There is hope for urban America, but only if we are willing to give more power to the people with the biggest stake in its success: neighborhood and community groups. Information can provide that power and information can be derived from data and GIS technology. The problem is that community groups don't have access to any of these resources. Information providers have sprung up as middle-men across the country, but they are finding the problem runs deeper: community groups do not desire information, they desire responses to immediate problems. Significant work is required both to get better information to these groups and to help groups take better advantage of this information.
Community groups are a critical part of American democracy. They provide a conduit between individual citizens and the external bodies that administer and control them. The conduit passes information between the two bodies, but too often operates inefficiently because the information is incomplete. Several years ago I contacted 31 major cities in North America to learn what current neighborhood-level data is distributed to citizens. The results were appalling, at least in the U.S. where the Census is taken only once per decade. Only crime and health data were regularly available. Total housing unit counts were available annually in a majority of cities. Most population, housing, and economic data were never summarized and distributed for subcity levels between vensus years--despite the fact that most of the raw data for such summaries exists as part of normal municipal operations.
Even census data is becoming less accessible to community groups as the Bureau looks to reduce expenditures. Recent announcements from the Census Bureau indicate that many useful reports will no longer be published, henceforth available in machine-readable form only. Neighborhood summaries will be published no more with the demise of the UDAP (User Defined Area Program); the rationale is that programmers can quickly build their own summaries from block and block-group data. All of these moves assume that all potential users of census data have high levels of technical expertise. This is certainly not the case with community groups, especially those serving low income areas.
In a growing number of places around the country the need for technical assistance is being provided by intermediaries. If an information revolution is to come, these are the quartermasters. They are a ragtag collection of academics, nonprofit groups, public libraries, and government offices. They are taking advantage of more readily available data and technology to supply the troops with the information needed to develop local programs and get the resources necessary for community vitality. Typically these groups collect and process data from the Census, local government, and agency records. Because all problems are local in nature, they often work with small area data or geocoded records of individual events. These data must be summarized, mapped, analyzed to show the rates within various distances, etc. All of these activities use GIS technology.
To learn more about these groups, I convened a day-long special session at the URISA 1994 conference where eleven premier data providers discussed their successes and frustrations. The successes were significant enough to convince anyone that providing information to community groups is a worthwhile activity. A bilingual Head Start program was located in an area with a concentration of Puerto Rican residents in Milwaukee, a large grant helped a neighborhood group restore a local theater in Pittsburgh, and so on.
Each frustration identified at this session is a potential area for new development and research. One obvious shortfall is the lack of information providers. The NCGIA is revising its GIS curriculum. One of the most useful GIS courses offered at the University of Minnesota--useful to both students and the community--requires students to tackle real world problems. We solicit projects from the community, assign one as a group project while the students learn the potential of the technology, then offer the other submissions as options from which the students can choose their required individual project. Students from this class have gone on to work as volunteers and paid workers with community groups across the state, so this training can provide a long-term solution by upgrading the capability of community groups to access and analyze data to their own ends.
A second frustration is lack of interest in data on the part of community groups. Crime might an important issue and the neighborhood organization would like it stopped, but they don't think about asking for a map of crime incidents--much less a map showing buildings owned by absentee landlords. On the other hand, the appetite for data is enormous. Everybody wants all the data he or she can get. Obviously, these groups are lacking in knowledge and sophistication about what data is relevant. The existing solution is for the data providers to work with individual groups, providing direct answers to initially stated needs, but suggesting further exploration and analysis. This is time-consuming and works only for those community groups with enough interest to visit a data provider. A more efficient solution is to create a "reverse directory" of data and analyses, something that shows what products might be relevant to particular community problems. A second approach is to share solutions among community groups. This approach is used by the Community Information Exchange, a national group based in Washington DC, but their list of experiences is limited. A better approach might build on local experiences, shared electronically, perhaps over local FreeNet. Such an effort is underway in the Twin Cities. The biggest problem with this approach is finding a way to entice those who have developed data-based solutions to take the time to document their experience.
The third frustration is getting useful data from organizations. NCGIA Initiative 9, Sharing Geographic Information, addressed this issue and I was pleased to be a part of that effort. Thankfully the federal government is moving toward putting information on web servers and a number of state and local government agencies are doing the same. This is good and it would be a valuable research project to see what factors and actors are pushing this development. For example, did neighborhood activists in city X push the city council to instruct city departments to publish data this way or did the initiative come from city staff. The problem with web-published data is that it predefines what type of analysis is possible. Better that community groups are able work with analysts as they choose the data and analytical techniques. No matter how well-meaning the analysts are, they will never understand the issues and concerns as well as the people who must live with them. For this reason, I am frustrated by the wording of I-19's third conceptual issue: "3) how the knowledge and needs of marginal social groups can be incorporated into GIS-based decision-making." The only correct answer is to give them equal footing and have them at the table.
The fourth frustration is knowing how to best use the results of the analysis. At the day-long session the providers said that community groups know how to do this, but I'm less sanguine. One answer given is that community groups often hold a newsworthy event, get their analysis published (or broadcast), then rely on that publication as substantiation of the merits of their case. Another approach is to feed the findings to a sympathetic policy-maker. This approach assumes that the community group is trusted by that person in power. In one case a data provider in Atlanta determined that a state law regarding enterprise zones, written to assist rural counties, would apply to poor urban tracts if only the county geographic limitation were relaxed. A sympathetic legislator agreed to help, but needed the data provider to find other qualifying urban tracts around the state to win support from enough legislators to alter the law. A third approach is to use superior information about your own community to buy access to key management and policy people; a Minneapolis neighborhood mapped crime data and shared the results with precinct officers. Such intriguing stories beg the question. We know too little about how information affects policy and what steps to recommend to community groups.
It is unfortunate that these issues are of such little concern to the GIS community and I am pleased that this NCGIA initiative is underway. As evidence of the lack of interest, an abstract about this topic submitted to the GIS/LIS'95 conference was rejected (but accepted as a keynote presentation at AGI'95). None of three recent proposals by data providers to URISA for its Exemplary Systems in Government (ESIG) Award won anything but honorable mention.
I have described numerous research opportunities. My
particular interests lie in the reverse data directory and
further study of how geographic information is used to affect
policy.