The Future of Participatory Approaches Using Geographic Information: Developing a Research Agenda for the 21St Century Steve Carver

Volume 15 • APA I • 2003 Journal of the Urban and Regional Information Systems Association

CONTENTS

REFEREED 5 Introduction to the Special Issues on Access and Participatory Approaches in Using Geographic Information Harlan J. Onsrud and Max Craglia, Co-Editors

9 Toward a Framework for Research on Geographic Information-Supported Participatory Decision-Making Piotr Jankowski and Timothy Nyerges

19 In Search of Rigorous Models for Policy-oriented Research: A Behavioral Approach to Spatial Data Sharing Uta Wehn de Montalvo

29 Cultural and Institutional Conditions for Using Geographic Information; Access and Participation W.H. Erik de Man

35 A New Era of Accessibility? Sarah Niles and Susan Hanson

43 World Status of National Spatial Data Clearinghouses Joep Crompvoets and Arnold Bregt

51 Access to Geographic Information: A European Perspective Max Craglia and Ian Masser

61 The Future of Participatory Approaches Using Geographic Information: developing a research agenda for the 21st Century Steve Carver

73 Transparency – Considerations for PPGIS Research and Development Christina H. Drew Journal

Publisher: Urban and Regional Information Systems Association

Editor-in-Chief: Stephen J. Ventura

Journal Coordinator: Scott A. Grams

Electronic Journal: http://www.urisa.org/journal.htm

EDITORIAL OFFICE: Urban and Regional Information Systems Association, 1460 Renaissance Drive, Suite 305, Park Ridge, Illinois 60068- 1348; Voice (847) 824-6300; Fax (847) 824-6363; E-mail [email protected].

SUBMISSIONS: This publication accepts from authors an exclusive right of first publication to their article plus an accompanying grant of non- exclusive full rights. The publisher requires that full credit for first publication in the URISA Journal is provided in any subsequent electronic or print publications. For more information, the “Manuscript Submission Guidelines for Refereed Articles” is available on our website, www.urisa.org, or by calling (847) 824-6300.

SUBSCRIPTION AND ADVERTISING: All correspondence about advertising, subscriptions, and URISA memberships should be directed to: Urban and Regional Information Systems Association, 1460 Renaissance Dr., Suite 305, Park Ridge, Illinois, 60068-1348; Voice (847) 824-6300; Fax (847) 824-6363; E-mail [email protected].

URISA Journal is published four times a year by the Urban and Regional Information Systems Association.

© 2003 by the Urban and Regional Information Systems Association. Authorization to photocopy items for internal or personal use, or the internal or personal use of specific clients, is granted by permission of the Urban and Regional Information Systems Association.

Educational programs planned and presented by URISA provide attendees with relevant and rewarding continuing education experience. How- ever, neither the content (whether written or oral) of any course, seminar, or other presentation, nor the use of a specific product in conjunction therewith, nor the exhibition of any materials by any party coincident with the educational event, should be construed as indicating endorsement or approval of the views presented, the products used, or the materials exhibited by URISA, or by its committees, Special Interest Groups, Chapters, or other commissions.

SUBSCRIPTION RATE: One year: $295 business, libraries, government agencies, and public institutions. Individuals interested in subscriptions should contact URISA for membership information.

US ISSN 1045-8077

2 URISA Journal • Vol. 15, APA I • 2003 URISA Journal • Vol. 15, APA I • 2003 3 EDITORS AND REVIEW BOARD

Francis J. Harvey, Department of Geography, URISA Journal Editor Article Review Board University of Minnesota Kingsley E. Haynes, Public Policy and Editor-in-Chief Geography, George Mason University Peggy Agouris, Department of Spatial Stephen J. Ventura, Department of Information Science and Engineering, University Eric J. Heikkila, School of Policy, Planning, and Environmental Studies and Soil Science, of Maine Development, University of Southern California University of Wisconsin-Madison Stephen C. Hirtle, Department of Information Michael Batty, Centre for Advanced Spatial Science and Telecommunications, University of Analysis, University College London (United Pittsburgh Kingdom) Thematic Editors Dr. Gary Jeffress, Department of Geographic Kate Beard, Department of Spatial Information Science and Engineering, Information Science, Texas A&M University- Editor-Urban and Regional Information Corpus Christi Science University of Maine Richard E. Klosterman, Department of Lewis Hopkins, Department of Planning, Yvan Bédard, Centre for Research in Geomatics, Geography and Planning, University of Akron University of Illinois-Champaign/Urbana Laval University (Canada) Robert Laurini, Claude Bernard University of Editor-Applications Research Barbara P. Buttenfield,Department of Lyon (France) Lyna Wiggins, Department of Planning, Geography, University of Colorado Rutgers University Thomas M. Lillesand, Environmental Keith C. Clarke, Department of Geography, Remote Sensing Center, University of Editor-Social, Organizational, Legal, University of California-Santa Barbara Wisconsin-Madison and Economic Sciences David Coleman, Department of Geodesy and Paul Longley, Ian Masser, Department of Urban Planning Centre for Advanced Spatial Analysis, Geomatics Engineering, University of New and Management, ITC (Netherlands) University College, London (United Kingdom) Brunswick (Canada) Xavier R. Lopez, Oracle Corporation Editor-Geographic Information Science David J. Cowen, Department of Geography, Mark Harrower, Department of Geography, David Maguire, Environmental Systems Research University of South Carolina University of Wisconsin Madison Institute Massimo Craglia, Department of Town & Editor-Information and Media Sciences John McLaughlin, Research and International Regional Planning, University of Sheffield Michael Shiffer, Department of Planning, Cooperation, University of New Brunswick (Canada) (United Kingdom) Massachusetts Institute of Technology Harvey J. Miller, Department of Geography, William J. Craig, Center for Urban and University of Utah Editor-Spatial Data Acquisition and Regional Affairs, University of Minnesota Integration Joel L. Morrison, Center for Mapping, Ohio Gary Hunter, Department of Geomatics, Robert G. Cromley, Department of Geography, State University University of Melbourne (Australia) University of Connecticut Atsuyuki Okabe, Department of Urban Editor-Geography, Cartography, and Kenneth J. Dueker, Urban Studies and Engineering, University of Tokyo (Japan) Cognitive Science Planning, Portland State University Jeffrey K. Pinto, School of Business, Penn State Erie David Mark, Department of Geography, Geoffrey Dutton, Spatial Effects Gerard Rushton, Department of Geography, SUNY-Buffalo University of Iowa Max J. Egenhofer, Department of Spatial Editor-Education Information Science and Engineering, University Jie Shan, School of Civil Engineering, Karen Kemp, Department of Geography, of Maine Purdue University University of California-Berkeley Bruce D. Spear, Federal Highway Administration Manfred Ehlers, Geoinformatics and Institute for Environmental Sciences, University of Vechta Jonathan Sperling, Policy Development & (Germany) Research, U.S. Department of Housing and Urban Development Section Editors Manfred M. Fischer, Economics, Geography & Geoinformatics, Vienna University of David J. Unwin, School of Geography, Birkbeck Software Review Editor Economics and Business Administration College, London (United Kingdom) Jay Lee, Department of Geography, Kent State (Austria) Stephen J. Ventura, Environmental Studies and Soil Science, University of Wisconsin-Madison University Myke Gluck, Department of Math and Book Review Editor Computer Science, Virginia Military Institute Nancy von Meyer, Fairview Industries Rebecca Somers, Somers-St. Clair Michael Goodchild, Department of Geography, Barry Wellar, Department of Geography, Literature Review Editor University of California-Santa Barbara University of Ottawa (Canada) Michael F. Worboys, Department of Computer Zorica Nedovic, Department of Urban and Michael Gould, Department of Science, Science, Keele University (United Kingdom) Regional Planning,University of Illinois- Experimentales Universitat (Spain) Champaign/Urbana Benjamin Zhan, Department of Geography, Daniel A. Griffith,Department of Geography, Southwest Texas State University Syracuse University

2 URISA Journal • Vol. 15, APA I • 2003 URISA Journal • Vol. 15, APA I • 2003 3 URISA Journal • Vol. 15, APA I • 2003 5 Introduction to the Special Issues on Access and Participatory Approaches in Using Geographic Information

Harlan J. Onsrud and Max Craglia, Co-Editors

Access issues and participatory approaches in using geographic The second primary research theme focuses on access as a information have been persistent and growing areas of research basis of wealth and power in society and addresses societal issues focus across the globe over the past decade. The proliferation of such as equity, ownership, and control. Those active in this re- spatial technologies, growth of geographic databases, pervasive search domain argue that the foundations of legal rights of citizens use of the web, and expanding use of mobile communication and scientists to access information are being undermined as we networks are providing exciting opportunities for increasing move into networked digital data environments. Thus, research- personal access to virtual and physical resources and enabling ers focused in this arena believe it is important to identify the broader participation in decision making. Yet these same changes processes by which losses in access are occurring, publicize that the are exacerbating existing social inequities and giving rise to new losses are occurring, explore alternatives for halting or reversing the access problems. losses, and investigate models for expanding access or providing There are at least two major research themes in address- more equitable access. There is also a need to observe society in ing issues of “access” within the geographic information science action through assessment of the impact of evolving policy and (GIS) research community. The first involves spatial concepts and legal trends on access to scientific and technical data generally theories that relate to access to goods and services generally. The and geographic data specifically. second involves the notion of access to geographic information Closely affiliated with concepts of access is the concept of as a source of wealth and power. participation. Geographic information technologies are seen Over time, geographers have developed theories about the alternatively as tools that empower communities versus tools way individuals and aggregates of individuals respond to the con- that invasively advantage some people and organizations while straints of time, cost, and effort in order to gain access to work, marginalizing others. Public Participation GIS research efforts shopping, recreation, and other spatially distributed activity op- largely adhere to the first and more positive view and seek to portunities. Many of the models developed from these theories develop approaches, mechanisms, technologies, and institutions have been very useful. However, the nature of accessibility is that aid self-determination by various forms of self-defining changing as many goods and services may be accessed without communities with full awareness of the second view that choices recourse to physical movement. Information, in combination with made may have negative as well as positive power ramifications. the infrastructure that carries it, is a new and expanding resource Although consistency in definition has yet to emerge, some of that often replaces labor, capital, and physical resources. Access the characteristics envisioned for participatory geographic infor- to physical space can be replaced or complemented by access to mation technologies have included: 1) emphasis on the role of virtual space in which traditional notions of distance, nearness, participants in creation and evaluation of data; 2) accommodation and spatial interaction lose meaning or, at the very least, must of equitable representation of diverse views, preserving contradic- be reassessed. In virtual environments the activities appear to be tion, inconsistencies, and disputes against premature resolution; more people-based than place-based. Where you physically are 3) system outputs redefined to reflect the standards and goals of may become less and less an indicator of what you may be doing. the participants; 4) capabilities for managing and integrating all Thus, the traditional assumption of a strong correlation between data components and participant contributions from one interface place and activity upon which many geographic models have been – technologically sophisticated but easy to use; 5) preservation based often may be unwarranted in virtual space. and representation of the history of its own development and be

URISA Journal • Vol. 15, APA I • 2003 5 more capable of handling time components than existing GIS, ing of how the costs of different models structure participant and 6) embedment of the condition that “Public Participation interaction. The power of their framework is in linking premises GIS” is not primarily enabling technology focused but is primar- with research questions and testable hypotheses and the empiri- ily an ongoing “process” of self-determination by self-defining cal testing of the hypotheses. This may lead to verification of the communities. theoretical framework and therefore a logical basis upon which to In the Fall of 2001, a workshop with participation by United base guidelines for designing and using geographic information States and European researchers was held in Spoleto, Italy to assess technologies in participatory decision making. the current state of research on access to geographic informa- The development of spatial data-sharing cultures is important tion and on geographic access theory, to evaluate the impact of to the successful implementation of geographic information tech- evolving policy and legal trends in the U.S. and Europe on access nologies among and within user groups and to the advancement to scientific and technical data generally and geographic data of GIS generally. After reviewing several theoretical models of specifically, to assess the current state of research on participa- decision-making, Uta Wehn de Montalvo demonstrates in her tory approaches surrounding the use of geographic information, article that the theory of planned behavior drawn from the field to explore commonalities and differences in U.S. and European of social psychology can be fruitful as an organizing principle for directions of research within these arenas, and to develop a joint research into spatial data sharing. That is, the theory has been used U.S./European research agenda on geographic information access successfully in identifying in a systematic fashion the incentives and participatory issues. and disincentives that determine the intention of key individuals The idea for a “Workshop on Access and Participatory Ap- within an organization to share spatial data. Results of her em- proaches in Using Geographic Information” arose from research pirical work are presented. The author suggests that the strength discussions among university members of the University Con- of the approach is in identifying context-specific incentives and sortium for Geographic Information Science (UCGIS) and the disincentives to spatial data sharing which in turn may provide a Association of Geographic Information Laboratories in Europe basis for policy makers in achieving desired sharing behaviors. (AGILE). The articles in this first and a second special issue of The third article by W.H. Eric de Man is yet another that the URISA Journal arose from papers presented at the Spoleto draws on theories from other scholarly domains to explore access Workshop as well as from a subsequent call for papers on the and participatory processes in geographic information settings. topics of the Workshop. The Workshop was jointly funded by the Approaches are needed to describe and analyze the predominant National Science Foundation and the European Science Founda- cultural conditions affecting access to and participation in the use tion, and the final report from the Workshop is contained in the of geographic information. This article introduces two existing second special issue. models used to describe and compare different cultures, namely the approaches of Geert Hofstede and Mary Douglas. Because access to and use of geographic information are embedded within Special Issue I the culture and institutional arrangements of a host society, the We begin the issue with three articles focused on methodological author argues that cultural and institutional values and factors and research framework issues. Next we present an article that must often be accounted for in order to more completely explain explores a framework for considering accessibility in the context and identify resistance or non-resistance to particular approaches of evolving technology and social structures. This is followed by to sharing or using spatial data. two articles that pursue a survey approach in exploring access is- Many of the accessibility models developed over time by sues. Finally, we present two articles that focus on participatory scholars have been very useful for understanding how people experiences in geographic information technology settings. are related spatially to their economic and social activities. This To establish a context for research on geographic informa- understanding has improved the quality of life of communities tion-supported participatory decision-making, Piotr Jankowski by enabling better planning and design of land development and Timothy Nyerges explore the multiple constructs of par- activities, transportation systems, and preservation projects. ticipatory situations and highlight how models of participatory However, the previous models and inputs to those models need situations can help organize our understanding of participatory to be reassessed in the light of different ways that people now settings in the use of GIS. The authors present a framework for communicate, learn, work, and conduct commerce brought comparing participatory GIS models. Eighteen empirical research about by the Internet. Sarah Niles and Susan Hanson illustrate strategies are compared to each other, and the authors suggest numerous ways that grounded geographies mark cyberspace and circumstances under which each strategy might contribute to examine the intersection of people’s use of information technol- empirical investigations of participatory GIS use activities. They ogy and grounded socio-spatial relations. They argue and present urge systematic comparisons by the research community in order evidence that physical access to the Internet closely resembles to gain an overall sense of how and under what circumstances a pre-Internet spatial patterns of access to goods and services. Their participation model might be more beneficial than others in terms examples illustrate how the social context in which online mate- of efficiency, effectiveness, and equity, and to gain an understand- rial is produced and consumed can greatly affect the accessibility of that information.

6 URISA Journal • Vol. 15, APA I • 2003 URISA Journal • Vol. 15, APA I • 2003 7 Although insufficient by itself, access to spatial data is a neces- the current use of geographic information to facilitate involvement sary precondition to be able to engage meaningfully in decision in decision making and then explores factors working against or making that depends on or could be affected by such data. Across for use of geographic information and technologies in communal the globe, various actors have been engaged in making spatial decision-making. A research agenda is outlined for building on data available to others through the establishment of a system strengths, addressing weaknesses, exploring opportunities, and of clearinghouse nodes. Sometimes, the nodes provide access to neutralizing threats in the use of geographic information tech- metadata so that others may know that specific spatial data exist nologies in participatory processes. while other nodes provide online access to actual spatial data. Specific groups are often tasked through legislation or other Joep Crompvoets and Arnold Bregt accomplished a survey of the authority to make decisions affecting large segments of the popula- status of such nodes in December of 2001. Their survey shows tion. Transparency in the data and processes used in such decision that nodes had been established in 59 countries as of that point in making is very important when the uncertainties of following time. Their article also highlights differences in the use, content, particular courses of action may be high or the potential nega- and management of various nodes across the globe. tive effects of decisions may be severe or widespread. Geographic An article by Max Craglia and Ian Masser next addresses is- information and Internet technologies may be used as positive sues of access to spatial data from a European perspective. They tools in increasing the transparency of decisions. In an article by review recent policy and legislative developments in Europe and Christina Drew, an example is detailed of how increased transpar- identify three main strands of research on access to geographic ency was facilitated through the use of a geographic information information that they believe deserve concerted effort by the re- system and the Internet. The example illustrates the complexity search community in Europe. In brief, the strands include a focus and challenges of measuring decision transparency and highlights on the role of governments in developing frameworks for sharing the need for research and methodological development focused spatial information, a focus on the organizational level, and a focus on such measures. on regulatory frameworks in national and trans-national contexts such as those that seek to protect intellectual property rights while Articles in this special issue are based partially upon work sup- promoting wider access and use of geographic information. ported by the National Science Foundation under Grant No. The article by Steve Carver focuses on participatory ap- 0098389 and the European Science Foundation. Any opinions, proaches using geographic information. It pays initial attention findings, and conclusions or recommendations expressed in this to general theories of participation and empowerment before material are those of the authors and do not necessarily reflect examining the role of geographic information in providing a the views of the National Science Foundation or the European framework for active public participation. The author reviews Science Foundation.

6 URISA Journal • Vol. 15, APA I • 2003 URISA Journal • Vol. 15, APA I • 2003 7 URISA Journal • Jankowski, Nyerges 9 Toward a Framework for Research on Geographic Information-Supported Participatory Decision-Making

Piotr Jankowski and Timothy Nyerges

Abstract: This article offers a framework for evaluating geographic information technology applied in the context of participatory problem solving and decision-making. The framework consists of constructs and detailed aspects describing significant issues of participatory decision-making. Constructs reflect the structure, while aspects address the content of participatory decision- making. Aspects are tied through premises, which present fundamental statements about the nature of convening, process, and outcome phases of decision-making. The power of the framework lies in linking premises with research questions and testable hypotheses, which can be formulated on the bases of research questions. Empirically based testing of hypotheses leads in turn to verification of theoretical framework and to design guidelines for future uses of geographic information technology in participatory decision-making.

Introduction in environmental decision-making called the “analytic-deliberative At a workshop on Access and Participatory Approaches in Us- process.” The model involves three consecutive steps: ing Geographic Information organized by the U.S. National 1) identification and selection of concerns and evaluation Science Foundation and the European Science Foundation in criteria; Spoleto, Italy, in December 2001, the development of place- 2) identification and measurement of impacts and consequences based methodologies and methods for more inclusive community related to different policy options; and participation in spatial decision-making was recognized as one of 3) conducting a discourse with randomly selected citizens as six core research areas in the context of geographic information jurors and representation of interest groups as witnesses. system (GIS) and society (Weiner et al. 2001). The plea for the development of methodologies implies some shortcomings in This model was applied with different results in Germany, the current state of knowledge on participatory decision-making Switzerland, and the United States. While participants in Ger- and begs the question: “What are the main obstacles to successful many and Switzerland were grateful for the invitation to par- implementations of participatory decision-making supported by ticipate, U.S. participants distrusted prefabricated participation geographic information technology?” models and suspected hidden agendas with such an approach. Many examples of community development projects in- This underscores the cultural differences that come into play in volving public participation from around the world suggest that participatory decision-making. Renn concluded that the success cultural and political context rather than hardware and software of public involvement in the U.S. would depend on securing are the main obstacles to successful public participation in deci- approval of the process by the affected communities before that sion-making (Craig et al. 2002). This may be true for participatory process is put into play. Based on multiple empirical studies in decision-making situations employing relatively simple informa- Europe and North America, Renn also concluded that discursive/ tion tools. Yet there are many other spatial decision situations participatory processes needed a structure that assures the integra- where the success of a participatory approach requires both a tion of technical expertise, regulatory requirements, and public good understanding of cultural and political context and more values. Other authors (for example, Pickles 1995) postulated that efficient, effective, and equitable information and communication participatory decision-making should become a tool of politi- tools. In these types of situations, there is a lack of operational cal discourse challenging the instrumental logic of the existing knowledge to guide the selection of information tools appropriate practice of decision-making. This, according to Pickles, may for the task at hand. Insufficient knowledge also extends to the require using a different logic than a normative approach to the influence of culture and political systems on shaping feasible forms decision-making process. Shiffer (1995) saw the promise of Inter- of participation. One example of the importance of knowing how net-based participatory decision-making, but at same time admits different cultural and political contexts influence what works and potential difficulties arising from involving computer-technology what does not work in adopting a participatory approach to a challenged people and differential understanding of information decision-making process is work by Renn et al. (1995), which is presented through virtual images and map-based representations. based on ideas articulated in a National Research Council report Weiner et al. (2001:13) expressed belief in potential contributions entitled Understanding Risk (1996). In that study, Renn presented of public participation geographic information systems (PPGISs) the results of evaluating one specific model of public participation stating that “Future PPGIS academic research can thus contribute

URISA Journal • Jankowski, Nyerges 9 Figure 1. EAST-2 framework consisting of convening, process, and outcome constructs (including example aspects) plus the respective premises provides a conceptual map for understanding a participatory decision process.

significantly to geography and to the social sciences in general.” The Need for a Framework This statement is somewhat tempered by their earlier concern Characterizing Participatory GIS “…that academics engaged with PPGIS will tire and fall back to their familiar role as researchers. In such cases, PPGIS has the Use potential to become another form of community exploitation” Any successful implementation of participatory decision-making (Weiner et al. 2001:9). This seeming inconsistency underscores requires the knowledge of people who are involved in the deci- the tension resulting from the interdisciplinary character of in- sion-making process, plus their values and expectations. Imple- quiries about the access to and the participatory use of geographic mentation requires data and information about the problem at information. Applications and research about participatory deci- hand – both scientific and non-scientific – representing individual sion-making are rooted both in social-behavioral sciences and in experiences and the collective wisdom of participants. It requires broadly defined information science and technology. This is a an understanding of socio-political influence on power and con- reflection of what geographic information-supported participa- trol. Last, but not least, implementation requires data processing tory decision-making is about – open and transparent access to and communication tools. spatially enabled data and information handling tools for people The range of tools that may be appropriate in a given decision interested in place-based problem solving and decision-making problem context ranges potentially from push-pin maps to such in a specific socio-political context. This ad hoc definition dem- high-tech structures as virtual reality visualization environments. onstrate the complexity of the subject. But how is one to know whether a given problem situation is a

10 URISA Journal • Vol. 15, APA I • 2003 URISA Journal • Jankowski, Nyerges 11 good candidate for a participatory approach and, if so, which issues for characterizing group decision-making (Jankowski and information and communication tools are appropriate for the Nyerges 2001a). The seven premises of EAST-2 describe the rela- task at hand? There is a clear need for a knowledge base about tions among the eight constructs (Figure 1; the seven premises when, in what context, and how one should apply participatory are numbered from P1 through P7 and the eight constructs are approaches to spatial decision problems and which information depicted in the respective eight boxes). The structuration process and communication technologies should be employed in support of what/who influences what/who is the embedding context for of such participatory approaches. EAST-2. Neither technological nor social character of an organi- Developing such a knowledge base requires a systematic zation predominates in change – they work together to structure approach to empirical studies of participatory approaches to and, hence, reconstruct each other – the fundamental idea un- place-based decision-making. Empirical studies are potentially derlying “adaptive structuration.” the main source of knowledge guiding further development and The organization of constructs in EAST-2 reflects an organi- applications of participatory decision-making. In order to develop zation of participatory decision-making process. The constructs the knowledge base, it is necessary to understand the social-behav- are grouped into convening, process, and outcome categories. ioral implications of information and communication technologies applied in a spatial problem-solving context. Unfortunately, Constructs about Convening a Participatory anecdotal evidence about the implementations of participatory decision-making supported by geographic information technol- Situation Three constructs consisting of 11 aspects characterize the con- ogy, although good for sharing experiences and telling stories, vening influence about a decision situation when information is not sufficient to understand the complexity of information technology is involved. technology intertwined with human perceptions and understand- ings of spatially represented information in various social-political Construct 1: Social-Institutional Influence. Social-institutional contexts. Without a systematic knowledge of participatory use of influence is usually based in law, mandate, policy, social norm, GIS and other information and communication tools based on or natural events (i.e., influences commonly outside the control social-behavioral research, poor designs of geographic information of any single individual). One aspect of social-institutional influ- technology-supported public participation are likely to be repro- ence is power and control, which refer to the entitlements that are duced again and again and to have (sometimes unintended) social granted by formal or informal mandate (e.g., laws and regulations implications for efficiency, effectiveness, and equity. If anything, or special interest group awareness). Whether specific participant an approach based on anecdotal evidence rather then on system- groups intend to exercise their power is a concern to all participant atic research framework may lead to community exploitation. groups (Susskind and Field 1996). Without a systematic approach to researching participatory A second aspect of social-institutional influence is subject decision-making, stories and experiences are difficult to integrate; domain. Sharing an interest in subject domain is what principally hence we are less likely to accrue “knowledge about use.” Nyerges brings people together into participatory decision processes. Many et al. (2002) proposed one such approach that may be helpful in decision problems from the domain of environmental and land developing empirical evidence about the use of participatory GIS management contain intangibles that cannot be easily quantified, (PGIS). The approach is based on social-behavioral research on and their structure is only partially known or is burdened by the uses of PGIS consisting of three stages: planning, implementa- uncertainties. These problems require the participation of people tion, and corroboration. Each stage involves a balance among three representing diverse areas of competence, political agendas, and research domains: substance, theory, and methods. From this ap- social values. Environmental and land management problems are proach, a rigorous methodological framework consisting of problem characteristic of at least three kinds of uncertainty, which adds articulation, treatment mode selection, data gathering strategy, data to the complexity. One type of uncertainty relates to knowledge analysis strategy, and reporting strategy can be derived for empiri- about the natural environment (i.e., what we don’t know about cal studies about substantive problems. A theoretical foundation natural processes and the influences that humans might have in to back up that methodological approach is Enhanced Adaptive such processes). A second type of uncertainty is about the inten- Structuration Theory (EAST). The theory provides constructs and tions in related fields of choice from a technical perspective (i.e., premises that help to interpret how people make use of PGIS in is one solution technically better than another). A third type of a given problem context and elucidates the elements involved in uncertainty is about values (i.e., which valued concerns in society structuring human–computer-human interaction. are those that should be pursued), often at full awareness that we cannot pursue all values of concern. Beck (1992) observed that Enhanced Adaptive Structuration major social conflict in western societies has become centered on Theory (EAST) the distribution and tolerability of risks for social groups, regions, The most recent version of the framework, called EAST-2, con- and future generations, leading to the label of “risk society.” With sists of a set of eight constructs detailed in terms of 25 aspects such interests running high, environmental problems thus lead (the elements of the conceptual domain) that describe significant to a democratic dilemma for at least two reasons. First, allocat-

10 URISA Journal • Vol. 15, APA I • 2003 URISA Journal • Jankowski, Nyerges 11 ing resources to address such problems require that we consider values, experience and trust, and knowledge and expertise that possible redistributions of such resources. Second, there are multi people align themselves into stakeholder groups; world views and loci of relevant knowledge with which to make such decisions values are more important to the alignment, experience and trust (Coenenet al. 1998). are next, and knowledge and expertise are less important. The A third aspect of social-institutional influence involves the difference in stakeholder perspective leads to different values, persons, groups, and/or organizations as convener of participants. objectives, and criteria being articulated as the basis of solutions The convener can potentially exert fundamental influence in set- of group-influenced problems. ting a topic and direction for discussion, and might also be the A third aspect concerns participants’ trust in the process. party that acts as the facilitator for the participatory decision Because in many cases the public is disillusioned with the politi- process. The influence of the convener can be amplified or di- cal process, participation in collaborative decision efforts is on minished, depending on the organizational and technological the rise. A fourth aspect of participant influence comes from arrangements of decision process. participants’ beliefs and feelings about technology. The emotive A fourth aspect of social-institutional influence is choosing issues for GIS technology are often under studied, but in some the number, type, and diversity of participants that are brought cases may be as important as the technology itself. Feelings and together to address a problem. Access to the discourse in terms of beliefs are likely an important aspect of reinforcing a person’s giving voice to all groups who are impacted by a complex problem experience with technology. Such experiences encourage and/or situation sets up an ironic dilemma. The larger the group with hinder expressing ones interest in considering new ways of ac- different interests being convened, the more opportunity for complishing tasks. conflict. Thus, it is perceived that more interests means longer solution times. However, when all interests are not convened at Construct 3: Participatory GIS Influence. One aspect of social- the beginning of a process, there is more opportunity for challenge technical information structuring deals with the combination at completion of the process (perhaps in the form of a law suit). of place, time, and channel of communications. Whereas rules Thus, even though convening more groups in the short-run might and norms for social structuring are social-institutional aspects appear to extend a process, not convening the appropriate groups of participation, different types of meetings structured in terms sets up the risk of having the deliberation fail through continual of place, time, and communication channels also have an impact challenges (Susskind and Field 1996). on who says what and when during participation in a decision A fifth aspect of social-institutional influence (not reflected situation. The physical (or virtual) setting of a place has a signifi- in Figure 1) comes from a collection of rules and norms as social cant impact on whether people attend a discussion. Being able to structures among participants. Such structuring influences the attend a meeting due to scheduling (distance and timing) con- modes of participation in a public decision process. The modes straints is a fundamental concern in participation. More local and of participation are in turn adopted as a matter of expectations more frequent meetings do not always enhance the opportunity for communication, cooperation, and collaboration. to participate, since more time away from some other activity (perhaps work) is not always as convenient. It is for that reason Construct 2: Group Participant Influence. One aspect of group that technology-supported meetings have been on the increase participant influence deals with participants’ expectations-based to open channels of communication. values. Values set the stage for participant perspectives about ex- A second aspect of social-technical information influence pected benefits and outcomes. Different perspectives on values involves the availability of social-technical structures as informa- lead to differentiating between the facts that get stored in GIS tion aids. These structures provide information aids to support and the social values used to interpret the facts. In participatory the participatory effort. Three types of information aids are of- decision processes, values get often exemplified as concerns. Par- ten integrated into a spatial decision support system. These are ticipants’ interpretations of the effectiveness in decision-making cartographic visualization tools, spatial and attribute data query depend on how their concerns are addressed during the decision- tools, and analytical models. Computer-mapping techniques making process. implement cartographic visualization tools. Spatially referenced A second aspect of participant influence, participants’ views/ database management systems implement spatial and attribute knowledge of the subject domain and each other, mainly involves data query tools. Spatial analysis techniques support analytical how participants approach the importance of the topic and how model development, and decision analysis techniques make use of they approach each other in terms of “friendship” or “enemy” the results of spatial analysis to introduce evaluation of multiple feelings. Stakeholder views develop as a result of experience and alternatives for decision-making. A spatial decision support system educational background with topics; people build for themselves integrates those techniques in a computerized, analytical environ- a frame of reference for particular issues and they sometimes share ment that supports participants in their search for solutions. them. Frames of reference are anchored by “reference points” (i.e., familiar information elements that are used as a basis for interpreting material and each other’s backgrounds). Often, it is because of the similarities and differences in world views and

12 URISA Journal • Vol. 15, APA I • 2003 URISA Journal • Jankowski, Nyerges 13 Constructs and Aspects of the Participatory Habermas concept of communicative competence to characterize Process the ability of participation strategies to support different types of The central box in Figure 1 represents an expanded view of a discourse, providing participants with an ability to exchange ideas. participatory decision process flowing from appropriation (con- Four types of discourse were characterized: explicative, theoretical, struct 4), decision process (task) management (construct 5), and practical, and therapeutic. Explicative discourse involves terms, emerging information (construct 6), and back to appropriation definitions, grammar, and the everyday use of language; the par- and so forth around. ticipatory process should allow conversations that make reference to worldly events in everyday language. Theoretical discourse Construct 4: Appropriation. Appropriation is the act of invoking involves references to scientific studies as in an objectified world; a structure, whether the act is one time or continual (DeSanctis the participatory process should also allow reference to the detail and Poole 1994). Continual appropriation of the same structure of the nuances of complex problems described in terms of techni- can be called “use” but does not include the act of continual use cal (discipline-based) language. Practical discourse involves social once invoked. Various appropriation acts structure the character of needs and the appropriate forms (norms) of social interaction; the the information use. One aspect of appropriation involves social- participatory process must support social interaction that develops institutional influences (i.e., appropriating at any time any one or out of conventions people know from their everyday experience. more of the five aspects of social-institutional influences). Therapeutic discourse makes reference to the subjectivity of a A second aspect of appropriation concerns appropriation of speaker in terms of sincerity and authenticity of claims. Conflict participant influence. When participants are recognized by others is undoubtedly going to arise in complex situations where values in the participation process, then they provide voice to certain differ. Coping with those conflicts can be handled by permit- concerns. These might be concerns important to interested and ting therapeutic discourse. How geospatial information tools affected parties, concerns of clarification from technical special- can support these different styles of communication in order to ists, and/or concerns of allocating a redistribution of resources encourage certain types of participatory process for certain types by managers or decision makers. Participants as the “agents of of problem tasks, effecting certain types of outcomes, is still a change” in conversation introduce information about concerns matter of empirical research. based on their trust with the process of getting a “fair voice.” A second aspect of group process focuses on participatory Sometimes that information is introduced through the use of task flow management, which concerns the structuring into stages, GIS technology based on their belief that such technology treats steps, or phases, either from a pre-determined agenda, an open information in a way that suits their need for information. agenda, or a mixture of both. Renn et al. (1995) characterized A third aspect of appropriation deals with appropriation of the fairness of participation strategies in terms of three criteria: participatory GIS influence. We know little about place, time, agenda and rule making, moderation and rule enforcement, and and communication channel influences on information use in discussion. Agenda and rule making deal with who participates geographic problem solving and decision-making. Most people in setting the agenda and the rules by which the group will in- recommend anytime, anywhere access, as this covers all situations. teract. Moderation and rule enforcement deal with whether the However, the advantages and disadvantages to the various forms group process is facilitated and whether a facilitator enforces the of place and time meetings when map information is being dis- rules that have been established. Discussion involves the degree cussed are not known. In addition to physical settings, we know to which all who are impacted by the decision have a voice in the little about how information technology can be put to best use process. A group process that is fair is a basic tenant in a direct under different types of participatory process methods. There is democracy. Few complex public-private problems are addressed an inherent tradeoff between sophistication (representational and through direct democracy; more of them are addressed through analytical power) and ease of use of decision support tools. More a representative democratic process. However, one of the major sophisticated tools are often more challenging to use, despite issues seems to be that representative democracy is not function- the enormous effort put into making tools “user friendly.” This ing as it once did. Thus, participatory decision situations appear tradeoff becomes even more significant when the users of deci- to be on the rise. sion support tools are people of various educational and cultural A third aspect of group process is the behavior of participants backgrounds. A non-specialist approaching GIS software often toward each other. This concerns the working relationships that naively expects to work with a set of virtual maps that portray develop as ideas are exchanged and the decision process proceeds. an “objective, shared understanding” about the world. However, Stakeholder behavior involving conflict has been studied, and idea GIS has roots in many disciplines and its effective use requires differentiation and integration are both important (Susskind and a considerable knowledge, suggesting that there are “multiple Cruikshank 1987, DeSanctis and Poole 1994). Obermeyer and realities” to be portrayed. Pinto (1994) see the introduction of GIS as encouraging more conflict between groups rather than less conflict. Jankowski and Construct 5: Group Process. One aspect of group process con- Nyerges (2001b) investigated group conflict in decision support cerns idea exchange as social interaction. Renn et al. (1995) used for habitat redevelopment site selection and found that maps are less likely to be associated with discussion conflict than tables

12 URISA Journal • Vol. 15, APA I • 2003 URISA Journal • Jankowski, Nyerges 13 – a table being more an analytic display for priority ranking of major concern is the stability of an outcome based on whether habitat sites than a map. it lasts beyond the duration of decision-making process. Several researchers suggest that “decision sustainability” is a pragmatic, Construct 6: Emergent Influence. One aspect of this construct is substantive criterion that could be used to evaluate the quality the emergence of social-technical information influence. Although of decision outcomes. Decision sustainability involves the abil- various technological capabilities are provided by software and ity to manage worldly events in such a way as to preserve the hardware as per the design of a system, certain other emerging “validity” of the decision without having to overturn the action structures might come to light during the treatment of informa- that was decided. tion. The emergence of social-technical structures such as new map designs or database designs might help a group with further Construct 8: Social Outcomes. One aspect that concerns social- information structuring. However, such emergence could make institutional outcomes deals with whether there is an opportunity information easier or more difficult to understand in the longer for challenge of the outcome. The degree to which any decision term. Consequently, the emergence of social-technical informa- issue is final and whether it can be changed through further con- tion structures has a rather significant impact on what information siderations is rather important to promote the results of a process. a group treats from activity phase to activity phase (DeSanctis Some of the participation processes facilitate the opportunity and Poole 1994). for challenge, whereas others prohibit it outright. Risky deci- A second aspect of this construct is the emergence of group sions should always be amenable to challenge if new and better participant influence. A better understanding of values, goals, information arises. objectives, and beliefs are bound to come to light through par- A second aspect of this construct concerns the reproduction ticipant conversation. Participants might clarify their own per- and temporality of the group participant structuring. This aspect spectives and/or the perspectives of others in regard to values, deals with the stability and longevity of the social relationships goals, objectives, and beliefs. Views of each other in regard to among group participants, particularly as promoted in multiple respectful opinion of what others have to say will undoubtedly meetings (DeSanctis and Poole 1994). Successful participation get refined. Trust in each other might change as a result of ideas can spread through decision processes far beyond the immediate being exchanged. Those who encourage use of technology can and direct. This is one place where the possibility of participa- have an impact on the feelings that participants develop in regard tion-based learning supported by GIS could be considered. The to its continual use. Feelings for people and technology might impact of participation-based learning about decision-making well be connected. would seem to be a fertile subject for more in-depth research. A third aspect of this construct is emergence of social-institu- A third aspect of this construct concerns the reproduction tional influence. This deals with how rules or norms are brought and temporality of social-institutional structuring. Such structur- into use and eliminated or reinforced during the decision process. ing in regard to changing mandates for power or control is only Clarifying mandates and the problem at issue can lead to refocus- likely to occur over repeated projects – either successes or failures. ing activity for any particular task. In regard to agendas, one can However, continued successes with projects encourage similar choose to make use of rules to keep the conversation on track projects to be addressed in the same way, whether this involves or to de-rail it. The emergence of new rules about how people the task domain and/or the way that participants are convened. communicate during the participation will change the course of Eight constructs with the respective 25 aspects presented the interaction. above likely constitute the single largest enumeration of issues concerning decision-making within a geographical context. In Constructs and Aspects about Participatory addition, each aspect can spawn several variables when making the EAST-2 operational, ultimately contributing to the compre- Outcomes hensive nature of EAST-2. The above constructs and aspects, Two constructs are part of the participatory outcomes in EAST-2: together with respective variables, are but concepts that in and of task outcomes and social outcomes. themselves can be used only for developing a “task description” of each macro-phase in a decision situation. It is the premises Construct 7: Task Outcomes. Two aspects appear to be fun- connecting these constructs, aspects, and variables that turn a damental in regard to task outcomes. One aspect is character conceptual framework into a theory. of decision outcome. Because decision outcomes tend to be so diffuse, depending on the circumstances of the participatory effort, a reasonable approach is to focus on process rather than Premises of EAST-2 outcomes. Multiple characteristics of decision processes such as Premises are fundamental statements that tie each aspect on one fairness, transparency, efficacy, and inclusiveness can contribute side of the premise to an aspect of the other side. Premises motivate to effectiveness of participatory decision-making. one or more research questions (as, for example, in Table 1) which, The second aspect linked to task outcome concerns deci- when phrased in terms of variables, can be considered hypotheses sion outcome and participant structuring dependence. The (propositions) about the dynamics of a decision situation. The research questions presented in Table 1 are a few examples of ques-

14 URISA Journal • Vol. 15, APA I • 2003 URISA Journal • Jankowski, Nyerges 15 Premises Research Question Motivated by Respective Premise Convening Premises Premise 1. Social-institutional influences affect the - In what way does the purpose of a decision task influence the types of geographic appropriation of group participant influences and/ information structures (e.g., maps, tables, diagrams) appropriated by the or social-technical influences. participants? - In what way does the organization that convened the decision situation in combination with the diversity of participants influence the type of group participant structuring?

Premise 2. Group participant influences affect the - How do the different perspectives such as those oriented to policy/decision maker, appropriation of social-institutional influences and/ technical/scientific specialist, and interested and affected party influence the types of or social-technical influences. geographic information structures appropriated? - What types of social-technical information structures appear to be linked to differences in participant structuring?

Premise 3. Participatory GIS influences affect the - How does each of four meeting venues influence the generation of information appropriation of social-institutional influences and/ structures (e.g., maps, animations, tables, and text narration) useful for or group participant influences. understanding spatial criteria that can be processed with a GIS? - How do the social-technical capabilities of software get appropriated across meetings in relation to participants’ trust in the group agenda which represents a plan for the process? Process Premises Premise 4. Appropriation of influences affect the - What types of geographic information structures are appropriated during the dynamics of social interaction described in terms different intensities of participation that seem to facilitate an analytic-deliberative of group processes. process and which information structures seem to hinder the process? - Appropriation of what types of group participant structures has what type of influence on group process?

Premise 5. Group processes have an affect on - What kinds of geographic information structures emerge during the different levels the types of influences that emerge during those of participation in an analytic-deliberative process? processes, and emergent influences affect the - What emergent structures influence the type of appropriation that is undertaken? appropriation of influences.

Outcome Premises Premise 6. Given particular influences being - Given that a group appropriates a particular type of information structure that has appropriated, if successful appropriation occurs been found to be useful in the past, and if the information structure is appropriated and group processes fit the task, then desired during “specific conditions,” can we expect the outcome from the process to be outcomes result. satisfactory to all participants? - What structure appropriation under what conditions of group process appear to affect the dependence of the decision outcome on group participant structuring?

Premise 7. Given particular influences being - In what way are various social-institutional structures together with group appropriated, if successful appropriation occurs participant structures linked to the opportunity to challenge the task outcome? and group processes fit the task, then reproduction - How do inter-organizational protocols and the social interaction during group of social-institutional influences result. process promote or discourage further group work?

Table 1. Example Research Questions Motivated by Premises in Enhanced Adaptive Structuration Theory

tions that could be posed. Each premise is a general statement, and understanding” about the impact of software designs on GIS each premise statement consists of a subject construct related to support of participatory decision-making. Addressing any one an object construct. Hence, each research question asks something or more questions among such a wide variety of questions is a about how a subject aspect relates to an object aspect, thus many considerable challenge. different questions could be posed. In the context of empirical, The challenge “begs” for a systematic approach to empirical social-behavioral research on participatory decision-making, we social-behavioral research so that we can better understand of could say: “how does one variable relate to another variable?” how empirical results relate to each other in our attempts to build The seven premises in EAST-2 presented in Table 1, together knowledge about the implications of PGIS use. Such a systematic with the respective example questions, indicate that a wide va- approach is particularly important when trying to understand the riety of interesting, empirical research opportunities exist with nuances of “participation models” (e.g., the three-phase participa- regard to the use of GIS in participatory settings. Articulating tion model set forth by Renn et al. (1995) as described above). social-behavioral explanations by way of the premises and the Jankowski and Nyerges (2001a) outline 18 research strategies research questions they motivate is intended to lead to a “deeper that could be used to unpack the “process relationships” in such

14 URISA Journal • Vol. 15, APA I • 2003 URISA Journal • Jankowski, Nyerges 15 a model. Choosing among research settings (field, lab, or field- About the Authors lab), among treatment modes, among data collection instruments, and among data analysis approaches are all important, and thus Piotr Jankowski is Professor of Geography at the Department of impact the nature of the empirical evidence derived. Geography, University of Idaho. He is also a Professor of Geo- informatics at the Institute for Geoinformatics, University Conclusion of Muenster. His teaching and research areas include models Participation models can help us organize the way we think people and methods for spatial problem solving and collaborative interact during work in participatory settings. Empirical studies decision-making. of participation models that help organize our understanding of Corresponding Address: the use of participatory GIS are an activity that can contribute Department of Geography, to “participatory geographic information science.” University of Idaho Although participation models abound in the literature, we Piotr Jankowski have little understanding about the efficiency, effectiveness, and Moscow, ID 83844-3021, USA equitable benefits and costs of how different models structure par- [email protected] ticipant interaction. When placed in the context of participatory GIS use, we have even less knowledge of such issues. An organizing Timothy Nyerges is Professor of Geography at the Department framework for arraying the multiple constructs (and correspond- of Geography, University of Washington. His teaching and ing detailed aspects) of a participatory situation is beneficial in research areas include GIS design and social-behavioral stud- helping to organize the kinds of research questions that can be ies of collaborative, geospatial decision problem solving. examined. The Enhanced Adaptive Structuration Theory as an organizing framework suggests that people structure situations; in Corresponding Address: turn, situations structure people’s interaction. Technology enables, Department of Geography but it also constrains – thus the need for a systematic approach University of Washington to empirical studies of such situations. Timothy Nyerges A wide array of empirical research strategies exists. The Box 353550 18 strategies outlined in Jankowski and Nyerges (2001a) are Seattle, WA, 98195, USA, [email protected] compared to each other to provide a sense of what each strategy might contribute to such empirical investigations. Having an understanding of how one strategy differs from another, and hence References why we might use one research design rather than another to examine a single participation model or to compare participa- Beck, U., 1992, From Industrial Society to the Risk Society: Ques- tion models would help us build knowledge in a systematic way. tions of Survival, Social Structure and Ecological Enlighten- Only through systematically comparing evidence from a variety ment. Theory, Culture and Society, 9(1):97-123. of research approaches will we gain an overall sense of how and Coenen, F.H.J.M, D. Huitema, and L.J. O’Toole, 1998, Participa- why one participation model in comparison to another provides tion and Environment. In Coenen, F.H.J.M, D. Huitema, us with efficiency, effectiveness, and equity advantages and dis- and L.J. O’Toole (Eds.) Participation and the Quality of advantages. As there is little empirical evidence at this time to Environmental Decision-Making (Dordrecht: Kluwer Pub- point to these advantages and disadvantages, there is considerable lishers), 1-20. opportunity for empirical research to help develop a geographic Craig, W.J., T.M. Harris, and D. Weiner (Eds.), 2002, Com- information science of PGIS use. munity Participation and Geographic Information Systems. (London: Taylor & Francis). DeSanctis, G. and M.S. Poole, 1994, Capturing the Complexity in Advanced Technology Use: Adaptive Structuration Theory. Organization Science, 5(2):121-147. Jankowski, P. and T. Nyerges, 2001a, GIS for Group Decision- Making (London: Taylor & Francis). Jankowski, P. and T. Nyerges, 2001b, GIS-Supported Collabora- tive Decision-Making: Results of an Experiment. Annals of the Association of American Geographers, 91(1):48-70. National Research Council, 1996, Understanding Risk: Inform- ing Decisions in a Democratic Society (Washington, D.C.: National Academy Press).

16 URISA Journal • Vol. 15, APA I • 2003 URISA Journal • Jankowski, Nyerges 17 Nyerges, T., P. Jankowski, and C. Drew, 2002, Data Gathering Schiffer, M., 1995, Issues of Collaborative Spatial Decision- Strategies for Social-Behaviour Research about Participatory Support in City Planning Context. In Densham, P.J., M.P. Geographic Information System Use. International Journal Armstrong, and K. Kemp (Eds.) Report from the Specialist of Geographical Information Science, 16(1):1-22. Meeting on Collaborative Spatial Decision-Making, Initia- Obermeyer, N. and J. Pinto, 1994, Managing Geographic Infor- tive 17. Santa Barbara, CA: National Center for Geographic mation Systems (New York: Guilford). Information and Analysis. Pickles, J. (Ed.), 1995, Ground Truth: The Social Implications of Susskind, L. and J. Cruikshank, 1987, Breaking the Impasse: Geographic Information Systems (New York: Guilford). Consensual Approaches to Resolving Public Disputes (New Renn, O., T. Webler, and P. Wiedemann, 1995, Fairness and York: Basic Books). Competence in Citizen Participation: Evaluating Models Susskind, L. and P. Field, 1996, Dealing with an Angry Public for Environmental Discourse (Dordrecht: Kluwer Academic (New York: The Free Press). Publishers). Weiner, D., T.M. Harris, and W.J. Craig, 2001, Community Participation and GIS. Workshop on Access and Partici- patory Approaches in Using Geographic Information in Spoleto, Italy. http://www.spatial.maine.edu/~onsrud/ Spoleto/Keynotes.htm

16 URISA Journal • Vol. 15, APA I • 2003 URISA Journal • Jankowski, Nyerges 17 URISA Journal • Wehn de Montalvo 19 In Search of Rigorous Models for Policy-oriented Research: A Behavioral Approach to Spatial Data Sharing

Uta Wehn de Montalvo

Abstract: This article introduces a theory from social psychology as an organizing framework for policy-oriented research on spatial data sharing. The article demonstrates how, using this approach, the incentives and disincentives that determine the intention of key individuals within organizations to share spatial data can be identified in a systematic fashion. With a view to methodological considerations, the article reviews research on spatial data sharing and examines the relevance of decision-making research for this field. It considers the contribution of the Theory of Planned Behavior to research on spatial data sharing and identifies the appropriate unit of analysis. Actual results that have been generated using the Theory of Planned Behavior for research on spatial data sharing are presented. These findings can form a profile of perceptions that provides a basis for addressing the issue of spatial data sharing more effectively at the policy-making level. Conclusions are drawn on the relevance of this behavioral approach to research on spatial data sharing. Introduction1 Behavioral Models and Research With the increasing use of the geographic information system on Spatial Data Sharing (GIS) in industrialized and developing countries, the availability The growing importance of spatial data availability and data ac- of spatial data has become an issue that affects many organizations. cess for the effective use of GIS has begun to generate a branch of Spatial data initiatives are reliant on the willingness of different research that focuses directly on the issue of spatial data sharing. organizations to engage in spatial data sharing in order to be Several contributions have been made to gain an understanding effective in overcoming bottlenecks in the availability of spatial why organizations may or may not wish to engage in spatial data data. An important role of spatial data initiatives involves more sharing. The factors that have been considered are mostly con- than the coordination and development of metadata, clearing- ceptualized in terms of facilitators and constraints (Stage 1995, houses, and standards components. The important aspect that Tosta 1995), costs and benefits (Alfelor 1995, Dangermond can be expected to vary considerably for each national initiative 1995, Dueker and Vrana 1995), or antecedents and consequences is the existence or the fostering of a data sharing culture (Clarke (Obermeyer and Pinto 1994, Pinto and Onsrud 1995) of spatial et al. 1998) that encourages participation in spatial data sharing. data sharing. Several models have been advanced in this field of The purpose of this article is to demonstrate that the Theory of research, and this section considers their strengths from a meth- Planned Behavior can be applied as an organizing framework for odological perspective. policy-oriented research to identify the motivations and obstacles The conceptual framework for spatial data sharing across of different groups of actors to engage in spatial data sharing. organizational boundaries proposed by Obermeyer and Pinto This article is organized into three parts. In search of rigor- (1994) combines a number of variables as facilitators and inhibi- ous models to better understand spatial data sharing, the first tors, and as outcomes. These variables consist of the antecedents section reviews research on spatial data sharing and examines that may improve the likelihood of creating positive, collaborative the relevance of decision-making research, and of the Theory relationships between organizations and of the consequences of of Planned Behavior in particular, for the issue of access to geospatial data sharing. Although Obermeyer and Pinto claim that the graphic information and spatial data sharing. The second section variables in the model have been drawn from organization theory, discusses the components of this theory, notes the contribution intergroup dynamics, exchange theory, and political-economy, a that its application can make to research on spatial data sharing, major limitation of this framework is the lack of a justification and considers the appropriate unit of analysis. The third section for the inclusion of these factors in the model and for the exclu- demonstrates the results that can be generated using the Theory sion of other factors. of Planned Behavior as an organizing framework for research on Kevany (1995) proposed a structure for observing spatial data spatial data sharing. Thus it is possible to identify–empirically–the sharing by exploring the factors and conditions that may create fundamental factors underlying a disposition to engage in spatial a conducive environment for sharing. These factors consist of: data sharing within a specific community of actors (i.e., national a) sharing classes; b) project environment; c) the need for shared or regional). For policy makers, these insights can provide a ba- data; d) the opportunity to share data; e) the willingness to share sis for the formulation of appropriate mechanisms to encourage data; f) the incentive to share data; g) the impediments to sharing; spatial data sharing behavior. h) the technical capability to share; and i) resources for sharing. A number of measures for each of the factors are proposed in order

URISA Journal • Wehn de Montalvo 19 to determine the probability of successful sharing. Although very Perceptions may also vary with respect to the costs of, or comprehensive, Kevany’s method of analyzing the environments obstacles to, sharing spatial data. The nature of these perceptions for spatial data sharing has several limitations. There is no explicit should not be generalized across organizations without recourse basis for the inclusion of the factors, nor is there a basis for con- to an empirical investigation that permits validation of some of ceptualizing the likely relationships between these factors. The these assumptions and that provides a basis for generalization. factors are derived on the basis of the author’s personal insights Rhind (1998) provided a summary of the views of different and experiences with GIS design, implementation, and operation players and stakeholders in spatial data and their agendas, which projects. Hence, there is no organized framework in which to suggests that the perceptions of various actors from the public analyze the relevance of these factors. Moreover, the framework and the private sector may differ substantially in terms of finance proposed by Kevany has not been verified empirically. and the dissemination of information, both in the short and in Azad and Wiggins (1995) proposed a research framework the long term. that focuses on inter-organizational dynamics. The central tenet Under “Research Initiative 9: Institutions Sharing Geographic of their framework is autonomy. Azad and Wiggins assume that Information” of the National Center for Geographic Information sharing results in a loss of autonomy and increased interdepen- and Analysis (NCGIA) in the United States, a focus group con- dence of organizations, and they argue that these consequences sidered the organizational aspects of sharing (Batty 1992, Onsrud are at odds with the organizational goals. These are the principle and Rushton 1992, 1996). The outcome is a two-dimensional obstacles to spatial data sharing that have to be addressed. Their matrix of organizational issues and organizational forms to mea- framework consists of a typology of inter-organizational spatial sure the degree and impact of spatial data sharing (Onsrud and data sharing and the reasons for the development of inter-orga- Rushton 1992). Research in this case is directed to investigating nizational relationships. In addition, Azad and Wiggins argue the complexities of dynamic interactions at work in sharing ac- that the likelihood of establishing inter-organizational relations tivities. Similarly, Nedovic-Budic and Pinto (1999) proposed a is related to the intensity of the relationship and the loss of orga- conceptual framework, based on Kevany (1995) and a literature nizational autonomy required by the relationship. Finally, they review, consisting of four general theoretical constructs, namely propose a process model to manage the development of inter- context, motivation, coordination mechanisms, and outcomes. organizational relations for spatial data sharing. In their subsequent research, they used case study and survey The starting point for the Azad and Wiggins framework methods. In particular, the empirical research on the nature of is open to question. The assumptions that, firstly, spatial data interorganizational sharing arrangements (Nedovic-Budic et al. sharing necessarily leads to a loss of autonomy and increasing 2001) focused on sharing “clusters” of organisations. However, interdependence among organizations and that, secondly, these in both the Research Initiative 9 and the Nedovic-Budic and consequences are inherently negative from an organization’s point Pinto framework, the focus on actual incidents of sharing (albeit of view have not been subject to empirical confirmation. Similar successful and unsuccessful sharing activities) limits the scope assumptions about the nature of the power of, and control over, of the research to an investigation of the views of the “sharers.” spatial data are embedded in much of the literature as indicated “Non-sharers” are excluded and, hence, important insights into by the following: why individuals within organizations may not be willing to share cannot be captured. Because geographic information has potential value to those with In summary, the above discussion has raised two essential effective access to it, this realization gives rise to the desire to exercise ownership rights over the information. Thus, the power that concerns about research on spatial data sharing. First, the deter- information provides is antipathetic to sharing. (Onsrud 1995:293) minants of whether and why organizations may be willing to engage in spatial data sharing with other organizations should If we agree that the possession of [spatial] information serves as a be established empirically, they should not be assumed a priori. source of control for individuals and organizations, then we are faced with questions about the ways in which organizations can be induced Second, the scope of empirical research should encompass the to relinquish this control. (Obermeyer and Pinto 1994:107) whole community of potential sharers involved in GIS and not just the actual “sharers.” If the fundamental factors underlying a In contrast, in this article it is argued that it is not possible to disposition to engage in spatial data sharing can be identified, this presume the obstacles to sharing a priori. Instead, they should be can provide a basis for the formulation of appropriate mechanisms the subject of empirical verification. As Pinto and Onsrud (1995: to encourage spatial data sharing behavior. In this article, it is 48) suggested; “... little is known, for instance, about the reasons argued that the Theory of Planned Behavior represents a robust why governmental agencies and other GIS-using organizations and rigorous model from social psychology that can be employed will or will not share GIS-related information.” to make the incentives for, and obstacles to, spatial data sharing The point of departure for research into whether and why the subject of empirical verification and that it will yield useful individuals within organizations may be willing to engage in spa- insights for policy. Furthermore, the empirical investigation based tial data sharing should be located one step back owing to: “... on the Theory of Planned Behavior can be designed to incorporate the potentially differing perceptions of the benefits from [spatial] all potential “sharers.” data sharing” (Sperling 1995:391).

20 URISA Journal • Vol. 15, APA I • 2003 URISA Journal • Wehn de Montalvo 21 Theory-Driven Models in Decision-Making of direct measurements of perceived preferences and constraints (Jones 1985, Ajzen 1996, Opp 1999) which have been developed Research since the 1930s (Hogg and Vaughan 1995)3. The expectancy value Research into factors that influence human judgment and deci- models of attitudes used by social psychologists are employed here sion-making provides a basis for exploring the issue of spatial data in order to provide a basis for analyzing the considerations that sharing from a perspective that focuses on the various positions may underlie real-life decisions to share spatial data. that individuals in different organizations take toward spatial data sharing and on why they take these positions. The field of human judgment and decision making is interdisciplinary, drawing on Attitude-Behavior Models contributions from economics, political science, organization and Research within the discipline of social psychology deals with management studies, and social psychology. The starting point for decision making in the general context of predicting and explain- much decision-making research is rational choice theory (Abelson ing behavior, and research on attitudes is a central consideration and Levi 1985, Medin and Bazerman 1999), and much of the within this body of work. Beliefs are understood as providing the research has focused on the comparison of actual decision making subjective basis for individual’s decisions. The source of beliefs with principles of rationality in decision making (Dawes 1998). may be logical processes as well as emotions or desires (Ajzen Rational choice theory assumes that preferences and constraints 1996). Attitudes are assumed to reflect the beliefs that the deci- affect behavior and that individuals optimize in some way (Opp sion-maker holds (Eagly and Chaiken 1998), and the relationship 1999). The narrow assumptions about a fully informed, rational between attitude and actual behavior is taken to be mediated by decision-maker have given way to the realization that cognitive as the intention to act (Fishbein and Ajzen 1975). well as noncognitive factors influence the decision-making process Several models of the attitude-behavior relationship have (Keren 1996, Mellers et al. 1998) and that decision-makers are been developed that examine the beliefs that influence attitude not necessarily fully informed. That perceived, subjective and not formation. The framework for examining this relationship (which merely objective tangible constraints may be relevant. Constraints has received widespread attention in the literature) is that which and preferences taken together, rather than individual constraints has given use to expectancy value models that characterize the on their own, may explain behavior (Opp). relationship between beliefs and attitudes. Expected values are said Distinguishing between the many different approaches to to be made up of subjective probabilities of outcomes and subjec- decision making are structural and process models (Abelson and tive evaluations (positive or negative) of outcomes of a behavior. Levi 1985)2. Structural models are concerned with what deci- The most salient, the Theory of Reasoned Action proposed by sion-makers choose, while process models analyze the intervening Fishbein and Ajzen (1975), is most appropriate when behavior is steps in cognitive processes. A further distinction, although less subject to volitional control and, apart from attitudes, it contains clear, can be drawn between normative and descriptive models a social norm component to capture situational constraints that (Stevenson et al. 1990, Keren 1996). The former model considers may influence decisions. Ajzen (1991) extended the Theory of how decision-makers should make decisions and the latter model Reasoned Action to account for planned, more complex behaviors how they actually do make decisions. In light of the overall goal to formulate the Theory of Planned Behavior. Neither, the Theory to examine the nature of the conditions under which the different of Reasoned Action nor the Theory of Planned Behavior assume actors involved with spatial data are willing to engage in spatial that decision-makers necessarily engage in elaborate cognitive data sharing, a structural, descriptive model is most appropri- processes prior to taking action (Ajzen and Fishbein 1980, Ajzen ate. This enables an investigation of what decisions individuals 1991); instead, an individual has formed an attitude by thinking within organizations actually do take with respect to spatial data about the consequences of a given behavior and those attitudes sharing. or intentions can be retrieved and acted upon at a later time. The Fitting this requirement are prospect theory and expected Theory of Reasoned Action and the Theory of Planned Behavior value (EV) models. The prospect theory developed by Kahneman are not restricted to explaining the determinants of behavior in a and Tversky (1979) takes into account contextual factors, but it specific behavioral domain; they can be used in a wide range of 4 assumes the existence of an idealized individual (Abelson and Levi practical situations. Among the contending theories, the Theory 1985) and suffers from ambiguity (Yates 1990). Expectancy value of Planned Behavior framework is the most widely applied and models make no assumptions about rationality and instead rely on tested with considerable proven explanatory and predictive value the internal consistency between the constructs included in the for many behaviors. models (Ajzen 1996). Expectancy value models are not restricted to cognitive elements and allow for the inclusion of noncognitive The Theory of Planned Behavior factors such as emotions and desires (Ajzen). as an Organising Framework for Much behavioral decision research relies on revealed, observed preferences where probabilities and values have to be in- Spatial Data Sharing Research The Theory of Planned Behavior represents a sound theoretical ferred from people’s judgments (Medin and Bazerman 1999). The framework to map out the belief structures underlying inten- contribution of social psychology to this field is the development tional behavior.5 Owing to the predictive power of the intention

20 URISA Journal • Vol. 15, APA I • 2003 URISA Journal • Wehn de Montalvo 21 construct, understanding the antecedents of intentions implies to provide the basis for perception. “It is at the level of beliefs that understanding the behavior. This section introduces the prin- we can learn about the unique factors that induce one person to ciples of the Theory of Planned Behavior. As stated above, the engage in the behavior of interest and to prompt another to follow Theory of Planned Behavior (Ajzen 1985, 1988, 1991, 1996) is a different course of action” (Ajzen 1991:206-207). an extension of the Theory of Reasoned Action (Fishbein and The Theory of Planned Behavior, like the underlying Theory Ajzen 1975, Ajzen and Fishbein 1980) and is most appropriate of Reasoned Action, focuses on subjective perceptions of individu- for the investigation of behaviors that are not assumed to be als rather than on objective observations to explain behavior. under volitional control (i.e., performance of the behavior is not only reliant on the intention to carry out the behavior but also Basic to this approach is the view that people use the information available to them in a reasonable manner to arrive at their on opportunities and resources (Ajzen 1988)). Spatial data shar- decisions. This is not to say that their behavior will always be ing among different organizations cannot be assumed a priori to reasonable or appropriate from an objective point of view. People’s be under volitional control because certain skills, resources, or information is often incomplete and at times also incorrect. But opportunities may play a role in determining whether they can we would argue that a person’s behavior follows quite logically and systematically from whatever information he happens to have engage in spatial data sharing. available. (Ajzen and Fishbein 1980:44)

Basic Principles of the Theory of Planned Although people are assumed to hold a great number of be- Behavior liefs about a particular behavior, only a small number, the so-called The basic model of the Theory of Planned Behavior consists of “salient” beliefs, are expected to be the predominant determinants five distinct components: a particular behavior under consid- of intention and action. According to the expectancy value prin- eration, the intention to act, and three determinants of inten- ciple, it is necessary for each belief and its strength (likelihood) tion. A distinction is made between the decision to engage in a to be rated and then combined. particular behavior, conceptualized as Behavioral intention, and Three types of beliefs can be distinguished: behavioral actual performed behavior. beliefs (influencing attitude), normative beliefs (determining so- It has been argued: cial norm), and control beliefs (underlying perceived behavioral control). Behavioral beliefs consist of the evaluation of specific Intentions are assumed to capture the motivational factors consequences and outcomes that may result from performing the that have an impact on a behavior; they are indications of how behavior in question. Normative beliefs constitute beliefs about hard people are willing to try, of how much of an effort they the important referents (individuals or groups) and their likely are planning to exert, in order to perform the behavior. These intentions remain behavioral dispositions until, at the appropriate approval or disapproval with regard to engaging in the behavior. time and opportunity, an attempt is made to translate the intention Control beliefs capture the perceived presence or absence of req- into action. (Ajzen 1988:113) uisite resources and opportunities (such as skills and capabilities) and the cooperation of other people deemed necessary to perform Because of the distinction between intention to act and actual the behavior. behavior, the intention construct provides the basis for investigat- A behavior is said to be explained once the determinants ing the disposition of organizations toward spatial data sharing. of intention and behavior have been traced to the underlying Hence, it is possible to measure the intention, or the willingness, beliefs. Detailed descriptions developed at the belief level of to engage in spatial data sharing which is, conceptually, expected analysis provide a systematic basis for inferring why individuals to be closely linked to actual sharing behavior. embedded within organizations may be willing, or resistant, to The first determinant of intention (i.e., the attitude toward share spatial data. the behavior) is a person’s positive or negative evaluation of performing the behavior. The second (i.e., the subjective norm) The Appropriate Unit of Analysis: Macro and captures the individual’s perception of social pressure to engage or not to engage in the behavior. Finally, perceived behavioral control Micro Considerations consists of the perceived availability of required opportunities and Pfeffer (1985) suggested that when trying to understand orga- resources to perform the behavior. nizational behavior and decisions, a focus on the individual as These components have been tested empirically to predict the unit of analysis often leads to a neglect of normative contexts intentions and behavior (e.g., Ajzen 1985, 1988, 1991, Ajzen and technological aspects. Yet Jones (1985:53) stressed that: “The and Madden 1986, Ajzen and Driver 1992). The relative weight individual must be seen as the intersection point of a variety of the attitudinal, normative, and control factors is expected to of pressures: immediate situational demands, conflicting social vary according to the behavior under investigation. As the Theory expectations, and internalized beliefs and values.” of Planned Behavior aims to explain, rather than merely predict, Using the Theory of Planned Behavior as an organizing a behavior, it can be used to trace the determining antecedents framework, the influence of these aspects may be explicitly ad- of attitude, social norm, and perceived behavioral control. The dressed by the inclusion of social norm and control components underlying foundation of beliefs about the behavior is considered within the model. Moreover, the suitability of psychological theo-

22 URISA Journal • Vol. 15, APA I • 2003 URISA Journal • Wehn de Montalvo 23 Figure 1: Model of the willingness to share spatial data across organizational boundaries (South Africa)

ries to understanding organizational behavior has been discussed for the research strategy advocated here. Rather than limiting by Staw (1991) and more specifically for the Theory of Planned the scope of the research to an in-depth study of a few organiza- Behavior, by Elliott et al. (1995). Staw (1991:812) suggested that tions,6 a range of individuals and organizations can be included, psychological theories that typically examine behavior at the micro with the goal being to examine their dispositions toward spatial level can also be used to understand action at the macro level data sharing. (i.e., organizational behavior) “...because it is possible to identify key actors in important organizational decisions, psychological Application of The Theory of research can be applied to these individuals in order to explain organizational actions.” Planned Behavior to Spatial Data According to Staw, the most fruitful approach is not to ex- Sharing amine in detail all individual behavior within an organization, but As argued above, the Theory of Planned Behavior provides a rather to study the key organizational decision-makers. Bacharach systematic way of combining qualitative and quantitative empiri- et al. (1995) argued that because actors making decisions are ac- cal research methods to arrive at a model of the propensity to countable for their decision, they seek decision criteria that can be engage in spatial data sharing across organizational boundaries. used to justify those decisions. These key decision-makers may be This framework was employed to integrate qualitative empirical assumed to have an accurate understanding of their organization’s evidence, with insights drawn from several bodies of literature position toward spatial data sharing and their perceptions may in a systematic fashion to arrive at a model of the willingness of be the best indicator of their organization’s behavior (Elliott et organizations to engage in spatial data sharing (Wehn de Mon- al. 1995). In this way, the focal situation can be expanded to a talvo 2001). This entailed two stages of empirical research. The larger arena (Markus and Robey 1988). qualitative stage involved semi-structured interviews, the results of This position with respect to the representation of individuals which were analyzed and complemented by insights drawn from as indicators of likely organizational behavior provides the basis the appropriate theoretical and empirical literatures to build a

22 URISA Journal • Vol. 15, APA I • 2003 URISA Journal • Wehn de Montalvo 23 conceptual model of the willingness of individuals within organi- Regarding the intention of key individuals within organiza- zations to share spatial data (see Figure 1). The resultant model was tions to share spatial data, the results suggest that willingness in operationalised using a questionnaire, and the second, quantitative South Africa cannot be taken for granted; this is contrary to the stage of research entailed the application of the questionnaire underlying assumption in the design of, and in many discus- instrument using a face-to-face interview method. The survey sions on, spatial data infrastructures. It was found that the actual included local, provincial, and national government, para-statal motivation of organizations in the sample was not in favor of organizations, the private sector, academia, and nongovernmental participating in spatial data sharing. Overall, the willingness organizations in South Africa.7 of the representatives of organizations in the sample to share Owing to the complexity of the behavior–spatial data shar- spatial data across organizational boundaries was low, with only ing–interviews alone would not have been sufficient to provide a slight increase expected in the near future. The majority of the necessary information about the full range of beliefs or a basis the respondents reported that their organization was undecided for translating them into questionnaire items. Therefore, based about whether to engage in spatial data sharing, and a small share on the analysis of the qualitative interviews in stage one, the indicated that their organizations were unwilling to share spatial development of the model of spatial data sharing also employed data across organizational boundaries. theoretical insights drawn from several related fields of enquiry The results of the analysis further indicate that the intention such as resource dependence theory and knowledge creation. The of organizations, as expressed by the respondents in the sample, construction of the model and its components, as well as detailed to engage in spatial data sharing did not vary significantly for aspects of the research methodology of the empirical research the different sectors in the GIS community.8 There were also no and of the analysis, are beyond the scope of this article and are significant differences in the willingness to share according to the considered elsewhere (Wehn de Montalvo 2001). perceived extent of self-sufficiency or dependence of organizations In this model, the intention construct refers to the willingness on spatial data. Contrary to what was expected, the willingness to engage in spatial data sharing activities across organizational to share did not differ with respect to the organizations’ spatial boundaries; “attitude” refers to the attitude toward spatial data data position (i.e., whether organizations were predominantly sharing; “subjective norm” is referred to as the social pressure to giving spatial data to, or receiving spatial data from, other or- engage, or not to engage, in spatial data sharing; and “perceived ganizations). behavioral control” is referred to as the perceived control over The results of the statistical analysis show that the inten- spatial data sharing activities that key individuals within orga- tion of key individuals within organizations to engage in spatial nizations perceive. data sharing can be predicted from corresponding measures of The verification of the model of the willingness to share “attitude” toward spatial data sharing and “social norm” from spatial data against empirical data demonstrated that, in terms important referents to engage in spatial data sharing. Somewhat of the reliability of the measures, the items in the questionnaire greater emphasis was placed on the “social norm” component. It had highly satisfactory levels of internal reliability. With respect was found that the technical aspects of sharing in terms of the skills to conceptual validity, the analysis of the empirical data con- and resources considered under the “perceived control” compo- firmed that the items in the questionnaire instrument tapped the nent could not be confirmed as an important factor in influencing concepts suggested by the model. Finally, concerning construct the willingness of individuals in organizations to engage in spatial validity, the relationships between the variables derived from the data sharing. This finding suggests that, while the technical aspects Theory of Planned Behavior were all in the expected direction that are currently the focus of attention in the context of spatial and confirmed the explanatory power of the model. data sharing initiatives (such as the interoperability of different These results offer strong statistical support for the model. GIS applications and spatial data sets, the establishment of stan- The attainment of the validity of the model constitutes a crucial dards, and the implementation of clearinghouses) are necessary, step in the approach which proposes to systematize the determi- it cannot be expected that their resolution will be sufficient to nants of organizations’ spatial data sharing behavior. Only after overcome the obstacles to spatial data sharing. a model has been demonstrated to be valid, can it be relied upon An additional finding was that the National Spatial Infor- to further the understanding of the incentives and disincentives mation Framework (NSIF) directorate did not emerge as an for key individuals within organizations to engage in spatial data important referent for organizations with respect to their en- sharing. gagement in spatial data sharing. The aim of the NSIF is to build a spatial data infrastructure in South Africa in pursuit of social Research Results–Determinants of Spatial Data and economic goals by promoting spatial data sharing among different organizations. In the context of diffusion of innovations Sharing research, the intended role of the NSIF can be conceptualized as Empirical data were used to assess the factors likely to influence that of a change agent.9 Change agents are seen as opinion lead- the willingness of key individuals embedded in organizations to ers who can influence the attitudes and behavior of individuals engage in spatial data sharing across organizational boundaries. with respect to a particular innovation (Rogers 1995). Since the The specific findings are summarized below. analysis could not directly confirm the NSIF as an important

24 URISA Journal • Vol. 15, APA I • 2003 URISA Journal • Wehn de Montalvo 25 referent for organizations regarding their engagement in spatial considered sharing partners. It would also be important to data sharing, this finding suggests that the capacity of the NSIF clarify the mechanisms that might be put in place to ensure to influence the decisions of key individuals within organizations that rules regarding the ownership of, access to, and the use of to engage in spatial data sharing is still limited and that it could the organization’s spatial data can be enforced. With respect to be strengthened. The additional results of the analysis revealed to the benefits of spatial data sharing, reference should be ways in which the effectiveness of the NSIF to encourage greater made to the potential benefits of spatial data sharing for an spatial data sharing could be improved. organization’s particular activities. A more detailed and substantive explanation of spatial data 4. Generally, promotion of the National Spatial Information sharing across organizational boundaries was obtained by distin- Framework in South Africa makes reference to the benefits of guishing between three distinct groups of respondents–those who establishing the NSIF and of spatial data sharing. However, perceived their organizations to be willing to engage in spatial the results of this study indicate that it would be beneficial to data sharing, those undecided, and those unwilling. This revealed articulate explicitly how organizations that engage in spatial areas important to take into account for the current spatial data data sharing are able to benefit, by focusing on their core sharing initiative in South Africa in order to foster a spatial data activity, to increase the usefulness of their GIS and to improve sharing “culture.” The six points are: the quality of their decision making. Furthermore, it would 1. As a key referent inside the organization for spatial data be important to mention the specific benefits of spatial data sharing activities, the management of GIS-using organizations sharing in terms of the savings in administrative efforts that (rather than mainly the technical GIS personnel, as is would otherwise be needed for data capture. currently the case) could be targeted by the spatial data 5. Furthermore, aside from the cost-benefit approach usually infrastructure initiative in order to increase their awareness adopted in the promotion of spatial data sharing initiatives, and understanding of spatial data sharing activities for the the results showed that more intangible consequences of effectiveness of their organizations’ GIS and so that they may spatial data sharing, such as the opportunity to gain new provide support and encouragement for such activities. insight, can act as incentives for spatial data sharing. They 2. Although they are key referents outside the organization for should be referred to in detail by providing information about spatial data sharing activities, the different sectors of the how organizations can benefit from spatial data sharing in GIS community do not yet seem to be providing clear and terms of these intangible outcomes with salient examples. encouraging stimuli for spatial data sharing. The results of Specific reference could be made to these by detailing how the analysis also suggest that, according the perception of organizations can benefit from spatial data sharing because the respondents, the NSIF is conceptually linked to the it may: 1) trigger dialogue and collective reflection across GIS national government departments, thereby indirectly organizational boundaries (such as joint problem solving) including the NSIF in the referents from the GIS community. and involve articulating and comparing spatial data-related Yet the role of the NSIF as an important referent for concepts; 2) allow access to ideas that are codified in organizations regarding their engagement in spatial data spatial data, enabling their translation, combination, and sharing could not be directly confirmed. In combination application in different organizations; 3) provide the means with the assessment of actual sharing behavior, which to observe spatial data skills through personal interaction found that the extent of sharing was still very limited, these and to create shared mental models and experiences across results suggest that a spatial data sharing “culture” has not organizational boundaries; and 4) enable the combination of yet developed. This adds further weight to the remaining new and existing spatial data sets from different organizations findings of the research that can be used to guide the fostering to create new knowledge. of such a sharing culture. 6. Finally, the results revealed that the South African 3. While it is important to specify particular benefits of engaging government initiative ‘Integrated Development Planning’ in spatial data sharing, the results show that the disincentives has emerged as another incentive for spatial data sharing. for spatial data sharing also need to be addressed. In particular, Since this has not previously been linked to the promotion the importance of reducing the fears associated with losing of spatial data sharing, specific reference to the benefits of control over spatial data has been highlighted. While much of spatial data sharing in terms of integrated development the promotion of the NSIF stresses the benefits of spatial data planning should be made in the context of the spatial data sharing across organizational boundaries, it would be helpful infrastructure initiative in South Africa. to reduce the fears associated with the negative consequences of such activities rather than solely emphasizing positive Conclusions outcomes. The threats to an organization’s strategic position The theory-based, behavioral approach to spatial data sharing would need to be addressed by specifying how unregulated presented in this article takes advantage of a robust model devel- access to an organization’s spatial data by other organizations oped in another field of enquiry within the social sciences (i.e., can be limited and how the use of an organization’s spatial in the existing body of knowledge about human behavior and data can be restricted to the particular organizations that are

24 URISA Journal • Vol. 15, APA I • 2003 URISA Journal • Wehn de Montalvo 25 decision making). Based on the validation and the analysis of References resultant model of the willingness to engage in spatial data sharing, it can be argued that: 1) the Theory of Planned Behavior has Abelson, R.P. and A. Levi, 1985, Decision Making and Decision been applied successfully as an organizing framework for research Theory. In Lindzey, G. and E. Aronson (Eds.) The Handbook on spatial data sharing; and 2) this model, as implemented us- of Social Psychology, 3rd ed (New York: Random House), ing the questionnaire instrument, could also be used in other 1, 231-309. national contexts to elicit the dispositions of actors in a specific Ajzen, I., 1985, From Intentions to Actions: A Theory of Planned GIS community. Behavior. In Kuhl, J. and J. Beckmann (Eds.) Action Con- From a policy perspective, the findings of the analysis re- trol–From Cognition to Behavior (Berlin: Springer-Verlag) ported here show that by using the Theory of Planned Behavior, 11-39. it is possible to generate specific insight for policy formulation by Ajzen, I., 1988, Attitudes, Personality, and Behavior (Milton revealing–in a systematic fashion–the incentives and disincentives Keynes, UK: Open University Press), 110. for decision-makers within organizations to share spatial data. The Ajzen, I., 1991, The Theory of Planned Behavior. Organizational results can provide a basis for specific guidance as to how policy Behavior and Human Decision Processes, 50, 179-211. makers may influence the actual behavior of spatial data sharing Ajzen, I., 1996, The Social Psychology of Decision Making. In more effectively. The strength of this approach is not to produce Higgins, E.T. and A.W. Kruglanski (Eds.) Social Psychol- generalisable truths but to identify context-specific incentives and ogy–Handbook of Basic Principles (New York and London: disincentives to spatial data sharing. The Guildford Press), 297-325. Ajzen, I. and B.L. Driver, 1992, Application of the Theory of Planned Behavior to Leisure Choice. Journal of Leisure About the Author Research, 24(3), 207-224. Ajzen, I. and M. Fishbein, 1980, Understanding Attitudes and Uta Wehn de Montalvo is researcher and advisor at the Insti- Predicting Social Behavior (Englewood Cliffs, N.J.: Pren- tute for Strategy, Technology and Policy at the Netherlands tice-Hall). Organisation for Applied Scientific Research. She previously Ajzen, I. and T.J. Madden, 1986, Prediction of Goal-Directed worked as a programmer for IBM UK Ltd. and as a research Behavior: Attitudes, Intentions, and Perceived Behavioral officer in SPRU–Science and Technology Policy Research. Control. Journal of Experimental Social Psychology, 22, She holds a B.Sc. in Computer Science, and a M.Sc. and 453-474. a Ph.D. in Science and Technology Policy. Her research Alfelor, R.M., 1995, GIS and the Integrated Highway Informa- focuses on access to geographic information and spatial tion System. In Onsrud, H.J. and G. Rushton (Eds.) Sharing data sharing, the social aspects of implementing spatial data Geographic Information (New Brunswick, NJ: Center for infrastructures, and the role of information and communica- Urban Policy Research), 397-412. tion technologies for development. She currently chairs the Azad, B. and L.L. Wiggins, 1995, A Proposed Structure for Data Policy Working Group of the Association of Geographic Observing Data Sharing. In Onsrud, H.J. and G. Rushton Information Laboratories for Europe (AGILE). (Eds.) Sharing Geographic Information (New Brunswick, Corresponding Address: NJ: Center for Urban Policy Research), 22-43. Dr. Uta Wehn de Montalvo Bacharach, S., P. Bamberger, and B. Mundell, 1995, Strategic TNO - The Netherlands Organisation for Applied Scientific and Tactical Logics of Decision Justification: Power and Research Decision Criteria in Organizations. Human Relations, Institute for Strategy, Technology and Policy 48(5), 467-488. Schoemakerstraat 97 Batty, M. (1992), Sharing Information in Third World Planning P.O. Box 6030 Agencies, National Center for Geographic Information and 2600JA Delft, NL Analysis, Buffalo, Technical Report No.92-8, February. email: [email protected] Callon, M., 1991, Techno-Economic Networks and Irrevers- ibility. In Law, J. (Ed.) A Sociology of Monsters: Essays on Power, Technology and Domination (London and New York: Routledge), 132-164. Clarke, D., E. Gavin, W. Honu, T. Krieg, M. Muller, H.J. Smith, T. Smith, and S. Vorster, 1998, Proceedings of the National Spatial Information Framework Workshop, Sinodale Sen- trum, Pretoria, February 11, 1998.

26 URISA Journal • Vol. 15, APA I • 2003 URISA Journal • Wehn de Montalvo 27 Dangermond, J., 1995, Public Data Access: Another Side of GIS Mellers, B.A., A. Schwartz, and D.J. Cooke, 1998, Judgement Data Sharing. In Onsrud, H.J. and G. Rushton (Eds.) Shar- and Decision Making. Annual Review of Psychology, 49, ing Geographic Information (New Brunswick, NJ: Center 447-477. for Urban Policy Research), 331-339. Mohr, L.B., 1982, Explaining Organizational Behavior (San Dawes, R.M., 1998, Behavioral Decision Making and Judge- Francisco: Jossey-Bass Publishers). ment. In Gilbert, D.T., S.T. Fiske, and G. Lindzey (Eds.) Nedovic-Budic, Z. and J. Pinto, 1999, Understanding Interorga- The Handbook of Social Psychology, 4th ed (New York: nizational GIS Activities: A Conceptual Framework, Journal McGraw-Hill), 1, 497-548. of the Urban and Regional Information System Association Dueker, K.J. and R. Vrana, 1995, Systems Integration: A Reans (URISA), 11(1), 53-64. and a Means for Data Sharing. In Onsrud, H.J. and G. Rush- Nedovic-Budic, Z., L. Warnecke, and J. Pinto, GIS Database ton (Eds.) Sharing Geographic Information (New Brunswick, Development and Exchange: Interaction Mechanisms and NJ: Center for Urban Policy Research), 149-171. Motivations. Submitted to Journal of the Urban and Regional Eagly, A. and S. Chaiken, 1993, The Psychology of Attitudes Information System Association (URISA), Version 09/24/01, (Fort Worth: Harcourt Brace College Publishers). www.urisa.org/journal/ Eagly, A.H. and S. Chaiken, 1998, Attitude Structure and Func- Obermeyer, N. and J. Pinto, 1994, Managing Geographic In- tion. In Gilbert, D.T., S.T. Fiske, and G. Lindzey (Eds.) formation Systems (New York and London: The Guildford The Handbook of Social Psychology, 4th ed (New York: Press). McGraw-Hill), 1, 269-322. Onsrud, H.J., 1995, The Role of Law in Impeding and Facili- Elliott, R., D. Jobber, and J. Sharp, 1995, Using the Theory of tating the Sharing of Geographic Information. In Onsrud, Reasoned Action to Understand Organisational Behaviour: H.J. and G. Rushton (Eds.) Sharing Geographic Information The Role of Belief Salience. British Journal of Social Psychol- (New Brunswick, NJ: Center for Urban Policy Research), ogy, 34, 161-172. 292-306. Fishbein, M. and I. Ajzen, 1975, Belief, Attitude, Intention, and Onsrud, H.J. and J.K. Pinto, 1991, Diffusion of Geographic In- Behavior: An Introduction to Theory and Research (Reading, formation Innovations. International Journal of Geographi- MA: Addison-Wesley). cal Information Systems, 5(4), 447-467. Hogg, M. and G. Vaughan, 1995, Social Psychology–An Intro- Onsrud, H.J. and G. Rushton, 1992, NCGIA Research Initiative duction (London: Prentice Hall). 9: Institutions Sharing Geographic Information, National Jones, E.E., 1985, Major Developments in Social Psychology Center for Geographic Information and Analysis, Report of During the Past Five Decades. In G. Lindzey and E. Aronson the Specialist Meeting, Technical Report 92-5, June 1992. (Eds.) The Handbook of Social Psychology, 3rd ed (New Onsrud, H.J. and G. Rushton (Eds.), 1995, Sharing Geographic York: Random House), 1, 47-107. Information (New Brunswick, NJ: Center for Urban Policy Kahneman, D. and A. Tversky, 1979, Prospect Theory: An Research). Analysis of Decision Making under Risk. Econometrica, Onsrud, H.J. and G. Rushton, 1996, Institutions Sharing 47, 263-291. Geographic Information, National Center for Geographic Keren, G., 1996, Perspectives of Behavioral Decision Making: Information and Analysis, Closing Report, Research Initia- Some Critical Notes. Organizational Behavior and Human tive 9, March 1996. Decision Processes, 65(3), 169-178. Opp, K.-D., 1999, Contending Conceptions of the Theory Kevany, M.J., 1995, A Proposed Structure for Observing Data of Rational Action. Journal of Theoretical Politics, 11(2), Sharing. In H. J. Onsrud and G. Rushton (Eds.) Sharing 171-202. Geographic Information (New Brunswick, NJ: Center for Pfeffer, J., 1985, Organizations and Organization Theory. In Urban Policy Research), 76-100. Lindzey, G. and E. Aronson (Eds.) The Handbook of Latour, B., 1987, Science in Action (Cambridge, MA: Harvard Social Psychology, 3rd ed (New York: Random House), 1, University Press). 379-440. Law, J. and J. Hassard (Eds.), 1999, Actor Network Theory and Pinto, J.K. and H.J. Onsrud, 1995, Sharing Geographic In- After (Oxford: Blackwell Publishers) formation Across Organizational Boundaries: A Research Masser, I. and H.J. Onsrud (Eds.), 1993, Diffusion and Use of Framework. In Onsrud, H.J. and G. Rushton (Eds.) Sharing Geographic Information Technologies (London: Kluwer Geographic Information (New Brunswick, NJ: Center for Academic Publishers). Urban Policy Research), 44-64. Markus, M.L. and D. Robey, 1988, Information Technology Rhind, D. (1998) ‘Public/private sector relationships in the and Organizational Change: Causal Structure in Theory creation, management and exploitation of geospatial data’ and Research. Management Science, 34(5), 583-598. in Conference Proceedings, XXI International Congress Medin, D.L. and M.H. Bazerman, 1999, Broadening Behavioral Developing the Profession in a Developing World - Com- Decision Research: Multiple Levels of Cognitive Processing. mission 3: Land Information Systems, vol.3, Brighton: The Psychonomic Bulletin and Review, 6(4), 533-546. International Federation of Surveyors, pp.201-215.

26 URISA Journal • Vol. 15, APA I • 2003 URISA Journal • Wehn de Montalvo 27 Rogers, E.M. (1995) Diffusion of Innovations, 4th ed, New York: Notes The Free Press. Sabini, J., 1995, Social Psychology, 2nd ed (New York: W.W. 1 The author is grateful for comments from the participants Norton and Company). of the ESF-NSF Workshop on Access to Geographic In- Sperling, J., 1995, Development and Maintenance of the TIGER formation and Participatory Approaches Using Geographic Database: Experiences in Spatial Data Sharing at the U.S. Information, Spoleto, Italy, 6-8 December 2001, where an Bureau of the Census. In Onsrud, H.J. and G. Rushton earlier version of this article was presented. (Eds.) Sharing Geographic Information (New Brunswick, 2 This parallels the distinction between variance and process NJ: Center for Urban Policy Research), 377-396. models in more general literature on organisational behavior Stage, D., 1995, A Multi-Agency Management Structure to (for example, Mohr 1982). Facilitate the Sharing of Geographic Data. In Florida. In 3 For a discussion of different measurement techniques, see Onsrud, H.J. and G. Rushton (Eds.) Sharing Geographic Hogg and Vaughan (1995). Information (New Brunswick, NJ: Center for Urban Policy 4 The current list of empirical research papers applying the Research), 426-447. Theory of Planned Behavior exceeds 240 articles (http: Staw, B.M., 1991, Dressing Up Like an Organization: When //www-unix.oit.umass.edu/~aizen). Psychological Theories Can Explain Organizational Action. 5 Modifications to the Theory of Planned Behavior have Journal of Management, 17(4), 805-819. been discussed, such as the inclusion of personal norm and Stevenson, M.K., J.R. Busemeyer, and J.C. Naylor, 1990, Judge- perceived moral obligation, self-identity, and past behavior ment and Decision-Making Theory. In Dunette, M.D. and variables, to improve the predictions of intention and be- L.M. Hough (Eds.) Handbook of Industrial and Organi- havior (Sabini 1995, Eagly and Chaiken 1993). It is argued zational Psychology, 2nd ed (Palo Alto, CA: Consulting that these additions may be drawn upon and included in Psychologists Press), 1, 283-367. the TPB model depending on the specific behavior under Tosta, N., 1995, The Evolution of Geographic Information Sys- investigation (Eagly and Chaiken 1998). tems and Spatial Data-Sharing Activities in California State 6 This would be the case, for example, for research based on Government. In Onsrud, H.J. and G. Rushton (Eds.) Shar- the concepts developed within actor-network theory (see, for ing Geographic Information (New Brunswick, NJ: Center example, Latour (1987), Callon (1991) and Law and Has- for Urban Policy Research), 193-206. sard (1999)) focusing on the negotiations and interactions Wehn de Montalvo, U., 2001. Crossing Organisational Boundar- among, and the motivations of, a small set of actors. ies: Prerequisites for Spatial Data Sharing in South Africa, 7 The sample includes just over 110 key individuals from some D.Phil. Thesis, SPRU–Science and Technology Policy Re- 70 different organisations. search, University of Sussex, Brighton. 8 The South African GIS community is made up of the fol- Yates, J.F., 1990, Judgement and Decision Making (Englewood lowing groups: local authorities, provincial government, Cliffs, NJ: Prentice Hall). national government, para-statal organisations, academic research institutions, GIS industry, private sector, and nongovernmental organisations. 9 For a conceptualisation of the diffusion of geographic information technologies, see Onsrud and Pinto (1991) and Masser and Onsrud (1993).

28 URISA Journal • Vol. 15, APA I • 2003 URISA Journal • de Man 29 Cultural and Institutional Conditions for Using Geographic Information; Access and Participation

W.H. Erik de Man

Abstract: Access to geographic information is a necessary and enabling but not a sufficient condition for participation in its use. Access and participation are both essentially behavioral in nature and therefore depend on social conditions, in particular, on culture and institutions. This article explores two existing models to describe different cultures (viz. well-known models proposed by Geert Hofstede and Mary Douglas, respectively) and the potentials of these models in analyzing cultural conditions for access to geographic information and participation in its use. Finally, the article proposes inclusion of cultural and institutional conditions in the emerging research agenda on access to and participatory approaches in using geographic information and suggests some steps to this end.

Introduction participation in general. What are critical conditions in this re- Taken together, the Workshop by the European Science Founda- spect? How will the information be integrated with folk perception-U.S. National Science Foundation on Access to Geographic tions about the world, which are generally in terms of narratives Information and Participatory Approaches in Using Geographic (stories) and images? In this article, I will argue that access to and Information, held in Spoleto, Italy, from December 6-8, 2001, participation in use of geographic information depend on social raised the question whether increased access to geographic infor- context, specifically on local culture and institutions. Participa- mation also increases public participation; if not: reinvigorates tory approaches in using geographic information – or any kind democracy, and benefits society at large? The obvious answer of information for that matter – are likely to have limited effects to that question is: it depends. Of course, the ultimate reason in societies that are not supportive to (popular) participation in for being concerned with the issue of participatory use of geo- general. The dependency on context does not ignore, however, graphic information – or of any information for that matter – is the positive – although limited – role that availability of and its potential support for some kind of “participatory develop- access to (geographic) information can play in increasing public ment.” However, there is evidence of both empowerment and participation. marginalization of people and communities due to differential access to information and to participation. (For a discussion of Access and Participation Depend these contradictory aspects related to geographic information, on Social Context see articles by Obermeyer and Pinto (1994:169-181) and Harris Access to (geographic) information and its usability are relevant and Weiner (1998).) concerns whenever people seek information. This will generally What do we mean by access to and participatory use of geo- be the case when they meet or anticipate problems. Many of these graphic information? I suggest we view information as a resource problems have a strong spatial dimension as in spatial analysis, that can be tapped but needs to be nurtured and maintained as spatial planning, and decision-making, implementation of these well. It follows that participants can both make use of the stock plans, and environmental monitoring and management. Access of information and add (local) knowledge, facts, interpretations, to and use of geographic information are schematically situated analysis, and the like to it. Participatory use of information is an between geographic information and its providing technology act of sharing, dialogue, and collaboration among individuals. on the one hand and spatial problems on the other. Access to Access to information would then enable these two-way flows. geographic information is important because the information Geographic information and its participatory use are particularly may be used because of solving spatial problems. significant when social groups (such as communities or societies at Access to and participatory use of (geographic) information large) collectively deal with development problems most of which are essentially behavioral in nature. Access to information is both having important spatial or locational dimensions. to be sought and seized by some, and to be enabled or facilitated The relationship between access to and participatory use of by others. The same is the case for participation. geographic information (GI) is not a linear process. Certainly, Although individuals exercise concrete problem-solving making use of information necessitates some form of access to actions, these individual actions generally do take place within it. But access itself may also be a drive into participatory use of groups or societies. And so do access to and the use of geographic information and this, in turn, may be a condition for popular information. Consequently, both are subject to social conditions.

28 URISA Journal • Vol. 15, APA I • 2003 URISA Journal • de Man 29 But both may contribute in shaping social conditions as well; of geographic information must be “rooted” within society in be it to a limited extent. Because of these social interactions, I order to be of real value. It is in this sense that one could view view geographic information as being socially constructed. Con- the provision of specific sets of geographic information itself as sequently, I view participatory approaches in using information being institutionalized within a group or society (see, for instance, as essentially social processes in the construction of information. De Man 2000). (The traditions that I follow in the article include “symbolic actionism” (Blumer 1969) and “social construction of technol- Access and Participation are Culturally and ogy” (Bijker and Law 1992, Bijker 1995) (see also, for instance, Harvey 2000). Institutionally Embedded We now can summarize the assumptions and speculations made so far (Figure 1). First, access to and use of geographic information Culture and Institutions are “between” spatial problems and geographic information tech- Which social conditions in particular influence access to and use nologies. Second, participatory approaches in using geographic of geographic information? Because culture and institutions guide information deal both with access to and use of geographic infor- individual actions and behavior, I would argue that both access to mation. Third, participation in the use of geographic information and use of geographic information are embedded in the culture is embedded in the culture and institutional arrangements of the and the institutional arrangements of the host society. But they host society. are not fully determined by cultural and institutional conditions only. The situation, for example, that laws in the United States allow greater access to government information and use of that information than is generally allowed in the nations of Europe Cultural Differences Culture differs from one society to another. How do we dif- cannot be explained by differences in cultural conditions alone. ferentiate between them? A classic distinction is between active Laws are only to a limited extent expressions of the dominant and passive societies. Active societies seek opportunities in their culture of the host society. But “the rule of law” as such might environment for improving their conditions and display a desire be culturally embedded. for attainment and to be in charge. Passive societies, on the contrary, seek to maintain their status quo and display a tendency Culture to be under the control of natural processes, of social waves and Culture is a major condition for access to and use of geographic developments, or of active others (Etzioni 1968). In other words, information because both rely ultimately on behavior. In social societies differ in they way they deal with uncertainties: do they sciences, the term “culture” refers to the shared ways and think- perceive them as opportunities or as threats? It is self-evident that ing that grow out of group experience and are passed from one seeking new opportunities requires fundamentally different sets generation to the next (Broom et al. 1981). Specifically, it refers of information than the maintenance of a status quo. to the deeper level of basic assumptions and beliefs that define in Notwithstanding its merits, the active-versus-passive dimen- a basic taken-for-granted fashion a group’s view of itself and its sion will not be sufficient to describe and explain the interaction environment. These assumptions and beliefs are learned responses between culture and the application of (geographic) information to the group’s problems of survival in its external environment technology. For example, it does not explain how different bureau- and its problems of internal integration (Schein 1985). cratic ways of management deal with information. To this end, Culture is of particular relevance for the use of information in problem solving. First, culture shapes what people think ought to be. Second, what solutions and remedies are acceptable in a given problem-situation depends to a considerable extent on cultural values. In both ways, culture determines information needs. Finally, the very processes of collecting and interpreting data are also influenced by culture.

Institutionalization The sociological notion “institution” is relevant as well. It refers to a recurrent social mechanism that is established and valued by a group, community, or society. Specifically, an institution has a normative impact on the behavior of individuals and may develop within groups or within society at large (see, for example, Broom et al. 1981 and Robertson 1982). In this sociological view, institutions are embedded within the host culture. Figure 1. Access, use, and participation are culturally and Access to and use of geographic information interact with institutionally embedded institutions. In this respect, one could argue that the provision

30 URISA Journal • Vol. 15, APA I • 2003 URISA Journal • de Man 31 the manner in which a society deals with differences in power and Power Distance. Culture’s way to accommodate human inequality. hierarchy and with uncertainty and risk has to be accounted for Large power distance cultures are hierarchical, authoritarian, as well. A more elaborate view on “culture” will be necessary. and elitist in the sense of accumulation of the good things Geert Hofstede’s well-known four-dimensional model to in life at the higher levels in the hierarchy and of the bad describe different cultures seems attractive and promising to things in life at the lower levels. Small power distance cultures this end. However, an alternative model to deal with different demonstrate flat organisations and value participation (e.g., cultural “biases” proposed by Mary Douglas and others is worth spread in the distribution of the good and the bad things to mentioning as well. In this section, both models are briefly in life). sketched. Individualism versus Collectivism. Culture’s way to accommodate the individual and the ‘group.” Individualist cultures are Four Dimensions of Culture (Geert Hofstede) composed of calculating citizens. In collectivist cultures, Hofstede’s study on cultural values of people in over 50 coun- group values are dominant. In such cultures, one’s sense of tries revealed common problems, but with solutions differing life is derived from one’s contribution to the common good from country to country (Hofstede 1980 and 1997). From (e.g., closely-knit crowds). Hofstede’s analysis of 116,000 questionnaires and more than Masculinity versus Femininity. Culture’s way to accommodate 20,000 interviews with employees of the multinational IBM in masculine and feminine “values.” Masculine cultures focus over 50 countries, he found that (national) cultures differed by: on achievements and success. They are aggressive and have 1) social inequality (including the relationship with authority); the visibility of success. Feminine cultures are caring cultures 2) the relationship between the individual and the group; 3) emphasizing quality of life, networking, and relationships as concepts of masculinity and femininity; and 4) ways of dealing social values (e.g., egalitarian and compassion). with uncertainty. These terms seem to reflect reasonably well Uncertainty Avoidance. Culture’s way to accommodate uncertainty. the basic (cultural) problems each society faces. Together, these Strong uncertainty avoiding cultures are characterized by dimensions form a four-dimensional model explaining, as Hof- little risk taking, minimal innovation, extensive institutions stede claims, around two-thirds of the value differences among to bring security and stability, conservative, and thorough national cultures. These cultural dimensions are crucial to issues planning. Weak uncertainty avoidance cultures are innovative as accessibility and sharing of information as well, and these are, and creative, and tolerant of differences in views and in turn, most relevant to any information and communication behavior. Risk and excitement are greater values for such technology (ICT). The latter is reflected in numerous applica- cultures than security and stability (e.g., play as it comes, tions of Hofstede’s work in research investigations within the incremental planning, and few contingency scenarios). field of ICT (see, for example, Grover et al. 1994 and Shore and Venkatachalam 1996). Cultures can be described in terms of various combina- The four dimensions of national culture in Hofstede’s model tions of these dimensions. Hofstede found that countries with can be briefly described as follows: a generally large power distance are also likely to be more col-

Aspects of Power Distance Uncertainty Avoidance Masculinity versus Femininity Participatory Use of Geographic Information Large Small Strong Weak Masculine Feminine Access to geographic L – Top H – H – Emphasis L – GI may L/H – GI is H – The possible information managers Accountability on control, be interesting needed only in contribution of assume they and contingency, but not really so far required GI in adjusting know best with transparency are and (“hard”) needed (“who to shine success various no need for valued sciences cares”) (Masculinity is interests and in further visibility defensive and protecting the aggressive at the underprivileged same time.) is valued Participation in L – Top H – Relatively L/H – GI L – GI may L – H – Networking, using geographic managers “flat” social needed only be useful in Communication establishing information assume they groups and if it provides involving others is not valued relationships, know best with organizations, assistance but little need and caring are no need to seek and involvement to perceived for full-fledged valued knowledge and of a wider circle security and (expensive) experiences of actors stability outputs from others

Table 1 Possible connotation of cultural dimensions vis-à-vis access to and participatory use of geographic information (Adapted from Van den Toorn and E. De Man 2000)

30 URISA Journal • Vol. 15, APA I • 2003 URISA Journal • de Man 31 lectivist, and countries with small power distance appear to be When conditions change, ways of life may change as well. Within more individualist. Hence, the dimensions of power, distance, one social group, different ways of life can be recognized and and individualism-versus-collectivism are correlated. This leads are in permanent dynamic imbalance; forming alliances though to three “independent” dimensions: power distance, uncertainty remaining competitors. avoidance, and masculinity vs. femininity. Armored with Hof- As compared with Hofstede’s model, this “way of life” model stede’s multidimensional model, we return to culture’s possible lacks a quantifiable foundation (operationalization). impact on access to geographic information and its participatory use. However, this can only be done speculatively at this stage Unanswered Questions (Table 1). At first sight, both models show some similarity. However, more L, H = low (L) and high (H) support for access to and par- elaborate analysis is required before any sound conclusion can ticipation in using GI from the respective cultural dimensions be drawn. For example, how does Hofstede’s observation that in his study “individualism” and “power distance” are inversely Cultural Biases as a Different Way of Life (Mary correlated across the full set of countries relate to “individualism” Douglas) and “hierarchy” as two distinct bur possible concurrent ways of Notwithstanding its merits, Hofstede’s model has some weak- life in the other model. nesses. For example, data for the model were restricted to those Both models reflect the understanding of “culture” as the countries with IBM establishments. Moreover, within national group’s learned responses to problems of survival in its external cultures, different sub-cultures can be identified (see, for in- environment and its problems of internal integration (Schein stance, Riggs 1962 and 1964). The application of the model at 1985). But how these models deal with spatial problems is not the level of these sub-cultures is embryonic at best. In addition, clear. For instance, is the “NIMBY syndrome” (Not In My Back Van Dongen (1997) argues that Hofstede is not analyzing culture Yard) a cultural response to a distinct category of spatial problems but producing it by assuming homogeneity as a basis for his fac- or is it a problem of internal integration itself? tor analysis, given that the assumption of heterogeneity would inhibit this analysis. Culture’s Possible Contribution to An alternative model is suggested by Thompson et al. (1990) a Research Agenda and describes different cultures as different ways of life. They Both (theoretical) literature and practice suggest that cultural build upon the grid-group typology proposed by Mary Douglas and institutional conditions are important for answering ques- (1978). Douglas argues that two dimensions of sociality can ad- tions regarding access to and use of geographic information, and equately capture the variability of an individual’s involvement in participatory approaches in this respect. However, to deal with social life: group and grid. Group refers to the extent to which this in concrete and operational terms lies still ahead and more an individual is incorporated into bounded units. The greater research will be needed. Here only a few steps – partly overlapping the incorporation, the more individual choice is subject to group – to this end can be proposed. determination. Grid denotes the degree to which an individual’s Operational framework for analyzing culture’s influence. life is circumscribed by externally imposed prescriptions. The Integrating different existing models (like the models proposed more binding and extensive the scope of the prescriptions, the by Hofstede and Douglas) and/or developing specific models for less of life that is open to individual negotiation. analyzing cultural conditions for access and participatory use of These two dimensions (group and grid) together produce geographic information. Preferably, this will be part of a larger four different “ways of life” (summarized by Carver 2001): framework emerging from the ESF-NSF Spoleto-meeting in Individualism (low group–low grid). Decisions present December 2001. opportunity, except for those that threaten freedom of choice Cultural connotation of spatial problems. For example, action (“the market” will provide control, so there is no need how does culture influence the definition and perception of spa- for other kinds of control). tial problems? How does culture influence coping with spatial Hierarchy (high group–high grid). Technological and problems? environmental decisions should be left to experts). Cultural connotation of institutional arrangements. To which Fatalism (low group–high grid). Fatalists feel that they have very extent are differential institutional arrangements governing access little control over decisions that affect them and accept to and use of geographic information culturally determined? whatever decisions are made on their behalf). Real-life case studies. Learning-by-doing involving real-life Egalitarianism (high group–low grid). Egalitarians fear risk to the case studies. What is the impact of participatory approaches in environment, the collective good, and future generations, using geographic information? What are the cultural conditions and believe that power and influence should be spread more in these real-life case studies? evenly within society.

Thompson et al. (1990) emphasize the dynamic character of these ways of life. Cultures are neither permanent nor singular.

32 URISA Journal • Vol. 15, APA I • 2003 URISA Journal • de Man 33 Conclusion Broom, L., P. Selznick, and D. Broom-Darroch, 1981, Sociology Central to the article is the question whether increased access (New York: Harper and Row). to geographic information also increases public participation; if Carver, S., 2001, Participation and Geographical Informa- not: reinvigorates democracy and benefits society at large. The tion: A Position Paper ESF-NSF Workshop on Access to obvious answer to this question that it depends still remains. Geographic Information and Participatory Approaches in Therefore, the article has concentrated on some major conditions Using Geographic Information, held in Spoleto (Italy), 6-8 in this respect. December, 2001. http://www.shef.ac.uk/~scgisa/spoleto/ Access to and participation in use of geographic information workshop.htm are important conditions when communities (or societies at large) De Man, W.H.E., 2000, Institutionalization of Geographic Infor- address common problems in their living environments. Access mation Technologies: Unifying Concept? Cartography and to (geographic) information is both a necessary and possibly an Geographic Information Science, 27(2), 139-151. enabling condition for participation in its use; but not a suffi- Douglas, M., 1978, Cultural Bias (Occasional Paper No. 35. Royal cient condition. Because of their essentially behavioral character Anthropological Institute of Great Britain and Ireland). both access to and participatory use in (geographic) information Etzioni, A., 1968, The Active Society (New York: The Free depend on social conditions; specifically on local culture and Press). institutions. Hence, approaches are needed for describing and Grover, V., A.H. Segars, and D. Durand, 1994, Organisational analyzing the predominant cultural conditions in concrete and Practice, Information Resource Deployment and System operational terms. Literature and practice suggest that such ap- Success: A Cross Cultural Survey. Journal of Strategic In- proaches can be found and developed. formation Systems, 3(2), 85-106. Finally, the article proposes to include cultural and institu- Harris, T., and D. Weiner, 1998, Empowerment, Marginaliza- tional conditions in the emerging research agenda on access to tion, and “Community-Integrated” GIS. Cartography and geographic information and participatory approaches in its use. Geographic Information Systems, 25(2), 67-76. Harvey, F., 2000, The Social Construction of Geographic Infor- mation Systems (Editorial Introduction). International Jour- About the Author nal of Geographical Information Science, 14(8), 711-713. Hofstede, G., 1980, Culture’s Consequences: International Dif- ferences in Work-related Values (Beverly Hills, CA: Sage W. H. Erik de Man is Senior Lecturer in Geographic Informa- Publications). tion Management and Institutional Development at the Hofstede, G.H., 1997, Cultures and Organizations: Software of International Institute for Geo-Information Science and the Mind (New York: McGraw-Hill). Earth Observation (ITC), Enschede, The Netherlands. He Obermeyer, N.J. and J.K. Pinto, 1994, Managing Geographic studied land surveying and – later – public administration. Information Systems (New York: Guilford Press). His current research interests are cultural and institutional Riggs, F.W., 1962, The “Sala” Model: An Ecological Approach to conditions for GIS, GIS and organizations, institutionaliza- the Study of Comparative Administration. Philippine Journal tion of GIS, GIS and governance. of Public Administration, 6, 3-16. Corresponding Address: Riggs, F.W., 1964, Administration in Developing Countries; The Social Sciences Division Theory of Prismatic Society (Boston: Houghton Mifflin). W. H. Erik de Man Robertson, I., 1982, Sociology (New York: Worth). International Institute for Geo-Information Science and Earth Schein, E.H., 1985, Organizational Culture and Leadership: A Observation (ITC) Dynamic View (San Francisco: Jossey-Bass). Enschede, The Netherlands Shore, B., and A.R. Venkatachalam, 1996, Role of National [email protected] Culture in the Transfer of Information Technology. Journal of Strategic Information Systems, 5(1) 19-35. Thompson, M., R. Ellis, and A. Wildavsky, 1990, Cultural Theory References (Boulder, CO: Westview Press). Van den Toorn, W. and E. De Man, 2000, Anticipating Cultural Bijker, W.E., 1995, Of Bicycles, Bakelites, and Bulbs: Towards Factors of GDI, In Groot, R. and J. McLaughlin (Eds.), a Theory of Sociotechnical Change (Cambridge MA: The Geospatial Data Infrastructure; Concepts, Cases and Good MIT Press). Practice (Oxford: Oxford University Press), 97-111. Bijker, W.E. and J. Law (Eds.), 1992, Shaping Technology/ Van Dongen, A.M., 1997, Culture as Method; Homogenizing Building Society: Studies in Sociotechnical Change (Cam- or Heterogenizing Strategies of Change; The Case of the bridge MA: The MIT Press). International NGO Forum on Indonesian Development Blumer, H., 1969, Symbolic Interactionism: Perspectives and (INFID) (Delft: Eburon Publishers). Method (Englewood Cliffs, N.J.: Prentice-Hall).

32 URISA Journal • Vol. 15, APA I • 2003 URISA Journal • de Man 33 URISA Journal • Niles, Hanson 35 A New Era of Accessibility?

Sarah Niles and Susan Hanson

Abstract: The authors explore some of the many ways that grounded social relations remain salient to the online world of the Internet. Because of the ability to distribute via the Internet information at a very low cost anywhere that is connected to a telephone line, many advocate the Internet as being a democratizing force in the modern world. Such claims, however, often overlook the constraints on this medium created by uneven social relations, which shape how we use technology and how we are able to use technology. This article explores examples of Internet use from recent research on the topic, examples that show how the social context in which online information is produced and consumed greatly affects the accessibility of that information.

Introduction mation—requires physical mobility. In both physical and virtual A recent front-page story in the Wall Street Journal asks whether access, one must know of the existence of a destination that will the city—as a high-density, highly centralized settlement pat- meet one’s needs, be aware of how such a destination might be tern—has a future (Wessel 2001). Citing many contemporary found, and be able to reach the destination. If connecting to the examples of the demise of centralized institutions and the concom- information superhighway from anywhere gives a person access itant success of dispersed, networked ones, the author wonders if to everywhere, the physical locations of the origin and destination the downtown can survive in the face of terrorist threats, which of that connection do not matter. What is germane to a person’s encourage dispersal, and information technologies, which enable ability to access online information is their knowledge of what it. Wessel concedes, however, that the kinds of accessibility and the information is available online and how to navigate to the place face-to-face interaction that cities foster are unlikely to be replaced where that information resides. or are replaceable to such a degree in the foreseeable future that In this article, we review research that considers accessibility the existence of cities will be undermined: “In short, the very via the Internet in light of grounded social relations. We draw on forces that are breaking up the old centralized institutions of the research from a wide range of disciplines to illustrate that physical 20th century—the ones that make possible the dissemination of access to infrastructure does not equate to accessibility and that information through decentralized networks and threaten rigid social and geographic context remain important to how and by military-style hierarchies—are often the result of old-fashioned whom information is produced and consumed. Specifically, we human interaction in crowded places.”1 By highlighting the de- provide examples showing that the use of both technology and pendence of the very creation of space-transcending information online information are context-specific; appreciating this specific- technologies (ITs) on face-to-face social interactions, this news ity is important in understanding the accessibility or final utility article nicely pointed to the enduring imprint of grounded social of the Internet in everyday life. Throughout our discussion, we relations2 on accessibility in an information age. relate questions of accessibility in an information age to changing In this article, we explore the concept of accessibility at the urban spatial structure and inequality. The 1999 Participatory GIS intersection of cyberspace and physical (or grounded) space. Early Conference held in Spoleto, Italy sought to discuss the liberaliza- prophecies hailing the ability of cyberspace to offer complete action of geographic information (GI) data and what that means for cess to everyone portrayed cyberspace as a sci-fi version of an iso- empowering people to participate in public decision-making. This tropic plain, where even the friction of distance would no longer review foregrounds the need to understand the various conditions hold sway. In fact, cyber-accessibility has proved to be remarkably under which people gain access to online information, including dependent on good, old-fashioned geography and grounded social GI, and the possible impediments to the general dissemination relations. Our focus here is on why physical places and grounded of this information. geographies remain salient to travel in cyberspace. Accessibility has always been important to retailers, politicians, and geographers, inter alia. Individual access refers to one’s ability to reach or obtain something (usually something desirable such as a paying job, medical care, or entertainment), and in the non-virtual world achieving access—often even access to infor-

URISA Journal • Niles, Hanson 35 Accessibility to Information in to understand the spatial patterns of flows of information. Ex- Physical and Virtual Space amining the distribution of physical infrastructure and patterns of digital information flows reveals that online communications are distributed to favor large cities in the U.S. Focusing on ag- Measuring Access to Online Information: gregate flows of information between places, geographers, and Technology Penetration others have determined that physical locations shape the form Access to the Internet in the United States has mostly focused on that the Internet takes—the information available on it as well access to the hardware necessary to connect to the Internet. Set up as patterns of traffic and connectivity. A number of researchers as an issue of uneven technology penetration, most surveys aimed at focus on measures such as communication flows (Castells 1996) measuring a population’s access to the Internet examine geographi- and information transfer capacities (Moss and Townsend 2000) cal and socio-economic disparities among those who have or don’t between places. By examining information transfer capacities (i.e., have a computer and modem in the home and consider those who the capacity of the physical infrastructure), Moss and Townsend lack these hardware items to fall on the wrong side of the “digital found that seven cities in the U.S. have the greatest capacity for divide” (National Telecommunications and Information Adminis- information transfer and, although these cities are also connected tration 2000). These important and influential studies equate access to the regions within which they exist, there is more capacity be- to the Internet with the physical presence of a home computer tween and among the seven large centers than between the centers equipped with a modem, the lack of which is considered the only and their surrounding regions. Castells claimed that instead of barrier to the unlimited benefits that the Internet has to offer. decentralization or the even distribution of information with the The results of these national studies of the digital divide Internet, a global urban network is being created where nodes are widely disseminated and have raised alarm that the U.S. is (major information producing and consuming cities) dominate becoming a nation of information haves (those with computers information production and dissemination. The ascendance of in their homes) and have-nots (those without computers in their cities to node status is usually explained through place-based homes). Although the overall penetration rates are rapidly increas- inertia and first-mover advantages that favor historically domi- ing across the U.S., the disparities persist and are even growing in nant economic and political centers (Sui 2000). This inertia, in some instances (National Telecommunications and Information which the spatial patterns of the IT economy mirror those of the Administration 2000). Rural populations, the inner-city poor, pre-IT world, has been demonstrated for everything from the minorities, young households, and female-headed households spatial distribution of information workers to the building of the remain disproportionately disadvantaged (McConnaughey physical infrastructure of the Internet atop existing rights-of-way and Lader 1998). Similar studies examining penetration rates that privilege cities (The Economist 2001c). The irony here is, of of computers and modems have shown that the digital divide is course, that the very places that have long benefited from excel- even more severe on a global scale; illiteracy and lack of physical lent physical access are now also benefiting from superb IT access, access to computer hardware explain why 88% of Internet us- so that instead of mitigating spatial inequalities, the Internet is ers come from countries that represent only 15% of the world’s exacerbating them. population (Charp 2001). Although access to appropriate hardware is a necessary condi- Place Matters: The Production and Regulation of tion for online activity, conceptualizing access to online informa- Internet Content tion as a function of uneven penetration of physical infrastructure The production and regulation of Internet content also increasing- encourages one to see physical space and physical access as the ly reflects the importance of geographic context. The production primary constraints on access in cyberspace; in this view, if you of Internet content is biased in the U.S. toward large established lack proximity to a telephone network, electricity, etc., you are cities, and the production of specialized Internet content often oc- offline. Yet, this sort of analysis denies the possibility that one’s curs in places where there was a previous specialization in activities on-the-ground social relations also shape opportunities for and associated with that content (e.g., entertainment in Los Angeles) constraints on virtual accessibility. Nor does framing the problem (Kellerman 2000). According to Kellerman, New York City, Los around physical access address whether closing the digital divide Angeles, and the Silicon Valley are the largest Internet content will erode inequalities in the access to information and improve producers, with concentrations of production within those places the general welfare of Americans. By this measure, accessibility (e.g., Manhattan in New York City, Hollywood in Los Angeles). would be ubiquitous if universal service could be assured for every Kellerman points to the accumulation of expertise and other so- citizen. Universal service in turn, would guarantee equal access cial and physical infrastructure in these places as reasons for their to welfare-improving information. dominance in the production of Internet content. As technological advances have increasingly made it possible Place Matters: Internet Infrastructure and to identity the location of Internet users, Internet content pro- Information Flows viders have used geography both to shape the content produced In studying the Internet, geographers have looked at the extent to on the Internet and to appeal to the place-based context within which Internet infrastructure is place-dependent and have sought which online information will be received. Firms can now target

36 URISA Journal • Vol. 15, APA I • 2003 URISA Journal • Niles, Hanson 37 Internet users by identifying the geographic location from which (Aoyama 2001). To overcome these barriers, e-commerce retail- they are connecting to the Internet and can then differentiate ers have joined forces with neighborhood convenience stores to users geographically based on socio-demographic data (The provide a set of dispersed, easily accessible physical locations for Economist 2001c). A new service offered by the company Quova ordering, paying for, and sending e-commerce goods, proving can determine the geographic location of Internet users down to that “interaction between real space and cyberspace need not take the zip code, thus allowing for online marketing and the local- a single form” (Aoyama 2001:133). As these examples illustrate, ization of Web sites (Singer 2001). Such technologies will allow retailers can enhance their accessibility via IT (i.e., by becoming the same geomarketing3—or differentiation of consumers across e-tailers as well as—not instead of—retailers) by capitalizing on space—that is a key component of non-Internet marketing. their pre-existing geographies. Content can also be differentiated by jurisdiction, for ex- The uneven, context-specific adoption and use of Internet ample to offer state specials or to reflect local regulations. InThe technologies is highlighted in one study of four different scientific Economist (2001d), it was predicted that the mobile Internet (or disciplines. Walsh (1997) surveyed scientists in physics, biology, mobile data services) will permit network operators to fine-tune chemistry, and mathematics about their use of Internet technolo- location-specific content to let users know, for example, of up- gies. He found that modes of Internet use differed in important to-date special offers or sales at stores within, say, 500 meters of ways depending on the social practices and technical limitations a user’s current location. of each discipline. For example, Walsh found that physics re- Nations have used this same technology to control the types searchers, who are often involved in large interdependent and of information that can be disseminated within national space on capital-intensive projects requiring large amounts of information the Internet. France has cited the ability to locate Internet users sharing and communication, use the Internet largely to coordinate in a court case that decided to regulate Yahoo! to prevent that their activities. On the other hand, chemists, whose research is company from marketing Nazi paraphernalia to French citizens often commodifiable, are less likely to share their early findings (The Economist 2001b). In Saudi Arabia and Singapore, the online for fear they will disseminate proprietary information. national governments enforce morality regulations by the selective For chemists, then, Internet use focuses on accessing databases screening of Internet content to prevent “immoral content” from of published research abstracts. Walsh’s study makes it clear that being downloaded in those countries (The Economist 2001b). the social context from which people seek information affects Geography therefore remains important in a number of ways how they use the Internet and the context from which people that conspire to deny the placelessness of cyberspace. From the disseminate (or not) information shapes the way these technolo- physical capacity of infrastructure, to the flow of information, to gies can be used. the production of Internet content, to what information can be Another illustration of the importance of grounded socio- viewed, grounded geographic contexts and the grounded social spatial context in Internet use comes from a pilot study that we relations that are unevenly distributed across those contexts govern undertook in Worcester, MA in 1999 to explore some of the access to GI. As pointed out in The Economist (2001c:20): ways in which employers use online recruiting to find workers (Niles and Hanson 2003). We had hypothesized that employers It is undoubtedly true that the Internet means that the distance between two points on the network is no longer terribly would use the Internet to broaden their search for employees, important. But where those points are [emphasis added] still both socially and spatially; that is, we thought employers would matters very much. Distance is dying; but geography, it seems, see the Internet as a way to gain access to a more varied pool of is still alive and kicking. talent than was available locally through traditional methods of recruitment such as word of mouth and news-article advertise- Context Matters: Adoption and Use of IT ments. The results of our small case study suggested, however, that How people create and make use of online information and ser- employers were often not seeking to broaden their applicant pool vices are also particular to the socio-spatial context within which socially or (especially) geographically. Instead they were advertis- such services are received. For example, the e-commerce-only ing jobs on Internet job boards like Monster.com in the hope of online grocer GroceryWorks was unprofitable as an online-only capturing the attention of a type of employee they had already retailer with dedicated warehouse space. When Tesco, the United identified—via their traditional recruiting methods—as desirable. Kingdom-based grocery chain giant, bought it and re-organized Perhaps most important, in terms of demonstrating the power of it to integrate online sales with its own bricks-and-mortar op- geography, the employers voiced a strong preference for hiring erations, the company was able to successfully capitalize on its employees who already lived nearby. network of existing stores, store design (easy picking), reputation, These examples show how on-the-ground social organiza- and existing supply chains (The Economist 2001a). The nature tion affects the ways that Internet and other technologies are of e-commerce in Japan similarly shows the importance of the adopted and the uses to which they are put. The social context social context in which ITs are embedded. In Japan, a tradition of technology and, in particular, the ways in which information of inexperience with catalogue shopping, a population with few and services that are available over the Internet are produced credit card users, and low penetration rates for home computers and consumed have an uneven geography that depends in part make for a distinctive social and physical context for e-commerce on pre-IT geographies. As the example of e-commerce in Japan

36 URISA Journal • Vol. 15, APA I • 2003 URISA Journal • Niles, Hanson 37 shows, this geographically uneven social context precludes singu- cover costs of production and operation, network operators have lar measures of accessibility. Measuring Internet access via home an incentive to please sponsors. Many search engines have adopted computer penetration rates in the U.S. flattens the variety of ways a system of “pay-for-placement” where “a rising number of [search in which online services may be accessed and thus ignores the engines] let companies pay to pop up prominently when people uneven social contexts of production and use. enter particular search terms” (Businesss Week 2001). Such a system may create little incentive for search engines to search out and index Place Matters: Understanding and Applying alternative, small-scale, highly specialized or local pages. In turn, the willingness to please sponsors might translate into less effort to Information find relevant information to fit the searcher’s particular needs. Sui Context and place affect not only what information is made avail- (2000) argues also that information available online is not always able online and what technological means are used to receive it; the best or the most accurate information available; one could add they also affect what online information is sought and how that that it is not always appropriate to suit the context from which it information can and will be applied. Virtual accessibility requires is sought. If content is produced disproportionately in the U.S., that people be able to find, make sense of, and apply informa- and within the U.S. disproportionately in its historically dominant tion on the Internet: considering access to information as good political and economic centers, there is a question of whether the in its own right overlooks the importance of the context within information produced in these places is really accessible, relevant, or which that information is received and the reason for which it is interesting to a person in for example the Southern Sudan. Having sought. An individual’s social and spatial location places her/him information structured in one place so that it suits a particular set in a context through which s/he learns how to use technologies of social relations does not mean it will necessarily make sense to and interpret information. The production of information in one those who do not share that social understanding. context with little or no knowledge of how it will be found and Second, the skills needed to harness the Internet in this way interpreted by people in other contexts compromises the notion are usually acquired through social processes that mostly occur of accessibility. in places. Most people learn how to use the Internet from other The lack of appropriate skills required to find online informa- people. In much the same way as having a car does not predict tion, as well as the lack of an appropriate context for understand- one’s ability to drive, simply putting a personal computer in ev- ing how to interpret and prioritize information according to one’s ery home does not ensure that everyone will be flying down the needs, can prevent many people from accessing the information information highway. they require. Research has shown that frustration with informa- An important study called “HomeNet” conducted in Pitts- tion overload and irrelevant downloads is a major stumbling block burgh provided a large number of households with a computer, to accessing desired information online (Hoftsetter 1998, Dodge modem, and Internet software and measured how, how often, 2000, Kibirige 2001). Another study of new Internet users found and for what purpose these were used. One of the most inter- that although initially many people had positive attitudes toward esting findings of the HomeNet study was that individuals in the Internet, often these people had only vague ideas of what households with at least one heavy Internet user were less likely purpose the Internet could serve (Kiesler et al. 1997). to have the otherwise frequent experience of having trouble get- Problems such as information overload and the difficulty ting started on the Internet owing to a lack of understanding of of locating relevant information are likely due to incompat- overall concepts and difficulty diagnosing problems (Kraut et al. ibilities between the context in which information is produced 1996). Another telephone poll in two U.S. states similarly finds and those within which it is consumed. Three points need to be that individuals in multi-person households, and especially in considered: households with children, are more likely to be frequent Internet First, information available online is a product of a person, users (Bucy 2000), presumably because of increased motivation group, or firm/agency that is embedded in a social organization and skills. These studies show how on-the-ground social relations and/or set of social relationships. Information is posted online for in the home provided a social form of access via motivation and a variety of reasons—from profit to public service to entertain- technical help. ment—that may involve motives not appreciated by the Internet Third, obtaining and learning how to use the requisite tech- surfer. The link between people and the online information that nology may be the first step in accessing information, but also they seek is usually through search engines (as evidenced by the necessary is the ability to translate information acquired online fact that search engines are the most popular ways of finding and so that one can use it for productive purposes. Seely-Brown and accessing information on the Internet) (Kellerman 2000, Kibirige Duguid (2000) argue that information is useful only when inter- 2001). However, studies of Internet users show that popular search preted with a set of knowledge; social contexts provide knowledge engines bias the information returned in an Internet search (be- that imbues information with meaning. Knowledge, they claim, cause these search engines specialize in different areas and never is a result of learning to be a practitioner, and practice shapes index more than one-third of the web on average) (Fornesca and the assimilation of how information can be dealt with, which is King 2000, Kibirige 2001). Furthermore, because for the most knowledge: “Both examples, the classroom and the workplace, part search engines are driven by a profit motive, or at least must indicate how the resources for learning lie not simply in informa-

38 URISA Journal • Vol. 15, APA I • 2003 URISA Journal • Niles, Hanson 39 tion, but in the practice that allows people to make sense of and between the socio-spatial context for information production use that information” (Seely-Brown and Duguid 2000:133). In and the context of its consumption. The degree to which people this vein, Kibirige and DePalo’s (2000) research examining the are able to interpret and filter Internet information to suit their online searching strategies of undergraduates, graduates, and specific contexts will greatly affect how accessible that informa- faculty found that those with more specialized and/or advanced tion really is. training had an easier time finding information online. The HomeNet study in Pittsburgh found that “local Pitts- The Ascendance of E-Mail: The Importance of burgh and neighborhood information and communication services have special appeal to participants” (Kraut et al. 1996:60). Information Tailored to the Individual In light of the importance of social and spatial context, it is not Kraut et al. attribute the appeal of local sites to: 1) searching for surprising that the most popular online activity—the one that information that is only useful at the local level (e.g., bus sched- keeps people logging on—is e-mail. U.S. studies cited by Bikson ules, information about social services), and 2) the idea that local and Panis (1997) show that e-mail is the Internet activity most sites appeal to “people’s social identities, [Where] [p]eople want to often used by the largest proportion of people. The HomeNet participate in and be part of groups, and geography provides one project in Pittsburgh similarly found that e-mail, and especially basis for group formation” (p. 60). People visit local sites because the desire to check if a new message had arrived, was the main local information is embedded in a familiar context that enables service that kept people logging on (Kiesler et al. 1997). Once people to understand and make productive use of information. In online to check e-mail, people may then go elsewhere on the addition, people might visit local sites frequently because they are Internet; in the words of the researchers, “For our sample, in- more heavily advertised in local areas and/or are more frequently terpersonal communication via E-mail was both more popular recommended by friends and/or family living locally. and more sustaining than information acquisition via the Web” Together, these results imply that the more aware people (Kiesler et al. 1997:3). are—before going online—of the information available online E-mail is a unique online activity in that e-mail messages are and how it is useful in their own context, the more efficient the almost entirely specific to an individual; people receive informa- search and the more accessible the information retrieved. tion over e-mail that is related directly to them. E-mail messages Information that appears online is first filtered and inter- have a context, people know why they are receiving an e-mail preted through the socio-spatial context of the producers of message, they know the origin of information, and therefore Internet content, and then that information is filtered through they are usually aware of the context in which the information is the socio-spatial context of the individual accessing the informa- being sent. In short, e-mails assume the knowledge and context tion. These filters may hamper the ability of information seekers that will be used in receiving the information in them; they are to apply Internet content to the specific need for which they embedded in a set of social relations that give people the con- sought that information. Research in the medical health field text with which to process information. As the HomeNet study has found that, despite the widespread dissemination of medical found “Compared to the Web, electronic mail is personalized, information on the Internet, individuals do not see the Web as spontaneous, and interactive; the content of a particular message the easiest way to access health information; they trust advice is usually tailored to the recipient and often takes into account from doctors and/or nurses far more than the information they their prior interactions” (Kiesler et al. 1997:3). The HomeNet receive online (Pennbridge et al. 1999). The unspecific, decon- study in Pittsburgh found that “e-mail sustains ongoing dialogues textualized medical advice available online is unlikely to replace and relationships” (www.homenet.andrew.cmu.edu/progress/), the advice of a trained doctor who knows the specifics of a case. and similarly Wellman et al. (2001) found that the Internet is For these reasons too, Wimmer et al. (2000) predict that while especially used to maintain friends met in more traditional, on- services and information on the Web may curtail the demand for the-ground ways. A survey by Wellman et al. of 39,211 visitors many market intermediaries (middle men), new intermediaries to the National Geographic Society’s Website found that not only will be needed. The Internet creates a demand for new kinds is e-mail used especially to maintain active social ties to existing of specialists, experts, and purveyors of quality information to friends, but people use e-mail to contact friends who live physi- disseminate and translate online information so that people can cally close more frequently (by three times) than they do friends use it to make competent decisions on the transactions that the who live at a distance. This study also found that the frequent Internet now allows them to make themselves. use of e-mail is often associated with frequent face-to-face and Information online, as elsewhere, is filtered through both the telephone interactions, leading Wellman et al. to conclude that socio-spatial context of the producers and the consumers of that e-mail supplements but does not substitute for face-to-face and information. Information available on the Internet is unique as it telephone communication. often comes without any traditional ways to identify the purposes The success of e-mail as the most popular online activity pro- and biases with which it is produced or the audience for whom vides an important clue about the importance of embedded social it is intended. As researchers of virtual accessibility, we may be relations and social context in making information accessible or overlooking social context as an important new frontier on the useful to the individual seeking or receiving online information. digital divide: the potential for fundamental incompatibilities

38 URISA Journal • Vol. 15, APA I • 2003 URISA Journal • Niles, Hanson 39 Because information available on the Web is embedded in social About the Authors or knowledge structures that are not always transparent to the person trying to access the information, it may be difficult for Sarah Niles is a Ph.D. Candidate in the Graduate School of that person to understand how to comprehend and apply that Geography at Clark University. She has a Master’s Degree information. Especially given the difficulty mapping and priori- in Planning from the University of Toronto and her current tizing information available online, it is likely that some special research addresses the role of grounded social relations in the knowledge is needed to make sense of the information that can function of labor markets for temporary workers. be identified. A person might acquire the special knowledge with which to make sense of online information via general knowledge Corresponding Address: gained through socio-spatial context, via an explicit statement School of Geography accompanying the online information which makes obvious its Clark University application, or via someone in the person’s social network refer- Sarah Niles ring her or him to online information and by doing so providing Worcester, Massachusetts 01610 a way to apply the information to the person’s needs. [email protected]

Conclusion Susan Hanson is the Jan and Larry Landry Professor and Chair of the School of Geography at Clark University. Dr. Hanson We have outlined some of the many ways that grounded geog- received her Ph.D. in Geography from Northwestern Uni- raphies mark cyberspace. Because pre-Internet geographies to a versity and her current research examines the difference that large extent shape Internet geographies, understanding patterns grounded, gendered social relations make to entrepreneurship of individuals’ access to information on the Internet requires a activities in place. close examination of the intersection of people’s use of IT and grounded socio-spatial relations. Corresponding Address: Grounded geographies shape the Internet by guiding the School of Geography placement of IT infrastructure such that—at regional and Clark University even intra-urban scales—physical access to the Internet closely Susan Hanson resembles pre-Internet patterns of spatial access to goods and Worcester, Massachusetts 01610 services. But physical access to Internet infrastructure alone does [email protected] not equate to access. Pre-Internet geographies shape constraints on individuals’ accessibility to usable information and knowledge on the Internet in other ways as well. As suggested in the Wall Acknowledgements Street Journal article mentioned at the outset of this article, the production of Internet content takes place (quite literally) in The authors thank Harlan Onsrud for the invitation to prepare selected urban locations, and the content itself often reflects its this piece as a keynote address at the Spoleto conference and we specific geographic origins. A great deal of information on the thank the Spoleto participants for their reactions to our article. Internet is also targeted to users in particular locations, and some The authors are also grateful to the National Science Foundation nation-states have begun to regulate the Internet content that for supporting this research through a Varenius seed grant. citizens within their boundaries can access. The ways in which people and institutions adopt and use the Internet also reflect This article was written while Susan Hanson was a Fellow at the pre-Internet geographies, such that one’s ability to obtain grocer- Center for Advanced Study at Stanford; she gratefully acknowl- ies, for example, via the Internet differs by geographic location. edges the fellowship support of the William and Flora Hewlett Finally, we note the importance of geographic context in people’s Foundation (Grant # 2000-5633). ability to understand and usefully apply information accessed via the Internet. We cite the example of e-mail as testimony to the importance of context in Internet access. Notes By describing the key ways in which Internet access inter- sects with grounded social relations, we have pointed to some of 1 The last report of the Office of Technology Assessment before the key constraints on Internet access that need to be addressed the U.S. Congress abolished it in 1995 reached the same if the initial promise of the Internet to provide equal access to conclusion (Office of Technology Assessment 1995). information is to be realized. 2 The term “grounded social relations” is used throughout to refer to interactions in physical, not virtual, space. 3 For a discussion of pre-Internet geomarketing, see Goss (1995).

40 URISA Journal • Vol. 15, APA I • 2003 URISA Journal • Niles, Hanson 41 References McConnaughey, J.W. and W. Lader, 1998, Falling Through the Net II: New Data on the Digital Divide (Washington, Aoyama, Y., 2001, Structural Foundations for E-Commerce D.C.: National Telecommunications and Information Ad- Adoption: A Comparative Organization of Retail Trade ministration). Between Japan and the United States. Urban Geography, Moss, M.L. and A.M. Townsend, 2000, The Role of the Real City 22(2), 130-153. in Cyberspace: Understanding Regional Variations in Inter- Bikson, T.K. and C.W. Panis, 1997, Computers and Connectivity: net Accessibility. In Janelle, D.G. and D.C. Hodge (Eds.), Current Trends. In S. Kiesler (Ed.), Culture of the Internet Information, Place, and Cyberspace (Berlin/Heidelberg/New (Mahwah, NJ: Lawrence Erlbaum Associates), 407-430. York: Springer-Verlag), 171-186. Bucy, E.P., 2000, Social Access to the Internet. Harvard Interna- National Telecommunications and Information Administration, tional Journal of Press/Politics, 5(1), 50-61. 2000, Falling Through the Net: Towards Digital Inclusion Business Week, 2001, Search Engines: Leading Us Astray?, (Washington, D.C.: National Telecommunications and i3744, 8. Information Administration). Castells, M., 1996, The Rise of the Network Society (Oxford: Niles, S. and S. Hanson, 2003, The Geographics of Online Job Blackwell). Search Preliminary findings from Worcester, MA Environ- Charp, S., 2001, Bridging the Digital Divide, T H E Journal ment and Planning A, in press (Technological Horizons in Education), 28(10), 10. Office of Technology Assessment, 1995, The Technological Re- Dodge, M., 2000, Accessibility to Information with the Internet: shaping of Metropolitan America (Washington, D.C.: U.S. How Can it be Measured and Mapped? In Janelle, D.G. Government Printing Office), September 1995. and D.C. Hodge (Eds.), Information, Place, and Cyberspace Pennbridge, J., R. Moya, and L. Rodrigues, 1999, Questionnaire (Berlin/Heidelberg/New York: Springer-Verlag), 187-199. Survey of California Consumers’ Use and Rating of Sources The Economist, 2001a, Surfing USA. June 30, 2001, 58-59. of Health Care Information Including the INTERNET. The The Economist, 2001b, The Internet’s New Borders. August 11, Western Journal of Medicine, 171(5), 302-305. 2001, 9-10. Seely Brown, J. and P. Duguid, 2000, The Social Life of Informa- The Economist, 2001c, Special Report: Geography and the Net. tion (Boston, MA: Harvard Business School Press). August 11, 2001, 18-20. Singer, M., 2001, Quova Acquires Realmapping, Gains More The Economist, 2001d, Special Report: The Internet Untethered. Internet Geolocation Technology. (http://siliconvalley.inte October 13, 2001, 12. rnet.com/news/article.php/3531_733021). Fornesca, T. and M. King, 2000, Incorporating the Internet Sui, D.Z., 2000, The E-Merging Geography of the Information into Traditional Library Instruction. Computers in Librar- Society: From Accessibility to Adaptability. In Janelle D.G. ies, 20(2), 38-44. and D.C. Hodge (Eds.), Information, Place, and Cyberspace Goss, J., 1995, We Know Who You Are and Where You Live: (Berlin/Heidelberg/New York: Springer-Verlag), 107-130. The Instrumental Rationality of Geodemographic Systems. Walsh, J.P., 1997, Computer Networks and Scientific Work. Economic Geography, 71(2), 171-198. In S. Kiesler (Ed.), Culture of the Internet (Mahwah, NJ: Hofstetter, F.T., 1998, Internet Literacy (Boston: Irwin/McGraw- Lawrence Erlbaum Associates), 385-406. Hill). Wellman, B., A. Quan-Haase, J. Witte, and K. Hampton, 2001, Kellerman, A., 2000, Phases in the Rise of the Information So- Does the Internet Increase, Decrease or Supplement Social ciety. Info, 2(6), 537-541. Capital? Social Networks, Participation and Community Kibirige, H.M. and L. DePalo, 2000, The Internet as a Source of Commitment. American Behavioral Scientist, 45(3), 436- Academic Research Information: Findings of Two Pilot Stud- 455. ies. Information Technology and Libraries, 19(1), 11-17. Wessel, 2001, Decentralization and Downtowns. Wall Street Kibirige, H.M., 2001, Internet Access in Public Libraries: Results Journal, October 21, 2001, 1. of an End User Targeted Pilot Study, 1997-2000. Informa- Wimmer, B.S., A.M. Townsend, and B. Chezum, 2000, Infor- tion Technology and Libraries, 20(2), 113-115. mation Technologies and the Middleman: The Changing Kiesler, S., R. Kraut, T. Mukhopadhyay, and W. Scherlis, Role of Information Intermediaries in an Information-Rich 1997, HomeNet Overview: Recent Results from a Field Economy. Journal of Labor Research, 21(3), 407-419. Trial of Residential Internet Use. Carnegie Mellon University, The HomeNet Project, August 1997. http: //homenet.hcii.cs.cmu.edu/progress/ovrview8697.html. Kraut, R., W. Scherlis, T. Mukhopadhyay, and J. Manning, 1996, The HomeNet Field Trial of Residential Internet Services. Communications of the ACM, 39(12), 55-64.

40 URISA Journal • Vol. 15, APA I • 2003 URISA Journal • Niles, Hanson 41 URISA Journal • Crompvoets, Bregt 43 World Status of National Spatial Data Clearinghouses

Joep Crompvoets and Arnold Bregt

Abstract: In December 2001, a survey of the World Wide Web was carried out to assess systematically the state-of-the-art of national spatial data clearinghouses around the world. The aim of this Web survey was to evaluate the status and the spatial distribution of the clearinghouse implementations and to find out the similarities and differences between them. Each clearinghouse was characterized by the following criteria: the year of first implementation; the number of data suppliers; the type of data accessibility; the metadata-standard; the number of datasets; the most recently produced dataset; the number of Web references; the number of monthly visitors; the frequency of Web updates; the language used; and the use of maps for searching and registration-only access. For 59 countries, a national clearinghouse has already been established. The results of this Web survey show a large variety in history, use, content, and management.

Introduction way. This article presents only the results of data collected the At present, of the 192 countries in the world, 120 are working month of December 2001. on their national spatial data infrastructure (NSDI) in order to create an efficient environment for the access of spatial data. One of the main components of an NSDI is a national clearinghouse Web Survey Methodology (McLaughlin 1991, Clinton 1994, Federal Geographic Data In order to assess national clearinghouse developments around Committee 1997, Onsrud 1998). A spatial data clearinghouse the world, a Web survey was undertaken. The main intention of can be defined as an electronic facility for searching, viewing, this survey was to examine the condition (status) of national clear- transferring, ordering, advertising, and disseminating spatial data inghouses. Added to this survey is the word “Web” to emphasize from numerous sources via the Internet. Such a clearinghouse that the needed information was mainly collected on the Web. usually consists of a number of servers that contain information Because of the easy access to the clearinghouse sites, the Web is (metadata) about available digital data. A national clearinghouse an excellent means by which to gather the needed information aims to become a kind of shopping mall for all national avail- quickly and objectively on a regular temporal basis. In this case, able spatial data as acquired by governmental agencies and/or a traditional research survey, which would collect information industrial bodies. by asking a set of pre-formulated questions in a predetermined The first national clearinghouse was established in 1994 in sequence in a structured questionnaire to individuals (Hutton the United States. From that moment, there has been fast develop- 1990), would not be a suitable approach to collect the needed ment of other clearinghouses throughout the world. Few studies information quickly, objectively, and easily. exist with information about the worldwide status of national The Web survey began in December 1999 and was conducted clearinghouses. Onsrud (1998) and Lance and Hyman (2001) in a systematic and periodical manner. The procedure consisted presented a list of the existing national clearinghouses; however, no of the following two steps: 1) making an inventory of all existing detailed description was included. To the best of our knowledge, national clearinghouses on the Internet; and 2) measuring several no systematic periodical research has taken place with regard to characteristics to describe each clearinghouse. the status of national clearinghouses. In order to fill this gap, in The inventory (step 1) was compiled by extensive browsing 1999 we started a survey of the World Wide Web, which was of the Internet, reading related literature, and contacting experts repeated every 6 months. The Web survey’s main objectives are and several Webmasters. Clearinghouses were characterized (step to assess the worldwide progress, the spatial distribution, and 2) based on the following criteria: ease of measurement, objective the similarities and differences between national clearinghouses. character, and clear presentation of history, content, and use and Additionally, this Web survey can be considered as a starting management of the clearinghouse. The following 12 character- point to gather information necessary for the analysis of the legal, istics were measured: 1) the year of first implementation; 2) the economical, cultural, technical, and institutional impacts on the number of data suppliers; 3) the type of data accessibility; 4) the development of clearinghouses. Moreover, since clearinghouses metadata-standard used; 5) the number of datasets; 6) the most are a key component of spatial data infrastructures, the evaluation recently produced dataset; 7) the number of Web references (Al- of the findings of this Web survey might improve the planning taVista and Google); 8) the number of monthly visitors; 9) the and investing of spatial data infrastructures in a more strategic frequency of Web updates; 10) the languages used; 11) the use of

URISA Journal • Crompvoets, Bregt 43 maps for searching; and 12) registration-only access. Almost all Results and Discussion of the above information was sourced from clearinghouse Web The main results of this “December 2001” Web survey for each pages. Additionally, in cases of uncertainty or missing data, the of the 12 mentioned characteristics are presented and discussed. Webmaster was contacted. The history of the clearinghouse is Some of the results are presented by region. This division in re- described by characteristic 1. The content is described by char- gions is derived from Dorling Kindersley’s World Atlas (1997). acteristics 2, 3, 4, 5, and 6. The use is described by characteristics 2, 7, 8, 10, 11, and 12. Finally, the management is described by 1. The Year of First Implementation characteristics 4, 6, 9, and 12. In the next section, each charac- This characteristic describes the history of the clearinghouse. teristic is explained and discussed in greater depth. Additionally, Beginning in 1994, the number of national clearinghouses has where available, information about funding and clearinghouse been steadily increasing (Figure 1). Currently, 59 countries strategy has been incorporated. have an implemented version on the Web (the year of first implementation is known for 52 of the clearinghouses). Additionally, it is known that nine countries have projects for implementation. However, the variety in number between the different regions is considerable (Table 1). For example, in Europe, North America, and South America, more than 50% of the countries have established a national clearinghouse, whereas in Africa this number is less than 5%. It is important to note that 124 countries have not conducted any initiative to build such a national facility.

In Figure 2, the global distribution of implementation status of national clearinghouses is presented. Here we can see that the main “hotspots” of implementation are the American continent, Europe (except Eastern Europe), Southeast Asia, and Australia. On the other hand, the main “holes” are the African continent and Middle East.

2. The Number of Data Suppliers This characteristic describes the number and diversity of data suppliers. The power of a clearinghouse is that several Figure 1 The first year of national clearinghouse implementation: per data suppliers can disseminate their products via this facility. year (columns) and cumulative (dashed line). The average number of data suppliers participating in a

Figure 2 Global distribution of status of national clearinghouses

44 URISA Journal • Vol. 15, APA I • 2003 URISA Journal • Crompvoets, Bregt 45 First Year of Total No. of No. of Asian No. of No. of No. of North No. of South Implementa- No. of African Countries Australian European American American tion Countries Countries Countries Countries Countries Countries

1994 1 1 1995 3 3 1996 6 1 3 1 1 1997 5 2 2 1 1998 14 1 3 1 6 2 1 1999 6 3 1 2 2000 10 1 3 4 2 2001 7 3 1 3 Date unknown 7 3 3 1 1994 – 2001 59 2 9 2 24 12 10

Building phase 9 4 2 0 1 2 0

No implementation 124 45 38 12 18 9 2

Table 1 The first year of national clearinghouse implementation (distributed per region).

Region Average Standard Median Maximum number Deviation Africa (2) 11 11 10 19 Asia (9) 7 6 5 16 Australasia & Oceania (2) 14 17 14 26 Europe (24) 33 49 12 133 North America (12) 204 551 6 1758 South America (10) 4 3 2 8

World (59) 54 239 6 1758

Table 2 The number of data suppliers of national clearinghouse per region. The number in parentheses is the number of clearinghouses analyzed per region.

Region Abstract Metadata Data (+ metadata) Africa (2) 1 1 0 Asia (9) 1 6 2 Australasia & Oceania (2) 0 1 1 Europe (24) 6 16 2 North America (12) 0 9 3 South America (10) 2 8 0

WORLD (59) 10 41 8

Table 3 The type of data accessibility of national clearinghouse per region. The number in parentheses is the number of clearinghouses analyzed per region.

44 URISA Journal • Vol. 15, APA I • 2003 URISA Journal • Crompvoets, Bregt 45 clearinghouse is high; however, there is great variety between 4. The Metadata-Standard Used the clearinghouses (Table 2). For Austria, the Czech Republic, This characteristic describes the metadata-standard used. With Slovenia, and the U.S., the number of data suppliers exceeds the diverse sources from which spatial databases are built, it 100. In Canada, there are 1758 data suppliers. This contrasts is extremely important to maintain information about the with the 35 clearinghouses that have fewer than 10 suppliers content, quality, source, and lineage of the data. A number (notably in South America and Asia with their powerful of standard organizations have developed (or are in the national mapping agencies). process of developing) standards for storing and maintaining metadata. The most mature of these have been developed by 3. The Type of Data Accessibility the Federal Geographic Data Committee (FGDC) (1995) This characteristic describes the presentation of the content. and the European Committee for Standardization (CEN/ Not all existing clearinghouses give access to data or metadata. 287 1996). These metadata-standards form the backbone For example, in some cases the clearinghouse presents only of national clearinghouses. The FGDC metadata-standard is a simple (not standardized) description of the datasets. For the most applied and distributed standard around the world this reason, three classes of accessibility are distinguished: (Table 4). The CEN standard is only applied in Europe. 1) abstract (simple/short description about the databases Recently, the International Organization of Standardization without using any formal meta-data description); 2) has created the ISO19115 standard (ISO/TC-211 2001). metadata; and 3) data (+metadata). In most clearinghouses, Currently, 10 countries have started a project to apply this the user has access to metadata (Table 3). However, in eight last-mentioned standard for their national clearinghouse countries (Australia, Canada, Dominica, Finland, Malaysia, Portugal, Singapore, and the U.S.), an option exists to access 5. The Number of Spatial Datasets the data itself. A means to quantify the content of a clearinghouse is the number of datasets. However, it does not represent the importance of the accessible datasets to the economic and social development of the country. The variety in the number of datasets is enormous (Table 5). For example, Region CEN FGDC Na- Others the U.S. federal clearinghouse can give access to almost tional 100,000 datasets (December 6, 2001), while the average Africa (1) 1 of the 24 European clearinghouses is 440. The difference Asia (8) 3 2 3 in the total number of accessible datasets between the Australasia & Oceania (2) 1 1 U.S. and Europe is easily noticed (100,000 vs. 10,000). In total, the clearinghouses describe 170,000 spatial datasets Europe (18) 7 4 7 together. 10 clearinghouses have more than 1000 datasets North America (12) 11 1 described (Australia, Austria, Canada, the Czech Republic, South America (8) 8 Japan, Mexico, South Africa, Switzerland, Uruguay, and the U.S.). WORLD (49) 7 23 8 11 6. The Most Recently Produced Dataset Table 4 Metadata-standards as used by national clearinghouses per This characteristic describes the up-to-date nature of content region. The number in parentheses is the number of clearinghouses analyzed per region. and the management of content in the clearinghouse. It is

Region Average Standad Total number Median Minimum Maximum number deviation of data sets Africa (2) 1561 2198 3122 1561 6 3115 Asia (6) 676 857 4056 260 9 1782 Australasia & Oceania (2) 15,031 21,249 30,062 15,031 5 30,056 Europe (22) 440 867 9768 118 8 3011 North America (10) 11,802 31,089 118,020 211 8 99,649 South America (8) 721 1646 5768 38 7 4735

WORLD (50) 3616 14,618 170,796 111 5 99,649 Table 5 The number of datasets described within national clearinghouses per region. The number in parentheses is the number of clearinghouses analyzed per region.

46 URISA Journal • Vol. 15, APA I • 2003 URISA Journal • Crompvoets, Bregt 47 Region Average Standard Median Mini-mum Maxi-mum Duration Deviation (months) (months) Africa (2) 31 41 31 2 60 Asia (6) 47 19 55 23 66 Australasia & Oceania (2) 2 1 2 1 2 Europe (22) 21 31 9 1 126 North America (10) 15 12 17 1 29 South America (8) 73 106 43 1 257

WORLD (50) 28 44 15 1 257

Table 6 The duration in months between date of the Web survey (December 2001) and the date of the most recently produced dataset. The number in parentheses is the number of clearinghouses analyzed per region.

Number of Web References Number of Web References “Google” “AltaVista” Region Ave-rage STD Median Maxi-mum Ave-rage STD Median Maxi-mum Africa (2) 50 56 50 89 41 43 41 71 Asia (9) 151 148 105 477 59 112 52 175 Australasia and Oceania (2) 3084 2851 3084 5100 1315 913 1315 1960 Europe (24) 320 792 42 3642 123 129 67 502 North America (12) 96 168 40 480 146 351 55 1080 South America (10) 112 140 48 428 76 78 53 213

WORLD (59) 312 857 50 5100 145 309 54 1960

Table 7 The number of Web references by AltaVista and Google search engines per region (STD = STANDARD deviation). The number in parentheses is the number of clearinghouses analyzed per region.

the difference in months between the date of Web survey Web site. The link popularity of national clearinghouse and the date of the most recently produced dataset described is high, which means that they are an excellent source of in the national clearinghouse. On average, the time of the consistent and targeted Web traffic. However, the variety production of the most recent dataset is more than 2 years is enormous (Table 7). The “Number of Web references” (Table 6). However, 22 national clearinghouses describe does not differ that much between the regions and so the spatial datasets produced within 1 year of the Web survey. popularity of a national clearinghouse can be considered as However, for 12 national clearinghouses, this duration is universal. The following national clearinghouses have high longer than 3 years (mainly countries located in South link popularity: Australia, Canada, Colombia, Finland, New America or Asia). Zealand, Norway, the U.S., and Venezuela.

7. The Number of Web References 8. The Number of Monthly Visitors This number can be interpreted as a means to measure This characteristic describes the use of national clearinghouses the popularity (use) of the clearinghouse site within the for accessing spatial datasets. This amount is related to the Internet network. The “Free Link Popularity Service” http: number of visitors who have visited the homepage of the //www.linkpopularity.com (The PC Edge, Inc.) is used, clearinghouse. The average number of visits of this page which measures the number of links to the home page of the exceeds the 5000 visitors. It is worth noting that the variety national clearinghouse that can be checked by the following between the implementations is high due to some particularly search engines: “AltaVista” and “Google.” A well-linked popular clearinghouses (Table 8). The following national popularity can dramatically increase traffic to the specific clearinghouses are visited the most: Canada, Finland,

46 URISA Journal • Vol. 15, APA I • 2003 URISA Journal • Crompvoets, Bregt 47 Region Average Standard Median Minimum Maximum Deviation Africa (1) 423 423 Asia (5) 1 055 382 1 120 618 1576 Australasia & Oceania (1) 4 378 4 378 Europe (10) 10 521 18 571 1 743 410 60 000 North America (6) 5 384 7 492 1 973 328 18 700 South America (3) 1 684 944 1 517 835 2 700

WORLD (26) 5 871 12 337 1334 328 60 000

Table 8 The monthly number of visitors per region. The number in parentheses is the number of clearinghouses analyzed per region. (note: the number of clearinghouses is much lower as illustrated in the other tables due to the fact that not all clearinghouse sites are able to count the number of visitors).

Region Average Standard Median (days) Clearinghouses Clearinghouses Duration Deviation Updated Within Updated more (days) 1 Day than 100 Days Ago Africa (2) 902 1270 902 0 1 Asia (9) 482 723 7 4 3 Australasia & Oceania (2) 12 13 12 0 0 Europe (22) 195 312 27 6 7 North America (9) 3 3 2 4 0 South America (10) 37 42 26 3 1

WORLD (54) 201 440 15 17 12

Table 9 The frequency of Web updates per region (days). The number in parentheses is the number of clearinghouses analyzed per region.

Portugal, Slovenia, and the U.S.; Portugal’s clearinghouse and Korean script). 29 clearinghouses use only their home has approximately 60,000 visits per month. language. These language problems reduce the accessibility to data (for English-speaking people). 9. The Frequency of Web Updates This characteristic describes the management of the 11. The Use of Maps for Searching content in the clearinghouse. One possible indication of a The use of this facility can improve the accessibility to data. well-managed clearinghouse can be seen by the frequency In 18 clearinghouses, maps can be used as an option to of updated information. The average number of days of last search for (meta)data. This relatively advanced alternative update is high for the whole population of clearinghouses for searching is popular in Europe and Asia. due to instances of poor management (with some updates exceeding 100 days) in Europe and Asia (Table 9). The variety 12. Registration-Only Access between clearinghouses is high as, alongside the poorer This characteristic describes the management and possible managed clearinghouse, numerous excellently managed limitations of use. Before accessing the data, users must facilities operate (update within 1 day). register themselves by entering personal details. This characteristic could have a negative impact on accessibility. 10. The Languages Used For eight national clearinghouses, the user is required to This characteristic describes the number and diversity of register to access metadata or data (Canada, El Salvador, users able to access data because of their familiarity and Finland, Hungary, Malaysia, Singapore, Spain, Canada, and knowledge of the given language. 30 clearinghouses do not Uruguay). have a search mechanism written in English (in addition, five of these are written in Arabic, Chinese, Greek, Japanese,

48 URISA Journal • Vol. 15, APA I • 2003 URISA Journal • Crompvoets, Bregt 49 Conclusions suitable national clearinghouse are El Salvador, Nicaragua, and Since 1994, the number of national clearinghouses has steadily Uruguay. Based on the above research, for all countries, it seems increased to a total of 59. Looking at the trend of implementation, that one of the keys for successful clearinghouse implementation countries can expect to see additional national clearinghouses is high political support and interest by means of funding and established. In fact, building clearinghouses is a global activity long-term strategy. (with the exception of Africa and the Middle East (as well as Australasia and Oceania)). Most existing clearinghouses are established in Europe, Southeast Asia, and North and South America. About the Authors The main initiatives for establishment come from Anglo-Saxon countries, such as the U.S., South Africa, and Australia. The U.S., Joep Crompvoets is Assistant Professor of Geo-Information in particular, which is supported by the FGDC, has stimulated Science at the Wageningen University in The Netherlands. many (American continent) countries to build a clearinghouse. From 1999, he has been monitoring the development of However, 124 countries have still not shown any initiative to build national clearinghouses and spatial data infrastructures one. There are several reasons for this. For example, a country around the world. may not have appropriate network architecture or there may be Corresponding Address: institutional bottlenecks for implementation. The differences in Wageningen University content, use, and management between the clearinghouses are Arnold Bregt broad. An example of such broad difference in content is the total Centre for Geo-Information number of accessible datasets described in a clearinghouse. In the P.O. Box 47, 6700 AA, Wageningen U.S. clearinghouse, this number is 10 times as high as the total The Netherlands number of all 24 European clearinghouses. The reason for such [email protected] difference is due to each country’s unique historical, institutional, economic, legal, technical, and cultural setting. Especially in Eu- rope, there are great contrasts in the number of datasets, suppliers, Arnold Bregt is Professor of Geo-Information Science at Wa- visitors, Web references, and frequency of Web updates, probably geningen University. Following more than 15 years of experi- as a result of the high institutional, economic, legal, technological, ence in the field of geographic information systems research and cultural diversity within this region. However, similarities and applications, his current areas of interest are spatial data between clearinghouses do exist (for example, the type of data quality, dynamic modeling of land use change, and spatial accessibility and the metadata-standard used). data infrastructures. From 1996 to 1998, he was one of the The most applied metadata-standard is the FGDC. However, project leaders to develop a national clearinghouse for spatial looking to the numerous projects to apply the ISO standard, it data in The Netherlands. is likely that ISO19115 will be the most applied standard in the Corresponding Address: future. This international consensus standard reflects FGDC, Wageningen University CEN, and other inputs. It provides detail that goes beyond Arnold Bregt FGDC and CEN metadata, including special coverage of raster Centre for Geo-Information and imagery information. Currently, there are several initiatives P.O. Box 47, 6700 AA, Wageningen to create implementable subsets and extensions of ISO19115 so The Netherlands that conversion of FGDC-support tools and implementations [email protected] to meet ISO conformance requirements is facilitated (Federal Geographic Data Committee Metadata Staff Coordinator 2001). Looking to the average number of data suppliers, Web references, Acknowledgements and visitors, we can conclude that national clearinghouses are a popular facility to distribute and access spatial data. The authors would like to thank Dr. Gert Jan Hofstede and Dr. Finally, in the future, it is highly probable that many national Monica Wachowicz of Wageningen UR for their critical com- clearinghouses will give access to spatial data itself and provide ments and suggestions on earlier versions of this article. Addition- complementary services such as online mapping. However, a con- ally, the authors appreciate very much the numerous reactions of cern could be the low frequency of Web updates of several clear- the Webmasters to all their questions. inghouses due to poor management. Therefore, special attention has to be given to keep clearinghouse managers motivated for having a well-managed clearinghouse. Based on the 12 characteristics used, we can conclude that Australia, Canada, Portugal, and the U.S. have the best existing national clearinghouses. Additionally, this Web survey shows that not only the richest countries have good clearinghouses. Examples of relatively poorer countries with

48 URISA Journal • Vol. 15, APA I • 2003 URISA Journal • Crompvoets, Bregt 49 References Hutton, 1990, Survey Research for Managers: How to Use Surveys in Management Decision-Making, 2nd edition (Basingstoke: CEN/TC 287, 1996, Geographic Information-Data Description- Macmillan) Metadata, Technical Report, prEN12657 (Brussels: CEN). ISO/TC-211, 2001, Geographic Information: Metadata, Inter- Clinton, W., 1994, Coordinating Geographic Data Acquisition national Standard 19115. and Access to the National Spatial Data Infrastructure. Lance, K. and G. Hyman, 2001, Adoption and Implementation Executive Order 12096, Federal Register 59, 17671-4, of National Spatial Data Infrastructure in Latin America (Washington, D.C.). and the Caribbean. 5th Global Spatial Data Infrastructure Dorling Kindersley, 1997, World Atlas (London: Dorling Kinder- Conference, Cartagena de Indias, Colombia, May 2001. sley Limited), 402 pp. http://codazzi4.igac.gov.co/gsdi5/m_sesion7.htm Federal Geographic Data Committee, 1995, Content Standards McLaughlin, J., 1991, Towards National Spatial Data Infrastruc- for Digital Geospatial Metadata Workbook (Washington, ture. Proceedings of the 1991 Canadian Conference on GIS, D.C.). Ottawa, Canada, Canadian Institute of Geomatics, Ottawa, Federal Geographic Data Committee, 1997, Metadata to Canada, March 1991, 1-5. Clearinghouse Hands-On Tutorial. http://www.fgdc.gov/ Onsrud, H.J., 1998, Compiled Responses by Question for Select- clearinghouse ed Questions. Survey of National and Regional Spatial Data Federal Geographic Data Committee (Metadata Staff Co- Infrastructure Activities Around the Globe. Global Spatial ordinator), 2001, Proposal for a National Spatial Data Data Infrastructure survey. http://www.spatial.maine.edu/ Infrastructure Standards Project, June 2001. http:// ~onsrud/gsdi/Selected.html www.fgdc.gov/standards/status/iso_metadata.doc

50 URISA Journal • Vol. 15, APA I • 2003 URISA Journal • Craglia, Masser 51 Access to Geographic Information: A European Perspective

Max Craglia and Ian Masser

Abstract: This article is offered to stimulate the formulation of a cross-Atlantic research agenda on access to geographic information. The article is organized in four sections. The first sets the context for the article. The second reviews at some length recent developments in Europe in relation to access to public sector information and the development of the Information Society. The geographic information dimension of these developments and the emergence of a strong environmental policy framework are also discussed. The third section focuses on developments in data policy at the national level using the United Kingdom as an example, as this country has often been depicted as being, in policy terms, at the antithesis of the United States federal policy. The fourth section concludes the article and identifies three main strands of research on access to geographic information that deserve a concerted effort by the research community in Europe.

Introduction quarters of the total European population (Russia excluded). Ne- The purpose of this article is to identify key issues that need to gotiations are almost complete for the enlargement of the Union be researched further to support policies aimed at maximising to another 10 countries in 2004, with a further three countries to access to geographic information (GI). To do so, it sets access follow once they have met the relevant criteria. By the end of the to GI into the wider policy developments in Europe which decade, the EU may have up to 28 member countries and a popu- today strongly focus on promoting the re-use of public sector lation of some 550 million inhabitants. The challenges entailed information through more transparent conditions on access and by such massive enlargement are far from trivial and consume a exploitation. These developments are also starting to affect those significant amount of political energy within the EU. parts of public sector information, such as geographic and meteo- The decision-making process in the EU is complex. The rological data, that have traditionally been seen as commercially key institutions are the European Council, which includes the more valuable and therefore exempt from open-access policies. Prime Ministers and Heads of State of the member countries, We are thus starting to move away from the polarization between and the European Parliament, which is elected by universal suf- the United States (federal) position in favor of open access and a frage and has co-decision powers on some issues. Sectoral policies European position in favor of cost recovery (Weiss 2002) toward are decided by the Council of Ministers (i.e., a collection of the greater convergence on both sides. With these considerations in competent ministers of the Member states) in co-decision with mind, the following sections will consider recent developments Parliament. The European Commission (EC) is the other key at the European level, then focus on national developments, and institution. The responsibilities of the EC are threefold: it initi- then explore recent policy changes through a case study of Great ates Community action; it ensures that the Treaties are correctly Britain. On the basis of this analysis, the article identifies key implemented; and it is the executive body for Community rules research areas that deserve a concerted effort. and programmes. In other words, the Council and Parliament must wait for a Commission proposal before they can legislate, and it generally falls to the Commission to apply or enforce the European Developments decisions taken and Community law. Therefore, the EC is not like a fully fledged national or federal government. Although powerful, it is essentially a relatively small administrative body, with a staff Background of 15,000, 20% of which are devoted to translate key documents Recent policy developments on access to GI in Europe cannot in the 11 languages of the EU, managing a budget of € 97 billion be separated from the much broader issues relating to access to (1999), which is just over 1% of the combined gross domestic public sector information and the development of an Informa- product (GDP) of the Member states (www.europa.int). tion Society. Before summarizing such developments, however, The main policy areas of the EU in financial terms are agri- it is useful to consider some of the key facts and challenges that culture policy (€ 43 bn. per annum) and regional policy (€ 31 characterize the decision-making process at the European Union bn.), representing 45% and 33%, respectively, of the EU budget. (EU) level. Given that some 4% goes to administration, all other internal The EU currently comprises 15 sovereign Member states, policy areas, external actions, and aid to the accession countries with a total population of almost 400 million, representing three combined absorb only 18% of the budget. Given the relatively

50 URISA Journal • Vol. 15, APA I • 2003 URISA Journal • Craglia, Masser 51 small size of the EU budget compared to national budgets, its key milestone was the publication in early 1999 of the Green role is one of coordination and support in leveraging public and Paper entitled Public Sector Information (Commission of the private sector investments in the Member States. The competences European Communities 1998). This consultation article has devolved from Member States to the Union are set in Treaties that played a major role in raising the debate across Europe on the are signed by the European Council and are subject to ratifica- opportunities created by the increased availability of public tion through the national Parliaments or national referenda. For sector information (PSI) in digital format for its reuse be- the purpose of this article, one of the key developments is the yond the purposes for which it was originally collected. The 1993 Maastricht Treaty, which gave the Union responsibility in article recognized existing barriers to accessing PSI, including matters of trans-European networks in the transport, energy, and different legal frameworks and pricing regimes, and posed telecommunications sectors, thus providing the legal basis for the pertinent questions on the extent to which such frameworks development of the Information Society. should be harmonized across Europe differentiating between administrative and non-administrative data, and “essential” Toward the Development of the Information versus value-added data. After extensive consultation, this Green Paper was fol- Society lowed-up by a Communication (Commission of the European The Commission’s White Paper on Growth, Competitiveness, Communities 2001a) and a draft Directive (Commission of the and Employment published in December 1993 (Commission of European Communities 2002a), which make the case for ac- the European Communities 1993) stressed the urgent need for a tion at the European level to remove the identified barriers and pan-European infrastructure to help boost economic growth and create a minimum level of harmonization on the commercial competitiveness at a time in which Europe was facing significant and noncommercial re-use of PSI. The draft Directive does not problems of industrial restructuring and long-term unemploy- address issues of access to data, arguing that these are best dealt ment. The development of a society strongly based on the creation with at national, regional, and local levels. Instead, it focuses on and use of information-related knowledge, products, and services ensuring a level playing field, transparency, and nondiscrimina- was seen as the key to the creation of new job opportunities in tory practices in the conditions for the re-use and exploitation Europe in the medium and long run. The term “Information of accessible data. Society” has since come to encompass the set of policies, initia- The increased emphasis on access to PSI has also ben- tives, and investments needed to achieve this goal. efited from the privatization of state-owned telecomms across Acting on proposals made in the White Paper, the Brussels Europe and the success of the European standard for mobile European Council of December 1993 asked a group of high-level telephony (GSM), which was promoted by the EC in 1994, experts and industrialists under the chairmanship of Commis- and has resulted in a multi-billion euros industry. This is a sioner Martin Bangemann to draft a report on the Information notable European success and Society suggesting practical ways in which its objectives could be its implications for future access achieved. The Bangemann Report (Commission of the European to the Internet and through this Communities 1994a) stressed the need to speed up the process of to a wide range of public and liberalization of the telecomm sector in Europe, hitherto largely in private sector information may the hands of state monopolies, and reinforce universal service. It well be significant. In particular, proposed a list of 10 initiatives to demonstrate the utility of new the high penetration of mobile Telematics applications, including: distance learning and tele- phones across all segments of working, traffic management, health care, trans-European public society in Europe, coupled with administrations networks, and city information highways. It is the development of third-gen- worth noting that the report stressed that financing the informa- eration phones providing high- tion infrastructure should come from the private sector, while speed Internet connections may the role of the EC would be to help target long-term investment give Europeans a different way in the exploitation of available technology. These proposals were of accessing information than incorporated into an Action Plan entitled Europe’s Way to the Americans. This is shown in Information Society (Commission of the European Communities Figure 1, which clearly demon- 1994b) and were also supported by a number of funding streams

strates not only the high level of 2001 13th October Economist, Source: in the Fourth Research and Development Framework Programme use of mobile phones compared 1994-98, with 3.6 billion allocated to the Information Society € to personal computers (PCs) in Technologies programmes. 1 Figure Europe, but also that the U.S. The European approach to the development of the and Canada stand out for being Information Society hinged on the twin track of liberaliza- the only countries of this group tion of the telecomm sector to achieve physical access and where PC use outstrips mobile the “liberalization” of public sector information to achieve phone use. The survey by The social access and economic objectives. Consequently, a key

52 URISA Journal • Vol. 15, APA I • 2003 URISA Journal • Craglia, Masser 53 Economist (2001) on the developments leading to a mobile this backdrop that the developments relating specifically to GI Internet explains these differences as follows. need to be understood. America’s enthusiasm for PCs, and lack of enthusiasm for mobile phones, is due to a combination of factors, but mainly to The GI Dimension the relative prices of fixed and mobile calls. In Japan, expensive Recent European policy developments relating to GI have been access to the fixed Internet drove users to mobiles; in America it reviewed by Craglia and Masser (1997), Masser (2001), and was mobiles that priced themselves out of the market. Local calls Longhorn (2001), among others. These centre on the initiative are free, and PCs are cheap. Mobile phones, on the other hand, taken by the European Commission in 1995-97 to develop a suffer the huge drawback that users are charged to receive calls, policy framework for GI in Europe (GI2000), which was intended so Americans tend to leave their mobile phones switched off to to take the form of a Communication to the European Council avoid having to pay for unwanted incoming calls. Mobile network and Parliament on the importance of developing core reference coverage is patchy and rates until recently were high. So teenagers data, metadata standards, and catalogues, and a mechanism for who want to keep in touch with their friends are given their own co-ordination through a High-Level Working Party. phone line, a PC and a dial-up account with AOL, the world’s These three main elements echo the main features of the largest Internet-access provider. Soon they are happily sending national spatial data infrastructure (NSDI) in the U.S., although and receiving instant messages through their PCs. the relatively small budget of the European Commission and the Their counterparts elsewhere in the world, in contrast, are lack of European agencies with a mandate to collect pan-European brandishing mobile phones and sending text messages. Mobile data make it difficult to take a strong lead. Hence, the develop- phones have become even more appealing to teenagers in Europe ment of a Strategic Defense Initiative (SDI) in Europe has to be since the advent of “pre-paid” phones, which can be replenished supported essentially by the Member States in partnership with with vouchers on sale at newsagents and corner shops. Pre-pay- the private sector, while the role of the EC is one of political ment, which originated in Italy to exploit a tax loophole, means coordination. that even people without bank accounts (such as children) can The discussions of the GI2000 drafts held with the various have mobile phones. It also enables users to control their spend- stakeholders in government agencies, industry, research, and user ing. (The Economist, October 13, 2001:11) communities were helpful in finding common ground and raising It is against this backdrop that there continues to be a awareness of the issues involved. Although the intended Com- significant political momentum behind the development of the munication did not materialize due to a perceived lack of support Information Society in Europe. This was clearly shown by the from industry and the low political visibility of this initiative, there adoption of a new “eEurope” political initiative by the European have been several projects funded or co-funded by the EC to sup- Council in 1999 (Commission of the European Communities port key elements of a future infrastructure. With respect to the 1999a) whose key objectives are to bring all citizens, business, development of policies for access and dissemination, the focus and administrations online, promote education, the availability shifted for a few years away from GI per se toward the broader of venture capital, and ensure that the whole process is socially context of public sector information, in which GI is singled out inclusive. as one of the major, and potentially more valuable, components The eEurope initiative has been followed by Action Plans for (Pira International et al. 2000). Very recently, however, we have 2002 (Commission of the European Communities 2001b) and started to see a renewed recognition that GI is crucial to informed for 2005 (Commission of the European Communities 2002b) policy-making, particularly in the context of environmental policy focusing on cheaper Internet access, education and skills, and key as reviewed below. application areas including e-commerce, health, and the delivery of government services and information. Of particular relevance for this discussion is the increased emphasis on e-government, The Environmental Dimension During the last 20 years, there has been an increased awareness with binding agreements among Member States to reach set across Europe of environmental issues. This is due to global targets for the delivery of public services on line by 2005, and as developments such as the Bruntland Report on sustainable a corollary, much increased access to public sector information development, the 1992 Rio Earth Summit, and the 1997 Kyoto by all citizens. The Action Plans are largely funded by national Protocol on Climate Change, local developments including the governments, with some core EU funding. increasing profile of environmental activists and pressure groups, The developments reviewed above indicate the concerted and European developments with the accession in 1995 of Swe- effort made in Europe for the development of the Information den, Finland, and Austria, which have traditionally held the local Society, key planks of which have been the privatization of the environment in high regard. In recognition of these pressures, state-owned telecomm monopolies, the steps toward a common environmental policy was identified as a competence of the EU legal framework including issues of Intellectual Property Rights in the 1993 Maastrich Treaty, and “sustainable development” (IPR) and data protection, and the agreement of common tech- was added to the core objectives of the EU in the Amsterdam nical standards such as GSM that have fostered a multi-billion Treaty of 1996. euro industry that is creating millions of new jobs. It is against

52 URISA Journal • Vol. 15, APA I • 2003 URISA Journal • Craglia, Masser 53 As a result, there has been a noticeable shift in emphasis thematic policies such as agriculture, transport, and regional in European policy from a sectoral, vertical approach to a more policy. In this way, a process has been launched to develop a new integrated horizontal one that pays much more attention to the legislative framework addressing “common base data collection, cumulative environmental impacts of policies such as agriculture, definition and use of standards, establishment of co-ordination transport, and regional development. Therefore, spatial planning bodies, information gaps, harmonizing community geographic at the regional scale has emerged as a powerful framework for information reporting requirements, metadata, and access con- analysis, co-ordination of intervention, and evaluation of impacts. ditions” (Perera 2001). These are, in essence, all the elements The formulation of the European Spatial Development Perspec- envisaged by the GI2000 proposals but this time there is strong tive (Commission of the European Communities 1999b) is the political support and a legal basis from which to operate. most clear embodiment of this approach, but its principles are also present in the other areas of policy. These include, for example, Summary the shift in agricultural policy from direct support to farmers “Europe works in mysterious ways” could be the title of this toward integrated rural development and the new requirements section, as it is often baffling to Europeans, let alone external for integrated plans for coastal zone management. commentators. What is clear from the above review is that The regional approach to planning, the increasing recogni- Member States remain the focal point for decision-making in tion of the importance of local issues and local stakeholders, and Europe, even in those areas where competence is granted to the a tightening of the requirements for monitoring and evaluation EU through international treaties. This is all the more clearly have also increased the importance of more focused interven- demonstrated when considering that EU funding is just over 1% tions, and hence the importance of geographic information to of the combined GDP of the EU 15, compared, for example, to assess needs, target intervention, and monitor effectiveness. The 18% of the GDP spent by the federal government in the U.S. increasing requirement to adopt GIS for policy monitoring and (http://www.whitehouse.gov/omb/budget/fy2002/balances.pdf). evaluation in fields such as agriculture and water management EU funding is therefore used for direction and coordination and are examples of this increasing importance. to leverage funding from the public and private sector in the As the need for better monitoring and evaluation in complex Member States. environments is recognized, so are the current deficiencies in With respect to access to information, the review has shown harmonized, consistent datasets and indicators at the appropriate that a key priority has been given to increasing physical and eco- geographic scale and time series. To address these data limita- nomic access through the privatization of the telecomm industry tions, we are starting to see a major shift in emphasis toward a and the development of common standards. More recently at- more decentralized approach to data management, leaving the tention has turned to public sector information, with a focus on data at the level at which it can be more easily collected and services to the citizen and government-to-consumer relationships. updated, with an attempt to integrate information flows from Although geographic information is part of this, it is not the local to global and vice versa more cohesively. Assuring access to primary focus at the European level, although changes may occur such geographic and environmental data becomes an absolute with the new emphasis on environmental information. pre-requisite in this scheme. Hence the initiative announced by The European success in relation to the use of mobile cell the Directorate General for the Environment toward the devel- phones may have major implications for future access to the opment of an Infrastructure for Spatial Information in Europe Internet by the majority of the population. Thus, many of the (INSPIRE) to be embedded in Community legislation (www.ec- research questions that are dominant in the U.S. (see, for example, gis.org/inspire). Sheppard et al. 1999) may need revisiting from a European per- The legal basis for the development of INSPIRE is in the spective based on the widespread access through mobile phones Water Framework Directive 2000/60/EC of the 23rd October rather than through PCs. 2000 which established a framework for community action in the When it comes to access to data, it is also clear that parallels field of water policy. This Directive has a strong spatial impact, as with the U.S. context need to be viewed with great caution as it defines river basins as the most appropriate spatial framework there is no such thing as European data in terms of ownership, for a comprehensive approach to water protection. All Member given that there are no European-wide agencies collecting data. States have to define such basins by 2003 and prepare manage- All data come from the Member States and, although EU agen- ment plans within the following 6 years. Such plans require the cies may do a significant amount of work in harmonizing such collection, integration, and analysis of a large body of data which data, the resulting IPR is at best mixed, and conditions of access the Directive specifies need to be in GIS format, including the are often unclear, given the variations in national policies. This physical, environmental, and socio-economic state of the basin. may also change in the future with an increasing convergence of Given the legal backing of this Directive, which is manda- national policies as discussed in the following section. tory on all Member States, and the extent of spatial information it requires in GIS format, it is clear that a coordinated framework is needed to avoid duplication and ensure compatibility of layers. Moreover, the data collected are seen as a foundation for other

54 URISA Journal • Vol. 15, APA I • 2003 URISA Journal • Craglia, Masser 55 National Policies On Data Access: whatever the pricing policy, price should not deter use of A Case Study of Great Britain data but on the contrary should encourage it. What we are witnessing therefore is significant variation in a Overview number of areas, the importance attributed to GI-related initia- In recognition of the rapidly evolving picture with respect to poli- tives, the conditions of access, pricing, overall awareness, but a cies for access to public sector information and GI in particular distinct sense that no government can afford to do nothing, and in Europe, the European Commission and European Umbrella that this realization is sinking in even among those countries that Organisation for Geographic Information (EUROGI) organized are less advanced. workshops in 1999 and 2000 to identify similarities and dif- The second meeting focused on 12 accession countries. The ferences in policy in the EU Member States and the accession findings (Craglia et al. 2001, Craglia and Masser 2002) show the countries, respectively. tremendous effort that all 12 countries are making in moving The first meeting looked at 10 of the 15 countries in the EU from the command economy that characterized the last 50 years [Austria, Belgium (Walloon Region), Finland, France, Germany, of their history to an open market economy benefiting the whole Ireland, Italy, the Netherlands, Portugal, and the United King- of society and not just small parts of it. It is a huge challenge that dom] plus Hungary and Norway. The key findings (Craglia et involves not only the adoption of new legislation, but also a major al. 2000) are summarized below: reorganization of administration with decentralization at regional 1. There are many national initiatives across Europe concerned and local levels, and the development of new skills, procedures, with the development of data policies in general and GI and ways of working to be able to benefit fully from integration policy in particular. In some cases (for example, in Portugal in the Union, administer wisely the funding opportunities that and the Netherlands), the GI dimension is very strong. arise from this process, and guide the whole of society through In other countries, such as France, GI is recognized as an this transition process. important element of public sector information, which With respect to the development of NSDI strategies, it is highlights the potential conflicts between economic and important to note that the vast majority of the countries analyzed social objectives. have taken a very strategic view of GI infrastructures by formulat- 2. Whilst in most countries the emphasis is on the national ing and adopting national strategies either on their own or as part level, it is important to recognize that in other countries of a National Information Strategy and/or an Information Society the regional/local dimension is particularly important, as in Strategy. These are very significant initiatives because they are not Germany, Italy, and Belgium. related to the level of economic development of the countries 3. Whilst some countries display a higher level of awareness involved, but show a genuine awareness of the strategic role of than others, there is a sense to which the very rapid government in enabling the development of an Information So- developments brought about by technology and the ciety, and the key role of geographic information for both public Internet are forcing governments to come to terms with policy and service provision. The degree of vision and political the opportunities and challenges that these developments commitment displayed by most of the countries analyzed is ad- create. At the present time, most countries seem to have mirable and significantly higher than that of many governments adopted policies in respect to access to public sector of the existing EU Member States (Craglia et al. 2000). information. These may in some cases take the form of Whilst these overviews give a snapshot of the situation Freedom of Information legislation. There is, however, a in different countries at the turn of the millennium, it is also quite different set of policies that needs developing in respect worth looking very briefly at the direction of change in policy to a pro-active dissemination of public sector information. by focusing on one country, Great Britain, which has often been The Mandelkern (1999) report in France is a good example portrayed as the other extreme to the U.S. approach to freedom of the difference between access and dissemination. of information. Disseminating information has major implications in respect to the organisation of work in the public administration, Policy Shifts in Data Access in Great Britain intellectual property, metadata (i.e., the first step in active The situation in the Great Britain in the mid-1990s was described dissemination is to declare what is available), relationships and evaluated by Masser (1998). The position at that time was still with the private sector, and pricing. very much coloured by the 1986 Tradable Information Initiative, 4. In respect to pricing, there are significant variations among which was taking a robust commercial line asking all government the countries analysed. In some cases, a distinction has been departments to evaluate their information holdings and to charge drawn between essential data that is free of charge (i.e., paid market price where there was already an established market and for through general taxation) and value-added data for which where a market was yet to be developed to charge only for the there is a fee. In other cases, a policy has yet to emerge, and costs incurred over and above those associated with handling the individual organisations act independently. Where a policy information for their own purposes. The other key policy initia- exists, a general principle that seems to emerge is that, tive, the Next Steps Programme in 1988 (Chancellor of the Duchy

54 URISA Journal • Vol. 15, APA I • 2003 URISA Journal • Craglia, Masser 55 of Lancaster 1996) was to increase the efficiency of the public An Office of the e-Envoy was created as part of the Prime sector by turning many government departments into Executive Minister’s Cabinet Office to coordinate the achievement of the Agencies with defined efficiency, financial, and customer service objectives set above (http://www.e-envoy.gov.uk/). Central to targets. These agencies include the Ordnance Survey (OS), HM its strategy is the one-stop government portal UKOnline (http: Land Registry, and the Central Statistical Office, which were given //www.ukonline.gov.uk/). Underpinning this portal is the re- cost recovery targets varying between 40% for the Meteorological quirement to all government organisations to document their Office to 100% for the Land Registry, the OS being somewhere information resources using an agreed-upon metadata standard in between at 80% (Masser 1998). Whilst the OS had made no- and register them on the Information Asset Register. A search table successes in completing the national topographic database in engine, Inforoute, then enables trawling through these assets 1995, its licencing system was cumbersome and geared to satisfy (http://www.inforoute.hmso.gov.uk/). A separate Government few large corporate users such as the utilities and local government Gateway portal is used for secure transactions between citizens rather than the mass market. This evaluation indicated many op- and government (http://www.gateway.gov.uk/). portunities but also significant variations in the policy framework Government pricing policy and licencing systems have also for access to government data, with tensions between the desire been greatly simplified as a result of the review of Crown Copy- to promote the use of information and to raise capital at the same right (HMSO 2001) and the Cross-Cutting Review of the Knowl- time. With respect to the development of an NSDI-equivalent edge Economy by the Treasury in 2000 (HM Treasury 2000). A framework in the U.K., Masser argued that, whilst the core data Green Paper on the Crown Copyright (1998) set out a series of were already largely in place, the metadata was rather limited options including abolishing Crown Copyright, retaining it but and the coordinating mechanism rather weak. The development with waivers in particular conditions, or waiving copyright as a of a National Geospatial Data Framework promoted by the OS general rule except for specified categories of information (http: appeared promising, but had yet to deliver at that time. //www.hmso.gov.uk/copy.htm). The outcome of the consultation The Labour government that came into power in 1997 was a new regime whereby copyright is retained as a statement started putting in place a number of major initiatives and re- of the authoritative nature of the information. For “essential” views that have significantly changed the approach to government information such as legislation, forms, and so on, copyright is information. The key ones are itemized below: waived and access is free. For basic or “raw” information, a sim- Freedom of Information Act (Home Office 2000). The Act plified user licence easily available on the web is needed and the was passed in November 2000 after 3 years of consultation and information is charged at marginal cost with no restrictions on debate. While the first draft of the Act in 1997 was welcomed for reuse. For value-added information, and information from the its liberal approach, successive drafts have significantly restricted Trading Funds, individual licences are needed and prices are set the scope of the Act, creating a long list of exemptions which by the individual organisations. caused public outcry (see, for example, The Economist 1999). The move toward a marginal cost charge for all “raw” infor- The final version passed by Parliament took on board some of mation (unless otherwise specified by Government) was recom- the comments, but by most European and U.S. standards is still mended by the Cross-Cutting Review of the Knowledge Economy a rather restrictive legislation. For example, Sections 36-2c and (HM Treasury 2000). This recognized that “The current policy 36-4 exempt government-held information, including statistical of average cost pricing creates a significant barrier to the re-use information, from disclosure if it “would otherwise prejudice, or of information because it requires parts of government, where would be likely otherwise to prejudice, the effective conduct of this is not core business, to make assessments and attributions public affairs.” This clause is potentially much more restrictive of relevant costs and negotiate individual contracts in an area in than the simple harm test proposed in the 1997 original proposal. which many departments and agencies are ill-placed to operate. Moreover, the Act will not be implemented in full until 2005. Marginal cost pricing would remove this burden from both the More progress has been made within the framework of the department concerned and the private sector” (ibid. para 1.15). Modernising Government initiative (Cabinet Office 1999), Additional reforms adopted by the Government as a result of which set out to increase collaboration across all government this review include: departments to deliver better and more efficient services, and set  Freedom for all government bodies to produce value-added the target for all government services, including those by local services charged at market prices provided that this can be government, to be available online by 2005. This was followed achieved in a transparent manner and with a level playing by Modernising e-Government in 2000 (Cabinet Office 2000), field with the private sector. which laid down the four basic principles of public services in  Reposition the HMSO as a regulatory body for government the information age: content to ensure fair competition, including the overseeing  Building services around the citizens’ choices; of trading funds.  Making government and services more accessible;  Request all public sector copyright owners other than  Social inclusion; and the Crown Copyright, such as local authorities, to review  Using information better. their charging, licencing, and access practices and align

56 URISA Journal • Vol. 15, APA I • 2003 URISA Journal • Craglia, Masser 57 them with the new reforms (http://www.hmso.gov.uk/ Summary and Conclusions keyreforms.htm). Tosta and Domaratz (1997) made the point that the success of the NSDI was not so much due to a splendid enlightenment of These changes have had two positive results. First, they American decision-makers on the value of GI, but to the NSDI provide a much clearer framework for all government agencies fitting the political agenda of the Clinton administration to at central and local levels, and more importantly for users on ac- “re-invent” the federal government (Gore 1993) and was thus cess, charges, and licencing. Second, they enable a much increased included in a packet of recommendations “focused on establishing provision of government information in electronic form free of accountability and control at the lower levels of Federal agencies, charge or at affordable costs than was hitherto the case and, equally as well as at state and local government agencies” (Tosta and important, a mechanism through which this information can be Domaratz 1997:21). easily found and accessed. In the two main sections of this article, we have indicated Whilst Trading Funds were exempted from the rule of mar- that, in Europe, access to GI is part of a much broader picture ginal cost recovery and are allowed to charge market prices for on access to public sector information and the development of their value-added products, there has been a significant move to- the Information Society. This was launched in the early 1990s ward developing a light-touch regulatory framework that ensures as a policy that was aimed at job creation in the face of massive transparency of procedures and pricing, promotes fair trading and structural changes in the European economy. Over time, the a level playing field, and reassures the market that dominant play- achievement of social and environmental objectives has also ers do not exploit their position to the detriment of others (Tullo strengthened the momentum behind this policy, thus providing 2002). The “Fair Trading Scheme” promoted by the HMSO in a virtuous alliance of interest. its new position as regulator requires the Chief Executives of the The policy environment is very dynamic and moving toward Trading Funds to publicly agree to these principles and publish a much increased provision of public sector information, often their pricing policy. This is backed-up by an independent audit- through paths that could not be envisaged even a few years ago ing mechanism to ensure adherence to these principles (http: as the case-study of Great Britain demonstrated. Moreover, the //www.hmso.gov.uk/regulation-outcome-anxb.htm). enlargement of the Union and the very proactive policies being Compared to the situation described by Masser in 1996, pursued by the accession countries in respect to the Information there is now a much more proactive engagement and leadership by Society and the development of spatial data infrastructures (see central government in respect to access to public sector informa- the Overview in the section entitled “National Policies on Data tion. While in the past, government information was primarily Access: A Case-Study of Great Britain”) suggests a number of seen as a commodity, there is a real sense to which the concept of important research strands that ought to be developed at the custodianship is gaining ground, particularly given the extent to present time, which fall under three main strands: institutional, which access to information underpins key government policies organisation, and economic/legal. for improving services delivery and departmental cooperation. The first strand should focus on the role of governments This process, however, is not without some setbacks and some in developing appropriate frameworks for sharing information resistance as shown in the case of the Freedom of Information horizontally across government departments and between public Act. On the whole, however, the policy changes described bring and private sector organisations and vertically from central to local the U.K. much more in line with the developments in other parts government and individual citizens. As this article indicates, the of Europe, and we can see clear convergence with the policies great variety of approaches at national and regional levels across formulated in e-Europe (Commission of the European Communi- Europe and the U.S. create an excellent laboratory on which to ties 1999a) and with other national governments in respect to a study such processes. The focus on governments is particularly distinction between essential, core, and value-added information appropriate, given that they are at the same time the major pro- (Craglia et al. 2000). ducers, consumers, and regulators of public sector information Accessibility to GI has also increased, not only through in general and geographic information in particular. the OS, but through other government departments such The second strand should focus at the organisational level as the Office for National Statistics (see, for example, http: because this is the level at which major challenges exist to engen- //www.statistics.gov.uk/neighbourhood/home.asp) and through der a culture of information documentation, sharing, and re-use. the repackaging of government information by the private sector Over and over again, we witness that the best-intentioned policies (see, for example, www.upmystreet.com). These developments flounder in their implementation not due to technical difficul- have largely been driven by the market, technology, and broader ties but organisational ones. At the same time, some of the rapid government policy, rather than through the development of a progress made in many central European countries appear to be specific NSDI. In fact, the closest policy that the U.K. had in this related to the greater ability to change organisational behaviour respect, the National Geospatial Data Framework, has now been and cultures during a time of rapid societal transition and change abandoned, while its metadata service is now managed by the AGI, (Pauknerowa and Corbin 2002). Therefore, the relationship be- the Association for Geographic Information (www.agi.org.uk). tween organisational cultures and access to GI, in both private and public sectors should be studied in greater depth relation to

56 URISA Journal • Vol. 15, APA I • 2003 URISA Journal • Craglia, Masser 57 the ability to adapt and change, and the incentives or disincentives Ian Masser is Professor of Urban and Regional Planning at the for data sharing. Rigorous models exist to undertake cross national ITC in the Netherlands and Head of their Urban Planning comparative studies in this area that should be built upon (e.g., and Management Division. He was the Founder Chairman Hofstede 1997 and Wehn de Montalvo 2003), as well as existing of the Association of Geographic Information Laboratories work in the field of information management. in Europe from 1998 to 1999 and was elected President of Finally, the emergence and impact of regulatory frameworks, the European Umbrella Organisation for Geographic Infor- such as the one in Great Britain (see the section “Policy Shifts in mation in March 1999. Data Access in Great Britain”) which seek to protect the intel- Corresponding Address: lectual property rights of data providers while promoting wider Ian Masser access and the use of GI, trust among stakeholders, and a level Division of Urban Planning and Management playing field need to be closely monitored and carefully evaluated International Institute for Aerospace Survey and the Earth in both national and trans-national contexts. This is important Sciences (ITC) because such frameworks are emerging throughout Europe, and P.O. Box 6 are heavily promoted by European Commission policies such as 7500AA Enschede, The Netherlands those on the environment (INSPIRE) and the exploitation of Public Sector Information (Commission of the European Com- munities 2002a). Therefore, an evaluation of their impacts is absolutely necessary. References In promoting these key areas of research, there are opportunities to be exploited in the forthcoming 6th Framework Cabinet Office, 1999, Modernising e-Government (London: for Research, Technology and Development of the European HMSO). http://www.official-documents.co.uk/document/ Union, which promotes the development of European-wide cm43/4310/4310.htm networks of excellence to help coordinate research effort and Cabinet Office, 2000, e-Gov: Electronic Government Ser- achieve critical mass. Therefore, it is important to consider how vices for the 21st Century (London: HMSO). http: existing geographic information research networks established by //www.cabinet-office.gov.uk/innovation/2000/delivery/ organisations such as the European Science Foundation and the foreward/frames.htm Association of Geographic Information Laboratories in Europe Chancellor of the Duchy of Lancaster, 1996, Next Steps Agen- can be utilised to create a European network of excellence in cies in Government Review 1995. Cm 3164 (London: the geographic information policy field and join forces with our HMSO). American counterparts to build a programme of international Commission of the European Communities, 1993, Growth, comparative research. Competitiveness and Employment: The Challenges and Ways Forward into the 21st Century (Brussels: Commis- sion of the European Communities). Commission of the European Communities, 1994a, Europe and About the Authors the Global Information Society: Recommendations to the European Council (Brussels: Commission of the European Massimo Craglia is a Senior Lecturer at the University of Shef- Communities). field, Sheffield, United Kingdom, and an adviser to the -Eu Commission of the European Communities, 1994b, Europe’s ropean Commission on data policy issues. He founded and Way to the Information Society: An Action Plan, COM(94) chaired the Data Policy Working Group of the Association 347 Final (Brussels: Commission of the European Com- of Geographic Information Laboratories in Europe (AGILE) munities). 1999-2002. Commission of the European Communities, 1998, Public Sector Corresponding Address: Information: A Key Resource for Europe COM(98) 585 Max Craglia Final (Brussels: Commission of the European Communi- Dept. of Town and Regional Planning ties). University of Sheffield Commission of the European Communities, 1999a, eEurope: Sheffield S10 2TN, U.K. An Information Society for All. COM(1999)687 (Brus- [email protected] sels: Commission of the European Communities). http: //europa.eu.int/information_society/eeurope/news_library/ documents/index_en.htm

58 URISA Journal • Vol. 15, APA I • 2003 URISA Journal • Craglia, Masser 59 Commission of the European Communities, 1999b, The Hofstede, G., 1997, Cultures and Organizations: Software of the European Spatial Development Perspective (Brussels: Mind (London: McGraw-Hill). Commission of the European Communities). http: HMSO, 2001, Licencing of Crown Copyright: HMSO Regulato- //www.inforegio.cec.eu.int/wbdoc/docoffic/official/space_ ry Framework (London: HMSO). http://www.hmso.gov.uk/ en.htm regulation-framework.htm Commission of the European Communities, 2001a, eEurope HM Treasury, 2000, Cross-Cutting Review of the Knowl- 2002: Creating a EU Framework for the Exploitation of edge Economy (London: HMSO). http://www.hm- Public Sector Information COM(2001) 607 Final. (Brus- treasury.gov.uk/sr2000/associated/knowledge/index.html sels: Commission of the European Communities). http: Home Office, 2000, Freedom of Information Act (London: //www.cordis.lu/econtent/psi/ HMSO). http://www.legislation.hmso.gov.uk/acts/acts2000/ Commission of the European Communities, 2001b, eEurope 20000036.htm 2002: Impact and Priorities. COM(2001)140 Final (Brus- Longhorn, R., 2001, The Impact of Data Access Policies on Re- sels: Commission of the European Communities). http: gional Spatial Data Infrastructure. Proceedings of the 7th //europa.eu.int/information_society/eeurope/news_library/ EC-GIS Conference, Potsdam, June 10-13, 2001 (Ispra: documents/index_en.htm JRC). http://www.ec-gis.org Commission of the European Communities, 2002a, Proposal for Mandelkern, D., 1999, Diffusion des Données Publiques et a European Parliament and Council Directive on the Re-Use Révolution Numérique (Paris: Commissatiat General du and Commercial Exploitation of Public Sector Documents. Plan). http://www.ladocfrancaise.gouv.fr/ COM(2002) 207 Provisional (Brussels: Commission of the Masser, I., 1998, Governments and Geographic Information European Communities). (London: Taylor & Francis). Commission of the European Communities, 2002b, eEurope Masser, I., 2001, Regional SDI’s: The European Perspective. Ar- 2005: An Information Society For All. COM(2002)263 Final ticle presented at the International Symposium on Spatial (Brussels: Commission of the European Communities). Data Infrastructure, Melbourne, November 19-20, 2001. Craglia, M., A. Annoni, and I. Masser, 2000, Geographic Infor- Pauknerowa, E. and C. Corbin, 2002, GI Capacity Building in mation Policies in Europe: National and Regional Perspec- the Accession Countries: Country Report. http://www.ec- tives (Ispra: European Commission). EUR19552en http: gis.org/ginie/documents //www.ec-gis.org Perera, P., 2001, EESDI Presentation at the First Eurogeograph- Craglia, M., and J.-F. Dallemand, 2001, Geographic Information ics General Assembly, Dublin, September 3-5, 2001. http: and the Enlargement of the European Union (Ispra: Euro- //www.eurogeographics.org/NewsAndEvents/PastEvents/ pean Commission). EUR19824en http://www.ec-gis.org GA2001/Docs/ec_environ.ppt Craglia, M., and I. Masser, 1997, A European Policy Framework Pira International, Ltd., University of East Anglia, and Knowl- for Geographic Information. Computers Environment and edge, Ltd., 2000, Commercial Exploitation of Europe’s Pub- Urban Systems, 21(6), 393-406. lic Sector Information (Luxembourg: EC DG INFSO). Craglia, M., and I. Masser, 2002, Geographic Information and Sheppard, E., H. Couclelis, S. Graham, J.W. Harrington, and the Enlargement of the European Union: Four National Case H. Onsrud, 1999, Geographies of the Information Society. Studies. URISA Journal, 14(2), 39-47. International Journal of GIS, 13(8), 797-823. The Economist, 1999, Secret Society. May 27, 1999. http: Tosta, N. and M. Domaratz, 1997, The U.S. National Spatial //www.economist.com Data Infrastructure. In Craglia M. and Couclelis H. (Eds.), The Economist, 2001, The Internet Untethered: A Survey of the Geographic Information Research: Bridging the Atlantic. Mobile Internet. October 13, 2001. www.economist.com (London: Taylor & Francis), 19-27. Gore, A., 1993, From Red Tape to Results: Creating a Gov- Tullo, C., 2002, UK Government Information Initiatives: En- ernment that Works Better and Costs Less. Report of the couraging Global Access and Re-Use. Proceedings of GSDI National Performance Review (Washington, D.C.: U.S. 6 Conference, Budapest, September, 16-19, 2002. AutoDesk Government Printing Office). (Hungary: Budapest). Her Majesty’s Stationary Office (HMSO), 1998. Crown Copy- Weiss, P., 2002, Borders in Cyberspace: Conflicting Public Sec- right in the Information Age. tor Information Policies and their Economic Impacts. http: HMSO, London, http://www.hmso.gov.uk/copyhome.htm //205.156.54.206/sp/Borders_report.pdf Wehn de Montalvo, U., 2003, In Search of Rigorous Models for Policy-Oriented Research: A Behavioral Approach to Spatial Data Sharing. URISA Journal. Volume 15, Number 2.

58 URISA Journal • Vol. 15, APA I • 2003 URISA Journal • Craglia, Masser 59 URISA Journal • Carver 61 The Future of Participatory Approaches Using Geographic Information: developing a research agenda for the 21st Century

Steve Carver

Position paper prepared for ESF-NSF Meeting on Access and Participatory Approaches in Using Geographic Information, Spoleto Italy, December 5–9, 2001

Abstract Geographical space is an important factor across a wide range of decision-making problems, not only because many human and environmental processes vary markedly from one place to another, but also because space determines people’s perception of the world and ultimately determines what we call ‘local’ and ‘global’. The ability of individuals and social groups to map out, in a true geographic sense, their vision for the future is a key prerequisite for informed and sustainable planning. Obtaining sufficient input from a representative sample of the broad population is often made more difficult by the lack of ‘connectivity’ (in both the technical and conceptual sense) between the different stakeholders involved. Technological approaches to improving participation, including the use of GIS and ICTs, are reviewed. The effects of space, place and locality are identified as key factors determining public interest in decision problems. Recommendations are made for further research directions in the field.

Introduction of the technology within a framework of more traditional means The current heightened research interest in participatory ap- of outreach and participation. proaches to decision making may well be a waste of time or, This position paper reviews the current situation in respect at best, misplaced confidence. The effort expended is done so to public participation and the use of geographical information on the possibly false surmise that the general public want to be to both encourage and facilitate wider involvement in the deci- more closely involved in decision-making, and perhaps more sion-making process. In doing so it is first necessary to give some importantly, that those in positions of decision-making power attention to general theories of participation and empowerment actually value and therefore encourage public input. A pessimistic within the democratic process before examining the role of GI, assessment, maybe, but it needs careful consideration if we are to and space itself, in providing a framework for active public par- progress further with research and development in this field. ticipation. The factors that work for and against this are analysed I have chosen to start the discussion from this point be- in some detail, before drawing up a list of recommendations for cause over ten years experience of developing participatory GIS further research effort in this field. has taught me that we, as academics, tend to credit the public with more knowledge, greater rationality and enthusiasm for participation in decision making than we perhaps ought. It is Democracy and Participation surprising how little even well educated people know about a Democracy, the government by the population through elected broad range of basic topics, often holding quite erroneous beliefs representatives, is perhaps the single most defining characteristic of and irrational ideas that appear to be based largely on here-say, western civilization. Public participation involves ordinary citizens sensationalist media reporting and poor understanding of basic in decisions about, and the implications of, social and economic facts. The willingness to become actively involved is also lacking change. Participation in democratic society has for most people across the population as a whole and is symptomatic of a wider until recently been limited to choosing elected representatives malaise in the democratic process; witness the poor turnouts through the voting process and then lobbying them over issues in many elections and the mistrust and contempt in which we of concern. However, as people have become ever more informed hold many politicians. The techno-optimists among us maintain about a variety of issues through mainstream media they have that the revolution in ICTs (Information and Communication become noticeably more critical of decisions made on their behalf. Technologies) brought about by the Internet and wireless com- Over the last decade ICTs have massively increased the amount munications will redress the balance. I must admit to sharing of information available and the speed at which it is transmit- some of this optimism, but would however countenance caution ted. Yet, despite technological advances, there are still relatively in that technological approaches need to be carefully grounded in few opportunities for the general public to actively participate in a good understanding of the issues involved and an application decision making. Current modes of participation, including vot-

URISA Journal • Carver 61 inequalities of access to information and participatory mechanisms (Slocum and Thomas-Slayter, 1995). For example, common practice in traditional methods of public participation involves the public, or at least those with a particular interest, in attend- ing planning meetings. These often take place in an atmosphere of confrontation that can discourage participation by an often less vocal majority resulting in such meetings being dominated by vocal individuals whose views may not necessarily represent the wider opinions of local people. Planning meetings are often held during the evening at specific times, limiting the numbers of people who can attend. Access to relevant information is also limited, meaning that fully informed decision makers are clearly at an advantage when compared to a partially informed public. Such inequalities of access, both to information and the meetings themselves, severely limit the level of empowerment gained through participation in such meetings.

New methods of participation Under traditional forms of participation, public involvement in decision making is only a few rungs up the public participation Figure 1. The public participation ladder ladder and extends little beyond the “right to object”, though (After Arnstein, 1969; Wiedemann and Femers, 1993) the level of placation varies from country to country. Several methods for encouraging more active participation have been developed. These include the use of opinion surveys, referendum, focus groups, deliberative polls, citizens’ panels and citizens’ juries ing in elections, protest activities, lobbying and public meetings, (Fishkin and Luskin, 1999; Finney, 2000; Petts, 2001; Thompson appear over-stretched and may not fully meet the democratic and Hoggett, 2001). ideals and requirements of the 21st Century (Bryan et al., 1998; An interesting example of geographically based public partici- Friedmann, 1995). pation is “Planning For Real” (PFR). This is an idea developed and patented by the Neighbourhood Initiatives Foundation (NIF), as The participation ladder a means of involving local people more closely in local environ- A useful analogy when considering levels of public participation mental planning problems and decision-making (Gibson, 1996). is that of a ladder; the base of which represents zero opportu- NIF is a charity whose main aim is to maximise the participation nity to participate with successively higher rungs representing of local people in decisions that affect their neighbourhoods and increased levels of participation and greater public empowerment quality of life. The founding director, Dr Tony Gibson, devised in the decision making process. The top of the ladder represents PFR in the 1970s as a technique of active participation based full public control and responsibility for the final decision. The around interaction with large-scale maps or physical models of the participation ladder is shown in figure 1. This public participa- area of interest. NIF has continued to develop and adapt this pri- tion ladder is used by Arnstein (1969) to describe the transferral mary tool to meet both local and strategic consultation needs and of political power from traditional power-holders having power as an essential process in community development programmes. over citizens, to citizens having the power to achieve their own The PFR approach has been adopted as method of improving requirements. Wiedemann and Femers (1993) adapt this theory participation in developing countries. For example, Integrated to their consideration of environmental decisions about hazard- Approaches to Participatory Development (IAPAD) has devel- ous waste management. oped similar tools for use in integrating the views of local people into conservation and development projects. Three-dimensional topographic models are used as a focal point in merging people’s Empowerment local knowledge with traditional spatial information (Rambaldi Empowerment is the process by which stakeholders identify and and Callosa-Tarr, 2001). shape their lives and the society in which they live through ac- In contrast to more traditional or non-digital methods, new cess to knowledge, political processes and financial, social and forms of participation based around ICTs are beginning to evolve. natural resources (Arnstein, 1969). Participation alone cannot, These include online discussion, web surveys and online decision however, guarantee socially just development since the process support systems. Together these form what might be termed the of empowerment through participation can be undermined by e-participation ladder as shown in Figure 2. the motives of the traditional power holder, power relations, and

62 URISA Journal • Vol. 15, APA I • 2003 URISA Journal • Carver 63 The bottom rung of this participation ladder represents online delivery of public services such as payment of rates and taxes, applications for licences and access to government information. Local government web sites and those of various national government departments are good examples. Selected examples from the UK include Brent and Wandsworth. Here the flow of information is essentially one-way; from server to client, and does not take full advantage of the possibilities for two-way communication. Fur- ther up the e-participation ladder, the communication becomes bi-directional making participation more interactive through the sharing of information, ideas and feedback. One of the earliest attempts at online dialogue was by MN-Politics email forum set up in 1995 in Minnesota. An example of a long-running web site dedicated to online discussion of ideas surrounding digital democracy is UKCOD (UK Citizen’s Online Democracy) that provides access to a range of web services relating to electronic democracy and runs its own forums to discuss political issues affecting the United Kingdom as well as access to research, best Figure 2. The e-participation ladder practice and training in electronic democracy. The use of the (After Smyth, 2001) web as a delivery mechanism for opinion surveys is also growing (Tsagarousiano et al., 1998). Bulmer and Coleman (2001) go so that society is composed of different groups each with different far as making the case for a “Civic Commons in Cyberspace” to world-views (Slovic and Peters, 1998). For instance, research on create “an open-ended institutionally backed extension of people’s nuclear power shows that pro-nuclear respondents see economic opportunities to make contributions to public policy on those aspects as most salient, while the anti group see accidents and matters that specially concern them” (p.5). consequences for the environment as most salient. Thus differ- Online participation promises a number of improvements. ent stakeholders may hold different world-views leading to focus Firstly, participation is not restricted by geographical location. on different aspects of the available information. This leads to Access to the relevant information and ideas of other stakehold- different perceptions and attitudes to the decision problem and ers is available from any location that has Internet access. This clearly affects the ways in which different people participate in the information is also available at any time of the day thus avoiding decision-making process. Cultural theory suggests there are four the problems associated with holding meetings in the evenings. stakeholder groups, each having a distinctive attitude (Douglas, The concept of “24/7” access (i.e. 24 hours a day, 7 days a week) 1992). These biases reflect the individual’s favoured institutional opens up opportunities for more people to participate in public arrangements, and predispose people as to their attitude to risk, consultations. When compared with the traditional method of the fairness of decisions made on behalf of society, and who to making a point verbally in front of a group of relative strangers, blame when things go wrong. Though used by Douglas (1992) an Internet-based system also allows people to make comments to describe attitudes to risk, the four types are adapted here for and express their views in a relatively anonymous and (usually) decision making in general: non-confrontational manner. Thus ICTs are seen as breaking  Individualist/Entrepreneurs - decisions present opportunity, down key barriers to participation, principally those concerning save those that threaten freedom of choice and action; ‘the accessibility and socio-psychological factors. While addressing the market’ will provide control, so there is no need for other physical barriers to participation by online access is important, un- kinds of control; derstanding the social aspects of participation is perhaps more so;  Hierarchists - believe technological and environmental particularly how people perceive decision problems and respond decisions should be left to experts; to them as individuals and as members of social groups.  Fatalists - feel that they have very little control over decisions that affect them, and accept whatever decisions are made on Some thoughts on individual and group decision- their behalf; and  Egalitarians - fear risks to the environment, the collective making behaviour good and future generations, and believe that power and In recent years researchers have recognised that to understand how influence should be spread more evenly within society. the public respond to and participate in a decision problem it is Recently a further type has been added that perhaps better necessary to focus on the social and cultural factors that govern this describes those of us seeking to employ ICTs and GI to process. Public perceptions and attitudes are shaped by the “world further our egalitarian objectives, that of: views” shared by the groups to which individuals belong, such as  Technological Enthusiast - who show a commitment to the work, social and cultural groupings (Dake, 1991). This recognises development of a high-technology society.

62 URISA Journal • Vol. 15, APA I • 2003 URISA Journal • Carver 63 Groups holding these different views will often come into GIS as facilitator conflict. For instance, we might normally expect to find individu- Although it is not uncommon for new technologies such as GIS alists and egalitarians in conflict about nuclear issues. to play an important part in the decision making process, these Although cultural theory provides a plausible typology, the tools are, in the main, beyond the reach of ordinary citizens with research evaluating these ideas is more ambiguous. It has been an interest in a particular decision problem. This is true in both shown individuals may not perceive and act consistently within in a material and a cognitive sense, since GIS and spatial data are one type in all decision-making domains. Rather, it is likely that expensive and require high levels of training for competent use. they may show characteristics of one type in one situation and In its traditional mode of employment, behind closed doors and another type in a different situation (Dake, 1991). Groups, on the operated by trained decision makers using ‘restricted’ datasets, other hand, may be formed specifically to reflect a particular view. GIS actually works against participation and empowerment. On The group norms that determine group perceptions and actions these grounds GIS has often been criticised as being an elitist tech- may therefore be more stable and more reliably categorised. nology and one that merely enhances existing power structures From the standpoint of this theory, the search for a single, (Pickles, 1995). Making GIS and appropriate datasets available universal metric for decision-making behaviour may be futile, to the public over the Internet, however, provides at least the since the perception of the decision problem and responses to it potential to redress this situation by placing all stakeholders on varies across groups and between individuals as a function of social an equal footing. This may ultimately help move public participa- and cultural factors. Nonetheless, a better understanding of indi- tion further up the participation ladder past the rung of restricted vidual and group decision behaviour is clearly needed, especially participation (Figure 1). when complicated by the addition of geographical space. Access to GIS alone does not, however, provide the public user with a satisfactory means of active participation in the The role of space decision process. GIS is far too complex a technology to allow Of the scientific issues involved, the role geographical space in effective use by the non-specialist with little or no previous train- focusing interest and shaping our response to decision problems ing or experience in this field. The following quote from Mark is central to this discussion paper. The effects of space, place Monmonier’s essay on “Ridicule as a weapon against GIS-based and locality are important in determining who is interested in a siting studies” highlights the problem very effectively: decision problem and why. People local to a particular problem or issue will, by the very virtue of their geographical position, be “…in the same way that ridicule can undermine an incomplete or otherwise flawed siting study, project opponents armed with a (in the main) interested enough to get involved or at least express GIS but lacking the savvy to use the system appropriately become a considered point of view if polled. As scale increases, a smaller vulnerable to sarcastic attacks from site advocates and sceptical proportion of the population affected will be interested enough to journalists.” seek involvement, such that at national scales the proportion of the (Monmonier, 1996) interested population is pitifully small, even though the absolute numbers may be quite large (Kingston et al., 2000). This is true Putting GIS on the Internet therefore does not in its own even for those issues that at first glance it would seem everyone right, constitute an effective participatory decision support solu- ought to hold a vested interest, like where to dispose of nuclear tion, if only because it is such a complex beast and the data difficult waste for example. Only when locational decisions are made about to interpret. Intelligent interfaces to specific problem areas perhaps national issues does the problem then become “local” and so the needed to allow effective interaction between individuals and the politics of scale kick in again to create an explosion of local inter- computer. Such interfaces would need to be intelligent enough to est. NIMBY (Not In My Back Yard) is a much-maligned public recognise the socio-cultural and educational background of the reaction to “invasive” siting decisions - a kind of geographical user and adapt themselves to their requirements accordingly. In version of personal space - but it does demonstrate the parochial addition, GIS-based decision tools need to be exploratory rather nature of public interest in decision-making very well, particularly than definitive. People need the space and freedom to ‘explore’ the when concerned with controversial facilities such as those for decision problem and create personal constructs (digital, spatial nuclear waste (Sjöberg and DrottzSjöberg, 2001). or otherwise) that represent their own outlook (vision) or opinion The role of participatory GIS is to help minimise conflict and on a decision problem. GIS-based decision tools should provide arrive at decisions that are acceptable to the majority of stakehold- the means by which stakeholders can explore a decision problem ers through consensus building approaches based on awareness using existing information, experiment with possible solutions, of the spatial implications of a decision problem. Participatory view other people’s ideas, formulate their own views, and share GIS is well placed as a technology to explore phenomena such as these with the wider community. NIMBY-ism and perhaps offer up some solutions. Developments in participatory GIS There has been a blossoming of interest in public use of GIS in recent years. A recent rash of papers and research initiatives are indicative of something of a paradigm shift in the application of

64 URISA Journal • Vol. 15, APA I • 2003 URISA Journal • Carver 65 GIS technology, yet the ethical and epistemological arguments to proprietary GIS packages, but there are still comparatively have been aired in public for some time. The long running volley few instances of real-life usage of this technology within public of articles and editorials between Stan Openshaw, Peter Taylor participation exercises though a few notable exceptions do exist and others are an excellent and entertaining example of this de- (e.g. Shiffer, 1995; Jordan, 1999; Kingston et al., 2000). The bate between the two camps of techno-positivist GIS-ers on the reasons for this can only be guessed at, but it is clear that they one hand and GIS-hating social theorists on the other (Taylor, are likely to include political ‘difficulties’, lack of resources and 1990; Taylor and Overton; 1991; Openshaw, 1991; 1992; Pickles, expertise, and the fact that the public at large are, perhaps, just 1995; Openshaw, 1996; 1997). Despite this rocky start, social not ready for it yet. science and GIS are nevertheless coming together to collaborate on participatory approaches and the societal implications in and Participation and GI – a SWOT on GIS. The US National Center for Geographic Information and analysis It is clear from the above discussion and the amount of research Analysis (NCGIA) has run two related initiatives; Collaborative effort being expended around the world on participation, GIS Spatial Decision Making (I-17) and The Social Implications of and related topics that there is considerable enthusiasm about How People, Space & Environment are Represented in GIS the potential for participatory GIS of one form or another. It (I-19); known simply as “GIS and Society”. The latter focused may be useful at this juncture to analyse its wider potential in research efforts in two key areas concerning the inter-relationship the hope that this may provide a framework for identifying key between GIS and society: how the spread of this technology is areas for further research. A SWOT analysis is given as a means affecting the political, economic, legal, and institutional structures of organising some of the issues and factors promoting or acting of society; and how societal process affect the form taken by the against participatory approaches. technology itself (NCGIA, 1993). The emphasis on PPGIS itself comes about from a merging of the re-evaluation of the social implications of GIS by the GIS community and its critics, and Strengths existing lines of research into public participation and decision- People, on the whole, want to be better informed about issues making, principally from the planning field. The recent Varenius that affect them. If those issues have a strong spatial component, initiative on “Empowerment, Marginalisation and Public Partici- then GIS or its derivatives would seem natural contenders for pation GIS” clearly focuses the issues discussed here and identifies conveying that information. a broad range of issues of relevance to community representations From an institutional perspective, GIS has matured as a and PPGIS. These include: technology and is now universally accepted as the best method  The implications of map-based models for community of handling and analysing spatial data. This level of technological representations; acceptance is not however universal, a fact that is picked up later  Possible distortion of grass-root perspectives by GIS; in considering issues of trust and understanding. Nonetheless,  Methods of representing the “community knowledge base” most decision-making organisations have adopted GIS and so within GIS; ought to be in a good position to extend their operations through  Impacts of inequalities of access to GIS technology and participatory approaches using GI as a focus for and facilitator of data; discussions between stakeholder groups.  Information needs of community groups and role of GIS in Huge volumes traditional spatial data are available world- meeting these; and wide at a variety of scales and resolutions and covering a range of  Empowerment in the decision making process through the themes. These form the ‘back drop’ of most participatory GIS, use of PPGIS. but lack local knowledge and community perspectives. In this context, one of the strengths of participatory approaches has to Up until the formation of AGILE in 1998 research on these be their aim (successful or otherwise) of incorporating local com- topics in Europe seems to have lacked a central focus although a munity level perspectives on a particular decision problem that number of early projects within the various units of the Regional may lead to different solutions than might otherwise have been Research Laboratories have addressed pertinent issues (e.g. Carver reached using purely traditional forms of data. The community et al., 1992) and many individual papers have appeared in the itself needs to be regarded as a form of database, unconventional literature (e.g. Doyle et al., 1998). AGILE has now identified GI in the IT sense, but wholly understandable from a social science policy and society as one of five research challenges including is- perspective. Local people usually know their local area better than sues of access to information and organisational change. Despite anyone else and so can reasonably be expected to provide detailed these initiatives in both North America and Europe and the work insights into local phenomena that are not normally available via of many individuals, there does still seem to be a significant gap ordinary GI datasets. Incorporation of local knowledge is clearly a between the experimental and the practical application of par- major strength of participatory approaches and may go some way ticipatory GIS. Sure, the number of mapping-based web sites has towards the Geographical Knowledge Systems (GKS) proposed exploded in recent years helped along by web mapping add-ons by Taylor (1990).

64 URISA Journal • Vol. 15, APA I • 2003 URISA Journal • Carver 65 From the outset it is clear that many decision problems, actually have changed (Althaus and Tewkesbury, 2000; Bulmer particularly those concerning the environment and the way we and Coleman, 2001). occupy it and utilise it, have a strong spatial component. Includ- This discussion leads on to issues surrounding the public ing space as a central element or framework in addressing such understanding of science and specific decision problems. Those decisions is highly beneficial in organising both our view of the in positions of decision-making power have often argued that problem and our engagement with it. Despite its limitations, because they have access to all the relevant information, have the map is the best way of organising spatial information and is been extensively trained in the art of decision making, and have a the best available tool for interacting with it. One of the main detailed understanding of the decision problem in question, then strengths of PPGIS therefore lies in its ability to handle spatial they are best placed to make the decisions. The public on the other information and communicate this to interested stakeholders, and hand may well have limited knowledge of the decision problem, in turn, accept, organise and reflect inputs (spatial or otherwise) partial access to the facts and little or no training. The profession- that users provide during the participation process. als could be right in their assertion that they are the best people Finally, anything that increases public access to information to make the final decision, but that does not necessarily mean and active participation in the decision making process should that the public view should hold less weight or even be ignored. be seen as an improvement over existing public/decision maker Nevertheless, there are significant problems in incorporating power relationships. In this instance current experiments in public opinion into the decision making process. For example, participatory GIS are steps in the right direction; that of citizen the public at large often conceptualise a decision problem in very empowerment through greater involvement and openness and different ways, using simplistic thinking routines for judging risk accountability on behalf of decision makers. This process has and uncertainty, that in turn lead to predictable errors and bias recently received a fillip from the United Nations in the form in their judgement (Bazerman, 1997; Kahneman et al., 1982). A of the Aarhus Convention, adopted in June 1998 and entering useful illustration of this is how people consistently overestimate into force in October 2001. This international law on access to dramatic causes of death such as floods and homicide, while more information, public participation in decision making and access mundane ones such as diabetes are underestimated (Lichtenstein to justice in environmental matters seeks to strengthen the role of et al., 1982). members of the public and environmental organisations in pro- Despite the inclusion of the spatial dimension being advanta- tecting and improving the environment for the benefit of future geous for our investigation of decision problems, the role of geog- generations. Through its recognition of citizen’s environmental raphy in participatory decision-making is not clearly understood. rights to information, participation and justice, it aims to promote The NIMBY syndrome demonstrates that space and locality are greater accountability and transparency in environmental matters important for controversial siting decisions such as for nuclear (UNECE, 2001). While the convention has been ratified mainly waste disposal. This is further complicated by the “sense of place” by eastern countries, it is expected that western nations will ratify often associated with particular localities. GIS data models may the convention in due course. be able to cope with the quantitative and deterministic aspects of space, spatial scale and distance, but may not be best suited to Weaknesses representing the more qualitative and perceptual effects of place Ease of access to relevant information and the tools with which since this is more of a personal construct. Traditional GIS data to use it are oft quoted strengths of ICT-based public participa- models and representations of space may therefore distort local tion. Equality of access creates a level playing field from which all and individual/group world-views. Forcing community level data stakeholders can operate and debate the issues on equal terms. This into Cartesian coordinates and vector/raster data models may may be somewhat of a utopian dream. Recent initiatives in many also not be appropriate because of corresponding distortions of countries have centred on ensuring equal access to information space and distance in an individual’s mental map of their locality via the Internet. The reasoning behind this is the clear, if rather (Aitken and Michel, 1995). simplistic notion, that if everyone has access, then everyone has the opportunity to be equally well informed. Better access to the Opportunities Internet will help, but it is maintained here that there will always The medium and long-term opportunities for participatory ap- be a significant proportion of the population who do not have the proaches and GI are significant. PPGIS should not be viewed as appropriate training or intelligence with which to use it effectively. a replacement for more traditional forms of participation in the It is these people - the ‘Information underclass’ - who always have, democratic process rather its role is seen more as augmenting and and maybe always will be, excluded from the decision loop by the supporting stakeholder dialogue and public involvement in deci- so called ‘Digital Divide’ (Carver et al., 2000; Oden and Lentz, sion making. As such PPGIS is a form of Computer Supported 2001). Although participatory approaches are firmly rooted in Real Life (CSRL). This should be viewed as an opportunity for an ethos of social inclusion, it is clear that the knife cuts both increasing involvement and accountability. ways; the majority may well be better empowered, but a small Greater accountability in decisions made is seen as an yet significant minority will become increasingly marginalized by important opportunity for participatory GIS. At present many comparison although their position in absolute terms may not decisions are still made behind closed doors, using restricted in-

66 URISA Journal • Vol. 15, APA I • 2003 URISA Journal • Carver 67 formation and with only minimal public input. As the kind of service, trust and the public good at its heart. Trust is essential participatory decision making procedures described here become to any participatory process. Why would anyone take the effort more common place, then the decision makers themselves will to get involved if they believed that their input would simply be become more accountable for their actions on the basis that the ignored or, worse, misused and distorted? The same principle decision making process is more transparent and based on freer applies to the data being used, much of which is collected by or dialogue between stakeholders. Perhaps the real opportunity lies on behalf of the state, and therefore to some people an object of in making the public itself more accountable for decisions made some suspicion (Curry, 1995). The potential for misinformation by giving over greater decision-making responsibility to individual within participatory systems is quite high, and people need to be stakeholders and stakeholder groups. able trust the data and tools they are given if they are to trust the Greater participation implies greater levels of social inclusion. whole participation process. Reasons for why this might not always be the case have already The issue of (mis)trust in the wider democratic process is been given under the discussion of potential weaknesses, and beyond the scope of this paper, but it remains an important issue. these apply just as much here. If we subscribe to the position of The relative positions of power in respect to different stakeholder ‘technological enthusiasm’ then developments in the information groups within the decision making process is also an issue that society such as Internet access for all, digital television, mobile threatens the potential of PPGIS. Issues surrounding relative wireless communications, etc. will mean greater accessibility of equity within the decision-making process need to be addressed. information for all, implying greater opportunity and higher It is perhaps unrealistic to imagine that the views of a single levels of social inclusion. individual could ever hold as much power as that of a govern- From a technological standpoint, and assuming access prob- ment minister or elected representative, but taken together the lems are solved, the greatest opportunities for PPGIS perhaps lie collective views of individuals and community groups need to in the development of better interfaces and data models. Some of be taken seriously. the main criticisms of GIS are levelled at impenetrable interfaces We might not always be able to trust the responses gained and data models that are unsuited to public access and handling through the employment of participatory systems either. There qualitative interpretations of space and personal ideas of place and are significant and, to date, unresolved problems concerning the locality (Aitken and Michel, 1995). It may be that from a strict representative nature of the data generated and the validation of GIS perspective such data models are not possible, but hybrids or individual responses. Internet-based systems are especially vul- pseudo-models might feasibly be developed to bridge the gap be- nerable to being hijacked by determined individuals or groups tween Cartesian coordinate space and mental/personal space. This intent on pushing a particular view. It is also difficult to distin- recognises that there are many different types of space - from the guish between genuine responses and those submitted by whim physical to the metaphysical – and as many personal interpreta- or fancy. tions of these as there are people. GIS cannot possibly incorporate Antipathy in the very people it aims to empower may well representations of all of these within its necessarily reductionist turn out to be the greatest threat facing participatory GIS. The view of the world, but opportunities present themselves for GIS general public may well be interested in a particular decision to at least provide something of a framework (or interface) by problem but when faced with the opportunity to take part, often which we can express ourselves and our opinions/feelings about adopt a antipathetic position. “Why should I bother? I can’t make particular issues and decision problems. This gives rise to the idea a difference.” This fits the “fatalistic” or “hierarchical” category of of “territory as interface” where the map (and derivatives of it) is decision-maker described above. It may sound trite, but “while central to stakeholder interaction and dialogue – an environment you can lead a horse to water, you can’t always make it drink”. that be explored, an ethereal space in which ideas can be tested Regardless of the amount of hype and incentive, people might and decisions formulated. just not be interested enough or have the time and energy to participate (Davies, 2001). This is especially true in times of Threats economic recession such as facing the world today, since people If we were blindly optimistic about participatory GIS we would tend to be more concerned with day-to-day economic necessities probably deny the existence of any serious threats to its develop- in times of hardship than with the luxuries of playing politics and ment and further adoption. It may well be that PPGIS or similar long-term planning. approaches might actually do much to help foster a new era of greater trust through more open and accountable decision mak- The Way forward? ing structures. However, some quite significant and pernicious The industry standard strategic response to any SWOT analysis threats do exist and we ignore these at our peril. is simply to build on your strengths, address your weaknesses, The current political climate and the power relations that exploit the opportunities and neutralise any threats. The same have built up around it have bred a certain cynicism and mis- can be said of participatory GIS. We can perhaps use the SWOT trust among the general electorate in the democratic process. analysis as a frame of reference in drawing up a research agenda Politics is often seen as corrupt and obsessed with power and that supplements those arising from the I-19 and the Varenius money rather than as a democratic institution with the aims of Projects.

66 URISA Journal • Vol. 15, APA I • 2003 URISA Journal • Carver 67 A new research agenda GIS we inevitably end up working with a range of multiples: Several research directions can be identified that are to some multiple stakeholders, multiple criteria, multiple objectives and extent ‘generic’ or non-location specific in the sense that they multiple scales, together with differential levels of access, training applicable as much in Europe and they are in the US. A draft and finance to differentials in spatial cognition, education and ‘generic’ research agenda based on the above SWOT analysis cultural background. In short, we need to consider “multiplicity” might be as follows: as a universal research agenda item. 1. Building on existing strengths: Perhaps the main route by which the GI research commu- a. Raising awareness among decision makers about the nity can take this agenda forward is through the development of potential of GI-based participation; example real-world applications of participatory GIS approaches. b. Further research into methods of incorporating local In other words, real people using real systems to address real prob- knowledge into GIS databases; lems. To do so would necessarily involve answering many of the c. Further research into methods of communicating GI other research objectives in the above list. To date there has been (and associated issues) to the lay public; and a great deal of theorising and conceptualising and deconstruction d. More practical real-world applications of participatory about the possible role of GIS in society. Perhaps now is the time to GIS to increase experience. start doing and learn by our actions. May be then, and only then, 2. Addressing weaknesses: will we start to find answers to support the theory and discover a. Increasing access to wide range of GI for use in just how far GI can help in promoting the ideals and facilitating public participation exercises (e.g. through copyright the mechanics of the new deliberative democracy. agreements for ‘local’ use); b. Further research on public perceptions of space Geographical addendum and understanding of the spatial aspects of decision Much of the above discussion, whilst recognising that space, place problems; and locality are of tremendous importance to participatory ap- c. Further research on public approaches to decision proaches in spatial decision-making, does not fully distinguish making when there is a strong spatial component; and between the local ‘truths’ that exist between different communities d. Research on how more qualitative perceptions of and the parts of the world they occupy. For instance, planning law space, place and locality may be incorporated in to differs from one country to another, while the sense of place that GIS databases and/or models. shapes the way one group of people feel about a particular loca- 3. Exploiting opportunities: tion also shapes the way in which they may approach a problem a. Investigate how participatory GIS approaches may and the decision-making process itself. augment traditional means of participation; Basic cultural differences, based on social and environmental b. Develop mechanisms for demonstrating accountability stimuli, often mean that there are fundamental differences in of decision made using public input via participatory the way the global ‘we’ may approach the same type of decision GIS; problem. Different cultures possess differing notions of hierarchy, c. Investigate possibilities of porting participatory patriarchy, honour and dependency. For example, in Asian coun- approaches to various new media such as digital tries there is a much stronger sense of hierarchy than is generally television and wireless communications; and found in North American and European countries. This manifests d. Research and develop adaptive user interfaces to enable itself in a greater deference to decision-making structures that content and methods of interaction to be customised to are based on respect for the views of ‘elders’, leaders and learned the user. ‘experts’. In other instances, local people may see the interven- 4. Neutralising threats: tion of other countries in giving aid or advice as meddling or to a. Investigate methods of increasing the transparency of accept such help as dishonourable. The short history of GIS and GI and communicating limitations to the user; and ICT applications in international aid and development is littered b. Develop methods of verification and validation that can with projects that have failed because outside organisations have be applied to responses gained through participatory not understood the basics of the local situation or heeded local GIS. opinion (Mather, 1997). The cultural dimension further complicates our understand- A research item that cuts across all four of these headings is ing of group and individual decision-making dynamics in a way the multiplicity of issues, factors and concerns that characterise that makes cross-cultural and international decisions particularly the whole topic of transforming GIS into a tool for the masses. awkward - a fact made only too obvious by recent world events. So far this paper has attempted to draw the reader’s attention While politics and religion, disputed resources and the blinkered to a range of issues from the social science of decision-making tribalism of the ‘other’ may fuel these there is often a common behaviour to the technicalities of serving GIS on the Internet. theme of geography behind many conflicts. Armed conflict is What is clear from this brief undertaking is the multi-faceted often the inevitable result in the presence of two ‘nation states’ nature of the subject matter. When developing participatory confined within one geographical area, contemporary examples

68 URISA Journal • Vol. 15, APA I • 2003 URISA Journal • Carver 69 being Northern Ireland and Israel/Palestine. GIS is beginning to it ought to be apparent that there is some common ground within be used as a tool of analysis and mediation in cases of interna- the participation and GI/ICT research agenda. It is also clear that tional conflict (Wood, 2000a). Two well-publicised examples are the geographical differences mentioned above will be responsible the use of GIS to redraw the boundaries of ethnic enclaves in the for fundamental differences in the way we apply the results of this Balkans (Bouchardy, 2000; Wood, 2000b) and map Palestinian research at a regional and local level. The key to this is perhaps and Israeli territories in the West Bank and Gaza Strip (Aronson, through developing a broad and detailed understanding of the 2000; Starr, 2000). Although this may seem at first glance to social aspects of decision making and being able to apply this be an area ripe for participatory GIS, feelings of the potential sensibly within the rapidly changing technological arena. participants are likely to be too entrenched to yield acceptable Human nature has been thousands of years in the making compromise solutions. Alternatively, the anonymity of Internet- and cultural identities have developed over hundreds of years. Our based approaches to participation and GIS communication of geographical sense of ‘belonging’ can change over generations or geographical ‘realities’ may help break the deadlock by allowing an individual lifetime, and yet the technology of GIS and ICTs is peoples’ true feelings to surface without fear of recrimination developing and changing so quickly that the social and political among neighbours and ‘allies’. dimensions are having difficulty in keeping pace. It is perhaps for Returning to Europe and North America, basic differences reason that some geographers and social scientists feel threatened between respective constitutions create the need to investigate by GIS and ICTs. Rather than stalking each other from either a range of participatory approaches; each tailored to the needs side of the epistemological fence we need to work together to of individual countries. What works well in Europe, might not realise the potential benefits of GI and ICTs, but temper any necessarily work well in North America, and visa versa. There are trend toward ‘blind’ technological positivism with a social real- also likely to be regional differences due to differences in local ism that understands the physical and metaphysical boundaries government, such as size, resources and politics, and differences of essential human nature. in the geodemographics of the target population. This is especially true in Europe where history has created a mosaic of very different About the Author political and social systems. For example, unitary authorities in the UK are considerably larger than their counterparts in France, Italy Steve Carver and Germany, while the systems of national and local government are similarly diverse. Citizen involvement in local government Corresponding Address: decisions is therefore more developed in some EU countries than School of Geography, University of Leeds, LS2 9JT, UK in others, perhaps reflecting these geographical differences and Email: [email protected] the political and social opportunities presented. This gives rise to a number of questions that we, as GI and ICT researchers, may wish to address. For example: References  Do national and local governments in small and densely populated countries associate more with participatory Aitken, S.C. and Michel, S. (1995) Who Contrives the ‘Real’ approaches? in GIS?: Geographic Information, Planning and Critical  Is there a north-south gradient in levels of participation Theory. Cartography and Geographic Information Systems. with greater opportunities for participation in northern 22(1), 17-29. countries? Althaus, S.L. and Tewkesbury, D. (2000) Patterns of Internet  How has the history of the EU and North America affected and traditional news media use in a networked community. opportunities for participation? Political Communication. 17(1), 21-45.  Does participation vary between central and marginal states, Arnstein, S.R. (1969) A ladder of citizen participation. American and between old and new members? Institute of Planners Journal, 35, 216-224.  How does this historical legacy of shifting borders in Europe Aronson, G. (2000) Land for peace: mapping the future of Pal- affect the social and cultural dimension of participatory estine. GeoEurope October 2000, 28-31. approaches? Bazerman, M. (1998) Managerial Decision Making. Wiley. 4th Edition. These differences (if they do in fact exist) make the possibil- Bouchardy, J-Y. (2000) Winning the peace in Kosovo. GeoEurope ity of developing generic approaches to participatory GIS within July 2000, 36-41. the European scene seem rather remote. Indeed, we may identify Bryan, C., Tsagarousianou, R. & Tambini, D. (1998) Electronic similar differences between US/Canadian States, for example be- democracy and the civic networking movement in context. tween California and, say, Tennessee or between native-Americans R.Tsagarousianou, D.Tambini & C.Bryan (Eds.) Cyberde- and European immigrants. mocracy: technology, cities and civic networks. Routledge, Whatever the answers to these and other questions on the London, 1-17. opportunities for and potential uptake of participatory approaches

68 URISA Journal • Vol. 15, APA I • 2003 URISA Journal • Carver 69 Bulmer, J.G. and Coleman, S. (2001) Realising Democracy Lichtenstein, S., Fischhoff, B. and Phillips, L. (1982) Calibration Online: a Civic Commons in Cyberspace. IPPR/Citizens of probabilities: The state of the art to 1980. D.Kahneman, Online Research Publication No.2, March 2001. P.Slovic, and A.Tverski (Eds.) Judgement under uncertainty: Carver, S., Myers, A. and Newson, M. (1992) The role of public Heuristics and biases. New York, Cambridge University perception in the response planning for major incidents: a Press. proposal for a GIS-based strategy. NorthEast Regional Re- Mather, C. (1997) Geographic Information Systems, mapping search Laboratory Report No.92/1. University of Newcastle and development practice. Science, technology and Develop- upon Tyne. ment 15(2-3), 291-302. Carver, S., Evans, A., Kingston, R. and Turton, I. (2000) Ac- Monmonier, M. (1996) Ridicule as a weapon against GIS- cessing Geographical Information Systems over the World based siting studies. Position paper I-19 GIS and Society Wide Web: Improving public participation in environmental meeting, NCGIA. http://www.geo.wvu.edu/i19/papers/ decision-making. Information, Infrastructure and Policy, 6, monmonier.html 157-170. National Centre For Geographical Information and Analy- Curry, M.R. (1995) GIS and the Inevitability of Ethical Incon- sis (1996) GIS and Society: Initiative 19 NCGIA http: sistency. J.Pickles (Ed.) Ground Truth: The Social Implica- //www.geo.wvu.edu/i19/ tions of Geographic Information Systems. New York: The Oden, M.D. and Lentz, R.G. (2001) Digital divide or digital Guilford Press, 68-87. opportunity in the Mississippi Delta region of the US. Dake, K. (1991) Orienting Dispositions in the Perception of Telecommunications Policy 25(5), 291-313. Risk: An analysis of contemporary World views and Cultural Openshaw, S. (1991) A View on the GIS Crisis in Geography, or, Biases. Journal of Cross-Cultural Psychology, 22, 61 - 82. Using GIS to Put Humpty-Dumpty Back Together Again, Davies, A.R. (2001) Hidden or hiding? Public perceptions of Environment and Planning A, 23, 621-628. participation in the planning system. Town Planning Review. Openshaw, S. (1992) Further Thoughts on Geography and GIS: 72(2), 193-216. A Reply, Environment and Planning A, 24, pp.463-466. Douglas, M. (1992) Risk and Blame. Routledge, London. Openshaw, S. (1996) GIS and Society: a lot of fuss about very Doyle, S. Dodge, M. and Smith, A. (1998) The potential of Web- little that matters and not enough about that which does. based mapping and virtual reality technologies for modelling Position paper I-19 GIS and Society meeting, NCGIA. http: urban environments. Computers, Environment and Urban //www.geo.wvu.edu/i19/papers/openshaw.html Systems, 22(2), 137-155. Openshaw, S. (1997) The truth about Ground Truth. Transac- Finney, C. (2000) Implementing a citizen-based deliberative pro- tions in GIS, 2(1), 7-24. cess on the Internet: the Buckinghamshire Health Authority Petts, J. (2001) Evaluating the effectiveness of deliberative pro- electronic citizen’s jury in the UK. Science and Public Policy cesses: waste management case studies. Journal of Environ- 27(1), 45-64. mental Planning and Management, 44(2), 207-226. Fishkin, J.S. and Luskin, R.C. (1999) Bring deliberation to the Pickles, J. (Ed.) (1995) Ground Truth. The Social Implications democratic dialogue. M.McCombs and A.Reynolds (Eds.) of Geographic Information Systems. Guilford Press, New The Poll with a Human Face. Lawrence Erlbaum, Hillsdale, York. NJ. Rambaldi, G. and Callosa-Tarr, J. (2001) Participatory 3-D Friedmann, J. (1995) Empowerment. Blackwell, London. Modeling: Bridging the Gap between Communities and GIS Gibson, T. (1996) The Power in our Hands: Neighbourhood- Technology. Paper presented at the International Workshop based world-shaking. Jon Carpenter Publishing, Charlbury, “Participatory Technology Development and Local Knowl- Oxfordshire. edge for Sustainable Land Use in Southeast Asia”, 6-7 June Jordan, G. (1999) A Public Participation GIS for Community 2001, Chiang Mai, Thailand. Forestry User Groups in Nepal: Putting People Before the Shiffer, M. (1995) Interactive Multimedia Planning Support: Technology. NCGIA Specialist Meeting on “Empower- Moving from Stand Alone Systems to the Web. Environment ment, Marginalization, and Public Participation GIS” http: and Planning B: Planning and Design, 22, 649-664. //www.ncgia.ucsb.edu/varenius/ppgis/papers/jordan.pdf Sjöberg, L. and DrottzSjöberg, B.M. (2001) Fairness, risk and Kahneman, D. Slovic, P. & Tverski, A (1982) Judgement under tolerance in the siting of a nuclear waste repository. Journal uncertainty: Heuristics and biases. New York, Cambridge of Risk Research. 4(1), 75-102. University Press. Slocum, R. and Thomas-Slayter, B. (1995) Participation, empow- Kingston, R., Carver, S., Evans, A. and Turton, I. (2000) Web- erment and sustainable development. R.Slocum, L.Wichart, Based Public Participation Geographical Information D.Rocheleau, and B.Thomas-Slayter (Eds.) Power, Process Systems: An Aid To Local Environmental Decision-Mak- and Participation: tools for change. Intermediate Technology ing. Computers, Environment and Urban Systems, 24(2), Publications, London. 3-9. 109-125.

70 URISA Journal • Vol. 15, APA I • 2003 URISA Journal • Carver 71 Slovic, P. and Peters, E. (1998). The Importance of World Views Tsagarousianou, R. Tambini, D. and Bryan, C. (Eds.) (1998) in Risk Perception. Risk, Decision & Policy, 3, 165 - 170. Cyberdemocracy: technology, cities and civic networks. Smyth, E. (2001) Would the Internet widen public participation? Routledge, London. Unpublished MRes Thesis, University of Leeds. United Nations Economic Commission for Europe (2001) Starr, H. (2000) Using Geographic Information Systems to re- The Aarhus Convention: Access to Information, Public vist enduring rivalries: the case of Israel. Geopolitics 5(1), Participation and Justice in Environmental Matters. http: 37-56. //www.unece.org/env/pp/ Taylor, P.J. (1990) GKS, Political Geography Quarterly, 9(3), Weidemann, I. and Femers, S. (1993) Public participation in waste 211-212. management decision-making: analysis and management of Taylor P.J, and Overton M. (1991) Further Thoughts on Geog- conflicts. Journal of Hazardous Materials 33, 355-368. raphy and GIS, Environment and Planning A, 23, 1087- Wood, W.B. (2000a) GIS as a tool for territorial negotiations. 1094. Boundary and Security Bulletin 8(3), 72-79. Thompson, S. and Hoggett, P. (2001) The emotional dynamics Wood, W.B. (2000b) Complex emergency response planning of deliberative democracy. Policy and Politics 29(3), 351- and coordination: potential GIS applications. Geopolitics 364. 5(1), 19-36.

70 URISA Journal • Vol. 15, APA I • 2003 URISA Journal • Carver 71 URISA Journal • Drew 73 Transparency – Considerations for PPGIS Research and Development

Christina H. Drew

Abstract: Notions of transparency – what it means, how to measure it, and how these may be space and/or time dependent – should be included in the Public Participation Geographic Information Science research agenda. Transparency of decisions is particularly important when the stakes are high and uncertainties are large – as is the case in complex environmental decisions. In this article, I suggest that decision transparency is a complex topic requiring multiple measures to capture its full breadth and depth. The “Decision Mapping System,” which was created for Hanford, a former plutonium production facility owned and operated by the United States Department of Energy, is described. The Decision Mapping System is an example of how to facilitate decision transparency using geographic information and Internet technology.

Introduction research that should be applied to the PPGIS research agenda Increasingly, the notion of “transparency” has appeared in aca- are highlighted. demic and popular literature – indeed, three of the four keynote papers for the “Workshop on Access and Participatory Approaches The Hanford Cleanup Context in Using Geographic Information” (Spoleto, Italy, December The Hanford site (580 square miles) is located in the southeastern 2001) mentioned “transparency” in some context. Transparency region of Washington State (Figure 1). The site includes nine is often touted as essential for democratic decision-making and retired plutonium production reactors, three inactive chemical public involvement. In the literature, transparency seems to mean separations facilities, 177 aging underground storage tanks filled that something can be “readily understood,” but a formalized with high level radioactive waste, and many other contaminated definition for transparency and ways to measure if something facilities. As a result of site operations, intentional “releases,” is transparent are only just beginning to emerge (Florini 1999, unplanned spills, and chronic leakages, there is widespread ra- Drew 2002, Drew and Nyerges 2004). In this article, I argue that dioactive and chemical contamination of soil, groundwater, and conceptions of transparency – what it means, how to measure the nearby Columbia River ecosystem. Hanford’s mission is now it, and how these may be time and/or space dependent – should entirely related to managing nuclear waste and cleaning up the be included in the Public Participation Geographic Information environment. The U.S. Department of Energy (DOE) manages Science (PPGIS) research agenda. the Hanford cleanup; however, despite spending over a billion The project discussed here is focused specifically on the trans- dollars per year for the past 10 years on the cleanup, progress has parency of complex environmental decisions. If people are meant been slow. Some of the contaminants involved are not only very to participate in a decision process, they must first understand several things, such as how decisions get made, what the technical issues are, and how to get information necessary for a decision. These needs are even more important when decisions are complex, when stakes are high (e.g., potential health risks and costs), and when uncertainties are large (e.g., technical difficulty). Such is the case at Hanford – a former plutonium production facility in the State of Washington – and the setting for my research. This article contains four major sections. First, the Hanford cleanup context is briefly described. Second, the concept of transparency is explored and the beginnings of a framework by which it might be measured are presented. Third, we describe the tool that we developed called the “Decision Mapping System,” which uses Geographic Information Science (GIS) and the Internet to promote two-way information exchange among the Hanford decision-makers, active stockholders, and the general public about the Hanford cleanup. Finally, a few key lessons from the Figure 1

URISA Journal • Drew 73 dangerous to humans, but they persist in the environment for long In addition, seven key concepts stood out in the literature periods of time (sometimes tens of thousands of years). Plus, there as having a particularly close association with transparency. The is no “easy solution” when it comes to radionuclide contamina- concepts include: clarity, accessibility, integration with other deci- tion – in many cases, we simply do not have the technology to sions, logic and rationale, accountability, truth and accuracy, and dispose of the materials “safely.” And of course, the very question openness (Figure 2). of “how safe is safe” is contested on many levels. These components of transparency, which can also be Many people are interested in learning about Hanford and thought of as goals, are further described in Table 1. The bullets getting involved in its cleanup decision processes, but it is difficult initiate a robust measurement framework for evaluating transpar- to understand what is happening there and why. For example, ency – each can be used to devise a specific question to help in complex federal and state regulations have resulted in a vast but an evaluation. For example, parts of this framework were used poorly organized collection of documents that are difficult to find, to evaluate the transparency of a document (Drew and Nyerges obtain, and read. Moreover, decisions are geographically and tech- 2004) and a Web site (Drew 2002). From this diverse list of po- nically complex, but they are often defined very narrowly, making tential transparency criteria, we conclude that transparency itself it a challenge to see how the pieces fit together across space and is a complex topic and multiple measures are needed to express its time. These difficulties boil down to a lack of transparency. full breadth and depth. Major sources influencing our framework include Lodge 1994, the U.S. DOE 1994, the IMF Working The Importance of Transparency Group 1998, Buiter 1999, the Council of the European Union 1999, Florini 1999, the IMF 1999, Issing 1999, Kern 1999, In Complex Decisions Stiglitz 1999, and Katz 1999. Fuller accounts of this literature I first became interested in transparency while working with a and how the framework was developed are available elsewhere group called the Hanford Openness Workshops (HOW) (Kern (Drew 2002, Drew and Nyerges 2004). 1998, 1999). This group of diverse stakeholders provided advice to the DOE about the task of sorting through and making “public” the millions of pages of information that were de-classified in the 1990s. The HOW participants frequently paired THE DECISION MAPPING and interchanged the concepts of transparency and openness. SYSTEM AS A TOOL TO The importance of transparency for this group led my colleagues PROMOTE TRANSPARENCY and me into a review of the environmental and decision-making Before we explore the concept of transparency, we thought about literature (Drew 2002, Drew and Nyerges 2004). how several information structures (some geographic, some not) We found that in many cases, organizations and the literature could be used to support decision processes. Much of the com- declare transparency to be essential for public decision-making in plexity of Hanford cleanup information relates to risks – what democratic societies. Specifically, transparency is praised for lead- are they, where are they, and what can/should we do about them? ing to more informed choices, permitting evaluation, strengthen- A key question for me has been: how can risk information be ing institutional credibility, and promoting efficiencies in terms structured in a way that makes it easier for people to participate of long-term cost savings, efficient resource allocations, and less in a decision process? duplication (Drew 2002, Drew and Nyerges 2004). Despite the prominence of the term, prior to our analysis, no one had proposed any performance measures or other criteria for its evaluation. In general, the literature suggested that a transparent decision was one that “allows people who are interested in a decision to understand what is being decided, why, and where.”

Figure 2 Figure 3

74 URISA Journal • Vol. 15, APA I • 2003 URISA Journal • Drew 75 Table 1: Measuring Transparency

Clear Logical and Rational • Comprehensible/Intelligible • Follows a rational defensible plan, clear to a broad • Unambiguous array of stakeholders • Easily detected • Committed to scientific credibility (sound, • Easily seen/heard dependable, leading edge) • Visible (no hidden meanings) • Processes are consistent, standardized, formalized, • Precise and simple flexible, expandable • Contains minimal jargon • Identifies clear decision points (and opportunities for involvement) Integrated • Able to track decisions and policies over time • Comprehensive (process fully laid out/full • Electronic information includes descriptive disclosure) information (metadata) so can be interpreted by all • Takes a “big picture” view - shows decision in • Uses available technologies to improve access to context to related decisions declassified or formerly classified information • Consolidated – described in a single document • Provides logical referencing system pointing Accountable users to additional information and source • Analyses subjected to independent assurances of documentation credibility (i.e., peer review) • Contains detailed table of contents and indices • Shows that activities meet goals of policies • Shows linkages between decisions and Accessible implementation; i.e. records milestones (activities), • Allows citizens to have access to important follows cost and schedule changes, provides meetings of government officials rationale for changes, etc. • Allows citizens to request and receive government • Provides rationale behind decisions documents • Reassures the public • Makes of detailed documentation and databases • Responds to stakeholders in timely fashion available • Provides adequate time for stakeholders to be • Promotes two-way access to information involved (stakeholders have access to government legislation, and government has access to Open/Involve stakeholders stakeholder values) • Allows concerned citizens to see openly into government activities Truthful and Accurate • Allows citizens to have input into government • Truthful and accurate, free from deceit decisions and rule making • Messages undiluted (information delivered • Undertakes budget preparation, execution, and effectively without altering content) reporting openly • Allows stakeholders consistent opportunities to make suggestions during decision process and to appeal decisions • Provides early notification of opportunities • Seeks wide ranging early advice on key proposals • Provides clear and coherent messages • Describes impact of public/stakeholder input • Provides user friendly interfaces • Promotes inter-institutional cooperation and coordination with the Hanford Tri-Party Agreement, internal organizations, and stakeholders

Working closely with stakeholders, we developed a prototype The DMS has been created specifically for describing deci- for a geographically based Internet information tool that we call sions in the Hanford “100 Area,” but the concepts could be the Decision Mapping System (DMS) (Figure 3). The purpose of adapted to the entire site, to other weapons production sites man- the DMS is to allow a better understanding of cleanup activities aged by the DOE, and to other types of environmental cleanup occurring at Hanford and thus to make it easier for people to decisions. The DMS includes six related information structures participate in Hanford decision processes. The DMS is designed intended to unpack the spatial, temporal, and socio-cultural to present cleanup decision information to stakeholders and solicit dimensions of a decision: their comments and feedback.

74 URISA Journal • Vol. 15, APA I • 2003 URISA Journal • Drew 75 Information Systems have long been used as a tool to integrate data but we now propose that, when combined with other information structures (such as those in the DMS), GIS can be used to integrate information. This integration can allow participants to synthesize knowledge in new ways and more easily than in the past. Similarly, Internet publication will foster greater physical accessibility to information – mainly because of the number of cross-links that are envisioned for the system. Where possible, we have simply linked to existing information, but have also organized it in several ways – spatially, temporally, and conceptually. We hope that this multi-pronged approach will make the decision information easier to understand (i.e., Figure 4 so that it is more conceptually accessible). Internet publication should also allow a broader audience to access the information, but we recognize that Internet publication is not  Decision Maps consist of several interactive Web pages that a panacea and it cannot replace the “grounded social relations” connect decision information to geographic features on a (Niles and Hanson 2001) essential to any decision process. map (spatial dimension). Decision maps are integrated An important contribution of this research is to study how with background information for geographic areas and people want complex environmental decision information to be decisions. structured so that they have both physical and conceptual ac-  Background Information provides contextual information, cess to it. Our research design has been strongly influenced by a allowing users to learn more about Hanford history and how conceptual framework called Enhanced Adaptive Structuration the Department of Energy makes decisions (socio-cultural Theory-2 (EAST-2) (Jankowski and Nyerges 2001), which pro- dimension). vides a broad outline for exploring how geographic and other in-  Decision Paths describe decision process information formation technology is used in a group decision-making process (decision steps) on a timeline (temporal dimension) and (see also Nyerges et al. 2002). The research design includes an provide direct links to online documentation. Each decision Internet survey that evaluates the ability of the DMS to provide has one decision path (Drew et al. 2002). information transparently. Concepts identified in the transpar-  The Geographic Library connects decision documents ency literature review form the basis for the survey questions, to geographic features on a map (spatial dimension). resulting in an innovative protocol to evaluate the transparency Geographic libraries can be developed for different spatial of decisions (Drew 2002). scales.  Value Trees provide rationale for the decision; ideally, broad Conclusions values and goals as well as specific objectives and criteria The issues introduced in this article are only a beginning. While are provided for both procedural and outcome aspects of a the DMS is perhaps better characterized as “public records” GIS decision (socio-cultural dimension). A particular value tree (Weiner et al. 2001) than PPGIS per se, we are excited by working could be associated with many decisions, and each decision hands-on with stakeholders to address the problem of organizing could be connected to a number of value trees. complex spatial information for broad consumption. Ideally, we  Commenting Features allow users to make remarks about the will be able to continue with a planned “build out” of the DMS decision or the DMS (socio-cultural dimension) and view into a robust database driven PPGIS. (Currently, the only images remarks from others. and graphics available in the system are static image maps – there is no capability in the system to generate images, charts, or tables The DMS is intended to provide a transparent alternative on demand.) Nevertheless, I believe that the deeper understand- to the current “public record” that documents decision infor- ing about the meaning of transparency we expect to bring will be mation. A robust public record is important for the Hanford highly applicable to PPGIS research agenda. With this in mind, context because some radioactive contaminants may persist in I’d like to make three summary points: the environment for tens of thousands of years and because the First, the participatory design model has been a key to our volume of relevant information is so vast. We believe that GIS success. Hanford has been designated as a national repository and Internet technology could be powerful tools to better man- for nuclear waste, giving it high priority as a local, regional, and age both documents and data important for the public record, national issue. The DMS was developed using a participatory but they are currently underutilized at Hanford. Geographic

76 URISA Journal • Vol. 15, APA I • 2003 URISA Journal • Drew 77 approach that incorporated several active local and regional Acknowledgements Hanford stakeholders – including representatives from the U.S. Department of Energy, Washington Ecology, the U.S. Environ- This research is supported by the Consortium for Risk Evalu- mental Protection Agency, several activist groups, and university ation with Stakeholder Participation (CRESP) through a grant researchers. National, regional, and local stakeholders will also with the U.S. Department of Energy (Award No. DE-FG26- be involved in the evaluation of the Decision Mapping System. 00NT40938). Any opinions, findings, conclusions, or recom- Preliminary reactions to the DMS have been overwhelmingly mendations expressed herein do not necessarily reflect the views positive (Drew 2001). I believe that this response directly reflects of the Department of Energy. I thank the many Hanford stake- the participatory approach used to develop the system. holders who donated time and energy to the project and to Tim Second, physical access is only a first step. Access is a major Nyerges and the anonymous reviewers for comments on earlier component of our working definition of transparency because it drafts of this article. is a major concern of Hanford stakeholders. The participants of the Hanford Openness Workshops, for example, looked closely at the notion of “access” and concluded that physical access to References documentation is only a first step. Tools to comprehend technical Buiter, W.H., 1999, Alice in Euroland. Journal of Common information contained in the documents, tools to see how docu- Market Studies, 37(2), 181-209. ments and decisions fit together, tools that allow stakeholders to Council of the European Union, 1999, Basic Texts on Trans- provide feedback on the documents and decisions, and tools that parency Concerning the Activities of the Council of the allow stakeholders to see how their contributions have influenced European Union/Council of the European Union, General decisions are needed. These concerns have been a driving force be- Secretariat, DG F, Information Policy, Transparency, Public hind the DMS and transparency research – particularly in regard Relations (Luxembourg: The European Council). to the system features that foster commenting and information Drew, C.H., 2001, Decision Mapping System and Transparency integration. However, much more work is needed to make these Project Preliminary Design Report: Combined Summary of goals attainable. Stakeholder Interviews (Seattle: Consortium for Risk Evalu- Third, place matters. The Hanford setting is extraordinarily ation with Stakeholder Participation). complex and our findings will certainly be colored by this context. Drew, C.H., 2002, The Decision Mapping System: Promoting Some argue that Hanford is an excellent testing area because it Transparency of Long-Term Environmental Decisions at often represents the worst problems – if you can make something Hanford. Ph.D. Dissertation, University of Washington, work there, you can make it work anywhere. Others suggest that Department of Geography, Seattle, June 2002. http:// the Hanford context is so complex that results are simply not nalu.geog.washington.edu/dms transferable elsewhere. Either way, our findings will only provide Drew, C.H. and T.L. Nyerges, 2004, Decision Transparency for a partial “truth” which must be evaluated in other contexts. Long-Term Stewardship: A Case Study of Soil Cleanup at the These three lessons suggest that a dialogue about what trans- Hanford 100 Area. Journal of Risk Research (in press). parency means, how it should be measured in different contexts, Drew, C.H., T.L. Nyerges, K. McCarthy, and J.A. Moore, 2002, and its relative importance compared to other needs (e.g., equity Using Decision Paths to Explore Three Environmental or efficiency) should be a core component of any research agenda Cleanup Decisions: A Cross-Case Analysis. International involving “communities” or the “public.” Journal of Environment and Pollution, 17(3), 171-201. Florini, A.M., 1999, Does the Invisible Hand Need a Transparent About the Author Glove? The Politics of Transparency. Annual World Bank Conference on Development Economics, Washington, DC. Christie Drew is a Research Scientist with the Institute for http://www.worldbank.org/research/abcde/washington_11/ Risk Analysis and Risk Communication at the University pdfs/florini.pdf of Washington in Seattle. The Decision Mapping System International Monetary Fund, 1999, Code of Good Practices on was the subject of her Ph.D. thesis in Geography. Drew Fiscal Transparency – Declaration of Principles. International has worked with the Consortium for Risk Evaluation with Monetary Fund, March 2001. http://www.imf.org/external/ Stakeholder Participation (CRESP) on U.S. Department of np/fad/trans/code.htm Energy cleanup issues since 1995. International Monetary Fund Working Group, 1998, Report of Corresponding Address: the Working Group on Transparency and Accountability Christina H. Drew, Ph.D. (Washington, DC: International Monetary Fund). University of Washington, Institute for Risk Analysis and Risk Issing O (1999). The Eurosystem: Transparent and Accountable Communication or ‘Willem in 4225 Roosevelt Way, NE Suite 100, Seattle, WA 98195 Euroland”. Journal of Common Market Studies 37(3): 503- Tel: 206-616-7413; Email: [email protected] 519.

76 URISA Journal • Vol. 15, APA I • 2003 URISA Journal • Drew 77 Jankowski, P. and T.L. Nyerges, 2001, Geographic Information Nyerges, T., P. Jankowski, and C. Drew, 2002, Data Gathering Systems for Group Decision Making: Towards a Participa- Strategies for Social-Behavioural Research about Participa- tory, Geographic Information Science (London: Taylor and tory Geographic Information System Use. International Francis). Journal of Geographic Information Science, 16(1), 1-22. Katz, E.M., 1999, Transparency in Government – How American Stiglitz, J.E., 1999, On Liberty, The Right to Know, and Public Citizens Influence Public Policy. U.S. Department of State, Discourse: The Role of Transparency in Public Life. Oxford International Information Programs. Washington, DC. http: Amnesty Lecture, The World Bank, January 27, 1999. http: //usinfo.state.gov/products/pubs/transgov/ //www.worldbank.org/html/extdr/extme/jssp012799.htm Kern, M. (Ed.), 1998, Hanford Openness Workshops October U.S. Department of Energy, 1994, DOE Public Participa- 1997 – May 1998: Final Report (Seattle: Hanford Openness tion Policy. U.S. Department of Energy, Policy: DOE P Workshops). http://www.hanford.gov/boards/openness/trac- 1210.1. Washington, DC. http://www.em.doe.gov/public/ 0818/index.html doepub.html Kern, M. (Ed.), 1999, Is Openness Working? A Progress Weiner, D., T.M. Harris, and W.J. Craig, 2001, Community Report (Seattle: Hanford Openness Workshops). http:// Participation and Geographic Information Systems. Work- www.hanford.gov/boards/openness/trac-0828/index.html shop on Access to Geographic Information and Participa- Lodge, J., 1994, Transparency and Democratic Legitimacy. Jour- tory Approaches Using Geographic Information, Spoleto, nal of Common Market Studies, 32(3), 343-368. Italy, December 6-8, 2001. http://www.spatial.maine.edu/ Niles, S. and S. Hanson, 2001, A New Era of Accessibility: Or Is It? ~onsrud/Spoleto/WeinerEtAl.pdf Workshop on Access to Geographic Information and Partici- patory Approaches Using Geographic Information, Spoleto, Italy, December 6-8, 2001. http://www.spatial.maine.edu/ ~onsrud/Spoleto/NilesHanson.pdf

78 URISA Journal • Vol. 15, APA I • 2003