of a survey of professionals working with EA in Flanders on the per-ceived effectiveness of EA in this context. The survey was part of theformal evaluation of the Flemish EA legislation, commissioned by theDepartment ofEnvironmental&SpatialDevelopment,and carried outby the authors of this paper.1The survey largely built on the samemethodology as earlier assessments of EA effectiveness in the Nether-lands, UK and Denmark (Arts et al., 2012;Lyhne et al., 2016, 2017;Runhaar et al., 2013) and in this way facilitates international com-parativeassessmentsofeffectivenessofEAandthefactorsthataccountforit,particularlycontextualfactors.Ourpaperaddressesthefollowingquestions:1. How effective is EA in Flanders (as perceived by people activelyinvolvedinEA)?2. WhatfactorsareperceivedtocontributetoEAeffectiveness?3. Whatnewinsightsdoesacomparisonoftheresultsofthesurveyonthe Flemish EA with earlier surveys in the Netherlands, UK andDenmarkyieldregardingthecontext-specificityofEAeffectiveness?Thepaperunfoldsasfollows.InSection2 weoutlineouranalyticalframeworkandexplainthesurvey.InSection3 weprovideabriefin-troductionintotheFlemishEAsystem.ResultsarepresentedinSection4.WewrapupourmainconclusionsandreflectonourstudyinSection5.2. Analyticalandmethodologicalframework2.1. AnalyticalframeworkIn the EA literature, effectiveness and performance are often em-ployedinterchangeably(VanDorenetal.,2013).Ingeneralterms,bothconcepts refer to the degree to which EA meets its purposes. Since avarietyofpurposesareassignedtoEA(includingnotonlyformalgoalsthat are formulated in EA legislation and policy documents, but alsopurposes that e.g. stakeholders and researchers assign to EA;RozemaandBond,2015),itisnotsurprisingthateffectivenessandperformanceareoperationalisedindifferentways.Acommonlymadedistinctionisbetween ‘procedural’ and ‘substantive’ effectiveness. While the formerreferstothefulfilmentofproceduralrequirementssuchasthequalityoftheassessment,offeringsufficientopportunitiesforpublicparticipationandthetimelydeliveryofanEAreport(e.g.Sadler,1996;Zhangetal.,2012),thelatterreferstothedegreetowhichEAcontributestoabetterconsideration of environmental concerns in the preparation and deci-sion-makingaboutprojectsandplansand,eventually,toareductioninenvironmentalpressures(Kolhoffetal.,2016).Intheirstateoftheartoverview of EA effectiveness studies,Loomis and Dziedzic (2018)dis-tinguishtwoadditionaldimensionsofeffectiveness:transactive,whichconcernsthecosts(financialandtemporal)associatedwithEAandthenormativeeffectiveness,whichrefersto“theextenttowhichthepolicymeetsitsidealpurpose”(ibid.,p.30).Papers like those ofVan Doren et al. (2013)and Loomis andDziedzic(2018)providerichoverviewsofthevariousinterpretationsofEA effectiveness as well as indicators employed or suggested. In thispaper we restrict ourselves to thesubstantiveeffectiveness of EA and,more specifically, the extent to which EA contributes to the in-corporation of environmental objectives and concerns into the devel-opmentandapprovalofprojectsandplans.ThisfocusedinterpretationofEAeffectivenessnotonlyaddressestheultimateaimofEA(Kolhoffet al., 2016) but also allows for a more comprehensive exploration offactors affecting EA effectiveness, which is complicated in the case ofmultipleeffectivenesscriteria.IntheIntroductiontothispaperwedistinguishedbetweenexanteandexposteffectivenessofEA,referringtotwomomentsinprojectandplan development where an EA can have effects. The ex ante effec-tiveness refers to the incorporation of environmental objectives, oftenbeyond the minimum legal norms, in anticipation of an EA (so evenbefore the actual assessment is conducted). This is also known as the‘preventive effect’ of EA (Ten Heuvelhof and Nauta, 1997). Ex posteffectivenessreferstotheadjustmentofprojectsandplansinresponsetotheEAreport(Artsetal.,2012).Bothformsofeffectivenessareaf-fectedbymanyfactorsthatexplaintheeventualeffectivenessofEA.Inlinewith Artsetal.(2012)andLyhneetal.(2017),whichbuiltonotherpapers, we distinguish between the following, interrelated, categoriesoffactorsthatexplainEAeffectiveness:•Factors associated with the instrumentitself, e.g. the legal require-menttoconductEAforparticularplansandactivitiesandthelegalresponsibilitiesofactorsinvolved(includingproponents,competentauthorities, advisors, consultancy companies and stakeholders),which is found to be a main explanation for the eventual effec-tiveness of EA. These types of factors represent governance me-chanisms,i.e.rulesthatinvolveparticularactorsintheEAprocessand that aim to influence their behaviour, and that may differ be-tween countries. Sometimes governance mechanisms may haveunintended effects; for instance,Lyhne et al. (2017: 248)suggestthat“GreaterresponsibilityforEIAforthecompetentauthorityprovidesanegativeincentivefortheprojectproponenttouseEIAproactivelyasatool to enhance the environmental performance of projects”, implyingthat(legal)ownershipofEIAatleastpotentiallyinfluenceshowthetoolisusedandwhatitachieves;•Factors associated withhowthe instrument is applied, e.g. thequality of the assessment in terms of accuracy of the assessment,scope of effects, readability etc. and opportunities for stakeholdersto participate in the EA process. These factors logically also varybetweencontexts;•Factors associated with thecontextin which the instrument is ap-plied: the actors involved and for instance their concerns for theenvironment.Manyscholarshaveacknowledgedtheimportanceofcontext for how EA is designed and for its effectiveness, but littleclarityexistsaboutwhatconstitutes‘context’exactly(RunhaarandDriessen, 2007). Contextual factors that were included in earlierstudiesofEAeffectivenessintheNetherlands,theUKandDenmarkincluded contextual factors such as openness of proponents orcompetentauthoritiestoenvironmentalconcernsorhowEAimpactsupon decision-making processes in terms of lead times and costs(Artsetal.,2012;Runhaar etal.,2013;Lyhneetal.,2017).Lyhneetal.(2017)providefurtherexamples,suggestingthattheinfluenceofthepublicandotheractorscanbeamplifiedbyanaffordableandaccessiblecomplaintssystemandthatthesizeofthecommunityofactorsprofessionallyorotherwiseinvolvedinEAhasnosubstantialinfluenceonitseffectiveness.For a more detailed description of these factors we refer to theabovepublications.2.2. ResearchdesignStudies into the effectiveness of EA have employed a variety ofmethods. Case studies have been conducted by for instanceRunhaarandDriessen(2007)andRozemaandBond(2015).DocumentanalysisisoftenemployedforassessingtheproceduraleffectivenessofEA(e.g.Ahmad and Wood, 2002), although this can be complemented withexpert interviews (Kolhoff et al., 2018). The paper byLoomis andDziedzic(2018)conductedametaanalysisofpublishedpapersonthesubject. For international comparative studies, often surveys are em-ployed, which facilitates the collection of relatively many data in acomparableway.SimilarasRunhaaretal.,2011(fortheNetherlands),Artsetal.,2012(fortheUK)andLyhneetal.,2016(forDenmark)we1A copy of the report (in Flemish;Tractebel and KENTER, 2018) can beobtaineduponrequesttothefourthauthorofthepaper.H.Runhaar,etal.Environmental Impact Assessment Review 76 (2019) 113–119114