The Role of Formalisation, Participation and Context in the Success of Public Involvement Mechanisms in Resource Management1 Jens Newig 1*, Veronika Gaube 2, Karin Berkhoff 3, Kai Kaldrack 4, Britta Kastens 4, Juliana Lutz 2, Bianca Schlußmeier 5, Heidelinde Adensam 2, Helmut Haberl 2 1 Institute for Environmental and Sustainability Communication, Leuphana University, Lüneburg, Germany 2 Institute of Social Ecology, IFF Vienna, Klagenfurt University, Vienna, Austria 3 Institute for Environmental Planning, Leibniz University, Hanover, Germany 4 Institute of Environmental Systems Research, Osnabrück University, Osnabrück, Germany 5 Natur, Kultur, Begegnung, Koblenz, Germany Email address of corresponding author: e-mail: newig@uni.leuphana.de Key words: Public Participation, Participatory Methods, Information Flows, Learning, Effectiveness, Case Study Comparison. Published as: Newig, Jens, Veronika Gaube, Karin Berkhoff, Kai Kaldrak, Britta Kastens, Juliana Lutz, Bianca Schlussmeier, Heidi Adensam and Helmut Haberl. 2008. The Role of Formalisation, Participation and Context in the Success of Public Involvement Mechanisms in Resource Management. Systemic Practice and Action Research 21(6), 423441. doi: 10.1007/s11213-008-9113-9 The original publication is available at http://www.springerlink.com/content/c81p16u56m57743v/ 1 This work was jointly funded as part of the ‘PartizipA’ project by the German Ministry of Education and Research and the Austrian Ministry of Education, Science and Culture under grant no. 07 VPS 10 -2- Abstract In the face of complex and uncertain issues, one important goal of public participation in resource management and research is to foster communication and the inclusion of non-expert knowledge – thus the effective flow of information between project organisers and stakeholders. We compare different methods (instruments, tools) that were employed in the German-Austrian ‘PartizipA’ project to structure information flows in participatory processes. Depending on their goals and context, more or less ‘formalised’ and ‘participatory’ methods were applied, the most important being guided interviews, focus groups, agent-based modelling, nutrient modelling, cognitive mapping and group model building as well as the development of a common document. Two regional case studies, both concerned with European-induced institutional change, are portrayed in which the specific participatory methods were embedded. The Austrian case study involved the analysis and modelling of agricultural land use in the region of St. Pölten against the background of the reform of the Common Agricultural Policy, while the implementation of recent European water policy was the issue in the German agricultural region north of Osnabrück. Presenting both cases in their regional context, the applied methods are first described according to the logic of the entire respective process. Subsequently, the specific methods are systematically analysed and compared according to their objective, context and degrees of participation and formalisation. Finally, we evaluate all methods regarding their effectiveness in terms of goal attainment and their potential generalisation, seeking to respond to the question of when a particular method might best be used. 1. Introduction Participatory processes are becoming increasingly important in the management of natural resources. Policy makers, ‘experts’, and ‘lay persons’ – citizens or representatives of organised interest groups – are collaborating increasingly in public decisions on issues such as water management or land use planning (see, e.g. Beierle and Cayford 2002; Renn 2004). Likewise, transdisciplinary research processes are opening up to include non-scientists, users and other stakeholders not only as research subjects but, moreover, in a role to shape research agendas or to provide various kinds of knowledge (Funtowicz and Ravetz 1993). Given the rising complexity of issues to decide upon and the uncertainties involved, one main reason for the involvement of non-state actors in public decisions is to incorporate different sources of knowledge and to foster social learning. By ensuring an effective flow of information between project organisers and stakeholders, it is expected that it will be possible to arrive at potentially better, i.e. more informed and creative, decision-making (Steele 2001; Pellizzoni 2003; Lee and Abbot 2003; Newig et al. 2005). Classical typologies of participatory mechanisms distinguish between ‘information’, ‘consultation’ and ‘participation’, depending on the way in which information is flowing or being shared (see, e.g. Gramberger 2001; Rowe and Frewer 2005). At the same time, these types also mark the degree of involvement, ‘participation’ be- Published as: Newig, Jens, Veronika Gaube, Karin Berkhoff, Kai Kaldrak, Britta Kastens, Juliana Lutz, Bianca Schlussmeier, Heidi Adensam and Helmut Haberl. 2008. The Role of Formalisation, Participation and Context in the Success of Public Involvement Mechanisms in Resource Management. Systemic Practice and Action Research 21(6), 423-441. doi: 10.1007/s11213-008-9113-9 -3ing labelled the most intensive form.2 In order to avoid terminological confusion, taking the perspective of project organisers,3 in this paper we will refer to an information flow from organisers towards stakeholders as ‘giving information’, from stakeholders to organisers as ‘extracting information’ and an interactive information flow as ‘mutual exchange of information’. A spectrum of methods (instruments, tools) is available to structure information flows in participatory processes. Depending on their goals and context, not only more or less ‘participatory’, but also and more or less ‘formalised’ (or structured) methods can be applied. In this case, formalisation refers to the extent to which information is channelled in a certain way, leaving more or less scope for open communication. Narrative interviews or open space meetings, for instance, are rather non-formalised methods, whereas standardised questionnaires, multi-criteria analyses and modelling exercises involve a high degree of formalisation. In abstract terms, the more formalised a method, the more it serves as a filter, allowing only information of a particular nature to pass while excluding others. The selection and aggregation of information is, of course, a vital element of all cognition, for it defines what is important and what is not, and thus constructs meaning. This can be a tightrope walk, though, for while highly formalised methods allow for a strict selection of relevant information, they risk insufficient information being collected and communicated. This calls for a combination of both formalised and less formalised methods. More specifically, formalised participatory methods have the advantage of eliciting, handling and delivering focused, clear and unambiguous information (or of making uncertainties and ambiguities transparent). What is more, they can counterbalance unwanted group dynamics such as the tendency towards the convergence of ideas (Hinsz et al. 1997: 54) and thus can also maintain the divergence of information. Finally, power asymmetries among participants tend to play a weaker role in more formalised procedures (Rauschmayer and Wittmer 2006: 114). Yet their use also involves important challenges. One is that the process of formalisation is usually more laborious, requires more time and specific expertise, or trained personnel, and often technical equipment and is thus more costly. Secondly, stakeholders may have difficulties in accepting some more formalised methods as they might not comprehend all the modelling steps and may thus distrust certain model implications or simply wish to avoid the efforts required. In the light of these considerations, the question arises as to which kinds of methods (more or less formalised, more or less participatory) are best used under which circumstances. We seek to contribute to a response with insights from the participatory processes conducted as part of the German-Austrian project ‘PartizipA’ (Participative Modelling, Actor and Ecosystem Analysis in Regions with Intensive Agriculture; www.partizipa.net), carried out at the University of Osnabrück and the IFF Social Ecology in Vienna from October 2003 to March 2007. Two case studies involving agricultural regions, both set against the background of European2 Typologies of public involvement mechanisms that build on information or uncertainty are offered by (in brackets: categorising variables): Catt and Murphy 2003 (information provision, contestation, synthesis); Renn 2004 (complexity, uncertainty, ambiguity); Newig et al. 2005 (normative and informational uncertainties); Rauschmayer and Wittmer 2006 (complexity, information, uncertainty). Similar typologies, some also including the nature of environmental conflict and the policy context, are provided by Fiorino 1990; Renn et al. 1995; Bishop and Davis 2002; Green and Hunton-Clarke 2003; Carter 2005. 3 Project organisers are those who are responsible for conducting the participatory process, e.g. competent public authorities, research teams, etc. Published as: Newig, Jens, Veronika Gaube, Karin Berkhoff, Kai Kaldrak, Britta Kastens, Juliana Lutz, Bianca Schlussmeier, Heidi Adensam and Helmut Haberl. 2008. The Role of Formalisation, Participation and Context in the Success of Public Involvement Mechanisms in Resource Management. Systemic Practice and Action Research 21(6), 423-441. doi: 10.1007/s11213-008-9113-9 -4induced institutional change, are portrayed in which a number of specific engagement methods were embedded. The most important methods were guided interviews, focus groups, agent-based modelling, nutrient modelling, cognitive mapping and group model building as well as the development of a common document. The Austrian case involved the analysis and modelling of agricultural land use in the region of St. Pölten against the background of the reform of the Common Agricultural Policy, while the implementation of recent European water policy was the issue in the agricultural region north of Osnabrück. Our analysis proceeds as follows. Presenting both cases in their regional context, methods are first described according to the logic of the entire project process (sections 2 and 3). Subsequently, the specific methods are systematically analysed and compared according to their objective, context and degrees of ‘participation’ and ‘formalisation’ (section 4). By participation we understand the level of information flow (from project organiser to stakeholder and vice versa and mutual exchange of information) as well as the potential influence of participants. Drawing on the model by Rowe and Frewer (2005), a method’s degree of formalisation will be defined by its openness towards participants’ response modes and the structuredness of information aggregation. We conclude by evaluating all methods regarding the effectiveness of their goal attainment and their potential for generalisation, seeking to respond to the question of when to best use which method (section 5). 2. Regional case study: St. Pölten Umland (Austria) 2.1 Background and research objectives The European Union’s Common Agricultural Policy (CAP) reform, implemented on 1 January 2006, shifts incentives for farmers from the current situation in which certain products have been subsidised to allocating resources according to market demand and prevailing natural conditions through decoupled transfer payments. The Austrian case study analysed the effects of the CAP reform on the social and economic situation of farm households, changes in farming practices and, ultimately, land use in a rural region of Lower Austria. The scientific purpose was to evaluate how social and political interventions, in this case the CAP reform, affect patterns of land use as well as socio-economic conditions in rural regions, using an agent-based model embedded in a participatory project design. 2.2 Study area The region of the Traisen valley in Lower Austria, as defined in this project, is made up of 29 municipalities along the Traisen river and around Lower Austria’s capital, St. Pölten. The study region has an area of 700 km2 and is characterised by diverse land use. In the north, a part of the region characterised by flatland, agricultural production is dominated by cropland and wine production. In the south, however, the alpine character of the landscape generally favours milk production, pastures and forestry. Forestry holdings occupy over 200 km2, milk production accounts for another 100 km2 and the remaining 400 km2 is distributed over all other production types. The farms situated in the study region include all farming types usually practiced in Austria. Farms commonly have a size of between 20 and 50 ha. Nearly half Published as: Newig, Jens, Veronika Gaube, Karin Berkhoff, Kai Kaldrak, Britta Kastens, Juliana Lutz, Bianca Schlussmeier, Heidi Adensam and Helmut Haberl. 2008. The Role of Formalisation, Participation and Context in the Success of Public Involvement Mechanisms in Resource Management. Systemic Practice and Action Research 21(6), 423-441. doi: 10.1007/s11213-008-9113-9 -5of all farms are full-time farms, with the remaining half being run by part-time farmers. The number of part-time farms is increasing from year to year. 2.3 Participatory process Originally, the intention was to analyse the social and environmental aspects of the implementation of the European Water Framework Directive (WFD). During the kick-off workshop with regional representatives from the local administration and the government, it became clear, however, that the local stakeholders were much more interested in: (1) an analysis of the impact of the common agricultural policy (CAP 2006) instead of the proposed focus on the WFD, (2) the impact of structural changes in agriculture on the quality of life of women living in agrarian households. For this reason, the research team reframed the main research questions towards an analysis of the impact of CAP reforms, with a focus on gender-related aspects. This shift in the project’s focus of course also fundamentally affected the remaining participatory process. Methodologically, the Austrian case study aimed to generate an agent-based model in close co-operation with the relevant actors. Throughout the research, the participatory process reflected the information requirements of designing the agent-based model. In turn, the agentbased model was applied as a guiding tool in the participatory process to help local stakeholders reflect on the future, developing strategies and policy priorities. The process of agentbased model (ABM) building allows different forms of participation at different stages of the modelling process. An ABM consists of a set of virtual agents that encapsulate the behaviour of the various individuals that make up the system. Hence, when designing the model concept it was necessary to analyse the behaviour of agents. At the beginning, 15 guided interviews were conducted with experts from the provincial government, the federal government, the agency of agriculture and regional stakeholders. In combination with a broad literature research on the CAP reform and the expected impact on agricultural activities, these interviews were an effective instrument for making contact with relevant stakeholders, gaining local knowledge related to our research questions, stimulating actors’ interests and making them feel represented in the ongoing research. A second element in the participatory process was a series of focus groups for female stakeholders that were designed to obtain an insight into women’s perspectives. The group met four times over a period of one-and-a-half years. Each meeting ran for about three hours. In this project, the setting of the focus group was designed together with the head of the Department of Education of the Agency of Agriculture. She invited six female farmers and three female stakeholders from different departments of the Agency of Agriculture. Important criteria for the selection of the female farmers were a good representation of different production sectors (cropland / grassland) and a balanced age pattern. The process of the focus groups was adapted to the whole participatory process. The intention of the first meeting was to test predefined hypotheses concerning the ratio between income and working time. These hypotheses were generated by the analysis of the above-mentioned guided interviews. The outcome of the first focus group meeting and the information about stakeholders’ motivation and behaviour derived from the interviews helped to develop the first version of the ABM design. At this stage of model building, the first simulation runs required evaluation. Several evaluation methods are available. One involves the discussion of the output with stakeholders, followed Published as: Newig, Jens, Veronika Gaube, Karin Berkhoff, Kai Kaldrak, Britta Kastens, Juliana Lutz, Bianca Schlussmeier, Heidi Adensam and Helmut Haberl. 2008. The Role of Formalisation, Participation and Context in the Success of Public Involvement Mechanisms in Resource Management. Systemic Practice and Action Research 21(6), 423-441. doi: 10.1007/s11213-008-9113-9 -6– if necessary – by the redesigning and ‘reformalisation’ of the ABM. Accordingly, the task of the second focus group meeting was to discuss the design and initial output of the agentbased model, where hitherto generated insight was translated into formalised language and diagrams. Concrete decisions on the relevant model output (e.g. labour time per farm, cropland area per farm, etc.) and adjustable model parameters required were taken together with the stakeholders. The third focus group concentrated on developing four ‘storyline’ scenarios for the region. The outcome of this focus group meeting was again incorporated in the ongoing modelling process. Finally, the well-designed and well-evaluated model was used in the fourth and final focus group meeting as a tool to demonstrate the impact of different actions under different framework conditions. Group meetings were moderated by one of the researchers. The most crucial aspect for the participatory process described in the Austrian part of the project was the application of an ABM as a guiding tool to develop scenarios for sustainable resource use (see Rothman and Coppock 1996) together with regional actors. The incorporation in the project team of a representative (Josef Breinesberger) of the regional consulting company Agrar Plus, an individual regarded highly by key regional stakeholders due to his huge practical expertise and local embeddedness, was crucial in terms of giving the stakeholders confidence in the project team. Consequently, stakeholders accepted the research team as ‘one of us’ and were more willing to contribute to the participatory research process. Besides organising contacts, the function of Josef Breinesberger in the project was to directly feed information into the research process and to take responsibility for the outcome of both the participatory process and the ABM. 3. Regional case study: Osnabrück (Germany) 3.1 Background and research objectives The European Water Framework Directive (Directive 2000/60/EC; WFD), which entered into force in December 2000, demands that all European waters achieve “good status” by the year 2015. The WFD, moreover, calls for various modes of public participation and involvement, which the Commission judges as key factors in supporting the successful implementation of policy goals (EU 2002; Newig 2005). Presently, a range of forms of public participation is being put in place (see Kastens and Newig accepted for German examples). Comparable to the process of stakeholder involvement as demanded by the WFD, the German project team established an actors’ platform in a region with intensive agriculture, aiming to investigate and discuss measures for the reduction of diffuse pollution. Methodologically, participatory methods were tested and partly developed which have the potential to be applied in the official processes within the WFD implementation. Published as: Newig, Jens, Veronika Gaube, Karin Berkhoff, Kai Kaldrak, Britta Kastens, Juliana Lutz, Bianca Schlussmeier, Heidi Adensam and Helmut Haberl. 2008. The Role of Formalisation, Participation and Context in the Success of Public Involvement Mechanisms in Resource Management. Systemic Practice and Action Research 21(6), 423-441. doi: 10.1007/s11213-008-9113-9 -7- 3.2 Study area The study area comprises the Hase river catchment north of Osnabrück. A sub-catchment of the Ems in northwest Germany, the Hase covers an area of 3000 km2, 68 percent of which is agricultural land. With 4.7 million livestock, 20 percent of all German pigs are kept in this small region (Klohn and Windhorst 2003). This high livestock density leads to a remarkable amount of organic fertilizer, which involves the risk of nitrogen emissions entering the groundwater. Moreover, according to vulnerability assessment performed in the catchment, the groundwater pollution potential is high in at least parts of the catchment (Berkhoff 2005). The first inventory that was required by the WFD states that 98 percent of the region’s groundwater is unlikely to attain a good status by 2015 due to nitrogen from diffuse sources (NLWKN 2005). On the basis of these facts, the project focused on diffuse nitrogen pollution of groundwater bodies as one of the main ecological threats to the region. 3.3 Participatory process The regional actors’ platform (AP) was established in September 2004, involving 14 representatives from agriculture, water management, environmental protection and administration. Seven meetings took place over a period of 20 months. The main objectives of the AP were:   to reduce uncertainties among stakeholders concerning the new regulations of the WFD in dialogue with other experts; to investigate distinct measures to mitigate diffuse groundwater pollution in the Hase / Osnabrück region, including their costs and potential for realisation. Information flows and discussions during the platform meetings were supported by different consultative and participatory methods that were flexibly adapted to the participants’ requests and discussion shifts. Since participants’ knowledge of the WFD and of groundwater protection varied, it was important to generate a common state of knowledge regarding the most central issues in order to enable all stakeholders to participate equally in the discussions. Hence, the first two meetings concentrated on providing actors with essential information on the WFD and its challenges. A central aspect of discussions in the AP was the individual perception of each stakeholder. In order to identify different interests, causes of conflicts and individual views on possible solutions towards groundwater protection, face-to-face interviews of about 90 minutes duration were conducted with each participant. The interviews were supported by the use of cognitive mapping methods (Vennix 1996, Hodgson 1992). Starting with the regional challenges due to WFD, the interviewees developed interdependencies, causal relations and solutions, all of which were visualised in a poster-like diagram (‘cognitive map’). In three small groups, each involving about five of the interview partners, the participants presented their cognitive maps in order to discuss similarities and differences in their perspectives and to merge their views in one synthesis poster. This forced the actors to be exact in their concepts and helped to build up a common understanding of all the viewpoints and factors involved. The actors’ implicit beliefs became explicit and could then be discussed. The main foci of the synthesis phase were to determine open questions, to collect a list of possible Published as: Newig, Jens, Veronika Gaube, Karin Berkhoff, Kai Kaldrak, Britta Kastens, Juliana Lutz, Bianca Schlussmeier, Heidi Adensam and Helmut Haberl. 2008. The Role of Formalisation, Participation and Context in the Success of Public Involvement Mechanisms in Resource Management. Systemic Practice and Action Research 21(6), 423-441. doi: 10.1007/s11213-008-9113-9 -8regional groundwater protection measures and to identify criteria for their evaluation in terms of suitability and effectiveness. In a subsequent AP session, the synthesis diagrams of the three small groups were presented, resulting in intensive and critical discussions which also raised further open questions. In order to respond to these, the participants agreed upon a common strategy to search for missing information by contacting further experts and to distribute working tasks between stakeholders and the research team. Moreover, the AP developed an evaluation matrix for the measures discussed in the three small groups in order to determine groundwater protection measures for the region. Besides ecological criteria, the matrix involved social and economic parameters that were based on scientific research as well as on evaluations of the stakeholders. This procedure structured the analysis across a range of aspects which the actors assumed to be important. The matrix also served as a concise discussion basis for the development of problem-oriented action recommendations. In order to estimate ecological criteria in the matrix, the computer model STOFFBILANZ (cf. Gebel et al. 2005) was introduced within the platform. STOFFBILANZ was used to evaluate the effectiveness of groundwater protection measures concerning their ability to reduce nitrogen charges. In order to gain the participants’ acceptance, the model was presented and discussed extensively in the AP before scenarios with a particular focus on the amount of organic fertilizer application were calculated and visualised in maps (Berkhoff 2006). Since the implementation of the model took longer than expected and the actors could not agree on quantitative descriptions of required parameters, model results, although discussed intensively in the platform, were not generally accepted by all participants and were therefore not used in the further discussion process. Finally, the actors agreed to summarise the evaluations of discussed measures in a common document (Berkhoff et al. 2006), based on protocols and working documents that had been discussed and adopted previously during the AP process. Corrections and additions to the first draft prepared by the research team were provided by the stakeholders through a questionnaire and were incorporated in a new version, which the actors discussed and adopted in the final AP session. Both stakeholders and project organisers judged this iterative process as highly fruitful in terms of clearly mapping out the common ground as well as shedding light on the remaining differences in perceptions and interests. At the actors’ special request, the document also portrays the methodological AP process and its appraisal. In order to communicate the results of the AP, the participants and the research team jointly presented the document to a broad regional audience in May 2006. The document was distributed as a hard copy and also published on the internet. 4. Comparative analysis In this section, the six different participatory mechanisms applied in the project are analysed with reference to five aspects: 1. The intention of method application refers to the idea that certain methods are assumed to be suitable for specific outcomes the project organisers have in mind. Intentions can be, for instance, to elicit or exchange knowledge, to gain acceptance of or among stakeholders, or to develop a common decision. Published as: Newig, Jens, Veronika Gaube, Karin Berkhoff, Kai Kaldrak, Britta Kastens, Juliana Lutz, Bianca Schlussmeier, Heidi Adensam and Helmut Haberl. 2008. The Role of Formalisation, Participation and Context in the Success of Public Involvement Mechanisms in Resource Management. Systemic Practice and Action Research 21(6), 423-441. doi: 10.1007/s11213-008-9113-9 -92. The second aspect concerns the degree of participation. Depending on the direction of information flow, methods differ in their modes of participation: by ‘giving information’, project organisers intend to inform actors, e.g. in the form of leaflets or internet sites, but in factual terms do not aim at gaining any response. By ‘extracting information’, information is proposed to be elicited from the actors, and project organisers usually prepare specified mechanisms to deal with this information (Rowe and Frewer 2005: 255). Finally, by ‘mutual exchange of information’, project organisers intend to exchange information with actors. A certain degree of dialogue takes place, which serves to transform the opinions of both organisers and actors (cf. ibid.: 256). Moreover, the degree of participation can vary in terms of the capacity for action provided to the actors. Depending on the context, the opportunities for stakeholders to influence or monitor a process and/or its outcome might be either wide or narrow (cf. UNESCO 2006: 46; Haugaard 2003: 95; Rauschmayer and Wittmer 2006: 114). 3. The degree of formalisation depends either on an open or closed response mode and more or less structuredness of information aggregation. While methods with an open response mode allow for free responses from actors to organisers, mechanisms with a closed response mode restrict stakeholders to choosing from among pre-set response options (Rowe and Frewer 2005: 269). Facilitation of aggregation is structured if clear rules for information elicitation and combination are set out (by organisers and/or actors) and followed during the process. Where no facilitation of elicitation and aggregation of information is given, the process is unstructured. But where facilitation is provided, it can still be structured or unstructured (ibid.: 273). 4. Context refers to the conditions under which the method was applied, such as the stage of application, resource needs and actor attributes (e.g. in terms of group heterogeneity, interest of actors, their knowledge, etc.). 5. Finally, an evaluation of the method takes place, addressing the question of goal attainment in the specific application context as well as the effects of the method on project organisers and/or actors. Furthermore and independently from the context, the generalisability of the method is evaluated. Published as: Newig, Jens, Veronika Gaube, Karin Berkhoff, Kai Kaldrak, Britta Kastens, Juliana Lutz, Bianca Schlussmeier, Heidi Adensam and Helmut Haberl. 2008. The Role of Formalisation, Participation and Context in the Success of Public Involvement Mechanisms in Resource Management. Systemic Practice and Action Research 21(6), 423-441. doi: 10.1007/s11213-008-9113-9 - 10 - An overview of the six methods concerning these aspects can be found in Table 1. Guided interviews Agent-based modelling (ABM) Cognitive mapNutrient ping (CM) / modelling Group model (STOFFBILANZ) building (GMB) Elicitation of individual perspectives; information reduction; development of shared problem perception Development of common document Summary of results of participation process; input for policy makers Elicit information from stakeholders, stimulate their interest Test hypotheses, educe information on time use and decision making processes, discuss scenarios Develop a tool for participatory scenario building, characterising and evaluating potential future developments Adjustment of resource flow model to regional conditions; evaluation of groundwater protection measures; model results as a basis for further discussion Direction of information flow Extraction of information Extraction of information and information flow between stakeholders Mutual exchange of information Mutual exExtraction of change of in- information formation and information flow between stakeholders Influence on the process Incorporation of results in further methods Incorporation of common knowledge in ABM Influence on complete model input and structure, not on participation process Influence on parts of model input, not on model structure; free decision on further use of model Influence on complete model input and structure, incorporation of results in further methods Influence on complete document input Response mode (open / closed) Open; free responses Open: free responses Model design: open; parameter estimation: closed Partly open: free responses restricted by predetermined model structure Mixed: predefined questions and result representations, free responses (GMB) Mixed: free responses and yes / nodecisions Group opinion is aggregated and performed by the stakeholders Model design: unstructured; parameter estimation: structured Stringent model structure, but unstructured due to free discussions Structured (but flexible): rules for aggregation, visualisation, discussion Structured: rules for aggregation, discussion, decision Intention Degree of participation Degree of formalisation Context Focus groups Facilitation of aggregation (structured / unstructured) Mutual exchange of information Resource needs (very low – very high) Interviewer /analyser qualification: high; time: very high Facilitator qualification: high; time, particularly for preparation: high Modeller / facilitator qualification: very high; time: very high Modeller / facilitator qualification: very high; time: very high Modeller / facilitator qualification: high; time: very high Author / facilitator qualification: very high; time; very high Actor attributes Heterogeneous group of interviewees, high expertise and interest Only partly heterogeneous group (only females) Homogeneous and heterogeneous groups; high engagement and openness; continuity Heterogeneous group re model expertise and openness; highly sensitive issue Heterogeneous group re. expertise, problem perception and interest; high openness Heterogeneous group re. interest, dedication and willingness to compromise Stage in process Beginning Second half phase of process Whole process Middle phase Beginning phase End of process Published as: Newig, Jens, Veronika Gaube, Karin Berkhoff, Kai Kaldrak, Britta Kastens, Juliana Lutz, Bianca Schlussmeier, Heidi Adensam and Helmut Haberl. 2008. The Role of Formalisation, Participation and Context in the Success of Public Involvement Mechanisms in Resource Management. Systemic Practice and Action Research 21(6), 423-441. doi: 10.1007/s11213-008-9113-9 Evaluation - 11 - Agent-based modelling (ABM) Cognitive mapNutrient ping (CM) / modelling Group model (STOFFBILANZ) building (GMB) Development of common document Guided interviews Focus groups Effectiveness Information elicited as input of the ABM Successful information elicitation from different perspectives Successful creation of model and successful in structuring and reducing of complex information Model adjusted to regional conditions (for research use), no further use as tool for evaluation of measures Successful information elicitation and aggregation; support for focussing discussions and conflict resolution Development of valuable document; support for focussing discussions; difficulties of merging perspectives Generalisability High, classical consultation method High, concrete design depends on the group Partly useful; depends on research focus and system under consideration. High (used by many German public authorities); in participatory processes: rather for less critical issues High if stakeholder perspectives differ High, as long as resources are available Acceptance High, since actors felt they were being taken seriously High, especially after first meeting, when scepticism diminished High, scepticism diminished gradually during process Very low: usefulness was not recognised; critical view on data input High, apart from difficulties of understanding complex individual models Mixed: high for most participants, very low for some others Table 1: Comparison of the methods used in the Austrian and German case studies. See main text for a detailed discussion of how the methods were conducted. 4.1 Guided interviews Guided or semi-structured interviews (Patton 1990; Froschauer and Lueger 1992) are not participatory techniques per se. Interviews can be assumed to be a classical consultation method as the flow of information goes primarily from interview partners to the research team. At the same time, interviews might influence the whole participatory process. Interviews can promote the arousal of actors’ interest in the ongoing research and might further help to determine the participants and issues to be involved in later group processes. To some extent, qualitative interviewing follows a certain interview structure to ensure that information about similar topics is obtained from each person (Kvale 1996; Hoepfl 1997). Otherwise, it utilises open-ended questions, seeking individual responses (Rowe and Frewer 2005: 269) which are not predetermined but flexible in terms of information input. The response mode is open, since stakeholders are allowed to give free input. Even though leeway is given to the interviewees to introduce further issues to be discussed, facilitation of aggregation is structured, since questions are set beforehand. Topics to be discussed during the interview provide at least an initial structure to aggregate stakeholders’ interests and to reveal which proportion of respondents holds certain views (cf. ibid.: 273). Besides expertise in the area of study as well as in qualitative interviewing, interpersonal skills are essential for researchers to draw out information (Patton 1990). Moreover, time resources are needed to post-process individual interviews. Interviews are generally well accepted by the actors, since they do not necessarily require involvement throughout the entire project process. Thus they serve as a beneficial starting point Published as: Newig, Jens, Veronika Gaube, Karin Berkhoff, Kai Kaldrak, Britta Kastens, Juliana Lutz, Bianca Schlussmeier, Heidi Adensam and Helmut Haberl. 2008. The Role of Formalisation, Participation and Context in the Success of Public Involvement Mechanisms in Resource Management. Systemic Practice and Action Research 21(6), 423-441. doi: 10.1007/s11213-008-9113-9 - 12 for participatory processes. A mediator who establishes initial contacts to individual stakeholders can be of particular assistance, thus supporting an atmosphere of trust between the actors and researchers. In the Austrian case study, guided interviews initially served to elicit information from the stakeholders. They further stimulated the stakeholders’ interests by visualising the significance of the project and by treating them as experts, able to tell their own 'story' on the issue at stake. This created an atmosphere of trust, additionally supported by a mediator person, who enabled the project organisers to use interviews as a tool to design subsequent participatory events and processes, such as focus groups and work shops. 4.2 Focus groups In general, the intention to organise focus groups is to provide a discussion forum, collecting various perspectives on a specific topic (see, for example, Littig and Wallace 1997)4. With regard to participatory processes, focus groups are suitable for involving actors in the ongoing research. In contrast to interviews, the moderator in focus groups only initiates the discussion by providing start-up questions and comments, but hardly influences the discussion during the group process unless it stagnates or goes astray. Thus, information input is very flexible, since respondents are influenced not only by the moderator but also by other respondents. This group process in turn causes the response mode to be open and spontaneous, since stakeholders are allowed to make free responses not only to the moderator’s initial input but particularly to the views and statements of other participants. Unless facilitated group aggregation, such as the Delphi technique, or comparable methods are applied in the focus groups (Rowe and Frewer 2005: 274), group opinion is aggregated and performed by the stakeholders, structuring the discussion according to their own intentions. Moderation skills are essential to guarantee a dense information flow and to encourage all participants to contribute to discussions. Organising a series of focus groups that will meet several times during the research process requires a high degree of motivation on the part of the actors involved. To enable this, both the topic of discussion and the group setting can be assumed to be of particular importance. As experienced in the Austrian case study, a mediator who facilitates these aspects (selection of topic, selection and invitation of participants) is supportive. The intention behind organising focus groups was to test several hypotheses that arose out of the interviews and to educe different scenarios regarding structural agricultural change. With regard to the participatory process, focus groups were the most important tool for involving actors in the ongoing research over a certain period of time. This was mainly achieved because the female farmers felt and behaved as a group of experts, sharing similar problems and visions and helping researchers to generate a computer model that would be close to reality. Furthermore, the focus groups provided a forum to visualise and discuss shared problems and solutions that might be useful for female farmers’ everyday lives and for further activities to be organised by the female farmers and/or by their representatives. 4 On focus groups as a qualitative research method, see: Dürrenberger et al. 1997; Jaeger et al. 1999; Stewart and Shamadasani 1990 and Morgan 1993. Published as: Newig, Jens, Veronika Gaube, Karin Berkhoff, Kai Kaldrak, Britta Kastens, Juliana Lutz, Bianca Schlussmeier, Heidi Adensam and Helmut Haberl. 2008. The Role of Formalisation, Participation and Context in the Success of Public Involvement Mechanisms in Resource Management. Systemic Practice and Action Research 21(6), 423-441. doi: 10.1007/s11213-008-9113-9 - 13 - 4.3 Agent-based modelling Agent-based modelling (ABM) is often used when interactions between agents are complex and non-linear (see Bonabeau 2002). In the Austrian case study, ABM both facilitates and is based on participatory processes. In this sense, agent-based modelling aims at eliciting and structuring information about the behaviour of agents and systems (see, e.g., Grimm and Railsback 2005) in cooperation with the relevant actors. With this, it develops a tool for participatory scenario building, characterising and evaluating potential future developments. Participative agent-based modelling could be defined as a process of mutual exchange of information. This is due to the iterative process in which information flows vary according to the different phases of research. First, in the design of the ABM, information is extracted from the relevant actors and processed into the first model version. Subsequently, this first model version is evaluated by the actors and additional information is elicited to correct the model. Finally, the model is used as a tool to define scenarios in cooperation with the actors. The ABM is per se a formalised tool, since actors, actions and interactions have to be characterised in a formalised, computer-based language. In contrast, the process of gaining information relevant to model building varies according to the phases of modelling and in order to parameterise the model as closely as possible with real-world data (see Berger et al. 2001). While in the first phase, open responses are required to obtain a broad overview of existing information, the later phases aim at reducing broad information to relevant information - relevant to both the model and the actors. In general, ABM is very resource-intensive in terms of working time, expertise and money. Especially the implementation of the model as a computer simulation requires specialised expertise and a great amount of working hours. Furthermore, as actors need to participate throughout the entire model-building process, a flexible and work-intensive participatory process has to be designed and accomplished. Moreover, the actors face the challenge of becoming accustomed to the method of modelling and being confronted with a formalised reduction of information. The Austrian case study shows that, in combining a high degree of formalisation with a high degree of participation, ABM serves as an adequate tool to structure and reduce complex information in a way that is useful for both stakeholders and researchers. 4.4 Nutrient modelling Nutrient modelling is not a participatory mechanism itself, but a tool that can be implemented within a participatory process. The aim is to discuss land use changes and different groundwater protection measures in the context of the practical experiences of stakeholders (cf. Kunst et al. 2004). Within two meetings of the Osnabrück actors’ platform, the STOFFBILANZ model (Gebel et al. 2005) was introduced and discussed. Mutual exchange of information took place by considering input variables for a reference scenario and the initial model output that helped stakeholders to decide upon suitable groundwater protection measures for the region. Likewise, project organisers received further input from practice to improve model input and outPublished as: Newig, Jens, Veronika Gaube, Karin Berkhoff, Kai Kaldrak, Britta Kastens, Juliana Lutz, Bianca Schlussmeier, Heidi Adensam and Helmut Haberl. 2008. The Role of Formalisation, Participation and Context in the Success of Public Involvement Mechanisms in Resource Management. Systemic Practice and Action Research 21(6), 423-441. doi: 10.1007/s11213-008-9113-9 - 14 put. Participation opportunities were provided by leaving the decision open as to whether or not the model would be used in further AP meetings. Moreover, the participants could choose among the land use measures to be modelled and were also able to suggest additional measures. On the other hand, STOFFBILANZ follows a stringent model structure. Accordingly, the sessions in which the model was used were organised in strict compliance with this structure in order to gather detailed answers on specific model-related questions. In terms of the degree of formalisation, the process was thus only partly open in the response mode by focussing the discussion on the model structure, but it also allowed free comments on the modelling procedure. The structuredness of information aggregation differed: According to the model structure, information aggregation had to be very formalised. On the other hand, free discussions and the responses of the stakeholders led to unstructured aggregation. The model application in the participatory process was designed as an iterative procedure. Hence, provisional model results were presented to the stakeholders, who had very different backgrounds on modelling practice. Within the time provided, it was impossible to establish confidence in the model and its results on the part of the actors. Instead of a participatory modelling process, the stakeholders asked for complete and validated model results. This was neither intended nor possible within the project, since the modelling process is very timeconsuming and demands detailed expert knowledge that partly had to be provided by the participants (Berkhoff 2006: 4). A further – but probably far more important – reason for the participants’ critical reaction lies in the general context of agricultural groundwater pollution, which is a politically sensitive issue in the region. 4.5 Cognitive Mapping and Group Model Building Cognitive mapping (Vennix 1996, Hodgson 1992) as a special form of guided interview was used in the German case study to elicit broad information about the individual stakeholders’ perspectives on causes of conflicts and possible solutions concerning agricultural groundwater pollution. The resulting visualisations (causal loop diagrams) formed the basis for a twostaged group model building exercise, which served to develop a common understanding among stakeholders and to reduce information to the relevant aspects. The participants persented their causal loop diagrams in small groups, discussed the different perspectives and merged them where possible. The resulting shared group models were then presented and discussed in the larger group and delivered the criteria for the above-mentioned evaluation matrix (see 3.3). In contrast to the ABM, these models were not implemented as a computer simulation. Both cognitive mapping and group model building are tools to extract information from the stakeholders. Their contributions formed the basis for the further participation process by influencing the topics to be discussed and the further methods to be used within the AP. A high degree of participation was combined with a high degree of formalisation. The interviewer / facilitator asked for specific variables (e.g. aims of WFD, potential measures) and required the participants to identify causal relationships between them. The group model building exercise was highly structured, demanding that stakeholders present individual models in a specific manner and limited time, choose a certain number of relevant variables, and classify them according to a prepared structure. Published as: Newig, Jens, Veronika Gaube, Karin Berkhoff, Kai Kaldrak, Britta Kastens, Juliana Lutz, Bianca Schlussmeier, Heidi Adensam and Helmut Haberl. 2008. The Role of Formalisation, Participation and Context in the Success of Public Involvement Mechanisms in Resource Management. Systemic Practice and Action Research 21(6), 423-441. doi: 10.1007/s11213-008-9113-9 - 15 Again, the interviewer / facilitator needed great expertise. In addition, cooperativeness on the part of the participants was vital because they were not used to such methods and needed to become accustomed to the high degree of formalisation, which often meant that discussions were interrupted. The presentation of their individual and often complex models also demanded special presentation skills on the part of the presenters and information processing skills on the part of the audience. Common work on the shared model led to fruitful discussions between stakeholders, who reported that this method helped them understand the perspectives of others. They highly appreciated the entire modelling process. For the project organisers it was advantageous that the need for important changes of process organisation had already become apparent at an early stage. Less participatory methods might have postponed this consequence, leading to more difficulties in the course of adapting to participants’ perspectives. According to our experience, the described combination of methods should be applicable in any other context. 4.6 Development of a Common Document In order to summarise the results of the Osnabrück actors’ platform and to support the official implementation process of the WFD, a common document was developed. During the writing process, conflict resolution turned out to be a prerequisite and an important side-effect. The whole document was based on stakeholders’ contributions, and both content and structure were revised several times according to the participants’ comments. The final version was jointly edited in an AP session by both stakeholders and process organisers. The high degree of participation was also characterised by the stakeholders’ ability to decide whether or not to write the document at all and whether or not to publish it. By deciding to do both, they retained the chance to influence the official implementation process of the WFD. The participatory phases of document development had to be highly structured in order to gather relevant information in a short period of time. The process included closed as well as open response modes. Concrete questions were posed to obtain stakeholders’ specific comments and their agreement to concluding statements was measured by ticking ‘yes’ or ‘no’. However, participants also had the opportunity to make free comments whenever they wanted. In the final editing session, the facilitator structured information aggregation. Stringent rules defined how many comments were allowed and how to decide on the final formulations. Altogether, the entire development process was rather formalised. Eliciting the common views from many documents produced earlier within the AP and incorporating the stakeholders’ statements turned out to be more complicated than expected, which prolonged the writing time for the project organisers. Moreover, many contradicting comments upon the first draft made a careful preparation of the closing discussion necessary. The method required very dedicated participants, because they had to spend additional time between platform meetings. The document was impressive in terms of showing the breadth of information that had been gathered, discussed and evaluated throughout the whole participation process and which might have been lost if it had not been thus summarised. The document has the potential to support other WFD and participation processes, since much can be learned from it, for example regarding evaluation of groundwater protection measures or faPublished as: Newig, Jens, Veronika Gaube, Karin Berkhoff, Kai Kaldrak, Britta Kastens, Juliana Lutz, Bianca Schlussmeier, Heidi Adensam and Helmut Haberl. 2008. The Role of Formalisation, Participation and Context in the Success of Public Involvement Mechanisms in Resource Management. Systemic Practice and Action Research 21(6), 423-441. doi: 10.1007/s11213-008-9113-9 - 16 cilitation tools. The work on a common product, be it a final document, a model or the like, turned out to be a very useful method for exchanging relevant information. 5. Discussion and conclusion Having reviewed six more or less formalised public involvement methods in resource management research, what can we learn from this experience? Generally, we find that the applied methods were adequately chosen and served their intended purposes rather well, with the exception of the nutrient modelling exercise, which was virtually rejected by the participants. While guided interviews and focus groups were mainly employed to extract information from stakeholders, the latter, together with cognitive mapping (CM) and group model building (GMB), also successfully served to foster a flow of information among the stakeholders. Furthermore, the joint construction of the agent-based model (ABM) and the development of a common final document were both well suited to fostering the mutual exchange of information between stakeholders and project organisers. On the basis of the information exchange, learning took place. In terms of effective information uptake, project organisers explored participants’ knowledge and views and accordingly and continuously adapted the (participatory) research design, while participants learned about the regulatory challenges at stake. Furthermore, the mutual information exchange both between participants and project organisers and among stakeholders themselves facilitated collective learning about the causal relations in the socio-environmental systems at stake. The chosen methods not only fostered a fruitful deliberative process but, moreover, managed to eventually ‘get things done’. Thus, the Austrian process yielded an executable and thoroughly validated agent-based model as a resource management tool, while the German actors’ platform produced a diligently crafted and balanced final document as a policy recommendation for the further regional implementation process of the Water Framework Directive. As regards formalisation, we observe in both case studies a shift from rather open and less formalised methods at the outset towards less open, more stringent and on the whole more formalised methods towards the end of the process. This holds both within specific methods and across the spectrum of methods. In general, it seems advisable to commence with rather unstructured methods to collect information that is as complete as possible. In the phases of information reduction and aggregation, more formalised methods turn out to be appropriate. In the case of the Osnabrück actors’ platform, where time pressures to finalise the common document became a problem at the end of the process, the stringent structuring of discussions and decisions turned out to be absolutely necessary. Here, the formalised approach was also able to counterbalance unwanted group dynamics in the production of the final document, which was eventually accepted and endorsed by all actors, although some had previously questioned the approach as a whole. Likewise, the ABM constructed in the Austrian case began with rather open discussions and moved in the implementation phase towards a more structured and formalised approach. The application of formalised computer models, such as agent-based and nutrient models in participatory processes, always involves risks for the project organisers. It is difficult to assess the actors’ acceptance of these methods. The experiences in the PartizipA project suggest, for once, that stakeholders’ acceptance could depend on the stage when computer models are apPublished as: Newig, Jens, Veronika Gaube, Karin Berkhoff, Kai Kaldrak, Britta Kastens, Juliana Lutz, Bianca Schlussmeier, Heidi Adensam and Helmut Haberl. 2008. The Role of Formalisation, Participation and Context in the Success of Public Involvement Mechanisms in Resource Management. Systemic Practice and Action Research 21(6), 423-441. doi: 10.1007/s11213-008-9113-9 - 17 plied in the participatory process. Computer models that are explicitly used to demonstrate the impact of different actions encounter resistance concerning the plausibility of the model. If stakeholders are strongly involved in the designing process of the model from the very beginning, and if the outcome of the model is regularly evaluated by the actors, acceptance seems to be higher. Firstly, the stakeholders in their role as ‘model designers’ are very much responsible for the model and its outcome. Secondly, they have a much better understanding of the internal structure of the model. Thus if stakeholders feel that the model was built or designed by themselves, it is easier for them to accept the results of the model – even if not everything in the computer model can be made visible to them. On the other hand, not all models suit such an open procedure. Where existing modelling frameworks are to be applied in a participatory setting, stakeholders’ input is restricted by the model structure. In relation to this, we should like to stress the crucial importance of the context in which the various available methods are applied. The willingness of stakeholders to participate and accept model results depends largely on their personal interests – and thus the policy setting of the process – their apprehension and expertise, the composition of the actor group and whether financial compensation is provided. The assumption that more formalisation correlates with less participation holds true to a certain extent, namely in terms of the stakeholders’ possibility to influence the process or the specific method. Concerning the direction of information flow, formalisation and participation seem to be independent of each other. Thus, a high degree of formalisation could be linked with the giving or the extraction of information (e.g. cognitive mapping) as well as with mutual exchange (e.g. ABM). In terms of generalisability, finally, we found that this is rather high for ‘established’ methods such as guided interviews and focus groups. We judge ABM, which implies very high resource needs and the willingness of stakeholders, as a partly useful tool, depending largely on the respective context. With nutrient modelling, although a useful a tool as such and scientifically acknowledged for the applied issue, we encountered severe difficulties in the participatory setting, largely due to the highly sensitive issue that did not, in the given context, allow for a high degree of formalisation and thus the precise determination of variables that could oblige certain actors in particular ways. Rather, a more ambiguous documentation of environmental states and desired measures as laid down in the common final document did prove possible and, as such, successful. Cognitive mapping and group model building can again be judged as usable tools, in cases where the stakeholders’ perspectives actually differ and need to be brought together. References Beierle, T.C. and J. Cayford. 2002. Democracy in Practice. Public Participation in Environmental Decisions. Washington, DC: Resources for the Future. Berger, T., M. Goodchild, M.A. Janssen, S.M. Manson, R. Najlis and D.C. Parker. 2001. Methodological Considerations for Agent-Based Modeling of Land-Use and Land-Cover Change. In Agent-Based Models of Land-Use and Land-Cover Change. Report and Review of an International Workshop, October 4–7, 2001, Irvine, California, USA. LUCC Report Series No. 6, eds. D.C. Parker, T. Berger and S.M. Manson. Louvain-la-Neuve: LUCC Focus 1 Office: 7-26. Published as: Newig, Jens, Veronika Gaube, Karin Berkhoff, Kai Kaldrak, Britta Kastens, Juliana Lutz, Bianca Schlussmeier, Heidi Adensam and Helmut Haberl. 2008. The Role of Formalisation, Participation and Context in the Success of Public Involvement Mechanisms in Resource Management. Systemic Practice and Action Research 21(6), 423-441. doi: 10.1007/s11213-008-9113-9 - 18 Berkhoff, K. 2005. Nutrient modelling in an area of intensive livestock husbandry - facing the demands of the WFD. Paper read at Proceedings of the International Conference "Multifunctionality of landscapes", May 18-19, 2005, at Justus-Liebig-University Giessen, Germany. Berkhoff, K. 2006. Application of a GIS-based groundwater vulnerability assessment in a participatory process. In EnviroInfo06: 20th International Conference on Informatics for Environmental Protection - Managing Environmental Knowledge. Graz. Berkhoff, K., K. Kaldrack, B. Kastens, J. Newig, C. Pahl-Wostl and B. Schlußmeier, eds. 2006. EGWasserrahmenrichtlinie und zukunftsfähige Landwirtschaft im Landkreis Osnabrück. Schlussdokument zum PartizipA-Akteursforum September 2004 – März 2006. Osnabrück. Bishop, P. and G. Davis. 2002. 'Mapping Public Participation in Policy Choices.' Australian Journal of Public Administration 61 (1): 14-29. Bonabeau, E. 2002. 'Agent-based modeling: Methods and techniques for simulating human systems.' Proceedings of the National Academy of Sciences of the United States of America 99: 7280–87. Carter, C. 2005. The Role of Participatory Processes in Environmental Governance: the example of agricultural GMOs. In Partizipation, Öffentlichkeitsbeteiligung, Nachhaltigkeit. Perspektiven der Politischen Ökonomie, eds. P.H. Feindt and J. Newig. Marburg: Metropolis-Verlag: 181-207. Catt, H. and M. Murphy. 2003. 'What Voice for the People? Categorising Methods of Public Consultation.' Australian Journal of Political Science 38 (3): 407–21. Dürrenberger, G., J. Behringer, U. Dahinden, A. Gerger, B. Kasemir, C. Querol, R. Schüle, D. Tabara, F. Toth, M. van Asselt, D. Vassilarou and W. N. 1997. Focus Groups in Integrated Assessment: A Manual for a Participatory Research. Darmstadt: Center for Interdisciplinary Studies in Technology, Darmstadt University of Technology. EU. 2002. Guidance on Public Participation in Relation to the Water Framework Directive. Active Involvement, Consultation, and Public Access to Information. Luxemburg. Fiorino, D.J. 1990. 'Citizen Participation and Environmental Risk: A Survey of Institutional Mechanisms.' Science, Technology, & Human Values 15 (2): 226-43. Froschauer, U. and M. Lueger. 1992. Das qualitative Interview zur Analyse sozialer Systeme. Vienna: WUV-Universitätsverlag. Funtowicz, S. and J. Ravetz. 1993. 'Science for the post-normal age.' Futures 25 (7): 739-55. Gebel, M., K. Grunewald and S. Halbfaß. 2005. STOFFBILANZ - Programmerläuterung, Stand November 2005. Dresden: Technische Universität Dresden. Gramberger, M. 2001. Citizens as partners: OECD handbook on information, consultation and public participation in policy-making, Governance. Paris: OECD. Green, A.O. and L. Hunton-Clarke. 2003. 'A typology of stakeholder participation for company environmental decision-making.' Business Strategy and the Environment 12: 292-99. Grimm, V. and F. Railsback. 2005. Individual-based Modeling and Ecology. Princeton, NJ: Princeton University Press. Haugaard, M. 2003. 'Reflections on Seven Ways of Creating Power.' European Journal of Social Theory 6 (1): 87–113. Hinsz, V.B., R.S. Tindale and D.A. Vollrath. 1997. 'The Emerging Conceptualization of Groups as Information Processors.' Psychological Bulletin 121 (1): 43-64. Hodgson, A.M. 1992. 'Hexagons for Systems Thinking.' European Journal of Operational Research 59: 220-30. Hoepfl, M.C. 1997. 'Choosing Qualitative Research: A Primer for Technology Education Researchers.' Journal of Technology Education Volume 9 (1): 47-63. Jaeger, C.C., R. Schüle and B. Kasemir. 1999. 'Focus Groups in Integrated Assessment: A MicroCosmos for Reflexive Modernization.' Innovation - The European Journal of Social Sciences 3 (99). Published as: Newig, Jens, Veronika Gaube, Karin Berkhoff, Kai Kaldrak, Britta Kastens, Juliana Lutz, Bianca Schlussmeier, Heidi Adensam and Helmut Haberl. 2008. The Role of Formalisation, Participation and Context in the Success of Public Involvement Mechanisms in Resource Management. Systemic Practice and Action Research 21(6), 423-441. doi: 10.1007/s11213-008-9113-9 - 19 Kastens, B. and J. Newig. accepted. 'Will participation foster the successful implementation of the WFD? The case of agricultural groundwater protection in North-West Germany.' Local Environment. Klohn, W. and H.-W. Windhorst. 2003. Die sektoralen und regionalen Strukturen der Rinder- und Schweinehaltung in Deutschland. Vechta. Kunst, S., C. Scheer and N. Panckow. 2004. Signifikante Nährstoffeinträge aus der Fläche. Edited by ATV-DVWK (Deutsche Vereinigung für Wasserwirtschaft Abwasser und Abfall) Arbeitsgruppe Diffuse Stoffeinträge, ATV-DVWK-Themen. Kvale, S. 1996. InterViews. An Introduction to Qualitative Research Interviewing. Thousand Oaks, London, New Delhi: Sage. Lee, M. and C. Abbot. 2003. 'Legislation: The Usual Suspects? Public Participation Under the Aarhus Convention.' The Modern Law Review 66 (1): 80-108. Littig, B. and C. Wallace. 1997. Möglichkeiten und Grenzen von Fokus-Gruppendiskussionen für die sozialwissenschaftliche Forschung, IHS, Reihe Soziologie Nr. 21. Vienna. Morgan, D.L., ed. 1993. Successful Focus Groups: advancing the state of the art. Sage focus edition ed. Newbury Park, California: Sage Publication Inc. Newig, J. 2005. 'Die Öffentlichkeitsbeteiligung nach der EG-Wasserrahmenrichtlinie: Hintergründe, Anforderungen und die Umsetzung in Deutschland.' Zeitschrift für Umweltpolitik und Umweltrecht 28 (4): 469-512. Newig, J., C. Pahl-Wostl and K. Sigel. 2005. 'The Role of Public Participation in Managing Uncertainty in the Implementation of the Water Framework Directive.' European Environment 15 (6): 333-43. NLWKN. 2005. Flussgebietseinheit Ems. B-Bericht Mittlere Ems, version of 16.03.2005. Patton, M.Q. 1990. Qualitative evaluation and research methods. Newbury Park: Sage. Pellizzoni, L. 2003. 'Uncertainty and Participatory Democracy.' Environmental Values 12 (2): 195224. Rauschmayer, F. and H. Wittmer. 2006. 'Evaluating deliberative and analytical methods for the resolution of environmental conflicts.' Land Use Policy 23: 108-22. Renn, O. 2004. The Challenge of Integrating Deliberation and Expertise. Participation and Discourse in Risk Management. In Risk Analysis and Society. An Interdisciplinary Characterization of the Field, eds. T.L. MacDaniels and M.J. Small. Cambridge: Cambridge University Press: 289–366. Renn, O., T. Webler and P. Wiedemann. 1995. The Pursuit of Fair and Competent Citizen Participation. In Fairness and Competence in Citizen Participation: Evaluating Models for Environmental Discourse, eds. O. Renn, T. Webler and P. Wiedemann. Dordrecht: Kluwer Academic Publisher: 339-68. Rothman, D.S. and R. Coppock. 1996. Scenarios of Sustainability: The Challenges of Describing Desirable Futures. In Climate Change and World Food Security, ed. T.E. Downing. Berlin, Heidelberg: Springer-Verlag. Rowe, G. and L.J. Frewer. 2005. 'A Typology of Public Engagement Mechanisms.' Science, Technology, & Human Values 30 (2): 251-90. Steele, J. 2001. 'Participation and Deliberation in Environmental Law: Exploring a Problem-solving Approach.' Oxford Journal of Legal Studies 21 (3): 415-42. Stewart, D.W. and P.N. Shamadasani. 1990. Focus groups: theory and practice. Vol. 20, Applied social research methods. Newbury Park, California: Sage Publications Inc. UNESCO. 2006. Water. A shared responsibility. The United Nations World Water Development Report 2. Paris, New York. Vennix, J.A.M. 1996. Group model building: facilitating team learning using system dynamics. Chichester et. al.: John Wiley & Sons. Published as: Newig, Jens, Veronika Gaube, Karin Berkhoff, Kai Kaldrak, Britta Kastens, Juliana Lutz, Bianca Schlussmeier, Heidi Adensam and Helmut Haberl. 2008. The Role of Formalisation, Participation and Context in the Success of Public Involvement Mechanisms in Resource Management. Systemic Practice and Action Research 21(6), 423-441. doi: 10.1007/s11213-008-9113-9 - 20 - Published as: Newig, Jens, Veronika Gaube, Karin Berkhoff, Kai Kaldrak, Britta Kastens, Juliana Lutz, Bianca Schlussmeier, Heidi Adensam and Helmut Haberl. 2008. The Role of Formalisation, Participation and Context in the Success of Public Involvement Mechanisms in Resource Management. Systemic Practice and Action Research 21(6), 423-441. doi: 10.1007/s11213-008-9113-9