(table5.2).Perhapsrespondentsarelimitedbytheworkenvironmentandreflectmainlytotheirownenvironmentorhavelittleexperiencewithotherworkenvironments.

5.2Classesofpreferredcharacteristics(latentclasslogitmodel)Inthissub-section,thefifthsub-question’’Candifferentusergroupsbeidentifiedbasedontheirpreferences’’’andthesixthsub-question‘’Whatarethedifferencesbetweentheseusergroupsbasedonusercharacteristics’’areanswered.Inordertofindlatentclassesinthemultivariatedataofthepreferencesofcoworkingspacecharacteristicsalatentclasslogitmodel isestimated.Withtheaidofchi-squaretestsand independentsamplesT-tests, itcanbedeterminediftherearesignificantdifferencesbetweentheestimatedclassesregardingdemographicandpsychographiccharacteristics.First,thegoodnessoffitanddescriptivestatisticsofthemodelwillbedeterminedandthereaftertheutilityvaluesandsignificanceofestimatedsegmentsinpreferencesofcoworkingspacecharacteristics.Finally,statementsabouttheresultsofthechi-squaretestsandtheindependentsamplesT-testswillbepresentedwhichreflectsthesegmentcharacteristics.GoodnessoffitandmeasuresoffitInordertodetermineifthelatentclasslogitmodelperformswell,thegoodnessoffithastobecalculated.Theunderlyingequationatthismethodispresentedinsection3.2.2.Thegoalofa latentclass logitmodel isto identifyclasses inthegathereddata. Thegathereddata is formedbya sampleof respondents (N)of a totalpopulation. The classes (S) areunobservableand isbetweenoneand the totalnumberof respondents in the sample. Thenumberof classeswill bebetween1and219.Respondentswhohavesimilarobservedvariabledistributionsaregroupedinthesameclassofdata.Theseclassesarecalledlatentsegments,latentclustersorlatentclassesandareundefinedandunknownápriori(Swait,1994; Nijënstein, 2012). ‘’Through a latent segment classification mechanism, the membership likelihood functionsdeterminethelatentsegmenttowhichanindividualbelongs’’(Swait,1994,pp.78).Inordertodeterminewhichestimatedmodelsuitsbest,thefollowingoverview(table5.5)hastobeanalyzed.Table5.5:DescriptivestatisticsLatentClassLogitModel

Three latentclass logitmodelswereestimated.Theestimatedclassthatgivesthemost informationshouldbeused infurtheranalyses.Thelikelihoodoftheestimatedparameters(LL(𝛽))becomeslessnegativewithanincreasingnumberofclasses.Foreveryadditionalmodel,18parametersmorewereestimated;17foreachcodedattributelevelandoneclassprobabilityparameter.TheAkaike InformationCriterion(AIC) looksatthepredictivepowerofamodel (Kemperman&Timmermans,2008;Nijënstijn,2012).Toidentifytheoptimalnumberofclasses,thispredictorhastobeanalyzedwhichispresentedinthefifthrowofthetable.ThemostoptimalmodelisthemodelwiththelowestAIC,whichisthemodelwithtwoestimatedclasses(2.524).TheMcFadden’srho-squareincreasedforallthefourclasses;however,thebiggestincreaseisfromclass1toclass2.Inaddition,themodelswiththreeandfourestimatedclassesshowveryhighstandarderrors(seeAppendixH).Accordingthedescriptivestatisticsofthemodelsandthehighstandarderrors,themodelwithtwolatentclassesinthepreferredcharacteristicsofcoworkingspacesischosenforfurtherdescription.

Twoclassesareestimatedinthedataofpreferredcharacteristics.Theloglikelihoodfunctionoftheestimatedparametersis-2451,916whichcanbefoundinAppendixH(greenhighlighted).Theloglikelihoodfunctionatthezeroparametersis-2732.386whichcanbefoundinAppendixH(yellowhighlighted).Theloglikelihoodfunctionatthezeroparameters (restricted log likelihood) is derived from the output of the latent class logitmodel. In section 3.2.2.wasmentionedthatwhentherho-squarevalueisabove0.1.Itcanbeconsideredasusable(Louviereet.al.,2000).Themodelwhichestimatedsegmentsinthepreferredcharacteristicsofacoworkingspacedidperformwellenough(twoclasses:ρ2=0.103>0.1;threeclasses:ρ2=0.109>0.1;fourclasses:ρ2=0.111>0.1)forvalidconclusions.

Loglikelihoodoftheestimatedparameters -2451,9164 -2434,8965 -2426,9888Loglikelihoodofthezeromodel -2732,3862 -2732,3862 -2732,3862McFadden'sRho-square 0,103 0,109 0,111Numberofobservation 1971 1971 1971AICcriteriumdevidedbynumberofobservations 2.524 2.525 2.535Numberofparameters 35 53 71Numberofclasses 2 3 4

60

UtilityvaluesandsignificanceoftheparametersTheoverviewpresentedintable5.6istheoutputofNLOGIT(presentedinAppendixH)whenestimatingalatentclasslogitmodelwithtwoclasses.Asinthecaseofthemultinomiallogitmodel,thelatentclasslogitmodelalsoestimatesthepart-worthutilitiesandcorrespondingsignificance.Thelargestdifferencewiththemultinomiallogitmodelisthatinthistypeofanalysis,latentclassesaregatheredinthemultivariatedata. Table5.6:Part-worthutility,totalutilityandsignificanceoftheparametersofthelatentclasslogitmodel

Intable5.6,thetotalutilityvalueofanattributeispresentedaswellasthepart-worthutilityofeachparameterperclass.Onlythesignificantattributelevels(*,**and***)haveunderlyingcausesfortheiroccurrence.Whatstandsoutthatnoneoftheparametersinclass1issignificantwhichmeansthattheycannotbeinterpretedasso.Anykindofinterpretationofclass1canbeomitted.Theonlythingthatcanbedeterminedisthatthisestimatedclassofcoworkersmakesrandomchoices so they have no stated preferences concerning the characteristics of a coworking space or are still veryheterogeneousintheirpreferences.Thistypeofmodeltriestofindhomogeneoussetsofdata.Infact,itcanbesaidthatalltheparametersofclass1areequaltozero.Inclass2,fiveparametersaresignificant.Sincenoneoftheparametersoftheattributesatmosphereandinterioraesthetics,layoutofthespaceandreceptionandhospitalityaresignificant,thoseattributeswillnotbediscussed.Thefollowingenumerationandfigure(5.2)shows,inorderofimportance,thecontributionofeachsignificantattributeinclass2:

1. Events(1.05)o Sometimes:0.41**;o None:0.12;o Often:-0.52;

2. Diversityinsupplyspaces(0.79)o Premiumcoworkingspace:0.40;o Basiccoworkingspace:0.06;o Standardcoworkingspace:-0.34**;

3. Accessibilityoflocation(0.55)o Bypublictransport:0.22;o Bypublictransportandcar:0.06;o Bycar:-0.28**;

4. Diversityoftenants(0.53)o Strongdiversity:0.26;o Moderatediversity:-0.01;o Nodiversity:-0,25**;

5. Typeofleasecontract(0.39)o Nocontract:0.19*;o Shortcontract:-0.06;o Longtermcontract:-0.13.

Attribute Attributelevel β-level PWU Significance TotalAttributeUtility PWU Significance TotalAttributeUtilityConstant Constant β0 0,07 0,880 0,07 -0,14 0,322 -0,14

Bycarandpublictransport β1 2,44 0,290 - 0,06 0,695 -Bycar(class2:**) β2 -2,14 0,249 - -0,28 0,021 -Bypublictransport β3 -0,29 4,87 0,22 0,55Industrial β4 -0,92 0,512 - 0,02 0,849 -Modern β5 -0,21 0,611 - -0,04 0,715 -Homey β6 1,13 2,26 0,02 0,08Openlayout β7 0,21 0,527 - 0,14 0,207 -Halfopenlayout β8 1,36 0,204 - -0,12 0,543 -Closedlayout β9 -1,57 3,15 -0,02 0,28Basiccoworkingspace β10 0,26 0,648 - -0,06 0,567 -Standardcoworkingspace(class2:**) β11 1,36 0,236 - -0,34 0,039 -Premiumcoworkingspace β12 -1,62 3,24 0,40 0,79Noreceptionandnohost β13 -0,52 0,458 - -0,14 0,235 -Receptionbutnohost β14 0,35 0,338 - 0,10 0,415 -Receptionandactivehost β15 0,17 1,04 0,04 0,27None β16 -0,26 0,664 - 0,12 0,288 -Sometimes(class2:***) β17 -0,18 0,679 - 0,41 0,000 -Often β18 0,44 0,88 -0,52 1,05Nodiversityoftenants(class2:**) β19 -0,69 0,412 - -0,25 0,041 -Moderatediversityoftenants β20 0,49 0,276 - -0,01 0,895 -Strongdiversityoftenants β21 0,21 1,38 0,26 0,53Nocontract(class2:*) β22 2,16 0,115 - 0,19 0,088 -Shortterm β23 1,78 0,363 - -0,06 0,715 -Longterm β24 -3,94 7,89 -0,13 0,38

Totalutility 24,79 3,79Probabilityclass1andclass2*** 0,45 0.000 0,55 0.000

Events

Diversityoftenants

Typeofleasecontract

Accessibilityoflocation

Atmosphereandinterioraesthetics

Class1 Class2

Layoutofthespace

Diversityinsupplyspaces

Receptionandhospitality

0 0,2 0,4 0,6 0,8 1 1,2

Typeofleasecontract

Diversityoftenants

Accessibilityoflocation

Diversityinsupplyspaces

Events

Figure5.2:Totalutilityofattributesclass2

Thefifthsub-questionisasfollows: ‘’Candifferentusergroupsbeidentifiedbasesontheirpreferences?’’.Twoclasseswith equal preferences are estimated by the latent class logit model. It can be concluded that class 1 is still veryheterogeneousorselectedthealternativesinthequestionnairescompletelyrandom.Nofurtherconclusioncanbedrawnonthepreferencesofclass1.Class2hassomewhatdistinctpreferences.Theattributeeventshavethehighesttotalutilityvalue(1.05;p<0.05),followedbythediversityinsupplyspaces(0.79;p<0.05)andtheaccessibilityofthelocation(0,55;p<0.55).Autilityvalueof0.26isassignedifthereissometimesanevent inthecoworkingspacea lowutilityvalueisassignedifthereisoftenanevent.Thecoworkersinthisclasspreferapremiumcoworkingspace(collaborativeworkspace+meetingrooms+kitchenarea+eventspaces+informalzones+fitnesscenterandbar)ratherthanastandardcoworkingspace(collaborativeworkspace+meetingrooms+kitchenarea+eventspacesandinformalzones.Thisclassofcoworkersprefersacoworkingspacewhichisaccessiblebypublictransportandrathernotbycar.Accordingthediversityoftenants,thisclassofcoworkerspreferastrongdiversityandlowutilityisassignedtotheparametersmoderatediversityandnodiversity.Whatisstrikingisthatthetypeofleasecontractistheleastpreferredattributeinthisclass,butstillthehighestutilityvalueisattachedtotheparameternocontract.ClasscharacteristicsTheprobabilitycertaincoworkers(N=219)belongtoclass1orclass2,canalsobecalculatedwiththeaidofalatentclasslogitmodelaspresentedintheendofAppendixH.Theexperimentaldesignofthisstudy,asshownintable3.6,consistsof27alternatives.Thesealternativeswerepresentedinninechoicesets,withfourchoices/alternativesperchoiceset.One choice/alternative consisted of eight attributes. In total, there were 36 choices. For every choice/alternative, aprobabilityisestimatedwhetherarespondentbelongstoclass1orclass2.Eventuallyitcanbedeterminedwhetherarespondentbelongstoclass1orclass2whichistheaverageofallthepossibilitiesof36choices.Arespondentbelongstotheclasswiththehighestprobability.Thelastsub-questionofthismasterthesisis‘’Whatarethedifferencesbetweentheseusergroupsbasedonusercharacteristics?’’.Withtheaidofchi-squaretestsandindependentsamplesT-tests,itcanbe determined if there are significant differences between the two estimated classes regarding demographiccharacteristicsandpsychographiccharacteristics.Whethercertainpreferencesaremoreassociatedwithcertaintypesofrespondents. To indicate a clear difference between the preferences per class, an overview is presented, in order ofattributeimportance,intable5.7.Table5.7:Numberofrespondentsandpreferredattributesperlatentclass

Intotal,class1contains82coworkersandclass2contains137coworkers.Inthefollowingfigures(5.3–5.12),overviewispresentedofthedifferencesbetweentheclasseswithregardtotheirdemographicandpsychographiccharacteristics.Thedemographiccharacteristicsgender(figure5.3),agegroups(table5.8andtable5.9),levelofeducation(figure5.4),usergroup(figure5.5),positioninorganization(figure5.6),sectoroforganization(figure5.7),income(figure5.8),transporttocoworkingspace(figure5.9),hoursincoworkingspace(table5.8andtable5.9)areanalyzed.Themotivationsforworkinginacoworkingspacearealsoanalyzedaccordingtheclasscharacteristics.Inorderofimportance(1;firstmotivation(figure5.10),2; secondmotivation (figure5.11),3; thirdmotivation (5.12)), thecoworkerhad to select theirmotivation.Thedemographiccharacteristicsnationalityandlandofcoworkingspacebedisregardingbecauseofthelittlevariance.AlmostallparticipantsareofDutchnationalityduetothelowresponseinothercountries.Multiplesectors(design,commerce,art,management,research,projectmanagementandwriting)arecombinedunder‘other’becausetheoccurrenceofthosesectorsistoolowtoperformchi-squaretest.Thefiveleastimportantmotivations(seefigure4.6)arealsocombinedunder‘other’becausetheoccurrenceofthosemotivationsisalsotoolowtoperformachi-squaretest. Inthestackedbarcharts,theoccurrenceofrespondentswithcorrespondingpercentageinacertainclassarementionedaswellas,inthetitleofthefigure,thechi-square(χ2)withcorrespondingprobability(p-value).Whenthep-valueislargerthan0.05,thenthereisnotasignificantdifferencebetweenclass1andclass2accordingthatparticularvariableandhasnounderlyingcausesfortheiroccurrence.Whenthep-valueislessthan0.05,thisindicatesthatthereisasignificantdifferencebetweenclass1andclass2regardingthatspecificvariable. Inchapter4,thevariablesageandhoursworkinginacoworkingspacearerecodedintoordinalvariablestocreateabetterandclearerviewofthevariables.A ratiovariablecontainsmore information, so for that reason the ratiovariablesareused in theanalysesof theclasscharacteristics.Fortheratiovariablesageandhoursworking inthecoworkingspace, independentsamplesT-testsareperformed.Frequencytablesofthenumberofrespondentsinclass1andclass2regardingthefollowingcharacteristicsarepresentedinAppendixJ.Thechi-squarevalueswithcorrespondingsignificancearealsopresentedinAppendixJ.For

Class1 Class2Numberofrespondents 82 137Preferredattributes Nopreferredattributes Events

DiversityinsupplyspacesAccessibilityoflocationDiversityoftenantsTypeofleasecontract

62

allthevariables,thenumberofrespondentsinclass2islargerthanclass1.Thisgivesadistortedviewaboutquantitieshenceonlysomethingcanbesaidaboutthelevelofsignificance.

Figure5.3:Gender(χ2=0,122;p=0,73)Figure5.4:Education(χ2=1,389;p=0,24)

Figure5.5:Usergroup(χ2=1,587;p=0,66)Figure5.6:Positionorganization(χ2=2,37;p=0,67)

Figure5.7:Sectororganization(χ2=2,439;p=0,66)Figure5.8:Income(χ2=3,807;p=0,28)

Figure5.9:Transporttocoworkingspace(χ2=1,143;p=0,77)

63

Figure5.10:Motivation1(χ2=6,476;p=0,26)

Figure5.11:Motivation2(χ2=10,471;p=0,063)

Figure2.12:Motivation3(χ2=2,732;p=0,75)

Table5.8:DescriptivestatisticsindependentsamplesT-test

Table5.9:IndependentsamplesT-test

In theprevious figuresand tables, thedifferencesaccording theuser characteristics (demographic andpsychographiccharacteristics)perestimatedclassaredescribed.Onesignificantdifferencebetweenthetwoestimatedclassesregardingusercharacteristicsisfound.Thecoworkersinclass2worksignificantly(p=0.025)morehoursperweekinthecoworkingspacethanthecoworkersinclass1.Onaverage,thecoworkersinclass1working18,5hoursaweekinthecoworkingspacewhile the coworkers in class 2working 23 hours aweek in the coworking space (see table 5.8). No significantdifferencesarefoundbetweentheclassesaccordingtheratiovariableage.Foralltheother(nominalandordinal)variables,thep-valueis largerthanthenecessary0.05whichindicatesthattherearenosignificantdifferencesbetweenthetwoestimatedclassesregardinggender,levelofeducation,usergroup,positioninorganization,sectoroforganization,income,transporttocoworkingspaceandmotivationtoworkinacoworkingspace.Besidesthat,anumberofthingsstandoutregardingthenumberofrespondents.Firstofall,class2consistsofmorecoworkerswithaboard/ownerpositionintheorganizationthanclass1.Moreover,class2consistofmorecoworkersthatworksintheconsultancysectorthanclass1.Finally,class2consistofmorecoworkersthatgoestothecoworkingspacebybike.Class2hascertainpreferencesfortheattributesevents,diversityinsupplyspaces,accessibilityoflocation,diversityoftenantsandthetypeofleasecontract.Thismeansthatcoworkerswhospendmoretimeinthecoworkingspacehaveacertainpreferenceforsometimesanevent inthecoworkingspace,apremiumcoworkingspace,accessibilitybypublictransport,astrongdiversityoftenantsandnorentalcontract.Apparently,thosecoworkingspacecharacteristicsbecomesmoreimportantwhenmoretimeisspentinthecoworkingspace.

5.3ConclusionInthischapteraboutthedataanalysisandresultsofthisstudy,thefourth,fifthandsixthsub-questionareanswered.Withtheaidofamultinomiallogitmodelthesub-question‘’whataretheuserpreferencesforcoworkingspacecharacteristics?’’isanswered.Therho-squareofthemodel,whichdeterminesthegoodnessoffit,is0.0944whichissmallerthan0.1.Thisindicatesthatthemodeldoesnotperformverywellbecausetherho-squareshouldbelargerthan0.1tobeuseable.Thismeansthatthereismuchheterogeneitybetweentherespondentsormanyrandomerrors(faultsandindifferences)inthechoicesmade.Theparametersofthemodel,theutilityandsignificanceoftheparameterscanbeinterpretedassuchbutsomecautionisadvisedbecauseofthegoodnessoffit.Themostimportantcharacteristicofacoworkingspaceisthetypeofleasecontract,whichposedaclearpreferenceforacoworkingspacewithoutaleasecontractorashort-termleasecontract.Along-termcontractofayearorlongerhasastrongnegativeinfluenceonthepreferencesofacoworker.Accessibilityofthelocationcomesinsecondplaceofmostpreferredcoworkingspacecharacteristics.Thecoworkersinthesamplepreferacoworkingspacethatisaccessiblebycarandpublictransportthemost.Thelayoutofthecoworkingspaceisimportantforcoworkers.Acombinationofspacesforcollaborativeworkandspacesforconcentrationworkisthemostpreferred.Thishalf-openlayoutismorepreferredthanafullyopenlayout.Thediversityoftenantsisalsoofimportanceinthecoworkingspace.Lowutilityisallocatedtothecoworkingspaces,whichhavenodiversityoftenants.Relativelyhighutilityisallocatedtothecoworkingspaces,whichhaveamoderateorstrongdiversityoftenants.Thepresenceofareceptionisappreciatedinacoworkingspace.Thelowestutilityisassignedtotheattributeseventsandatmosphereandinterioraesthetics.Significanteffectsareshownfortheattributelevelsforsometimesaneventandforoftenanevent.Contrarytotheexpectation,relativelynovalueisattached

N Mean Std.Deviation Std.ErrorMeanClass1 82 34,33 10,550 1,165Class2 137 34,77 11,553 ,987Class1 82 18,50 13,465 1,487Class2 132 23,00 14,538 1,265

Age

HoursCWS

Lower UpperEqualvariancesassumed

,651 ,421 -,284 217 ,776 -,444 1,562 -3,524 2,635Equalvariancesnotassumed

-,291 182,884 ,771 -,444 1,527 -3,457 2,568

Equalvariancesassumed

,304 ,582 -2,264 212 ,025 -4,500 1,988 -8,419 -,581Equalvariancesnotassumed

-2,305 181,831 ,022 -4,500 1,953 -8,353 -,647

Std.ErrorDifference

95%

Age

HoursCWS

IndependentSamplesTestLevene'sTestforEqualityofVariances t-testforEqualityofMeans

F Sig. t dfSig.(2-tailed)

MeanDifference

65

totheaestheticsofacoworkingspace.Thefollowingcoworkingspacecharacteristicsarethemostpreferredcharacteristicsinorderofimportancewithbetweenbracketsthemostpreferredattributelevel(s):

• Typeofleasecontract(noorshortleasecontract);• Accessibilityofthelocation(bycarandpublictransport);• Layoutofthespace(half-openlayout);• Diversityoftenant(moderateorstrongdiversity);• Receptionandhospitality(receptionbutnohost);• Events(sometimes);and• Atmosphereandinterioraesthetics(homey);

Thefifthsub-questionofthismasterthesisis‘’Candifferentusergroupsbeidentifiedbasedontheirpreferences?’’.Inordertoanswerthissub-question,alatentclasslogitmodelisestimatedtofindusergroupswithcorrespondingpreferredcoworkingspacecharacteristics.Twoclasseswerefoundinthemultivariatedataofthedependentfactorinwhichclass1isstillveryheterogeneousandhasnosignificantpreferencesaccordingthecharacteristics.Class2hassomewhatdistinctpreferencesforsometimesanevent,apremiumcoworkingspace,accessibilitybypublictransport,astrongdiversityoftenantsandno leasecontract.Withtheaidofmultiplechi-squaretestsand independentsamplesT-tests, the lastsub-question ‘’What are the differences between these user groups based on user characteristics?’’ is answered. Only asignificant difference is found between the two estimated classes regarding the ratio variable hours working in thecoworking space. The coworkers in class 2 work significantlymore hours per week in the coworking space than thecoworkersinclass1.Thismeansthatcoworkerswhospendmoretimeinthecoworkingspacehaveacertainpreferenceforsometimesaneventinthecoworkingspace,apremiumcoworkingspace,accessibilitybypublictransport,astrongdiversityoftenantsandnorentalcontract.Apparently,thosecoworkingspacecharacteristicsbecomesmoreimportantasmoretimeisspentinthecoworkingspace.Thefollowingchapteristhefinalchapterofthismaster’sthesisinwhichtheconclusionsaredrawnandrecommendationsaccordingtheliterature,methodologyandthepracticalrelevance.Furthermore,thepossibilitiesoffutureresearchinthecoworkingframeworkaregivenaswell.

66

67

Chapter6:ConclusionsandRecommendations Inthelastchapterofthisresearch,theconclusionsandrecommendationsarepresented.Firstinsection6.1,theresearchresultsarepresentedwhereinthesixsub-questionsserveasaguidance.Thereflectionontheliteraturetakesplaceinthissectionaswell.Insection6.2,thelimitationsofthismaster’sthesisandthepossibilitiesforfutureresearcharepresented.Finally,thepracticalrelevanceofthisstudyisdiscussedinsection6.3.

6.1ResearchresultsThecoworkingphenomenonisobviouslyrisenonaglobalscale.Therewere7,800coworkingspacesworldwidein2015incomparisonwith310 in2009,andthenumberofcoworkers in2015was510,000 incomparisonwith43,000 in2011.Despitethe increasingpopularity,theconceptofcoworkinghasalmostcompletely ignoredbytheacademic literature.Sincecoworkingisarelativelynewconceptofmulti-tenantoffice(originatedin2005byBradNeuberg),littleresearchhasbeen done in this field of user characteristics and coworking space characteristics. With the aid of this study, thegeographic, demographic and psychographic user characteristics of coworkers in the Netherlands became clear. Thetypicalcharacteristicsofcoworkingspacesareanunder-exposedsubjectaswellandwerediscussedinthisstudy.The aimof thismaster’s thesiswas to identify user groupsbasedonuserpreferences and to analyze thedifferencesbetweentheseusergroups.Thecorrespondingresearchquestionofthisresearchobjectiveisasfollows:

‘’Candifferentusergroupsbeidentifiedbasedontheirpreferencesforcoworkingspacecharacteristicsanddotheydifferonusercharacteristics?’’

Thisresearchquestionwasansweredbasedonthefollowingfivesub-questions:

Whatarethetypicalcharacteristicsofcoworkingspaces? Whichtypesofuserscanbeidentifiedincoworkingspacesandwithwhichcharacteristicscantheybedescribed? Whataretheusercharacteristicsofcoworkersingeneral? Whataretheuserpreferencesforcoworkingspacecharacteristics? Candifferentusergroupsbeidentifiedbasesontheirpreferences? Whatarethedifferencesbetweentheseusergroupsbasedonusercharacteristics?

Inthisresearch,aquestionnairewascomposedtocollectdataontheusercharacteristicsofcoworkingspacesandtheuser preferences for coworking space characteristics. Geographic, demographic and psychographic characteristics areanalyzed in relation to six characteristics of coworking spaces. In total, 219 coworkers with a wide range of usercharacteristicshaveindicatedwhattheirpreferencesfortheaccessibilityofthelocation,layoutofthespace,diversityinsupplyspaces,receptionandhospitality,eventsinthecoworkingspace,diversityoftenantsandthetypeofleasecontract.Datafortheusercharacteristicsarecollectedwithmultiplesurveyquestions,datafortheuserpreferencesarecollectedwith the aid of anattribute based stated choicemethod. In this type ofmethod, a coworker is placed in a particularframeworktocomparealternativesofcoworkingspacesthatweredescribedbymultipleattributes.Therespondenthadtochoosetheirmostpreferredcoworkingspace/alternative.Twenty-fivecoworkingspacesintheNetherlandswerewillingtocooperateinthestudy.Theexistingliteratureshowedthatthereisrelativelylittleknowledgeofusercharacteristicsofcoworkers,especiallyonanationallevel.Sofar,fewstudiesfocusedonusercharacteristics,butthesestudieswereperformedonagloballevel.Nostudiesfocusedontheusercharacteristicsofcoworkersonanationallevel.Onthebasisofthesampleofthisstudy,insightisgivenintheusercharacteristicsofthecoworkersintheNetherlandsingeneral.Anumberofdemographiccharacteristicsareremarkable.Firstofall,thecoworkersarehighlyeducated(almost90%ofthesample)whichreflectsahighervocationaleducationorauniversitydegree(bachelor,mastersorPhD).ComparedtotheworkforceoftheNetherlands,inwhich28%ishighlyeducated,thispercentageisquitehigh.Coworkingisoftenassociatedwithfreelancersandself-employedworkers,which is clearly confirmed by several authors. The coworkers in the sample aremostly (53%) self-employedworkers,freelancerorentrepreneurswhich isalmostsimilar(44%)totheworldpopulationofcoworkers.Theriseofcoworkingspaceswasareactiontotheabsenceofanappropriateworkplaceforthisgroupofworkers.Inaddition,almost70%ofthisusergrouphasaboard/ownerpositionwhichindicatesthatalmosteverycoworkerinthatusergroupownsorcontrolsanorganization.Anotherremarkabledemographiccharacteristicisthatalmostthehalfofthecoworkersinthesampleisactiveinthesectorconsultancy(25%)orIT(21%).

68

Thetopthreemostimportantmotivations(psychographiccharacteristics)forworkinginacoworkingspaceisonthefirstplacetoseparateworkandprivatelife.Thefrontiersbetweenprivateandprofessionallifebecomesfuzzywhenworkingfromhomeandisolationanddistractionislurking.Themostmentionedmotivationinthisstudyisclearlyinlinewiththereasonwhycoworkingspacesemergedinthefirstplace.Coworkingspacesareenergeticandcreativemulti-tenantofficeswherepeoplecan interact, share,buildandco-create.Thevibrantandcreativeatmosphere in thecoworkingspace isappealing to most of the coworkers and ends in second place. Third most important motivation is the affordableaccommodation.Sincethesamplemostlyconsistsofself-employedworkers,freelancersandentrepreneursisitlikelythatthisusergrouphasdifficultiestorentanofficespaceatthebeginningoftheircareer,astheylacktherequiredcapitalorarenotcredit-qualityratedforalong-termleasecontract. InthestudiesofDeskmag(Deskmag,2012;Deskmag,2014;Deskmag,2016),itcomesforwardthatthecommunityandinteractionwithpeopleisthemostimportantmotivationtogotoacoworkingspace,whichendedinfourthplace(socialinteractionwithcoworkers)andseventhplace(thefeelingofbeingpartofacommunity)inthisstudy.OnereasonmaybethatthecoworkersintheNetherlandsfinditmoreimportanttoworkonanexternal(remote),relativelyaffordableandenergetic/creativelocationandattachlessvaluetothepresentcommunityinthecoworkingspace.Thethreemostmentionedmotivationsconcerninparticularthecharacteristicsofthecoworkingspace,andtoalesserextentthecommunitypresentinthecoworkingspace.Withtheaidofamultinomiallogitmodel,themostpreferredcharacteristicsofacoworkingspaceareestimated.Asidenote is that there ismuch heterogeneity between the respondents ormany randomerrors (faults and indifferences)becausethegoodnessoffitisslightlylessthanrequired(0,094<0.1).Theparametersofthemodel,theutilityvaluesandsignificanceof theparameterscanbe interpretedassuchbutsomecaution isadvised.Thefollowingcoworkingspacecharacteristics are themost preferred characteristics of a coworking space, in orderof importancewithbetween thebracketsthemostpreferredattributelevel:

1. Typeofleasecontract(noleasecontractorashortleasecontract(day,weekormonth));2. Accessibilityofthelocation(bycarandpublictransport);3. Layoutofthespace(half-openlayout);4. Diversityoftenants(moderateorstrongdiversity);5. Receptionandhospitality(receptionbutnohosttostimulateuserinteraction);6. Events(sometimesaneventinthecoworkingspace);and7. Atmosphereandinterioraesthetics(homey).

Locationisoftenmentionedasmostimportantcharacteristicinrealestate.Capdevila(2013)statedthatlocationisalsothemostpreferredcharacteristicofacoworkingspace.Consideringtheresultsofthisresearch,locationisnotthemostimportant characteristic of a coworking space but the type of lease contract. Location still comes in second place.Coworkersattachvaluetonocontractorashortcontractwhichincludesaleasecontractforaday,aweekoramonth.Coworkersare inneedofflexibility in leasecontractandthecurrentmarketofcoworkingspacesprovidesthatkindofflexibility,basedonthevisitedcoworkingspaces intheNetherlands.Sykes(2014)andDeijl (2011)mentionedthattheattractiveness of coworking spaces is due to the reflected flexibility and mobility, which is desirable for startingorganizations.Therentpricesareoftenlowandtheperiodofleaseisflexible.Mobilityreferstothe‘plugandplay’principleofcoworkingspacesinwhichthecoworkerscanimmediatelystartworking.Alargenumberofcoworkingspaceprovidershavecoworkingspacesatmultiplelocation.Thisgivestheusertheflexibilitytochoosewheretowork.Bizzarri(2014)andDeskmag (2012) stated thateconomic reasonsare themost important reasons for the rapidemergenceof coworkingworldwide.Thelowbarrierofoccupancyduetotherentlevelisdecisive.Deskmag(2012)foundthat47%ofrespondentsstatedthatrentcostswerethemostimportantreasonforcoworking.Theliteratureandtheresultsofthisstudyindicatethatthelowthresholdandflexibilityofthecoworkingspaces(rentprice,leaseperiod,multiplelocations)areattractivetocoworkersandcontributepositively to thepopularityof this typeofmulti-tenantoffice.Manycoworkingspaceshavespecificappearanceandgreatattentionispaidtotheinteriordesignwhichisoneoftheimportantfactorsofthecreativeandenergeticatmosphereinacoworking(Fuzi,2015).However,justlittleutilityvalueisassignedtothisattribute.Lessattentionneedstobepaidtotheappearanceofthespaceandmoreattentiontothestructureoftheleasecontract(periodandprice).Alsomoreattentionshouldbepaidtothelayoutofthespace.Thecoworkerspreferahalfopenlayoutinthecoworkingspace,whichisacombinationofworkspacesforcollaborationandclosedworkspaces.Thecoworkerspreferamoderateorstrongdiversityoftenants,somultiplesectorsoforganizationsarepreferred.Thefifthsub-questionofthismaster’sthesisisfocusedonidentificationofdifferentusergroupsinthedataofpreferredcharacteristics.With the aid of a latent class logitmodel, two latent classes are estimated containing coworkerswithcorrespondingpreferredcoworkingspacecharacteristics.Whatisstrikingisthatnosignificantparameterswerefoundintheestimationofclass1whichcouldindicatethatthisclassisstillveryheterogeneousorselectedrandomlyandhasnosignificantpreferredcoworkingspacecharacteristics.Class2hassomewhatdistinctpreferences.Thefollowingcoworking

69

spacecharacteristicsarethemostpreferredcharacteristicsofacoworkingspaceofclass2,inorderofimportancewithbetweenthebracketsthemostpreferredattributelevel:

1. Events(sometimesanevent);2. Diversityinsupplyspaces(premiumcoworkingspace);3. Accessibilityoflocation(bypublictransport);4. Diversityoftenants(strongdiversity);and5. Typeofleasecontract(nocontract).

Thecoworkersinclass2clearlywantssomeextrasinthecoworkingspace.Withtheaidofmultiplechi-squaretestandindependentsamplestest,thedifferencesbetweenthetwoestimatedclassesbasedonusercharacteristicsisdetermined.Onlyasignificantdifferenceisfoundbetweenthetwoestimatedclassesregardingtheratiovariablehoursworkinginthecoworking space. The coworkers in class 2 work significantlymore hours per week in the coworking space than thecoworkersinclass1.Thismeansthatcoworkerswhospendmoretimeinthecoworkingspacehaveacertainpreferenceforsometimesaneventinthecoworkingspace,apremiumcoworkingspace,accessibilitybypublictransport,astrongdiversityoftenantsandnorentalcontract.Apparently,thosecoworkingspacecharacteristicsbecomesmoreimportantasmoretimeisspentinthecoworkingspace.Whatisstrikingisthatthemostimportantcharacteristicofacoworkingspaceingeneral,thetypeofleasecontract,endsinlastplaceofthepreferencesinclass2.Inconclusion,twousergroupsare identifiedbasedontheirpreferencesforcoworkingspacecharacteristics.Oneusergrouphasnosignificantpreferencesforcoworkingspacecharacteristicsandtheotherusergroupcertainlywantsomeextrasinthecoworkingspace.Thisusergroupsspendsmoretimeinthecoworkingspace.

6.2 LimitationsofthestudyMultiplemethodsareusedinthisstudy.Thedataonthepreferredcharacteristicsofcoworkingspacesarecollectedwiththeaidofanattributebasedstatedchoicemethodandmultiplesurveyquestionsareaskedtocollectdataoftheusercharacteristics.Thedataonthepreferredcharacteristicsofcoworkingspacesisanalyzedwithamultinomiallogitmodel,latentclasslogitmodelandmultiplechi-squaretestsandindependentsamplesT-tests.Adamowiczetal.(1998)statedthatadesignofnineattributesisdeemedfeasible.Inthisstudy,eightattributeswereused.Duringthedistributionofthequestionnairethereweremanycomplaintsaboutthecomprehensivenessofthechoicesets,asanexampleisshownintable3.9.Thisquestionnairehasshownthatthecomprehensibilityoftheattributelevelsisevenasimportantasthenumberofattributes,perhapsevenmoreimportant.‘Only’eightattributeswereusedinthechoicesets, perhaps relatively too difficult attribute levelswere applied. Before the nine choice setswere presented to therespondent,anexplanationwasgivenoftheattributelevelsoftheattributeslayoutofthespace,diversityinsupplyspaces,reception and hospitality and diversity of tenants (seeAppendix C). These levels needed further explanation becausewithoutexplanationtheywerehardtointerpret.Especiallytheattributelevelsoftheattributediversityinsupplyspaces(basiccoworkingspace,standardcoworkingspaceandpremiumcoworkingspace)weredifficulttointerpret,accordingmultiplerespondents,withoutaclearexplanation.Perhaps,thismightbeareasonwhythisattributehasnosignificantutilityvaluesintheestimatedmultinomiallogitmodel.Whatcanbelearnedfromthis,isthattheattributelevelsshouldbeunderstandableandclearataglance.Besides the comprehensiveness of the choice sets, there is also something to note about the sequence in thequestionnaire.First,thepreferredcoworkingspacecharacteristicsarequestionedwiththeaidofanattributebasedstatedchoice method, thereafter survey questions were asked according the geographic, demographic and psychographiccharacteristicsoftherespondent.Regardingthereliabilityofthestudy(seesection3.5),thesequencewasdeliberatelychosen.Ithasbeennoticedthatthequestionnairefarmorestartedthanfinished.Forinstance,theDutchquestionnairewasmorethan400timesstartedofwhich219respondentsactuallyfinishedit.Thereisaremarkablylargediscrepancy(almost50%).Perhapstheattributebasedstatedchoicemethodhadadeterredeffectontherespondents,duetothecomplexityandcomprehensibilityofthechoicesets.Thisprobablycouldbepreventedbyfirstaskingthesurveyquestionsabouttheusercharacteristicsandthereafterthequestionsaboutthepreferredcharacteristicsofcoworkingspaces.Thequestionsabouttheusercharacteristicsoftherespondentsarelesscomplex.Inthisway,therespondentswerealreadymoreadvanced(‘warmedup’) inthequestionnaireandfinishingwouldbelesseffortbecausetherespondentsalreadyinvestedtimeinit.Whatcanbelearnfromthisisthatitmightbebettertoaskquestionswiththeuseofanattributebasedstatedchoicemethodfurtherdownthequestionnaireinordertogetahighernumberofrespondentswhichfinishesthecompletequestionnaire.

70

Another limitationaccordingtheattributebasedstatedchoicemethod istheexternalvalidity.Themethodisbasedonstatedpreferencesinwhichchoicesarehypothetically.Thequestionrisesifrespondentswouldmakethesamechoicesinrealworldsituations(revealedpreferencemethod).Inreallife,usersofcoworkingspacesperhapspreferotherattributeswhicharenotmentionedthestudyor theirchoices foraspecificcoworkingspacesarenotbasedon listedattributes.Futureresearchcouldbedonetwoseewhetherhypotheticalchoiceswillalsobemadeinrealworldsituations.Thiswouldinfluencetheexternalvalidityinapositiveway.However,itwillbeverydifficulttotestspecificattributessincerealworldsituationswillbringmoreexternalinfluences.Moreover,futureresearch,withtheuseofanattributebasedstatedchoicemethod,canalsobedonewiththeuseofothertypicalattributessincetherearemorecoworkingspacecharacteristicsidentified(seetable2.4).Mostof thedata iscollectedbypersonalapproachingpotential respondents in thecoworkingspace.Different foreignacademicscollaboratedinthecollectionofthedata.Unfortunately,thenumberofrespondentsinthosecountrieswastoolowto include inthesamplewhichresultedotherwise ina ‘contaminated’sample(possibilityofaveryheterogeneousgroups).Thelownumberofrespondentsintheforeigncountrieshadperhapsontheonehandtodowithnotapproachingthepotentialrespondentspersonally,ontheotherhandwiththedegreeofcomplexityofthequestionnaire.Whentheseaspectswouldhavebeenbetter(personalapproaching/lesscomplexquestionnaire),theprobabilityofalargernumberofrespondentswouldhavebeenhigher.Datafromdifferentcountriesarenownotincludedinthisstudyandfutureresearchcouldbedoneoncoworkingspacepreferencesinforeigncountries.Whendifferentcountriesareincludedinthesample,multiplecomparisonscanbemadebetweencountriesaccordingthepreferencesofcoworkingspacecharacteristics.Thepossibilitiesincoworkingspacesbecameclearandmultiplecoworkingspacesareanalyzedontheircharacteristics.Thevisitedcoworkingspacesinthisresearcharequitelimited,relativetothetotalnumberofcoworkingspacesintheNetherlands. Inaddition,theinterpretationoftheattributelevels iscompletedbyownconception.Overall,thevisitedcoworkingspacesscoredreallyhighsincethelowestutilityvalueis0,31whichisclearlyabovetheconstantof0.16whichreflectsthatacoworkerratherwouldworkathomeorsomewhereelse.Thelimitationinthisoverviewisthatrespondentsarelimitedandreflectmainlytotheirownenvironment.Thisgivessomewhatadistortedviewofthecoworkingspacesbecausethevisitedspaceslieveryclosetothemostpreferredcharacteristicsresultingfromtheestimation.The coworkerswhich spendmorehours perweek in a coworking space attach value to certain characteristics.Othersignificantdifferenceswerenotfoundbetweentheestimatedusergroups.Possiblyotherusercharacteristicshavetobeapplied inordertofindsignificantrelationsbetweentheestimatedusergroups.Futureresearchcouldbedone inthecontext of other geographic, demographic and psychographic characteristics of coworkers in order to find possiblerelations.Whenotherusercharacteristicsareasked,possiblymoredifferencesbasedonusercharacteristicscouldbefound.Futureresearchcouldalsobedoneonusercharacteristicsofcoworkersinothercountriesovertheworld,onanationallevel,sincethiskindofresearchisquitelimited.Withalargersamplesize,morerelationsbetweenusercharacteristicsanduserpreferencescouldhavebeenexplored.Alarger samplesize,especially in theNetherlands, shouldhave increased thevalidityof thestudyand thepossibility togeneralizethesampletothewholepopulation.Thesampleofthismaster’sthesisistoosmall(219)tosaysomethingaboutthetotalpopulationofcoworkersintheNetherlands.Itisverylikelythatalargersamplesizewouldalsogeneratebettermodels(multinomiallogitmodelandlatentclasslogitmodel)andthereforbetterconclusions.Thegoodnessoffitofthemultinomiallogitmodelscoredtoolow(0.094<0.01),whichindicatedthatthereismuchheterogeneityormanyrandomerrors(faultsandindifferences)betweentherespondents.

6.3PracticalimplicationsTomeet the future demand, real estatemarketsmust respond to trends in order to prevent a continuously growingmismatchbetweenofficespacesupplyanddemand.Intheintroductionofthismaster’sthesiswasstatedthatthisstudyisrelevantforprovidersanddevelopersofcoworkingspaces.Thepreferredcoworkingspacecharacteristicsbecameclearandtheworkenvironmentcanbeadjustedtotheuserpreferencesresultingfromtheresearch.Inthisway,thecoworkingspace can be developedmore effectively. Providers and developers of coworking spaces canmakemore thoughtfuldecisionsaboutthedevelopingandestablishingofcoworkingspaces,forbothphysicalandnon-physicalcharacteristics.Themorepreferredattributes,derivedfromthisresearch,areappliedinthecoworkingspace,themorevaluecoworkersattachtothecoworkingspaceandthecloseritcomestothe‘idealcoworkingspace’.Locationisoneofthemostimportantcharacteristicsofrealestateingeneral.However,ithasbecomeclearthatthetypeofleasecontractisamorepreferredcharacteristic regardingcoworkingspaces.Coworkersattachvalue tonocontractora shortcontract,which reflectsa

71

rentalperiodofaday,aweekoramonth.Theliteratureandtheresultsofthisstudyindicatethatthelowthresholdandflexibilityofthecoworkingspaces(rentprice,leaseperiod,multiplelocations)areattractivetocoworkersandcontributepositivelytothepopularityofthistypeofmulti-tenantoffice.Inordertoattractmorecoworkers,itisofimportancetopayattentiontothestructureoftheleasecontract,preferablynocontract.Noleasecontractisperhapsnoteasyachievableforcoworkingspaceoperatorssincenorentalincomewillbederived.Anaccessibleshort-termcontractisalsostronglypreferred.InthepresentmarketofcoworkingspacesintheNetherlands,thisleasestructureisalreadywellimplemented.Alongrentalcontract(yearorlonger)goesagainstthe(flexibility)principlesofcoworkingspacesandshouldnotbeapplied.Whenacoworkingspaceisaccessiblebycarandpublictransport,mostutilityisassigned.Forthecoworkingspaceswhichare located in the inner cities, it could be difficult to offer both forms of accessibility due to little place to park anddifficult/poor accessibility. The practical implication regarding the location is that it is key to be accessible by publictransport.Evenasalongrentalcontract,accessibilitybycarisagainsttheprinciplesofcoworkingspaces.Locationandtherentalcontractarethemostimportantfactorsonwhichcoworkingspaceshavetofocus.

With regard to the layout of the space, the coworkers prefer to have a half-open layout. The practicalrecommendationisthatdevelopershavetomakeacombinationofcollaborativespacesandconcentrationspaces(closedspaces)wherecoworkersareabletowithdraw.Itisofimportanceforoperatorstoofferamoderateorstrongdiversityoftenants.No diversity of tenants, inwhich coworkers are active in only one sector is clearly not preferred. Therefore,developingacoworkingspaceforaspecifictargetgroup,regardingsectoroforganization,isnotappealingfortenants.Themore variation in sectors, thebetter. Regarding the literature, the coworkinghostplays an important role in thecollaborationbetweencoworkers.Thestudyshowedthatthisisnotparticularpreferredandonlyareceptionissufficient.Inmostofthecoworkingspaces intheNetherlands,anonlineorganizationalplatformisappliedwherecoworkerscanmeetandaskquestions.Apparently,suchplatformmakesthecoworkinghostunnecessaryinconnectingcoworkers.Theresultsshowedthatcoworkersprefertohavesometimesanevent,butsincethischaracteristic finished innextto lastplace,nottoomuchattentionneedtobepaidontheattributeevents.Finally,alotofattentionispaidtotheatmosphereandinterioraestheticsincoworkingspaceswhilelittleutilityvalueisassignedtothisattribute.Thisattributeendedinlastplaceofthepreferredcharacteristics,coworkingspaceprovidershavetofocuswaylessontheappearanceofthespace.This research also distinguished two classes of respondents with equally preferred coworking space characteristics.Coworkers who spend more time in the coworking space have a certain preference for sometimes an event in thecoworkingspace,apremiumcoworkingspace,accessibilitybypublictransport,astrongdiversityoftenantsandnorentalcontract. Apparently, those coworking space characteristics becomes more important as more time is spent in thecoworkingspace.Thepracticalimplicationfortheprovidersanddevelopersofcoworkingspacesisthattheyhavetofocusonthosecharacteristicstoattractandsatisfythecoworkerswhichspendrelativelymoretimeinthecoworkingspace.This research is also of interest to current and future users of coworking spaces. Multiple coworking spaces in theNetherlandsareanalyzedbasedonmultiplecoworkingspacecharacteristics.Usersandfutureuserscandeterminewhichcoworkingspacecharacteristicssuitstheminordertodecidewheretheywanttowork.Thishelpsthecoworkerstosearchmorespecificwhatfitstheirpersonalneeds.Furthermore,thisstudyprovidesanoverviewoftheuser’smotivationstoworkinacoworkingspace(seetable2.2).ThepracticalimplicationfortheusersandfutureusersisthatthemajorityofthecoworkersintheNetherlandsattachvaluetoanexternal(remote),relativelyaffordableandenergetic/creativelocationandattachlessvaluetothepresentcommunity.Theusersknowwhattheaveragecoworkers’intentionsaretousethistypeofmulti-tenantofficeinordertoseeifitcorrespondstotheirpersonalneedsandrequirements.

72

73

Referencelist Adamowicz,W.,Louviere,J.,&Swait,J. (1998). IntroductiontoAttribute-BasedStatedChoiceMethods.Retrievedfrom

https://www.researchgate.net/profile/Jordan_Louviere/publication/242494897_Introduction_to_Attribute-Based_Stated_Choice_Methods/links/0046352a511270551c000000.pdf

Ahmad,R. (2003).BenefitSegmentation:Apotentiallyuseful techniqueofsegmentingandtargetingolderconsumers.

InternationalJournalofMarketResearch,45(3),337-388.Appel-Meulenbroek,H.A.J.A.(2013).HowtomeasureaddedvalueofCREandbuildingdesign(Dissertation).Retrieved

fromhttps://pure.tue.nl/ws/files/3800554/762833.pdfBaarda, D. B., & De Goede, M. P. M. (2006). Basisboek Methoden en Technieken. Groningen, Nederland: Noordhoff

UitgeversB.V.Barber,C.,Laing,A.,&Simeone,M.(2005).Globalworkplacetrends:ANorthAmericanandEuropeancomparison.Journal

ofCorporateRealEstate,7(3),210-221.Berg,J.vanden.&M.Stijnenbosch(2009).Atlasbedrijvencentra:eeninventarisatievanbedrijvencentrametvoorzieningen.

Stogoonderzoek+Advies.Binge,E.,Gnoth,J.&Andreu,L.(2008),“Advancedtopicsintourismmarketsegmentation”,inWoodside,A.andMartin,

D.(Eds),TourismManagementAnalysis,Behavior,andStrategy,Elsevier,Barking,Vol.3,129-150.Bizzarri,C. (2014).TheemergingPhenomenonofCoworking.ARedefinitionofJobMarket inNetworkingSociety, inK.

Müller,S.RothandM.Žák,eds.,SocialDimensionofInnovation.Prague:Lindenakladatelstvi.Borgers,A.W.J.,&Vosters,C.(2010).AssessingpreferencesforamegashoppingcentreintheNetherlands:Aconjoint

measurementapproach.InProceedingsoftheEuropeanInstituteofRetailingandServicesStudiesconference(RASS).(pp.21-).EIRASS.

Budie,B.(2016).Theemployeeinthemodernworkenvironment:aneed-basedapproachtodeterminetheeffectofpersonal

characteristicsonsatisfactionwiththephysicalworkenvironmentinconventionalandactivity-basedoffices(Masterthesis).Retrievedfromhttp://alexandria.tue.nl/extra2/afstversl/bwk/Budie_2016.pdf

BusinessCentreCapitalCompanyLtd.(2002).Theperformanceofbusinesscentresoveracompletebusinesscycle.Londen,

England:BusinessCentreCapitalCompanyLtd.Calder,M.,&Courtney,S.(1992).Businesscenters:theUKpicture.PropertyManagement,10(2),106-117.Capdevila, I. (2013).Knowledgedynamics in localizedcommunities: coworking spacesasmicroclusters.Accessed from

http://papers.ssrn.com/sol3/papers.cfm?abstract_id=241412Capdevila, I. (2015).Co-workingspacesand the localiseddynamicsof innovation inBarcelona. International Journalof

InnovationManagement,19(3),1-25.CBS. (2013, October 3). Onderwijsniveau bevolking gestegen. Retrieved from https://www.cbs.nl/nl-

nl/nieuws/2013/40/onderwijsniveau-bevolking-gestegenChrzan,K.(1994).ThreeKindsofOrderEffectsinChoice-BasedConjointAnalysis.MarketingLetters,2,165-172.Cirillo,C.(n.d.).Statedpreferencemethod[College].Retrievedfromhttp://ocw.nctu.edu.tw/course/ests021/Lect7.pdfColleoniE.andA.Arvidsson(2014).Knowledgesharingandsocialcapitalbuilding.Theroleofco-workingspacesinthe

knowledgeeconomyinMilan’,UnpublishedReport,OfficeforYouth,MunicipalityofMilan.

74

Danielis,R.,&Rotaris,L.(1999).Analysingfreighttransportdemandusingstatedpreferencedata:asurveyandaresearchprojectfortheFriuli-VeneziaGiuliaRegion.TransportiEuropei,13(1),30-38.

Davies,A.,&Tollervey,K.(2013).Thestyleofcoworking.NewYork,UnitedStatesofAmerica:Prestel.Deijl,C.M.(2011).Twoheadsarebetterthanone.Acasestudyoftheco-workingcommunityinTheNetherlands(master

thesis).Rotterdam,Nederland:ErasmusUniversityRotterdam.Deskmag.(2015,04June).Thehistoryofcoworking.Accessedfromhttp://www.tiki-toki.com/timeline/entry/156192/The-

History-Of-Coworking-Presented-By-DeskmagDeskmag.(2012).TheSecondAnnualCoworkingSurvey.Accessedfromhttp://reseau.fing.org/file/download/128857Deskmag. (2013). The 3th Global Coworking Survey. Accessed from

https://communityjelly.files.wordpress.com/2012/11/3rdglobalcoworkingsurvey-121108034918-phpapp02.pdfDeskmag.(2014).The4thGlobalCoworkingSurvey.Accessedfromhttps://www.youtube.com/watch?v=hutCeXQ9Z6YDeskmag.(2015).The5thGlobalCoworkingSurvey.Accessedfromhttp://www.deskmag.com/en/first-results-of-the-new-

global-coworking-survey-2015-16Deskmag. (2005, August 9). Co-Working: Independent Workers Unite [Video file]. Retrieved from http://www.tiki

toki.com/timeline/entry/156192/The-History-Of-Coworking-Presented-By-Deskmag#vars!panel=1505040DTZ Zadelhoff. (2015). Nederland Compleet. Geraadpleegd van

https://www.dtz.nl/media/308927/dtz%20nl%20compleet%202015%20nl.pdfEeuwijk, J. J. H. (2010). Leren over parkeren (Master thesis). Retrieved from

http://alexandria.tue.nl/extra2/afstversl/bwk/693402.pdfErler, J. (2010).DasKonzeptCoworking.EineempirischeAnalysederNutzungvonCoworking-Zentren (master thesis).

Chemnitz,Germany:TechnischeUniversitätChemnitz.Everitt,B.S.,Landau,S.,&Leese,M.(2001).ClusterAnalysis.NewYork,UnitedStatesofAmerica:ArnoldPublishers.Ewer, T. (n.d.). The Difference Between Freelancers, Solopreneurs, and the Self-Employed. Retrieved from

https://blog.bidsketch.com/everything-else/freelancer-solopreneur-self-employed/Ferguson,J.(2013,July23).Coworkingvalue:Sustainability.Retrievedfromhttp://www.coworkfrederick.com/coworking-

value-sustainability/Fiebig,D.G.,Keane,M.P.,Louviere,J.,&Wasi,N.(2010).TheGeneralizedMultinomialLogitModel:AccountingforScale

andCoefficientHeterogeneity.MarketingScience,29(3),393-421.Florida,R. (2002)Theriseof thecreativeclass.Andhow it’s transformingwork, leisure,communityandeveryday life.

Cambridge,England:BasicBooks.Fuzi,A.(2015).Coworkingspacesforpromotingentrepreneurshipinsparseregions:thecaseofSouthWales.Regional

studies,regionalscience,2(1),462-469.Fuzi,A.,Clifton,N.,&Loudon,G.(2014).Newin-houseorganizationalspacesthatsupportcreativityandinnovation:the

co-working space. Paper presented at R&D Management Conference, Stuttgart, Germany. Retrieved fromhttps://www.academia.edu/7430165/New_inhouse_organizational_spaces_that_support_creativity_and_innovation_the_co-working_space

Gandini,A.(2015).Theriseofcoworkingspaces:Aliteraturereview.Ephemera:theory&politicsinorganization,15(1),

193-205.

75

Genevieve, G. V., & Tang, A. I. (2011). Working in the ''UnOffice''. San Fransisco, United States of America:NightOwlsPress.com.

Ghose, S. (n.d.). Marketing segmentation: definition, criteria and other details. Retrieved from

http://www.yourarticlelibrary.com/marketing/marketing-segmentation-definition-criteria-and-other-details/50877/Gibson,V.A.,&Lizieri,C.M.(1999).Theroleofservicedofficespaceinofficemarketsandcorporatepropertyportfolios.

Reading,England:TheUniveristyofReading.Goyat, S. (2011). The basis ofmarket segmentation: a critical review of literature. European Journal of Business and

Management,3(9),45-55.Green,R.(2014).Collaborateorcompete:Howdolandlordsrespondtotheriseincoworking?CornellRealEstateReview,

12,52-59.DeGroot,M.,VanDijk,V.,&Pijman,R.(n.d.).Watzijndeverschillendetussenfreelancer,zzp'ereneenmanszaak?Retrieved

fromhttp://www.lancelots.nl/starten/veelgestelde-vragen/freelancer-versus-eenmanszaakGuillet,B.D.,&Kucukusta,D.(2016).Spamarketsegmentationaccordingtocustomerpreference.InternationalJournal

ofContemporaryHospitalityManagement,28(2),418-434Hamilton, M., Lim, L. C., & McCluskey, W. (2006). The changing pattern of commercial lease terms. Property

Management,24(1),31-46.Hartog,L.M.(2015).Usersatisfactioninmulti-tenantoffices(masterthesis).Eindhoven,Nederland:TUEindhoven.Hensher,D.A., Rose, J.M.,&Greene,W.H. (2015).AppliedChoiceAnalysis. Cambridge,UnitedKingdom:Cambridge

UniversityPress.Hillman,A.(2008,November7).6ReasonstoStartCoworking.Retrievedfromhttp://mashable.com/2008/11/07/reasons-

to-start-coworking/#TZYAZkeDzEqAHillman,A.(2011,October11).Coworkingcore-values.Retrievedfromhttp://dangerouslyawesome.com/tag/core-values/Holmes,T.P.,&Boyle,K. J. (2005).DynamicLearningandContext-Dependence inSequential,Attribute-Based,Stated-

PreferenceValuationQuestions.LandEconomics,81(1),114-126.Howells,J.(2012).Thegeographyofknowledge:Neverso,closebutneversofarapart.JournalofEconomicGeography,

12(1),1003-1020.Hurst,A.(2014).Thepurposeeconomy.Boise,UnitedStatesofAmerica:Elevate.Institution For Digital Research and Education. (n.d.). What is effect coding? Retrieved from

http://www.ats.ucla.edu/stat/mult_pkg/faq/general/effect.htmJansen,J.(2009).Segmentatievankantoorgebruikersopbasisvanbedrijfsstijl.Amserdam,Nederland:AmsterdamSchool

ofRealEstate.Jones,D.,Sundsted,T.,&Bacigalupo,T.(2009).I'mouttahere!.Austin,UnitedStatesofAmerica:MBAPress.Kalla,S.(2009,March01).StatisticallySignificantResults.Retrievedfromhttps://explorable.com/statistically-significant-

resultsKaufman,L.,&Rousseeuw,P.J.(2005).FindingGroupsinData:AnIntroductiontoClusterAnalysis.NewJersey,United

StatedofAmerica:WileyInterscience.

76

Kemperman,A.(2016,March4).Researchmethods:DiscretechoicemodelsIIStatedChoiceExperiments[College-slides].Retrievedfromhttp://oase.tue.nl

Kemperman, A.D.A.M. & Timmermans, H.J.P. (2008). Influence of socio-demographic and residential environment on

leisureactivityparticipation.Leisuresciences,30(4),306-324.Kerkman,K.(2010).Routepreferencesofcardriversandtheuseofin-carnavigationsystems:aStatedChoiceexperiment

(Masterthesis).Retrievedfromhttp://alexandria.tue.nl/extra2/afstversl/bwk/711203.pdfKetting, J. R. (2014). Het bedrijfsverzamelgebouw: een onderzoek naar de toegevoegde waarde van

bedrijfsverzamelgebouwen(masterscriptie).Delft,Nederland:TUDelft.Kløjgaard,M.A.,Bech,M.,&Søgaard,R.(2012).DesigningaStatedChoiceExperiment:TheValueofaQualitativeProcess.

JournalofChoiceModelling,5(2),1-18.Kojo, I.,&Nenonen, S. (2014).Userexperience inanacademic coworkingplace: the caseofAaltoUniversity’sdesign

factory.InP.A.Jensen(Red.),CIBFacilitiesManagementConference2014(pp.341-353).Lyngby,Denmark:TechnicalUniversityofDenmark.

Kojo,I.,&Nenonen,S.(2016).Typologiesforco-workingspacesinFinland.Facilities,34(5/6),302-313.Kónya, V., Matic, D., & Pavlovic, J. (2016). The Influence of Demographics, Job Characteristics and Characteristics of

OrganizationsonEmployeeCommitment.ActaPolytechnicaHungarica,13(3),119-138.Kotler,P.(1994).Marketingmanagement:analysis,planning,implementation,andcontrol.EnglewoodCliffs,NewJersey,

UnitedStatesofAmerica:PrenticeHall.Kwiatkowski,A.,&Buczynski,B.(2011).Coworking:BuildingCommunityasaSpaceCatalyst.FortCollins,Colarado,USA:

CohereCoworking.Kwiatkowski,A.,&Buczynski,B.(2011).Coworking:HowFreelancersEscapetheCoffeeShopOffice.FortCollins,Colarado,

USA:CohereCoworking.VanLaarhoven,W.(2016).Consumers'preferencesregardingdepartmentstores.Eindhoven,Nederland:TUEindhoven.Laterveer,M.(2011).ServicedOffices;eendynamischemarktinopkomst(masterscriptie).Utrecht,Nederland:Universiteit

Utrecht.Leesman. (2013). 100,000+ A workplace effectiveness report. Retrieved from http://leesmanindex.com/wp-

content/uploads/Leesman-100k-data-report-MAR16-INT-SP1.pdfLeforestier,A.(2009)‘Theco-workingspaceconcept’.Lesáková,L.(2012).TheroleofbusinessincubatorsinsupportingtheSMEStart-up.ActaPolytechnicaHungarica,9(3),85-

95.Lew, D. K. (2011). Overview of Stated Preference Methods. Retrieved from

https://www.st.nmfs.noaa.gov/Assets/economics/publications/RecFisheriesDataandModel/Overview%20of%20Stated%20Preference%20Models_Lew.072611.pdf

Liebregts,M.M.M.(2013).Themodernworkenvironmentindifferentcountries.Eindhoven,Nederland:TUEindhoven.Louviere,J.J.,Flynn,T.N.,&Carson,R.T.(2010).DiscreteChoiceExperimentsAreNotConjointAnalysis.JournalofChoice

Modelling,3(3),57-72.Louviere, J. L., Hensher, D. A., & Swait, J. D. (2000). Stated ChoiceMethods. Cambridge, United Kingdom: Cambridge

UniversityPress.

77

Nijënstein,S.(2012).Determiningtheroleofvaluesinstudents’housingchoicebehaviourwithlatentclassandmixedlogitconjoint analysis methods. (Master thesis). Retrieved fromhttp://alexandria.tue.nl/extra2/afstversl/tm/Nijenstein_2012.pdf

Newport,C.(2016).DeepWork.NewYork,UnitedStatesofAmerica:LittleBrown.McAllistar,P.(2002).Fromrentstorevenues:Canpropertybecomeaserviceindustry?Reading,England:TheUniversityof

Reading.Meel,J.van&Vos,P.(2001).Funkyoffices:Reflectionsonofficedesigninthe'neweconomy'.JournalofCorporateReal

Estate,3(4),322-334.Mensen,A.H.H.M.,&VanRijt-Veltman,W.V.M. (2005).MKB-locaties: onderzoeknaardeaard en kwaliteit vande

bedrijfslocatiesvanMKB-ondernemingen.DenHaag,Nederland:AtelierRijksbouwmeester.Merkel,J.(2015).Coworkinginthecity.Ephemera:theory&politicsinorganization,15(1),121-139.Moriset,B.(2013).Buildingnewplacesofthecreativeeconomy.Theriseofcoworkingspaces.Proceedingsofthe2nd

GeographyofInnovationInternationalConference2013,UtrechtUniversity.Moscardo,G.,Pearce,P.,&Morrison,A.(2001).EvaluatingDifferentBasesforMarketSegmentation.JournalofTravel&

TourismMarketing,10(1),29-49.Odhiambo,F. (2013,August10).Differencebetween inputs, activities,outputs,outcomesand impact.Retrieved from

https://evaluateblog.wordpress.com/2013/06/10/difference-between-inputs-activities-outputs-outcomes-and-impact/

Oldenburg,R.(1989).Thegreatgoodplace:Cafés,coffeeshops,bookstores,bars,hairsalonsandtheotherhangoutsat

theheartofacommunity.Massachusetts,USA:Cambridge.Orel, M. (2015). Working in co working spaces: the social and economic engagement of European youth. In Youth

Partnership(Ed.),Perspectivesonyouth-vol2-Connectionsanddisconnection(pp.133-139).Strasbourg,Germany:CouncilofEuropePublishing.

Parinno,L.(2015).Coworking:assessingtheroleofproximityknowledgeexchange.KnowledgeManagementResearch&

Practice,13,261-271.Peltier,S.(2001).Analysisofthesupplyofservicedofficespace(masterthesis).NewJersey,USA:

NewJerseyInstituteofTechnology.Remøy,H.T. (2010).OutofOffice:Astudyonthecauseofofficevacancyandtransformationasameanstocopeand

prevent (Dissertation). Retrieved fromhttps://books.google.nl/books?id=_Toa4ag6rEcC&pg=PA32&lpg=PA32&dq=mismatch+supply+demand+office+space&source=bl&ots=Hx2kCh_jIY&sig=HlPwO_0GQrnr3d0xZs_I_ZqwDAw&hl=nl&sa=X&ved=0ahUKEwizg5rr_YvOAhWMcRQKHUl5By4Q6AEIJzAB#v=onepage&q=mismatch&f=false

Rohlf, J. (2011, December 23). Coworking Core Values – Openness. Retrieved fromhttp://thearmoryworkspace.com/blog/coworking-core-values-%E2%80%93-opennessPohler, N. (2011, August 22). Coworking 101: A new definition [Blog post]. Retrieved from

http://www.deskmag.com/en/coworking-spaces-101-a-new-definitionRothe, P., Lindholm, A., Hyvönen, A., & Nenonen, S. (2011). User preferences of office occupiers: investigating the

differences.JournalofCorporateRealEstate,13(2),81-97.Rovers,C.J.T.(2016).Satisfactionofend-userswiththephysicalworkenvironmentinmulti-tenantofficebuildings(master

thesis).Eindhoven,Nederland:TUEindhoven.

78

Sanko,N.(2001).GuidelinesforStatedPreferenceExperimentDesign(Dissertation).Retrievedfromhttp://www.b.kobe-u.ac.jp/~sanko/pub/Sanko2001_1.pdf

Sapsford,R.(1999).Surveyresearch.London,England:SagePublications.Saurin,R.,Ratcliffe, J.,&Puybaraud,M. (2008).Tomorrow'sworkplace:a futuresapproachusingprospective through

scenarios.JournalofCorporateRealEstate,10(4),243-261.Shuttleworth,M.(2008,October20).ValidityandReliability.Retrievedfromhttps://explorable.com/validity-and-reliabilitySmith,E.(1995).CreatingProductiveOrganizations.BocaRaton,Florida,UnitedStatesofAmerica:CRSPress.Somsen,D.J.(2002).TheInstantOffice:eenverkenningvanbusinesscenters.Amstelveen,Nederland:SBV-MRE.Spinuzzi, C. (2012). Working alone together: Coworking as emergent collaborative activity. Journal of Business and

TechnicalCommunication,29(4),399-441.Spreitzer,G.,Bacevice,P.,&Garrett,L.(2015).Whypeoplethriveincoworkingspaces.HarvardBusinessReview,93(9),

28-35.Suárez,R.(2015).Hetcoworkinghandbook:theguide.Washington,UnitedStatesofAmerica:SerendipityAcceleratorTM.Sykes,K.(2014).Coworking:aworkplaceparadigmshift.Contract,55(6),140-145.Swait, J. (1994). A structural equation model of latent segmentation and product choice for cross-sectional revealed

preferencechoicedata.JournalofRetailingandConsumerServices,1(2),77-89.Thecoworkingwiki.(2013,September09).Historyofcoworking-atimeline[Forumpost].RetrievedJune01,2016,from

http://wiki.coworking.org/w/page/68852527/History%20of%20Coworking%20-%20a%20timelineThe Fact Club. (n.d.). De Nederlandse Arbeidsmarkt. Retrieved December 4, 2016, from http://thefactclub.nl/fact-

reports/arbeidsmarkt/Train,K.(2002).DiscreteChoiceMethodswithSimulation.Cambridge,UnitedKingdom:CambridgeUniversityPress.VanSusante,P.J.M.(2014).Differencesinemployeesatisfactioninnewversustraditionalworkplaceenvironments(master

thesis).Eindhoven,Nederland:TUEindhoven.VandeVoordt,T.,Jensen,P.A.,Hoendevanger,J.G.,&Bergsma,F.(2016).Astep-by-stepplantomanageandmeasure

addingvaluebyFM/CREM. InS.BalslevNielsen,&P.A. Jensen(Eds.),EuroFMResearchPapers2016 (pp.40-50).Lyngby,Denmark:PolytekniskForlag.

Vermunt, J. K., & Magidson, J. (2002). Latent Class Cluster Analysis. Geraadpleegd van

http://members.home.nl/jeroenvermunt/hagenaars2002b.pdfVries, N. E. de, & Besselaar, R. van der. (2013). Alles flex, is dat een mismatch? Accessed from

http://stadsregio.nl/sites/stadsregio.nl/files/pagina-bestanden/Weijs-Perrée,M.,Appel-Meulenbroek,H.A.J.A.,DeVries,B.,&Romme,A.G.L.(2016).DifferencesbetweenbusinesscenterconceptsintheNetherlands.PropertyManagement,34(2),100-119.Weijs-Perrée,M.,&Appel-Meulenbroek,H.A.J.A.(2015).Verschillentussentypenbedrijfsverzamelgebouwen.RealEstate

ResearchQuarterly,14(3),23-32.

79

AppendicesAppendixA-Valuedaspectsinmulti-tenantoffices

80

AppendixB–Codebookofquestionnaire

Geographicvariable Levelofmeasurement Typeofquestion TypeofitemCountrycoworkingspace Nominal Multiplechoice(64) 1;TheNetherlands,2;Afghanistan,3;America4;Australia,5;

Belgium,6;Brazil,7;Bulgaria,8;Canada,9;China,10;Cyprus,11;Denmark12;Germany,13;England,14;Eritrea,15;Estonia,16;Philippines,17;Finland,18;France,19;Ghana,20;Greece,21;Hongary,22;Ireland,23;India,24;Indonesia25;Italy,26;Iraq,27;Iran,28;Japan,29;Korea,30;Croatia,31;Luxembourg,32;Norway,33;Nigeria,34;Malta,35;Marocco,36;Austria,37;Pakistan,38;Poland,39;Portugal,40;Romania,41;Russia,42;Thailand,43;CzechRepublic,44;Turkey,45;Schotland,46;Slovenia,47;Slovakia,48;Suriname,49;Spain,50;South-Africa,51;Sweden,52;Other..,53;Argentinia,54;Colombia,55;Egypt,56;Serbia,57;Singapore,58;Switzerland,59;Tunisea,60;UnitedArabEmirates,61;Vietnam,62;Vietnam,63;Lithuania,64;Albania

Demographicvariables Levelofmeasurement Typeofquestion TypeofitemGender Nominal Multiplechoice(2) 1;Male,2;FemaleAge Ratio OpenquestionNationality Nominal Multiplechoice(64) 1;TheNetherlands,2;Afghanistan,3;America4;Australia,5;

Belgium,6;Brazil,7;Bulgaria,8;Canada,9;China,10;Cyprus,11;Denmark12;Germany,13;England,14;Eritrea,15;Estonia,16;Philippines,17;Finland,18;France,19;Ghana,20;Greece,21;Hongary,22;Ireland,23;India,24;Indonesia25;Italy,26;Iraq,27;Iran,28;Japan,29;Korea,30;Croatia,31;Luxembourg,32;Norway,33;Nigeria,34;Malta,35;Marocco,36;Austria,37;Pakistan,38;Poland,39;Portugal,40;Romania,41;Russia,42;Thailand,43;CzechRepublic,44;Turkey,45;Schotland,46;Slovenia,47;Slovakia,48;Suriname,49;Spain,50;South-Africa,51;Sweden,52;Other..,53;Argentinia,54;Colombia,55;Egypt,56;Serbia,57;Singapore,58;Switzerland,59;Tunisea,60;UnitedArabEmirates,61;Vietnam,62;Vietnam,63;Lithuania,64;Albania

Education Ordinal Multiplechoice(9) 1;Noeducation/elementaryschool,2;Preparatorysecondaryvocationaleducation,3;Seniorgeneralsecondaryeducation,4;Pre-universityeducation,5;Intermediatevocationaleduction,6;Highervocationaleducation,7;University(bachelor),8;University(master),9;University(PhD)

Usergroup Nominal Multiplechoice(5) 1;Self-employedworker,freelancerorentrepreneur,2;Employeeofacompany(2-10employees),3;Employeeofacompany(11-50employees),4;Employeeofacompany(morethan50employees),5;Student

Positioninorganization Nominal Multiplechoice(5) 1;Supportingstaff(deskattendant,receptionistetc.),2;Regularemployee,3;Manager,4;Board/owner,5;Doesnotapply

Sectororganization Nominal Multiplechoice(12) 1;Consultancy,2;Design,3;Commerce,4;IT,5;Art,6;Management,7;Research,8;Education,9;Projectmanagement,10;PR,marketing,sales,advertising,communication,11;Writing,12;Other

Income Ordinal Multiplechoice(6) 1;Lessthan20000ayear,2;20001-30000ayear,3;30001-500001ayear,4;40001-50000ayear,5;Morethan50000ayear,6;Idon'tknow/I'drathernotsay

Hoursincoworkingspace Ratio OpenquestionTransportcoworkingspace Nominal Multiplechoice(4) 1;Car,2;Bike,3;Byfoot,4;Publictransport

81

Psychographicvariables Levelofmeasurement Typeofquestion TypeofitemMotivation1,2and3 Nominal Rankingquestion(10) 1;Affordableaccommodation,2;Thefeelingofbeingpartofa

community,3;Vibrantandcreativeatmosphereinthecoworking

space4;Professionalappearanceforthecompany,5;

Professionalsupportiveworkenvironment(supportservices),6;

Theopportunitytonetworkwithcoworkers(possiblenew

projects),7;Socialinteractionwithcoworkers,8;Thepossibil ity

forwork-relatedconversationswithothercoworkers(knowledge

sharing,knowledgecreation),9;Iwaslookingforaworkplace

outsidethehome(separatingworkandprivatelife),10;Flexibility

(rentalperiod,numberofsquaremeters)

82

AppendixC-Questionnaire(English,Dutch,German,Italian)

GraduationresearchoncoworkingspacesDearuserofacoworkingspace,In the context ofmy thesis for theMaster Real EstateManagement & Development at the Technical University inEindhovenI’manalyzingpreferredcharacteristicsofusersofcoworkingspaces.Thesurveyasksontheonehandforyourpreferredcharacteristicsconcerningcoworkingspacesbyusingninechoicesetsandontheotherhandforyourpersonalcharacteristics.Thesurveyshouldonlytake5to10minutes,andyourresponsesare completely anonymous. If you have questions regarding the survey or the questionnaire, please contactviajaspervandekoevering@gmail.com.Thankyouinadvance!Sincerely,JaspervandeKoeveringINTRODUCTIONCHOICESETSThefirst9questionseachshowthreedifferentalternativecoworkingspaceswithregardtotheircharacteristics.Youareaskedtochooseineachcasewhichcoworkingspaceyouwouldprefertoworkat.Ifyouwouldprefertoworkathome,ratherthaninanyofthethreealternatives,pleasechoose"noneoftheseoptions."All9questions/pagesareindependentofeachother,soeachtimeindicatethemostappealingalternativeforyou.Anexampleofthesequestionsisshownhere:EXAMPLECHOISESET

*Layoutofthespace*:

• Openlayout(largeopenspaces)• Half-openlayout(combinationofopenspacesandconcentrationrooms)• Closedlayout(enclosedandseparatespaces)

**Diversitysupplyspaces**:

o Basiccoworkingspace(collaborativeworkspace+meetingroomsandkitchenarea)o Standardcoworingspace(collaborativeworkspace+meetingrooms+kitchenarea+eventspacesandinformal

zones)o Premiumcoworkingspace(collaborativeworkspace+meetingrooms+kitchenarea+eventspaces+informal

zones+fitnesscenterandbar)

***Receptionandhospitality***:o Noreceptionandnohosto Receptionbutnohosto Receptionandactivehost(activecoworkinghostthatconnectscoworkerstoeachother)

83

****Diversityoftenants****:o Nodiversityoftenants(tenantsinthesamesector)o Moderatediversityoftenants(tenantsinafewbusinessfields)o Strongdiversityoftenants(alotofdifferentbusinessfieldspresentinthecoworkingspace)

Nowninechoicesetswillfollow(oneexamplepresentedbelow)

THEFOLLOWING4QUESTIONAREABOUTPERSONALCHARACTERISTICSWhatisyourgender?

o Maleo Female

Whatisyourage(inyears)?Whatisyournationality?

o AfghanAlbanianAmericanArgentinianAustralianAustrianBelgianBrazilianBulgarianCanadianChineseChechCypriotCroatianColombianDanishDutchGermanEgyptianEnglishEritreanEstonianFilipinoFinnishFrenchGhanaianGreekHongarianIrishIndianIndonesianItalianIraqiIranianJapaneseKoreanLithuanianLuxembourgishNorwegianNigerianMalteseMaroccanPakistaniPolishPortugeseRomarianRussianScottishSerbianSingaporeanSlovenianSlovakSurinameseSpanishSouth-AfricanSwedishSwissThaiTunisianTurkishUnitedArabEmiratesVietnameseOther

Whatisyourhighestdegree?

o Noeducation/elementaryschoolo Preparatorysecondaryvocationaleducationo Seniorgeneralsecondaryeducationo Pre-universityeducationo Intermediatevocationaleducationo Highervocationaleducationo University(bachelor)o University(master)o University(PhD)

84

THEFOLLOWING4QUESTIONSAREABOUTWORKRELATEDCHARACTERISTICSWhichofthefollowingdescriptionsbestrepresentsyoursituation?

o Self-employedworker,freelancerorentrepreneuro Employeeofacompany(2-10employees)o Employeeofacompany(11-50employees)o Employeeofacompany(morethan50employees)o Student

Whatisyourcurrentpositioninyourorganization?

o Supportingstaff(deskattendant,receptionistetc.)o Regularemployeeo Managero Board/ownero Doesnotapply

Inwhichsectordoesyourcompanyoperatesmost?

o Consultancy(legaladvice,organizationaladviceetc.)o Design(graphic,web,product,gamesetc.)o Commerce(buyer,salesman,etc.)o T(softwareengineer,webdeveloperetc.)o Art(filmmaker,painter,photographer,music,etc.)o Management(accountmanagement,riskmanagement,highermanagementetc.)o Research(scientist,analyst,researcher,etc.)o Education(coaching,training,teaching,etc.)o Projectmanagement(events,community,cultureetc.)o PR,marketing,sales,advertising,communicationo Writing(journalist,writer,copywriter,blogger,etc.)o Other,namely:

Whatisyourannualnetincome?

o Lessthan€20.000ayearo €20.001-€30.000ayearo €30.001-€40.000ayearo €40.001-€50.000ayearo Morethan€50.000ayearo Idon'tknow/I'drathernotsay

THELAST4QUESTIONSAREABOUTCOWORKINGSPACESInwhichcountryisyourcoworkingspacelocated?

o AlbaniaAmericaArgentinaAustraliaAustriaBelgiumBrazilBulgariaCanadaChinaColombiaCroatiaCyprusCzechRepublicDenmarkGermanyEgyptEnglandEritreaEstoniaFinlandFranceGhanaGreeceHongaryIrelandIndiaIndonesia Italy Iraq Iran Japan Korea Lithuania Luxembourg TheNetherlandsNorwayNigeriaMaltaMaroccoPakistanPhilippinesPolandPortugalRomaniaRussiaScotlandSerbia Singapore Slovenia Slovakia Suriname Spain South-AfricaSwedenSwitzerlandThailandTunisiaTurkeyUnitedArabEmiratesVietnamOther

Howmanyhoursdoyouworkatthecoworkingspaceperweek(onaverage)?Howdoyouusuallymovetothecoworkingspace?

o Caro Bikeo Byfooto Publictransport

85

Whatareyourthreemostimportantmotivationstogotoacoworkingspace?Pleaseindicateinorderofimportance.Motivation1isthemostimportant.YoumaynotgivetwicethesamemotivationMotivation1

o Affordableaccommodationo Thefeelingofbeingpartofacommunityo Vibrantandcreativeatmosphereinthecoworkingspaceo Professionalappearanceforthecompanyo Professionalsupportiveworkenvironment(supportservices)o Theopportunitytonetworkwithcoworkers(possiblenewprojects)o Socialinteractionwithcoworkerso Thepossibilityforwork-relatedconversationswithothercoworkers(knowledgesharing,knowledgecreation)o Iwaslookingforaworkplaceoutsidethehome(separatingworkandprivatelife)o Flexibility(rentalperiod,numberofsquaremeters)

Motivation2

o Affordableaccommodationo Thefeelingofbeingpartofacommunityo Vibrantandcreativeatmosphereinthecoworkingspaceo Professionalappearanceforthecompanyo Professionalsupportiveworkenvironment(supportservices)o Theopportunitytonetworkwithcoworkers(possiblenewprojects)o Socialinteractionwithcoworkerso Thepossibilityforwork-relatedconversationswithothercoworkers(knowledgesharing,knowledgecreation)o Iwaslookingforaworkplaceoutsidethehome(separatingworkandprivatelife)o Flexibility(rentalperiod,numberofsquaremeters)

Motivation3

o Affordableaccommodationo Thefeelingofbeingpartofacommunityo Vibrantandcreativeatmosphereinthecoworkingspaceo Professionalappearanceforthecompanyo Professionalsupportiveworkenvironment(supportservices)o Theopportunitytonetworkwithcoworkers(possiblenewprojects)o Socialinteractionwithcoworkerso Thepossibilityforwork-relatedconversationswithothercoworkers(knowledgesharing,knowledgecreation)o Iwaslookingforaworkplaceoutsidethehome(separatingworkandprivatelife)o Flexibility(rentalperiod,numberofsquaremeters)

Youhavereachedtheendofthisquestionnaire.Thanks for your time! If you have any question/comments regarding this survey or research project, please contactviajaspervandekoevering@gmail.com.Pressthebuttonbelow(end),sothatthesurveywillbesavedcorrectly.

86

AppendixC2-Dutchquestionnaire

AfstudeeronderzoeknaarcoworkingspacesBestegebruikervaneencoworkingspace,InhetkadervanmijnafstudeeronderzoekvoordemasterRealEstateManagement&DevelopmentaandeTechnischeUniversiteitinEindhovenvoerikmomenteeleenonderzoekuitnaarcoworkingspaces.Deenquêtevraagtuenerzijdsnaaruwgeprefereerdekarakteristiekenbetrefendecoworkingspacesaandehandvannegenkeuzesetsenanderzijdsnaaruwpersoonlijkekarakteristieken.Deenquetezal5tot10minutenvanuwtijd inbeslagnemenenaluwantwoordenzullenanoniemblijven.Heeftuvragenmetbetrekkingtothetonderzoekofdezequestionnaire,kuntucontactopnemenviajaspervandekoevering@gmail.com.Alvastbedankt!Metvriendelijkegroet,JaspervandeKoeveringINTRODUCTIEKEUZESETSDevolgende9pagina'sgeventelkensdriecompletealternatieventenaanzienvandekenmerkenvancoworkingspaces.Uwordtgevraagdomtelkensaantegeveninwelkecoworkingspaceuhetliefstzouwerken.Alsulieverthuiswiltwerkendanineenvande3genoemdealternatieven,kiestuvoor'geenvandezeopties'.Geefperpaginatelkensopnieuwuwmeestaansprekendekeuze.Allepagina'szijnonafhankelijkvanelkaar.Hierondervolgteenvoorbeeldvraag:

*Layoutvanderuimte*:

o Openlayout(groteopenruimtes)o Half-openlayout(combinatievanopenruimtenenconcentratieruimten)o Geslotenlayout(afgeslotenenseparateruimtes)

**Diversiteitaanbodruimtes**:

o Basiscoworkingspace(collaborativeworkspace+vergaderruimtenenkeuken)o Standaard coworing space (collaborative workspace + vergaderruimten + keuken + evenement ruimten en

informelezones)o Premium coworking space (collaborative workspace + vergaderruimten + keuken + evenement ruimten +

informelezones+fitnessruimteenbar)

87

***Receptieenhospitality***:o Geenreceptieengeenhosto Receptiemaargeenhosto Receptieeneenactievehost(actievecoworkinghostdiecoworkersmetelkaarinverbindingbrengt)

****Diversiteitaanhuurders****:

o Geendiversiteitvanhuurders(huurdersactiefindezelfdesector)o Matigediversiteitvanhuurders(huurdersineenaantalverschillendesectoren)o Sterkediversiteitvanhuurders(huurdersinveelverschillendesectorenaanwezigindecoworkingspace)

Nuvolgen9keuzesets(éénvoorbeeldwordthierondergepresenteerd)

DEVOLGENDE4VRAGENHEBBENBETREKKINGOPPERSOONLIJKEKARAKTERISTIEKENWatisuwgeslacht?

o Mano Vrouw

Watisuwleeftijd(injaren)?Watisuwnationaliteit?

o NederlandsAfghaansAmerikaansAustralischBelgischBraziliaansBulgaarsCanadeesChineesCypriotischDeensDuitsEngelsEritreseEstlandsFilipijnsFinlandsFransGhaneesGrieksHongaarsIersIndiaasIndonesischItaliaansIraaksIraansJapansKoreaansKroatischLuxemburgsNoorsNigeriaansMalteesMarokkaansOostenrijksPakistaansPoolsPortugeesRoemeensRussischThaisTsjechischTurksSchotsSloveensSlowaaksSurinaamsSpaansZuid-afrikaansZweedsAnders

Watisuwhoogstbehaaldediploma?

o Geenonderwijs/basisonderwijs/lagereschoolo VMBO(kader,beroepsgerichteentheoretischeleerweg)o HAVOo VWO(atheneumengymnasium)o MBOo HBOo WO(bachelor)o WO(master)o WO(PhD)

88

DEVOLGENDE4VRAGENHEBBENBETREKKINGOPWERKGERELATEERDEKARAKTERISTEIEKENOnderwelkegebruikersgroepvaltu?

o Zelfstandigezonderpersoneel(zzp'er),freelancerofentrepreneuro Medewerkervaneenbedrijf(2-10medewerkers)o Medewerkervaneenbedrijf(11-50medewerkers)o Medewerkervaneenbedrijf(meerdan50medewerkers)o Student

Watisuwhuidigepositiebinnenuworganisatie?

o Ondersteunendpersoneel(baliemedewerker,receptionist(e)etc.)o Regulierewerknemero Managero Directie/bestuur/eigenaaro Nietvantoepassing

Inwelkesectorisuwbedrijfactief?

o Consultancy(juridischadvies,organisatorischadviesetc.)o Design(grafisch,web,product,gamesetc.)o Handel(inkoper,verkoperetc.)o IT(softwareengineer,webdeveloperetc.)o Kunst(filmmaker,schilder,fotograaf,muzieketc.)o Management(accountmanagement,risicomanagement,hogermanagementetc.)o Onderzoek(wetenschapper,analist,onderzoekeretc.)o Onderwijs(coaching,training,lesgevenetc.)o Projectmanagement(evenementen,community,cultuuretc.)o PR,marketing,sales,reclame,communicatieo Schrijfwerk(journalist,schrijver,copywriter,bloggeretc.)o Anders,namelijk:

Watisuw(netto)jaarinkomen?

o Minderdan€20.000perjaaro €20.001-€30.000perjaaro €30.001-€40.000perjaaro €40.001-€50.000perjaaro Meerdan€50.000perjaaro Datweetikniet/zegiklieverniet

DELAATSTE4VRAGENHEBBENBETREKKINGOPCOWORKINGSPACESInwelklandisuwcoworkingspacegelocationeerd?

o NederlandAfghanistanAmerikaAustraliëBelgiëBraziliëBulgarijeCanadaChinaCyprusDenemarkenDuitslandEngelandEritreaEstlandFilipijnenFinlandFrankrijkGhanaGriekenlandHongarijeIerlandIndiaIndonesiëItaliëIrakIranJapanKoreaKroatiëLuxemburgNoorwegenNigeriaMaltaMarokkoOostenrijkPakistanPolenPortugal Roemenië Rusland Thailand Tsjechië Turkije Schotland Slovenië Slowakije Suriname Spanje Zuid-AfrikaZwedenAnders

Gemiddeldhoeveeluurperweekwerktuineencoworkingspace?Hoegajemeestalnaardecoworkingspacetoe?

o Autoo Fietso Tevoeto Openbaarvervoer

89

Watzijnvoorudedriebelangrijkstemotivatiesomnaareencoworkingspacetegaan?(a.u.b.aangeveninvolgordevanbelangrijkheidwaarbijmotivatie1vooruhetbelangrijkstis.Umagniettweekeerdezelfdemotivatiegeven.)Motivatie1

o Betaalbarehuisvestingo Hetgevoelonderdeeltezijnvaneencommunityo Levendigeencreatievesfeerindecoworkingspaceo Professioneleuitstralingvoorhetbedrijfo Professioneelondersteunendewerkomgeving(facilitairediensten)o Demogelijkheidomtenetwerkenmetcoworkers(mogelijkenieuweprojecten)o Demogelijkheidtotsocialeinteractiemetanderecoworkerso Demogelijkheidtotwerk-gerelateerdegesprekkenmetanderecoworkers(kennisdeling,kenniscreatie)o Ikzochteenwerkplekbuitenshuis(scheidingwerkenprivé)o Flexibiliteit(huurperiode,aantalvierkantemeter)

Motivatie2

o Betaalbarehuisvestingo Hetgevoelonderdeeltezijnvaneencommunityo Levendigeencreatievesfeerindecoworkingspaceo Professioneleuitstralingvoorhetbedrijfo Professioneelondersteunendewerkomgeving(facilitairediensten)o Demogelijkheidomtenetwerkenmetcoworkers(mogelijkenieuweprojecten)o Demogelijkheidtotsocialeinteractiemetanderecoworkerso Demogelijkheidtotwerk-gerelateerdegesprekkenmetanderecoworkers(kennisdeling,kenniscreatie)o Ikzochteenwerkplekbuitenshuis(scheidingwerkenprive)o Flexibiliteit(huurperiode,aantalvierkantemeter)

Motivatie3

o Betaalbarehuisvestingo Hetgevoelonderdeeltezijnvaneencommunityo Levendigeencreatievesfeerindecoworkingspaceo Professioneleuitstralingvoorhetbedrijfo Professioneelondersteunendewerkomgeving(facilitairediensten)o Demogelijkheidomtenetwerkenmetcoworkers(mogelijkenieuweprojecten)o Demogelijkheidtotsocialeinteractiemetanderecoworkerso Demogelijkheidtotwerk-gerelateerdegesprekkenmetanderecoworkers(kennisdeling,kenniscreatie)o Ikzochteenwerkplekbuitenshuis(scheidingwerkenprive)o Flexibiliteit(huurperiode,aantalvierkantemeter)

Ubentaanheteindegekomenvandezevragenlijst.Bedankt voor uw tijd! Mocht u nog vragen/opmerkingen hebben met betrekking tot deze enquête of hetafstudeeronderzoek kunt u contact opnemen via jaspervandekoevering@gmail.com. Graag op onderstaande button(einde)drukkenzodatdeenquêtejuistwordtopgeslagen.

90

AppendixC3-Germanquestionnaire

StudienabschlussForschungzuCoworkingSpacesLieberNutzereinesCoworking-Space,imRahmenmeinerMasterarbeitfürdenStudiengang„RealEstateManagement&Development“anderTechnischenUniversitätinEindhoven,analysiereichbevorzugteEigenschaftenvonNutzernvonCoworkingSpaces.Mit Hilfe von neun Choice Sets soll diese Umfrage einerseits Ihre Präferenzen zur Nutzung von Coworking Spacesfeststellen.ZumanderensollenIhrepersönlichenEigenschaftenerfragtwerden.DieseUmfragesolltelediglich5bis10Minuten dauern und Ihre Antworten werden vollständig anonym erfasst. Rückfragen zu der Umfrage oder zu demFragebogen,richtenSiebitteperE-Mailanjaspervandekoevering@gmail.com.VielenDankimVoraus,MitfreundlichenGrüßen,JaspervandeKoeveringEINFÜHRUNGFRAGEBOGEN(CHOICESETS)Dieersten9FragenzeigenjeweilsdreiverschiedeneCoworking-SpacesimHinblickaufderenqualitativeEigenschaften.SiewerdengebetenfürjedenFallzuwählen,inwelchemCoworking-SpaceSieesbevorzugenwürdenzuarbeiten.WennSieeseherbevorzugenwürdenvonzuHauseauszuarbeitenalsineinerderdreiAlternativen,wählenSiebitte„KeinedieserOptionen“.AlleneunFragen/Seitensindunabhängigvoneinander.DahergebenSiebitteimmerdieammeistenaufSiezutreffendeWahlmöglichkeitan.EinBeispielfürsolcheineFrageistFolgende:

*LayoutdesRaumes*:

o OffenerLayout(großzügigeFreiflächen)o HalboffenerLayout(KombinationausFreiflächenundKonzentrationsräumen)o GeschlossenerLayout(abgeschlosseneundseparateRäume)

**VielfaltdesFlächenangebots**:

o BasicCoworkingSpace(GemeinsamerArbeitsbereich+KonferenzräumeundKüchenbereich)o Standard Coworing Space (Gemeinsamer Arbeitsbereich + Konferenzräume + Küchenbereich +

Veranstaltungsräumeundinformelle(Pausen-)Bereiche)o Premium Coworking Space (Gemeinsamer Arbeitsbereich + Konferenzräume + Küchenbereich +

Veranstaltungsräume+informelle(Pausen-)Bereiche+FitnesscenterundTheke)

91

***EmpfangundGästebetreuung***:

o KeinEmpfangundkeineGästebetreuungo EmpfangaberkeineGästebetreuungo EmpfangundaktiveGästebetreuung(aktiverCoworkingBetreuerderdieCoworkermiteinandervernetzt)

****VielfältigkeitderMieter****:

o KeineVielfältigkeit(Mieter/PächterindergleichenBranche)o MäßigeVielfältigkeit(Mieter/PächterineinigenwenigenGeschäftsfeldern)o StarkeVielfältigkeit(vieleverschiedenGeschäftsfelderimCoworkingSpacevertreten)

NunfolgenneunChoiceSets(EinBeispielistimFolgendendargestellt)

DIEFOLGENDEN4FRAGENBEZIEHENSICHAUFPERSÖNLICHEEIGENSCHAFTENWelchenGeschlechtshabenSie?

o Männlicho Weiblich

WasistIhrAlter(inJahren)?WasistIhreNationalität?

o AfghanAmericanAustralianAustrianBelgianBrazilianBulgarianCanadianChineseChechCypriotCroatianDanishDutchEnglishEritreanEstonianFilipinoFinnishFrenchGermanGhanaianGreekHongarianIrishIndianIndonesianItalianIraqiIranianJapaneseKoreanLuxembourgishNorwegianNigerianMalteseMaroccanPakistaniPolish Portugese Romarian Russian Thai Turkish Scottish Slovenian Slovak Surinamese Spanish South-AfricanSwedishOther

WasistIhrhöchsterBildungsabschluss?

o KeinAbschluss/Grundbildungo AllgemeinbildenderSchulabschluss(Hauptschule)o Sekundarabschluss(Realschule,MittlereReife)o WeiterführendeSchulenallgemeinbildenderoderberufsbildenderArt(Ausbildung)o Fachhochschulreife,AbschlusseinerFachoberschule(FOS,BOS12)o Abitur,allgemeineoderfachgebundeneHochschulreife(Gymnasium,BOS13)o Universität(bachelor)o Universität(master)o Universität(Doktortitel)

92

DIEFOLGENDEN4FRAGENBEZIEHENSICHAUFARBEITSBEDINGTEEIGENSCHAFTENWelchederfolgendenBeschreibungentriffammeistenaufSiezu?

o Selbstständiger,FreiberufleroderUnternehmero AngestelltereinesUnternehmens(2-10Mitarbeiter)o AngestelltereinesUnternehmens(11-50Mitarbeiter)o AngestelltereinesUnternehmens(mehrals50Mitarbeiter)o Student

WasistIhreaktuellePositioninIhrerOrganisation?

o UnterstützenderMitarbeiter(Sekretär,Rezeptionist,etc.)o FestangestellterMitarbeitero Geschäftsführero Vorstand/Eigentümero Trifftnichtzu

InwelchemBereichoperiertIhreFirmahauptsächlich?

o Unternehmensberatung(Rechtsberatung,Organisationsberatung,etc.)o Design(Graphik,Web,Produkt,Spiele,etc.)o Handel(Käufer,Verkäufer,etc.)o IT(Softwareentwicklung,Webentwicklung,etc.)o Kunst(Filmemacher,Zeichner,Fotograf,Musik,etc.)o Management(Kundenbetreuung,Risikomanagement,oberesManagement,etc.)o Forschung(Wissenschaftler,Analytiker,Forscher)o Bildung(Coachings,Trainings,Lehrgänge,etc.)o Projektmanagement(Veranstaltungen,Gemeinde,Kultur,etc.)o PR,Marketing,Sales,Werbung,Kommunikationo Journalist,Schriftsteller,Werbetexter,Blogger,etc.o Andere,nämlich:

WiehochistIhrjährlichesNettoeinkommen?

o Wenigerals20.000€proJahro 20.001€-30.000€proJahro 30.001€-40.000€proJahro 40.001€-50.000€proJahro Mehrals50.000€proJahro Ichweißesnicht/IchmöchtekeineAngabemachen

DIELETZTEN4FRAGENBETREFFENCOWORKINGSPACESInwelchemLandliegtIhrCoworkingSpace?

o Afghanistan America Australia Austria Belgium Brazil Bulgaria Canada Chinese CzechRepublicCyprusCroatiaDenmarkEnglandEritreaEstoniaFinlandFranceGermanyGhanaGreeceHongaryIreland India Indonesia Italy Iraq Iranian Japan Korea Luxembourg TheNetherlandsNorwayNigeriaMaltaMaroccoPakistanPhilippinesPolenPortugalRomaniaRussiaThailandTurkeyScotlandSloveniaSlovakiaSurinameSpainSouth-AfricaSwedenOther

WievielStundenIhrerArbeitszeitverbringenSieamCoworkingSpaceproWoche(imDurchschitt)?WiekommenSienormalerweiseandenCoworkingSpace?

o Autoo Fahrrado ZuFußo ÖffentlichenVerkehrsmitteln

93

WassinddiedreiHauptgründedieSiemotivierenaneinemCoworkingSpacezuarbeiten?BittenachWichtigkeitangeben.Motivation1 istdieWichtigste.BeiMöglichkeitnichtzweimaldiegleicheMotivationangeben.Motivation1

o ErschwinglicheUnterkunfto DasGefühlTeileinerGemeinschaftzuseino DynamischeundkreativeAtmosphäreamCoworkingSpaceo ProfessionellerAuftrittderFirmao ProfessionellesunterstützendesArbeitsumfeld(Unterstützungsdienste)o DieChancesichmitanderenCoworkernzuvernetzen(möglicheneueProjekte)o DieMöglichkeitsozialeUnterstützungvonanderenCoworkernzuerhalteno Die Gelegenheit arbeitsbezogene Konversationen mit andern Coworkern zu haben (Wissensaustausch,

Wissensschaffung)o IchhabenacheinemArbeitsplatzaußerhalbvonzuHausegesucht(TrennungvonArbeitundPrivatem)o Flexibilität(Mietzeit,AnzahlanQuadratmetern)

Motivation2

o ErschwinglicheUnterkunfto DasGefühlTeileinerGemeinschaftzuseino DynamischeundkreativeAtmosphäreamCoworkingSpaceo ProfessionellerAuftrittderFirmao ProfessionellesunterstützendesArbeitsumfeld(Unterstützungsdienste)o DieChancesichmitanderenCoworkernzuvernetzen(möglicheneueProjekte)o DieMöglichkeitsozialeUnterstützungvonanderenCoworkernzuerhalteno Die Gelegenheit arbeitsbezogene Konversationen mit andern Coworkern zu haben (Wissensaustausch,

Wissensschaffung)o IchhabenacheinemArbeitsplatzaußerhalbvonzuHausegesucht(TrennungvonArbeitundPrivatem)o Flexibilität(Mietzeit,AnzahlanQuadratmetern)

Motivation3

o ErschwinglicheUnterkunfto DasGefühlTeileinerGemeinschaftzuseino DynamischeundkreativeAtmosphäreamCoworkingSpaceo ProfessionellerAuftrittderFirmao ProfessionellesunterstützendesArbeitsumfeld(Unterstützungsdienste)o DieChancesichmitanderenCoworkernzuvernetzen(möglicheneueProjekte)o DieMöglichkeitsozialeUnterstützungvonanderenCoworkernzuerhalteno Die Gelegenheit arbeitsbezogene Konversationen mit andern Coworkern zu haben (Wissensaustausch,

Wissensschaffung)o IchhabenacheinemArbeitsplatzaußerhalbvonzuHausegesucht(TrennungvonArbeitundPrivatem)o Flexibilität(Mietzeit,AnzahlanQuadratmetern)

SiehabendasEndedesFragebogenserreicht.VielenDankfürIhreZeit!WennSieFragenoderAnregungenzuderUmfrageoderdemForschungsprojekthabensollten,wendenSiesichbitteperMailan jaspervandekoevering@gmail.com.KlickenSieuntenauf(Ende),damitdieUmfrageordnungsgemäßgespeichertwerdenkann.

94

AppendixC4-Italianquestionnaire

TesisuspazidicoworkingGentilefruitoredispazidicoworking,PerlamiatesidilaureaspecialisticainRealEstateManagement&Developmentpressol’UniversitàTecnicaDiEindhoven,stoanalizzandolepreferenzedeifruitoridispazidicoworking.Questobrevequestionariohal'obiettivodiindagare,daunaparte,lecaratteristichepersonalidegliutenti,edall’altralepreferenzerelativeaglispazidicoworking.Lacompilazionedelsondaggiorichiede5-10minutielerispostesarannodeltutto anonime. Per qualsiasi domanda riguardo la mia ricerca o questo questionario, potete contattarmi per emailall’indirizzojaspervandekoevering@gmail.com (in lingua inglese), oppure scrivere alla nostra referente presso ilPolitecnicodiMilanoall'indirizzo:chiara.tagliaro@polimi.it(inlinguaitaliana).Grazieinanticipo!CordialmenteJaspervandeKoeveringIntroduzioneaisetdisceltaLeprossime9domandepresentanoognuna3spazidicoworkingconcaratteristichedifferentil'unodall'altro.Tivienechiestodiscegliereinqualediquesti3spazialternativipreferirestilavorare.Sepreferirestilavoraredacasa,anzichéinunodeitrespazialternativiproposti,selezional'opzione"preferiscolavoraredacasa".Tuttele9domande/paginesonoindipendentil'unadall'altra,quinditivienechiestodiindicareognivoltal'alternativapertepiùattraente.Unesempiodiquestedomandeèmostratoqui:

*Strutturadeglispazi*:

o Strutturaaperta(ampispaziaperti,deltipoopen-space)o Strutturasemi-aperta(combinazionedispaziapertideltipoopen-spaceesaletteperlaconcentrazione)o Strutturachiusa(ufficichiusiespaziconfinati)

**Varietàdispaziadisposizione**:

o SpazidicoworkingBase(Spaziocollaborativo+saleriunioni+spaziocucina)o SpazidicoworkingStandard(Spaziocollaborativo+saleriunioni+spaziocucina+spazioeventiezonarelax)

95

o SpazidicoworkingPremium(Spaziocollaborativo+saleriunioni+spaziocucina+spazioeventi+zonarelax+centrofitnessebar)

***Receptioneaccoglienza**:o Nessunareceptionenessunserviziodiaccoglienzao Reception,manessunservizioaccoglienzao Receptionconservizioaccoglienzaattivo(servizioattivopercreareinterrelazionitrailavoratori)

****Diversitàdiaffittuari****:

o Nessunadiversità(affittuariimpegnatinellostessosettoredibusiness)o Diversitàmoderata(affittuariimpegnatiinpochidiversisettoridibusiness)o Fortediversità(tantidiversisettoripresentinellostessospaziodicoworking)

Seguononovepagine/setdiscelta(Unesempioèpresentatodiseguito)

LESEGUENTI4DOMANDERIGUARDANOLETUECARATTERISTICHEPERSONALIQualèiltuogenere?

o Maschioo Femmina

Qualèlatuaetà(inanni)?Qualèlatuanazionalità?

o ItalianAfghanAmericanAustralianAustrianBelgianBrazilianBulgarianCanadianChineseChechCypriotCroatianDanishDutchGermanEnglishEritreanEstonianFilipinoFinnishFrenchGhanaianGreekHongarianIrishIndianIndonesianIraqiIranianJapaneseKoreanLuxembourgishNorwegianNigerianMalteseMaroccanPakistaniPolish Portugese Romarian Russian Thai Turkish Scottish Slovenian Slovak Surinamese Spanish South-AfricanSwedishOther

Qualèiltuolivellopiùavanzatodiistruzione?

o Nessunaformazione/scuolaelementareo Scuolamediao Istitutotecnicooprofessionaleo Liceoo Istruzioneeformazioneprofessionale(IFP)–percorsitriennaliequadriennalio Istruzione e formazione post-secondaria non terziaria – percorsi post-qualifica e post-diploma, istruzione e

formazionetecnicasuperioreIFTSo Laureatriennaleo Laureaquinquennale,specialisticaomasterdiIeIIlivelloo Dottorato,PhD

96

LESEGUENTI4DOMANDERIGUARDANOINFORMAZIONIRELATIVEALTUOLAVOROQualedeiseguentitipidiinquadramentoprofessionalerappresentameglioiltuoprofilo?

o Lavoratoreautonomo,liberoprofessionistaoimprenditoreo Dipendentediunasocietà(2-10dipendenti)o Dipendentediunasocietà(11-50dipendenti)o Dipendentediunasocietà(piùdi50dipendenti)o Studente

Qualèlatuaattualeposizioneall'internodellatuaorganizzazione?

o Assistenza,supporto(front-desk,reception,ecc)o Dipendenteregolareo Managero Direttore/Proprietarioo Altro/nessunodeiprecedenti

Inqualesettoreoperalatuaazienda?

o Consulenza(consulenzalegale,consulenzaorganizzativa,ecc.)o Design(grafica,web,prodotti,giochi,ecc.)o Commercio(buyer,addettoallevendite,ecc.)o IT(ingegneriadelsoftware,sviluppoweb,ecc.)o Arti(produzionevideo,pittura,fotografia,musica,ecc.)o Management(accountmanagement,riskmanagement,topmanagementecc.)o Ricerca(scienziato,analista,ricercatore,ecc.)o Istruzione(istruttore,formazione,didattica,ecc.)o Projectmanagement(eventi,comunità,cultura,ecc.)o PR,marketing,vendite,pubblicità,comunicazioneo Scrittura(giornalista,scrittore,copywriter,blogger,ecc.)o Altro,cioè:

Qualèiltuoredditonettoannuale?

o Menodi€20,000l’annoo €20,001-€30.000l'annoo €30,001-€40,000l’annoo €40,001-€50,000l’annoo Piùdi50,000€l'annoo Nonloso/Preferisconondirlo

LEULTIME4DOMANDERIGUARDANOGLISPAZIDICOWORKINGINCUILAVORIInqualepaesesitrovalospaziodicoworkingcheutilizziinquestomomento?

o Italy Afghanistan America Australia Austria Belgium Brazil Bulgaria Canada Chinese CzechRepublicCyprusCroatiaDenmarkGermanyEnglandEritreaEstoniaFinlandFranceGhanaGreeceHongaryIreland India Indonesia Iraq Iranian Japan Korea Luxembourg TheNetherlandsNorwayNigeriaMaltaMaroccoPakistanPhilippinesPolenPortugalRomaniaRussiaThailandTurkeyScotlandSloveniaSlovakiaSurinameSpainSouth-AfricaSwedenOther

Mediamentequanteoreallasettimanalavoriinunospaziodicoworking?Comearriviallospaziodicoworking,abitualmente?

o Inautoo Inbiciclettao Apiedio Conimezzipubblici

97

Qualisonoletremotivazionipiùimportantiperlequaliutilizziunospaziodicoworking?Sipregadiindicarleinordinediimportanza.Lamotivazione1èlapiùimportante.Nonpuoidareduevoltelastessamotivazione:Motivazione1

• Postazionedilavoroaprezzoaccessibile• Sensazionediesserepartediunacomunità• Atmosferavibranteecreativachesirespira• Presentazioneprofessionaledellapropriaattività• Supportoprofessionalealpropriolavoro(servizidisupporto,es.accoglienza)• Opportunitàdi"farerete"conaltricoworker(possibilinuoviprogetti)• Interazionesocialeconaltricoworker• Possibilitàdidiscutereditematichedilavoroconaltricoworker(condivisioneecreazionediconoscenze)• Disponibilitàdiunapostazionedilavorofuoricasa(necessitàdiseparareglispazidilavorodaquellidellavita

privata)• Flessibilitàd'uso(periododiaffitto,metriquadri)

Motivazione2

• Postazionedilavoroaprezzoaccessibile• Sensazionediesserepartediunacomunità• Atmosferavibranteecreativachesirespira• Presentazioneprofessionaledellapropriaattività• Supportoprofessionalealpropriolavoro(servizidisupporto,es.accoglienza)• Opportunitàdi"farerete"conaltricoworker(possibilinuoviprogetti)• Interazionesocialeconaltricoworker• Possibilitàdidiscutereditematichedilavoroconaltricoworker(condivisioneecreazionediconoscenze)• Disponibilitàdiunapostazionedilavorofuoricasa(necessitàdiseparareglispazidilavorodaquellidellavita

privata)• Flessibilitàd'uso(periododiaffitto,metriquadri)

Motivazione3

• Postazionedilavoroaprezzoaccessibile• Sensazionediesserepartediunacomunità• Atmosferavibranteecreativachesirespira• Presentazioneprofessionaledellapropriaattività• Supportoprofessionalealpropriolavoro(servizidisupporto,es.accoglienza)• Opportunitàdi"farerete"conaltricoworker(possibilinuoviprogetti)• Interazionesocialeconaltricoworker• Possibilitàdidiscutereditematichedilavoroconaltricoworker(condivisioneecreazionediconoscenze)• Disponibilitàdiunapostazionedilavorofuoricasa(necessitàdiseparareglispazidilavorodaquellidellavita

privata)• Flessibilitàd'uso(periododiaffitto,metriquadri)

Ilquestionarioèfinito.Grazieperiltuotempo!Sehaidomandeocommentiapropositodiquestoprogettodiricerca,puoicontattarmiviamailall'indirizzo: jaspervandekoevering@gmail.com (in lingua inglese), oppure scrivere alla nostra referente presso ilPolitecnicodiMilanoall'indirizzo:chiara.tagliaro@polimi.it(inlinguaitaliana).Premereilpulsantequisotto(Fine!)persalvarecorrettamenteilquestionariocompilato.

98

AppendixD–Partnersindatacollection

Provider Address City Cooperate? Visitedtospread MailedoperatortospreadALab Overhoeksplein2 AmsterdamAmsterdamConnected Paasheuvelweg25 AmsterdamB.Amsterdam JohanHuizingalaan763A AmsterdamBouncespace Bogert1 EindhovenBouncespace Overtoom141 AmsterdamBovendeBalie KleineGartmanplantsoen10 AmsterdamDéPlek Kanaalpark157 LeidenDéPlek Waarderweg19 HaarlemDéPlek NoordBrabantlaan303 EindhovenHackers&Founders Herengracht504 AmsterdamHetNieuweKantoor VanDiemenstraat20-200 AmsterdamHetNieuweKantoor Boogschutterstraat1–43 ApeldoornHetNieuweKantoor Europalaan DenBoschHetNieuweKantoor OudeMiddenweg17 DenHaagHetNieuweKantoor BurgemeesterdeRaadtsingel93C DordrechtHetNieuweKantoor Bennekomseweg41 EdeHetNieuweKantoor Antareslaan65-81 HoofddorpHetNieuweKantoor Zernikepark12 GroningenHetNieuweKantoor Vasteland78 RotterdamHetNieuweKantoor Westblaak180 RotterdamHetNieuweKantoor WegderVerenigdeNaties UtrechtHetNieuweKantoor ArthurvanSchendelstraat650 UtrechtHouseofFashion Singel146 AmsterdamIgluu LouisCouperusplein DenHaagIgluu Lichttoren32 EindhovenIgluu Jansdam2a UtrechtInterpolisCarrousel Spoorlaan298 TilburgMakerversity Kattenburgerstraat5 AmsterdamMeet&Discover Stadsring2 AmersfoortMeetBerlage Oudebrugsteeg9 AmsterdamMeetingPlaza Godebaldkwartier20 UtrechtMixtup EersteJacobvanCampenstraat59 AmsterdamNewBabylonGardens AnnavanBuerenplein41 DenHaagNomadz Binckhorstlaan36 DenHaagPanoramaWest Baarsjesweg224 AmsterdamPeer! Maanlander14A AmersfoortQlubhouse Tivolistraat6 TilburgROOOMS LageFrontweg2A MaastrichtROOOMS Westhoven7 RoermondSeats2Meet Torenallee24 EindhovenSeats2Meet Stationsplein161 s-HertogenboschSeats2Meet Moreelsepark65 UtrechtSeats2Meet Stationsplein49 AmersfoortSpaces Herengracht124-128 AmsterdamSpaces BarbaraStorzzilaan101-201 AmsterdamSpaces Vijzelstraat68-72 AmsterdamSpaces Hofplein20 RotterdamSpaces ZuidHollandlaan7 DenHaagStartdock Herengracht430 AmsterdamTheRotterdamCollective Stadhuisplein15 RotterdamTheSpringHouse DeRuijterkade128 AmsterdamTheStartupOrgy Reguliersdwarsstraat73 AmsterdamTheThinkingHut Mauritskade55C AmsterdamTribes Papendorpseweg95 UtrechtTribes FlightForum840 EindhovenTribes Kalvermarkt53 DenHaagTribes RiviumQuadrant141-151 CapelleaandenIjsselTribes Wilhelminaplein1-40 RotterdamTribes GustavMahlerplein28 AmsterdamTribes PrinsBernhardplein200 AmsterdamTribes Leidseveer2-10 UtrechtUtrechtIncGarage Padualaan8 UtrechtWeWork Weteringschans165 AmsterdamWeWork Weesperstraat61-105 AmsterdamWickedgrounds TweedeSchinkelstraat26 AmsterdamWickedgrounds GeneraalVetterstraat55 Amsterdam

99

AppendixE–Conceptualmodeloftheindependentvariable

Userpreferencesforcoworkingspacecharacerstics

Usercharacteristics

Geographiccharaceristics

Countryofcoworkingspace

Demographiccharacteristics

Gender

Age

Nationality

Education

Usergroup

Positioninorganization

Sectoroforganization

Income

Hoursincoworkingspace

Transporttocoworkingspace

Psychographiccharacteristics

Motivationtogotoacoworkingspace

100

AppendixF–Conceptualmodelofthedependentvariable

UsercharacteristicsUserpreferencesforcoworkingspacecharacerstics

Location Accessibilityoflocation

Bypublictransportandcar

Bycar

Bypublictransport

Office decor Atmosphereandinterioraesthetic

Industrial

Modern

Homey

Officeexterioranddivision Layoutofthespace

Openlayout

Half-openlayout

Closedlayout

Officeexterioranddivision

Diversityinsupplyspaces

Basiccoworkingspace

Standardcoworkingspace

Premiumcoworkingspace

Facilitiesandservices

Receptionandhospitality

Noreceptionandnohost

Receptionbutnohost

Receptionandactivehost

Communityand

sustainabilityEvents

None

Sometimes

Often

Accessibility Diversityoftenants

Nodiversityoftenants

Moderatediversityoftenants

Strongdiversityoftenants

Accessbility Typeofleasecontract

Nocontract

Shortcontract(dayorweeror

month)

Longcontract(yearorlonger)

101

AppendixG–OutputNLOGITMultinomialLogitModel ---------------------------------------------------------------- Discrete choice (multinomial logit) model Dependent variable Choice Log likelihood function -2474.42529 Estimation based on N = 1971, K = 17 Inf.Cr.AIC = 4982.9 AIC/N = 2.528 Model estimated: Jan 02, 2017, 21:26:35 R2=1-LogL/LogL* Log-L fncn R-sqrd R2Adj Constants only -2708.2549 .0863 .0837 Response data are given as ind. choices Number of obs.= 1971, skipped 0 obs --------+-------------------------------------------------------------------- | Standard Prob. 95% Confidence KEUZE| Coefficient Error z |z|>Z* Interval --------+-------------------------------------------------------------------- ICONST| .15562*** .05848 2.66 .0078 .04100 .27024 ALOC| .48459*** .04253 11.39 .0000 .40122 .56795 BLOC| -.50041*** .04791 -10.44 .0000 -.59432 -.40650 ASFEER| -.12530** .04873 -2.57 .0101 -.22082 -.02979 BSFEER| -.00900 .04633 -.19 .8460 -.09981 .08180 ALAYOUT| .02749 .04773 .58 .5647 -.06606 .12103 BLAYOUT| .27553*** .04410 6.25 .0000 .18910 .36195 ADIV_RUI| -.03827 .04525 -.85 .3977 -.12697 .05042 BDIV_RUI| .06774 .04685 1.45 .1482 -.02408 .15956 AREC| -.16860*** .04884 -3.45 .0006 -.26432 -.07288 BREC| .12742*** .04297 2.97 .0030 .04320 .21165 AEVEN| -.15460*** .04717 -3.28 .0010 -.24704 -.06216 BEVEN| .15014*** .04459 3.37 .0008 .06274 .23753 ADIV_HUU| -.24084*** .04472 -5.38 .0000 -.32850 -.15318 BDIV_HUU| .14238*** .04448 3.20 .0014 .05520 .22956 ATYPCO| .50808*** .04355 11.67 .0000 .42272 .59344 BTYPCO| .20426*** .04687 4.36 .0000 .11241 .29612 --------+-------------------------------------------------------------------- Note: ***, **, * ==> Significance at 1%, 5%, 10% level.

102

AppendixH–OutputNLOGITLatentClassLogitModel(2,3and4classesestimated) Latent Class Logit Model(2 classes) Dependent variable KEUZE Log likelihood function -2451.91637 Restricted log likelihood -2732.38619 Chi squared [ 35 d.f.] 560.93964 Significance level .00000 McFadden Pseudo R-squared .1026465 Estimation based on N = 1971, K = 35 Inf.Cr.AIC = 4973.8 AIC/N = 2.524 Model estimated: Jan 04, 2017, 14:43:54 R2=1-LogL/LogL* Log-L fncn R-sqrd R2Adj No coefficients -2732.3862 .1026 .0973 Constants only -2708.2549 .0947 .0893 At start values -2474.4450 .0091 .0032 Response data are given as ind. choices Number of latent classes = 2 Average Class Probabilities .450 .550 Number of obs.= 1971, skipped 0 obs --------+-------------------------------------------------------------------- | Standard Prob. 95% Confidence KEUZE| Coefficient Error z |z|>Z* Interval --------+-------------------------------------------------------------------- |Utility parameters in latent class -->> 1 ICONST|1| -.06699 .44450 -.15 .8802 -.93820 .80421 ALOC|1| 2.42644 2.29196 1.06 .2897 -2.06571 6.91860 BLOC|1| -2.14412 1.85797 -1.15 .2485 -5.78568 1.49744 ASFEER|1| -.91931 1.40102 -.66 .5117 -3.66526 1.82663 BSFEER|1| -.21029 .41297 -.51 .6106 -1.01970 .59912 ALAYOU|1| .21469 .33931 .63 .5269 -.45035 .87972 BLAYOU|1| 1.36326 1.07400 1.27 .2043 -.74173 3.46826 ADIV_R|1| .26244 .57498 .46 .6481 -.86451 1.38939 BDIV_R|1| 1.35876 1.14726 1.18 .2363 -.88982 3.60734 AREC|1| -.52249 .70343 -.74 .4576 -1.90119 .85621 BREC|1| .35093 .36645 .96 .3382 -.36729 1.06916 AEVEN|1| -.26433 .60819 -.43 .6638 -1.45636 .92771 BEVEN|1| -.17746 .42924 -.41 .6793 -1.01876 .66384 ADIV_H|1| -.69249 .84462 -.82 .4123 -2.34792 .96294 BDIV_H|1| .48731 .44716 1.09 .2758 -.38910 1.36372 ATYPCO|1| 2.16308 1.37243 1.58 .1150 -.52684 4.85299 BTYPCO|1| 1.78125 1.95800 .91 .3630 -2.05636 5.61886 |Utility parameters in latent class -->> 2 ICONST|2| -.14265 .14394 -.99 .3217 -.42477 .13948 ALOC|2| .05987 .15259 .39 .6948 -.23920 .35894 BLOC|2| -.27741** .11980 -2.32 .0206 -.51222 -.04261 ASFEER|2| .02196 .11539 .19 .8490 -.20420 .24813 BSFEER|2| -.04012 .10983 -.37 .7149 -.25539 .17515 ALAYOU|2| .13971 .11077 1.26 .2072 -.07739 .35682 BLAYOU|2| -.11853 .19481 -.61 .5429 -.50035 .26328 ADIV_R|2| -.06113 .10678 -.57 .5670 -.27041 .14815 BDIV_R|2| -.33613** .16282 -2.06 .0390 -.65525 -.01702 AREC|2| -.13729 .11549 -1.19 .2345 -.36364 .08906 BREC|2| .09568 .11733 .82 .4148 -.13429 .32564 AEVEN|2| .11815 .11124 1.06 .2882 -.09988 .33618 BEVEN|2| .40523*** .10212 3.97 .0001 .20509 .60538 ADIV_H|2| -.24855** .12175 -2.04 .0412 -.48718 -.00991 BDIV_H|2| -.01438 .10870 -.13 .8948 -.22743 .19867 ATYPCO|2| .19379* .11367 1.70 .0882 -.02900 .41659 BTYPCO|2| -.05612 .15362 -.37 .7149 -.35721 .24497 |Estimated latent class probabilities PrbCls1| .44980*** .06594 6.82 .0000 .32056 .57903 PrbCls2| .55020*** .06594 8.34 .0000 .42097 .67944 --------+-------------------------------------------------------------------- Note: ***, **, * ==> Significance at 1%, 5%, 10% level. -----------------------------------------------------------------------------

103

Latent Class Logit Model(3 classes) Dependent variable KEUZE Log likelihood function -2434.89653 Restricted log likelihood -2732.38619 Chi squared [ 53 d.f.] 594.97931 Significance level .00000 McFadden Pseudo R-squared .1088754 Estimation based on N = 1971, K = 53 Inf.Cr.AIC = 4975.8 AIC/N = 2.525 Model estimated: Jan 04, 2017, 14:44:24 R2=1-LogL/LogL* Log-L fncn R-sqrd R2Adj No coefficients -2732.3862 .1089 .1008 Constants only -2708.2549 .1009 .0928 At start values -2474.4621 .0160 .0071 Response data are given as ind. choices Number of latent classes = 3 Average Class Probabilities .444 .150 .407 BHHH estimator used for asymp. variance Number of obs.= 1971, skipped 0 obs --------+-------------------------------------------------------------------- | Standard Prob. 95% Confidence KEUZE| Coefficient Error z |z|>Z* Interval --------+-------------------------------------------------------------------- |Utility parameters in latent class -->> 1 ICONST|1| -2.99328 2948.184 .00 .9992 -5781.32758 5775.34102 ALOC|1| 4.09487 4379.622 .00 .9993 -8579.80695 8587.99669 BLOC|1| -3.90120 4379.488 .00 .9993 -8587.54026 8579.73787 ASFEER|1| -1.93504 1473.940 .00 .9990 -2890.80523 2886.93516 BSFEER|1| -1.14800 1474.401 .00 .9994 -2890.92035 2888.62434 ALAYOU|1| .13932 .24128 .58 .5636 -.33357 .61221 BLAYOU|1| 4.19399 4422.699 .00 .9992 -8664.13656 8672.52454 ADIV_R|1| -.75368 1474.360 .00 .9996 -2890.44675 2888.93939 BDIV_R|1| 3.43420 2948.420 .00 .9991 -5775.36323 5782.23162 AREC|1| -.59897 1431.996 .00 .9997 -2807.25962 2806.06167 BREC|1| -.56786 1474.347 .00 .9997 -2890.23483 2889.09912 AEVEN|1| -.51643 .32388 -1.59 .1108 -1.15121 .11836 BEVEN|1| .06297 .27817 .23 .8209 -.48223 .60817 ADIV_H|1| -.92085 1474.285 .00 .9995 -2890.46642 2888.62471 BDIV_H|1| 3.03713 2948.313 .00 .9992 -5775.54940 5781.62366 ATYPCO|1| 4.66379 4422.777 .00 .9992 -8663.82047 8673.14805 BTYPCO|1| 3.90132 4379.572 .00 .9993 -8579.90255 8587.70519 |Utility parameters in latent class -->> 2 ICONST|2| 2.32937 112.2326 .02 .9834 -217.64241 222.30115 ALOC|2| .30146 1.90578 .16 .8743 -3.43381 4.03673 BLOC|2| 2.45465 8.46574 .29 .7719 -14.13789 19.04719 ASFEER|2| 3.92703 113.3928 .03 .9724 -218.31884 226.17290 BSFEER|2| -1.25565 5.81101 -.22 .8289 -12.64502 10.13373 ALAYOU|2| 6.18500 113.8260 .05 .9567 -216.90982 229.27981 BLAYOU|2| -1.99577 3.06239 -.65 .5146 -7.99796 4.00641 ADIV_R|2| -2.93677 112.3107 -.03 .9791 -223.06171 217.18816 BDIV_R|2| -2.50395 6.27125 -.40 .6897 -14.79537 9.78747 AREC|2| -1.30773 5.62476 -.23 .8162 -12.33206 9.71660 BREC|2| 4.59435 113.7554 .04 .9678 -218.36221 227.55091 AEVEN|2| 4.10113 113.3940 .04 .9711 -218.14696 226.34922 BEVEN|2| -.37177 2.59039 -.14 .8859 -5.44884 4.70529 ADIV_H|2| -1.37587 5.82507 -.24 .8133 -12.79280 10.04105 BDIV_H|2| 1.53445 3.73112 .41 .6809 -5.77842 8.84731 ATYPCO|2| 2.47216 8.42693 .29 .7692 -14.04432 18.98863 BTYPCO|2| -7.01599 114.9063 -.06 .9513 -232.22816 218.19617 |Utility parameters in latent class -->> 3 ICONST|3| -.36039 .35989 -1.00 .3166 -1.06576 .34499 ALOC|3| .35027 .28326 1.24 .2162 -.20491 .90545 BLOC|3| -1.32997* .69293 -1.92 .0549 -2.68809 .02815 ASFEER|3| -.32987 .33789 -.98 .3289 -.99213 .33239 BSFEER|3| .46129 .55775 .83 .4082 -.63189 1.55447 ALAYOU|3| -.98943 .69602 -1.42 .1552 -2.35360 .37473 BLAYOU|3| .44040 .42758 1.03 .3030 -.39765 1.27845 ADIV_R|3| .20001 .37096 .54 .5898 -.52706 .92707 BDIV_R|3| -.73502 .64055 -1.15 .2512 -1.99047 .52043

104

AREC|3| -.84332 .59429 -1.42 .1559 -2.00811 .32146 BREC|3| .43195 .39632 1.09 .2758 -.34482 1.20872 AEVEN|3| -.48571 .38651 -1.26 .2089 -1.24326 .27184 BEVEN|3| .91910** .45462 2.02 .0432 .02805 1.81014 ADIV_H|3| -.95446* .57391 -1.66 .0963 -2.07930 .17037 BDIV_H|3| .31754 .32698 .97 .3315 -.32332 .95840 ATYPCO|3| .12867 .33255 .39 .6988 -.52311 .78046 BTYPCO|3| .55224 .43007 1.28 .1991 -.29069 1.39517 |Estimated latent class probabilities PrbCls1| .44379*** .15199 2.92 .0035 .14589 .74168 PrbCls2| .14953*** .03528 4.24 .0000 .08039 .21867 PrbCls3| .40668** .16991 2.39 .0167 .07366 .73971 --------+-------------------------------------------------------------------- Note: ***, **, * ==> Significance at 1%, 5%, 10% level. ----------------------------------------------------------------------------- Latent Class Logit Model (4 classes) Dependent variable KEUZE Log likelihood function -2426.98884 Restricted log likelihood -2732.38619 Chi squared [ 71 d.f.] 610.79469 Significance level .00000 McFadden Pseudo R-squared .1117695 Estimation based on N = 1971, K = 71 Inf.Cr.AIC = 4996.0 AIC/N = 2.535 Model estimated: Jan 04, 2017, 14:44:42 R2=1-LogL/LogL* Log-L fncn R-sqrd R2Adj No coefficients -2732.3862 .1118 .1010 Constants only -2708.2549 .1039 .0930 At start values -2474.4727 .0192 .0073 Response data are given as ind. choices Number of latent classes = 4 Average Class Probabilities .490 .157 .165 .189 BHHH estimator used for asymp. variance Number of obs.= 1971, skipped 0 obs --------+-------------------------------------------------------------------- | Standard Prob. 95% Confidence KEUZE| Coefficient Error z |z|>Z* Interval --------+-------------------------------------------------------------------- |Utility parameters in latent class -->> 1 ICONST|1| -.77589 .60712 -1.28 .2013 -1.96583 .41405 ALOC|1| 4.83189 38.73731 .12 .9007 -71.09185 80.75562 BLOC|1| -4.98174 41.66716 -.12 .9048 -86.64787 76.68440 ASFEER|1| -2.63674 23.31418 -.11 .9100 -48.33170 43.05822 BSFEER|1| -1.61802 17.37989 -.09 .9258 -35.68198 32.44595 ALAYOU|1| 1.57863 16.82274 .09 .9252 -31.39334 34.55059 BLAYOU|1| 2.42054 23.68780 .10 .9186 -44.00670 48.84778 ADIV_R|1| -.70567 11.89768 -.06 .9527 -24.02468 22.61335 BDIV_R|1| 2.67318 23.39391 .11 .9090 -43.17804 48.52440 AREC|1| -.55411 1.41238 -.39 .6948 -3.32232 2.21411 BREC|1| .67938 .49986 1.36 .1741 -.30032 1.65909 AEVEN|1| -.34388 5.70664 -.06 .9519 -11.52869 10.84092 BEVEN|1| 1.53483 13.18785 .12 .9073 -24.31289 27.38255 ADIV_H|1| -1.36140 13.19772 -.10 .9178 -27.22845 24.50566 BDIV_H|1| -.59024 2.19151 -.27 .7877 -4.88553 3.70505 ATYPCO|1| 5.41645 40.66226 .13 .8940 -74.28012 85.11302 BTYPCO|1| 2.81585 13.29707 .21 .8323 -23.24592 28.87762 |Utility parameters in latent class -->> 2 ICONST|2| -4.60337 1111.986 .00 .9967 -2184.05534 2174.84859 ALOC|2| 4.28305 2304.424 .00 .9985 -4512.30418 4520.87029 BLOC|2| -4.38237 1178.489 .00 .9970 -2314.17883 2305.41409 ASFEER|2| 1.82159 1767.138 .00 .9992 -3461.70583 3465.34900 BSFEER|2| 1.74445 1762.731 .00 .9992 -3453.14521 3456.63411 ALAYOU|2| 3.63775 1153.120 .00 .9975 -2256.43611 2263.71161 BLAYOU|2| -1.86434 1138.650 .00 .9987 -2233.57724 2229.84855 ADIV_R|2| -4.77597 2705.969 .00 .9986 -5308.37844 5298.82649 BDIV_R|2| 1.75321 1593.444 .00 .9991 -3121.33904 3124.84546 AREC|2| 2.84886 1116.981 .00 .9980 -2186.39417 2192.09190 BREC|2| 3.83027 1099.881 .00 .9972 -2151.89688 2159.55742

105

AEVEN|2| -4.99641 2304.776 .00 .9983 -4522.27482 4512.28199 BEVEN|2| -.07760 1153.952 .00 .9999 -2261.78170 2261.62650 ADIV_H|2| 3.68552 1267.242 .00 .9977 -2480.06305 2487.43409 BDIV_H|2| .77248 2364.918 .00 .9997 -4634.38111 4635.92606 ATYPCO|2| -.08325 1137.599 .00 .9999 -2229.73586 2229.56936 BTYPCO|2| -1.13410 1155.540 .00 .9992 -2265.95156 2263.68337 |Utility parameters in latent class -->> 3 ICONST|3| -1.25710 26594.89 .00 1.0000 *********** 52123.77170 ALOC|3| 1.39043 20.92346 .07 .9470 -39.61879 42.39966 BLOC|3| -1.83764 27.73055 -.07 .9472 -56.18853 52.51325 ASFEER|3| 2.45857 26600.26 .00 .9999 *********** 52138.00642 BSFEER|3| 2.78062 26577.31 .00 .9999 *********** 52093.34566 ALAYOU|3| 2.10683 37.64109 .06 .9554 -71.66836 75.88202 BLAYOU|3| -1.19611 23.61488 -.05 .9596 -47.48042 45.08820 ADIV_R|3| -4.45950 53172.92 .00 .9999 *********** *********** BDIV_R|3| 3.69867 26585.71 .00 .9999 *********** 52110.72681 AREC|3| -.28699 26615.38 .00 1.0000 *********** 52164.89156 BREC|3| .82517 26592.65 .00 1.0000 *********** 52121.45946 AEVEN|3| -.37159 10.54191 -.04 .9719 -21.03335 20.29018 BEVEN|3| .21095 4.53642 .05 .9629 -8.68027 9.10217 ADIV_H|3| -1.25375 26610.20 .00 1.0000 *********** 52153.78752 BDIV_H|3| .07223 53204.85 .00 1.0000 *********** *********** ATYPCO|3| 1.59068 25.46750 .06 .9502 -48.32471 51.50607 BTYPCO|3| -2.89351 42.31173 -.07 .9455 -85.82297 80.03596 |Utility parameters in latent class -->> 4 ICONST|4| 2.46359 6208.656 .00 .9997 *********** 12171.20623 ALOC|4| -4.24071 9147.722 .00 .9996 *********** 17924.96540 BLOC|4| 6.04579 9179.779 .00 .9995 *********** 17998.08228 ASFEER|4| -2.73127 6174.621 .00 .9996 *********** 12099.30365 BSFEER|4| -1.21385 3118.672 .00 .9997 -6113.69769 6111.26999 ALAYOU|4| -7.22278 9146.359 .00 .9994 *********** 17919.31076 BLAYOU|4| 7.68397 9189.457 .00 .9993 *********** 18018.68817 ADIV_R|4| 5.25658 3097.789 .00 .9986 -6066.29746 6076.81061 BDIV_R|4| -5.30918 6203.769 .00 .9993 *********** 12153.85491 AREC|4| -4.79722 6081.778 .00 .9994 *********** 11915.26832 BREC|4| -1.99122 6223.333 .00 .9997 *********** 12195.51793 AEVEN|4| .60138 67.82059 .01 .9929 -132.32453 133.52729 BEVEN|4| 3.19628 9151.030 .00 .9997 *********** 17938.88572 ADIV_H|4| -1.23320 3076.776 .00 .9997 -6031.60285 6029.13644 BDIV_H|4| -2.44736 6094.717 .00 .9997 *********** 11942.97793 ATYPCO|4| -6.50267 9191.501 .00 .9994 *********** 18008.50745 BTYPCO|4| 4.97762 9268.901 .00 .9996 *********** 18171.68927 |Estimated latent class probabilities PrbCls1| .48969*** .12875 3.80 .0001 .23734 .74203 PrbCls2| .15652 .35991 .43 .6636 -.54889 .86193 PrbCls3| .16524 .45080 .37 .7140 -.71831 1.04879 PrbCls4| .18855*** .03017 6.25 .0000 .12943 .24768 --------+-------------------------------------------------------------------- Note: ***, **, * ==> Significance at 1%, 5%, 10% level. ----------------------------------------------------------------------------- Commandtocreateclassprobability:|-> CREATE ; p1 = 0 ; p2 = 0 $ |-> NAMELIST ; cp = p1,p2$ |-> DISCRETECHOICE;Lhs = Keuze ;Choices = 1,2,3,4 ;Rhs = iconst,Aloc,Bloc,Asfeer,Bsfeer,Alayout,Blayout,ADiv_rui,BDiv_rui,Arec,Brec,Aeven,Beven, ADiv_huu,BDiv_huu,Atypco,Btypco ;lcm ;classp=cp$

106

AppendixI–Part-worthutilitiesofthedependentvariable

UsercharacteristicsUserpreferencesforcoworkingspacecharacerstics

Location Accessibilityoflocation

Bypublictransportandcar

Bycar

Bypublictransport

Office decor Atmosphereandinterioraesthetic

Industrial

Modern

Homey

Officeexterioranddivision Layoutofthespace

Openlayout

Half-openlayout

Closedlayout

Officeexterioranddivision

Diversityinsupplyspaces

Basiccoworkingspace

Standardcoworkingspace

Premiumcoworkingspace

Facilitiesandservices

Receptionandhospitality

Noreceptionandnohost

Receptionbutnohost

Receptionandactivehost

Communityand

sustainabilityEvents

None

Sometimes

Often

Accessibility Diversityoftenants

Nodiversityoftenants

Moderatediversityoftenants

Strongdiversityoftenants

Accessbility Typeofleasecontract

Nocontract

Shortcontract(dayorweeror

month)

Longcontract(yearorlonger)

1,42

1,0

0.27

0.61

0,36

0,31

0,48

-0,71

0,20

0,10

0,14

-0,30

0,28

0,03

0,13

-0,01

-0,14

0,02

-0,50

0,48

-0,24

0

0,15

-0,15

0,04

0,13

-0,7

0,51

107

AppendixJ–Frequencytablesoftheusercharacteristicsofthelatentclasses Demographiccharacteristics

Positioninorganization Class1 % Class2 % Total % Chi-square P-valueSupportingstaff 2 33% 4 67% 6 3%Regularemployee 21 43% 28 57% 49 22%Manager 6 35% 11 65% 17 8%Board/owner 30 32% 63 68% 93 42%Doesnotapply 23 43% 31 57% 54 25% 2,369 0,668

Usergroup Class1 % Class2 % Total % Chi-square P-valueSelf-employedworker,freelanceror 45 38% 72 62% 117 53%Employeeofacompany(2-10employees) 11 29% 27 71% 38 17%Employeeofacompany(11ormoreemployees) 16 42% 22 58% 38 17%Student 10 38% 16 62% 26 12% 1,587 0,662

Agegroups Class1 % Class2 % Total % Chi-square P-value24yearoryounger 11 31% 24 69% 35 16%25-34year 35 41% 51 59% 86 39%35-44year 21 38% 34 62% 55 25%45yearorolder 15 35% 28 65% 43 20% 0,791,063

Gender Class1 % Class2 % Total % Chi-square P-valueMale 55 37% 95 63% 150 68%Female 27 39% 42 61% 69 32% 0,122 0,726

Levelofeducation Class1 % Class2 % Total % Chi-square P-valueLow-mediumeducationlevel 9 28% 23 72% 32 15%Higheducationlevel 73 39% 114 61% 187 85% 1,389 0,239

Income Class1 % Class2 % Total % Chi-square P-valueLessthan20.000ayear 22 37% 38 63% 60 27%20.001-50.000ayear 35 45% 43 55% 78 36%Morethan50.000ayear 11 35% 20 65% 31 14%Idon'tknow/I'drathernotsay 14 28% 36 72% 50 23% 3,807 0,283

Hoursworkingincoworkingspace Class1 % Class2 % Total % Chi-square P-value0-8hours/week 32 51% 31 49% 63 29%8-16hours/week 11 33% 22 67% 33 15%16-24hours/week 14 40% 21 60% 35 16%24-32hours/week 11 28% 29 73% 40 19%32-40hours/week 11 37% 19 63% 30 14%Morethan40hours/week 3 23% 10 77% 13 6% 7,83 0,166

Transporttocoworkingspace Class1 % Class2 % Total % Chi-square P-valueCar 15 32% 32 68% 47 21%Bike 42 38% 70 63% 112 51%Byfoot 5 38% 8 62% 13 6%Publictransport 20 43% 27 57% 47 21% 1,143 0,767

Sectoroforganization Class1 % Class2 % Total % Chi-square P-valueConsultancy 20 36% 35 64% 55 25%IT 18 51% 28 80% 35 16%Education 10 53% 9 47% 19 9%PR,marketing,sales,advertising,communication 8 32% 17 68% 25 11%Other 26 35% 48 65% 74 34% 2,439 0,656

108

Psychographiccharacteristics

Motivation1 Class1 % Class2 % Total % Chi-square P-valueIwaslookingforaworkplaceoutsidethehome 35 47% 40 53% 75 34%Vibrantandcreativeatmosphereinthecoworkingspace

12 32% 26 68% 38 17%

Affordableaccommodation 7 25% 21 75% 28 13%Socialinteractionwithcoworkers 6 40% 9 60% 15 7%Theopportunitytonetworkwithcoworkers 2 20% 8 80% 10 5%Other 20 38% 33 62% 53 24% 6,476 0,263

Motivation2 Class1 % Class2 % Total % Chi-square P-valueIwaslookingforaworkplaceoutsidethehome 8 21% 31 79% 39 18%Vibrantandcreativeatmosphereinthecoworkingspace

16 35% 30 65% 46 21%

Affordableaccommodation 13 54% 11 46% 24 11%Socialinteractionwithcoworkers 13 52% 12 48% 25 11%Theopportunitytonetworkwithcoworkers 7 32% 15 68% 22 10%Other 25 40% 38 60% 63 29% 10,471 0,063

Motivation3 Class1 % Class2 % Total % Chi-square P-valueIwaslookingforaworkplaceoutsidethehome 10 36% 18 64% 28 13%Vibrantandcreativeatmosphereinthecoworkingspace

16 41% 23 59% 39 18%

Affordableaccommodation 9 36% 16 64% 25 11%Socialinteractionwithcoworkers 12 43% 16 57% 28 13%Theopportunitytonetworkwithcoworkers 5 23% 17 77% 22 10%Other 30 39% 47 61% 77 35% 2,732 0,751

109