Information Use in the Application of SDSS to Land Use Planning Debates

Jonathan Gottsegen
University of California, Santa Barbara

How does the presentation of spatial information facilitate or inhibit productive discourse in environmental or land use issues? This question is a specific instance of the general application of spatial decision support systems (SDSS) to multiparty decision-making, and it is additionally important with the rapid proliferation of geographic information systems (GIS) to support decision making in environmental agencies and nongovernmental environmental organizations. Answers to the question of the role of information in land use and environmental decisions have significant implications for the way spatial information systems are developed and used. This paper raises issues critical to designing multiparty decision support systems for land use and environmental policy-making and suggests methods for beginning to address these issues. It concentrates on the adoption of information in the social and often contentious environment of land use policy-making and suggests research that will lead to improvements in the design of multiparty SDSS based on a more complete under-standing of the factors influencing the adoption of information in such an environment. It derives from my experience in applying GIS to land use planning and from research on information use and decision-making in policy environments.

Planners or policy-makers who produce spatial information often presume that additional information will necessarily enlighten the dialogue so that all of the parties in a debate will arrive at a rational resolution of the con-flict. This presumption requires that actors in the policy process behave at least partly according to precepts of rational choice models of decision-making. However, environmental policies are developed through a social discursive process that often seems chaotic and irrational (Hajer, 1993). Policies tend to be the outgrowth of social interaction between opposing sides with different values and views of the issue. This interaction is often antagonistic resulting in highly charged conflict. Thus the practice of policy-making is often a process of conflict resolution and directing the interaction to productive results rather than of model building and rational analysis (Fisher and Forester, 1993).

The knowledge that is acted on by decision makers is constructed from values, experiences, affect, and meanings associated with a place as well as objective information. The composite of all of these components has been called "commonsense knowledge" (Kuipers, 1983), and it is this commonsense knowledge that underlies policy debates and decisions. It is becoming accepted among the GIS community that the effectiveness of the traditional approach of presenting scientific information in decision-making is limited by the consideration of other knowledge held by decision-makers. Rushton (1995) has described this more personal knowledge of places in locational decisions as a consideration of geographic space as compared to the solution space of locational models. Thus, even given the same objective information, the response to the information will vary among parties involved in a land use debate because of the different commonsense or geographic knowledge resulting from the involved parties' differing subjective components.

Because of the social nature of policy-making, the use of information in the policy process is often indirect (Zwart, 1991), so simply throwing addi-tional data at parties involved in the policy debate will often have no additional benefit (O'Hare, 1987). When considering information systems for resolving land use conflicts, it is essential to understand the discursive nature of the debate and the role of differing beliefs and values in influenc-ing stakeholders' response to information in the policy process. Data provid-ers and analysts must be more aware of how their potential clients respond to the data. They must also develop new models of the role of information in the policy process (e.g. Innes, 1988; Dutton and Kraemer, 1985).

Armstrong and Densham (1990) describe SDSS as systems to help decision-makers solve semistructured problems, and several articles have enumerated the necessary functions and components of SDSS (Densham and Rushton, 1988; Densham and Goodchild, 1989). However, there are several important questions that must be answered if the concepts and technology of SDSS are to extended to highly contested multiparty decision environments. Armstrong's (1993, 1994) research agenda for expanding SDSS to multiparty environments concentrates on the technical and conceptual aspects of sharing views of information although he briefly introduces the role for SDSS in conflict resolution. Other practitioners have also begun to take advantage of the dynamic nature of GIS or SDSS to assist in environmental conflict resolution (Godschalk, et al., 1992; Maguire and Boiney, 1994; Brown, et al., 1994). Maguire and Boiney (1994) make the useful distinction between decision analysis and conflict resolution. Following this division, it seems that many SDSS applications concentrate on the decision analysis component. In the area of conflict resolution, both Godschalk, et al. (1992) and Maguire and Boiney refer to concepts of principled negotiation developed by the Harvard Negotiation Project (Fisher and Ury, 1981). These include 1) separating people from the problem, 2) focusing on interests not positions, 3) inventing options for mutual gain, and 4) using objective criteria.

It is clear how SDSS can accomplish the first and fourth objectives, i.e. separating the people from the problem and using objective criteria. In addition, the third objective, i.e. inventing options for mutual gain, is the professed purpose of SDSS. However, the means for realizing a focus on interests rather than positions is not evident in the current application of SDSS. Some implementations of SDSS have applied the revealed preference technique to ascertain the interests of the debating parties. This approach presumes that 1) the survey method in terms of perceived values and results of behavior is adequate at eliciting the interests of the parties, 2) that people will behave predictably in the negotiation according to these elicited values, and 3) that information presented in the SDSS will promote a change in the values equally among all of the parties.

The first assumption may be less than obvious, but the third is the most problematic. If opposing sides of a debate bring divergent values, experiences and beliefs to a debate, can one set of information that is interpreted according to these divergent backgrounds bring consensus or even support a common foundation for debate on the issue? It is not clear that representing data or alternative practices in one constant way in a SDSS will change the usable body of knowledge on which the stakeholders in a debate base their interests.

The three assumptions underlying the common applications of SDSS limit the flexibility required for the systems to be effective in the typical land use policy environment. This flexibility should include creative but structured methods for eliciting negotiable interests of stakeholders and different representations of data that are effective in actually creating socially agreed upon knowledge for the stakeholders in the debate. For instance, Armstrong (1993) alludes to the latter in mentioning other representations of data such as "delta" maps.

The need for these aspects of flexible systems is exemplified in the experience of the New Jersey state planning process. This process included cooperation with local government entities in the development of a statewide growth management plan. This involved trading maps in an attempt to define a common database on which growth management zones could be based. However, the maps from the localities reflected only their negotiating positions. For example, the counties' maps often showed no farmland because counties did not want any farmland preservation policies applied within their jurisdictions. In addition to the noncommunicative aspect of the maps, new maps prepared by the State's GIS did not seem to expand the knowledge or willingness to negotiate of the stakeholders in the state planning debate. The maps consistently elicited the same response, i.e.,"On which side of the line is my house?" This example is useful as a caveat regarding the possible limitations that SDSS must overcome, and combined with the argument presented in this paper points to two areas of research:

First, what types of representations of data and alternative solutions of the issue are most effective in communicating information to stakeholders in the debate? These representations may be ones most consistent with the commonsense knowledge of the issues and may be more meaningful to stakeholders in a debate than cartographic representations. Delta maps, while informative, are an expansion of traditional cartographic representations; data visualiza-tion through animation is receiving attention and may be useful for SDSS. Representations could also emulate photographs or other ostensibly natural presentations. It is easy to see how these somewhat trivial examples can be integrated technically into SDSS. Other dramatically different methods of representation may be useful as well. The important question is what type of representations communicate information and expand new knowledge effectively.

Second, how can such additional representations or manipulations of the data in the database of the SDSS be used to identify the interests of the stakeholders on which productive negotiation can be based. Related to this question is the question of how alternative representations can be used to enhance the comparison and evolution of interests in the negotiation process. Rather than asking stakeholders to state their priorities explicitly, which depends on their abilities to identify these priorities (more possible for experts in the field than nonexpert members of an impacted community) and the validity of these priorities in expressing interests (it is possible that the priorities represent publicly expressed positions rather than true interests), it may be possible that data can be presented in varying ways that transcend public posturing. For example, in the New Jersey state planning process, rather than including the management areas and their policies in the discussion of the mapped data, presenting information regarding easement purchase programs and programs that have protected farmers' agricultural viability may have established the interests of compensation for the value of their land and the economic viability of their farms as bases for negotiation.

The research that I propose to undertake will address these two questions by 1) analyzing a typical land use debate to evaluate the responses to informa-tion introduced into the debate and identify the association between interests and spatial knowledge used in the debate, 2) proposing innovative representa-tion mechanisms that may be more compatible with the commonsense conceptions of issue, and 3) comparing the response to the different representations among stakeholders in the debate to determine whether some are more effective in communicating information than others and whether they can be used to illumi-nate the stakeholders' interests in the debate. Clearly there are numerous ways to accomplish these tasks, and space restrictions limit the detail of my descriptions here. The first task will entail a content analysis of hearing testimony and other dialogue surrounding the issue. The second task will extend current research in spatial data models and visualization, and the third will involve experiments with subjects involved in the debate. The results will be a clearer understanding of the relationship between the use of spatial information and interests in a debate and an expanded set of methods of representing data to promote greater use of information. These methods can then be included in the suite of tools associated with multiparty SDSS.


Armstrong, M. P. 1993. Perspectives on the Development of Group Decision Support Systems for Locational Problem-Solving. Geographical Systems, 1:69-82.

Armstrong, M. P. 1994. Requirements for the Development of GIS-Based Group Decision-Support Systems. Journal of the American Society for Information Science, 45:669-677.

Armstrong, M. P. and Densham, P. J. 1990. Database Organization Strategies for Spatial Decision Support Systems. International Journal of Geographical Information Systems, 4:3-21.

Brown, S., et al. 1994. Linking Multiple Accounts with GIS as Decision Support System to Resolve Forestry/Wildlife Conflicts. Journal of Environmen-al Management, 42:349-364.

Densham, P. and Goodchild, M. 1989. Spatial Decision Support Systems: a Research Agenda. Proceedings, GIS/LIS `89. 707-716.

Densham, P. and Rushton, G. 1988. Decision Support Systems for Locational Planning in Behavioural Modelling in Geography in Planning, R. Golledge and H. Timmermans (eds). pp. 56-90.

Dutton, W. and Kraemer, K. 1985. Modeling as Negotiating: The Political Dynamics of Computer Models in the Policy Process. Norwood: Ablex Publishing Corporation.

Fischer, F. and Forester, J. 1993. Editors' Introduction in The Argumentative Turn in Policy Analysis and Planning. F. Fisher and J. Forester (eds.). Durham: Duke University Press. pp. 1 20.

Fisher, R. and Ury, W. 1981. Getting to Yes. New York: Penguin Books.

Godshcalk, D. R., et al. 1992. Using GIS for Computer-Assisted Dispute Resolution. Photogrammetric Engineering and Remote Sensing, 58:1209-1212.

Hajer, M. 1993. Discourse Coalitions and the Institutionalization of Practice: The Case of Acid Rain in Great Britain in The Argumentative Turn in Policy Analysis and Planning. F. Fisher and J. Forester (eds.). Durham: Duke University Press. pp. 43 76.

Innes, J. 1988. Effects of Data Requirements on Planning: Case Studies of Environmental Impact Assessment and Community Development Block Grants. Computers, Environment and Urban Systems, 12:7-88.

Kuipers, B. 1983. The Cognitive Map: Could It Have Been Any Other Way? In Spatial Orientation: Theory, Research, and Application. H. Pick and L. Acredolo (eds.). New York: Plenum.

Maguire, L. A. and Boiney, L. G. 1994. Resolving Environmental Disputes: a Framework Incorporating Decision Analysis and Dispute Resolution Techniques. Journal of Environmental Management, 42:31-48.

O'Hare, M. 1987. Improving the Use of Information in Environmental Decision Making. Environmental Impact Assessment Review, 1:229 250. Reprinted in Resolving Locational Conflict. R. Lake (ed.). New Brunswick: Center for Urban Policy Research. pp. 209-231.

Rushton, G. 1995. Personal communication.

Zwart, P. 1991. Some Indicators to Measure the Impact of LIS in Decision Making. Proceedings of the Annual Conference of the Urban and Regional Information Systems Association, 4:77-89.

Biography for Jonathan Gottsegen

Jonathan Gottsegen is currently pursuing his Ph.D. in Geography at the University of California, Santa Barbara (UCSB). He is affiliated with the NCGIA as a graduate student researcher and has worked on several projects for the NCGIA including a conceptual data model for Intelligent Transportation Systems, the Global Demography Project and the Alexandria Digital Library. He received a Master's Degree in Regional Planning in 1986 at the University of Pennsylvania where he studied GIS and its use in regional planning. After completing his Master's Degree, Jonathan worked for five years with the Office of State Planning in New Jersey implementing a GIS to support statewide growth management. During his tenure with the State of New Jersey and between his work there and entering the Ph.D. program at UCSB, he taught courses in GIS at the university level and to professional planners and consulted for municipali-ties interested in implementing GIS.

Jonathan's Ph.D. research is in the area of the use of spatial information in the development of land use policy. The ultimate objective of his research is to enhance to effective use of GIS and its data in policy-making by improving the tools that GIS offers.