3
Data Dissemination and Software Tools

To provide relevant and accurate data on housing and urban issues requires balanced attention to a number of related efforts that will encourage access and use of the data. These issues include promoting broad public access to data and information by building local capacity to use data, and supporting the development of analytical tools and skills (in-house and among the agency’s clients) that enable spatial analytical research on the complexities of neighborhood and urban issues.

Privacy considerations are also critical for data dissemination and application. Chapter 2 discusses the need to incorporate local data into national databases and to get that data back to users for comparative analysis. Unless methods are developed to disseminate data at required resolution while maintaining privacy, local groups will be unwilling to share their data. This chapter describes the wide array of HUD data users and discusses methods for gathering and disseminating these data, and for developing related tools to support decision making and spatial analysis.

HUD DATA USERS

HUD data are important to a broad spectrum of users. Users include: the agency’s staff at national and regional offices and public housing authorities; policymakers at federal, state and local levels; community-based and advocacy organizations; private researchers; university faculty and students;



The National Academies | 500 Fifth St. N.W. | Washington, D.C. 20001
Copyright © National Academy of Sciences. All rights reserved.
Terms of Use and Privacy Statement



Below are the first 10 and last 10 pages of uncorrected machine-read text (when available) of this chapter, followed by the top 30 algorithmically extracted key phrases from the chapter as a whole.
Intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text on the opening pages of each chapter. Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.

Do not use for reproduction, copying, pasting, or reading; exclusively for search engines.

OCR for page 51
3 Data Dissemination and Software Tools To provide relevant and accurate data on housing and urban issues requires balanced attention to a number of related efforts that will encourage access and use of the data. These issues include promoting broad public access to data and information by building local capacity to use data, and supporting the development of analytical tools and skills (in-house and among the agency’s clients) that enable spatial analytical research on the complexities of neighborhood and urban issues. Privacy considerations are also critical for data dissemination and application. Chapter 2 discusses the need to incorporate local data into national databases and to get that data back to users for comparative analysis. Unless methods are developed to disseminate data at required resolution while maintaining privacy, local groups will be unwilling to share their data. This chapter describes the wide array of HUD data users and discusses methods for gathering and disseminating these data, and for developing related tools to support decision making and spatial analysis. HUD DATA USERS HUD data are important to a broad spectrum of users. Users include: the agency’s staff at national and regional offices and public housing authorities; policymakers at federal, state and local levels; community-based and advocacy organizations; private researchers; university faculty and students;

OCR for page 51
tenant associations; and community residents and businesses. The data are used to plan community development and services, prepare grant applications, conduct research, implement programs, and support advocacy efforts. HUD serves low-income and minority populations in the nation and it is important to include these groups in decision making about their community. How can HUD get the tools of GIS into the hands of citizens so they get involved in their local planning processes? How can HUD link available analytical and decision-making tools with local knowledge to address local needs and priorities in housing and urban development? How can HUD promote spatial analysis of urban and housing issues in its client communities and among its in-house staff? These are among the agency’s challenges in which PD&R plays a central role. GIS can be used in HUD’s 81 field offices throughout the United States in a number of ways: internally by HUD for many purposes including linking its far-flung field offices, by recipients of HUD grants, and by HUD-related advocacy groups. At present, however, the use of GIS by HUD grant recipients (for instance, entitlement communities) is limited despite efforts to make the technology available (e.g., Community 20/20). Often, these groups are disadvantaged by being disconnected from both the planning and the information technology divisions of local government (Michael Martin, U.S. HUD Milwaukee Field Office, personal communication, 2002). HUD grantees have tended to use GIS to identify the spatial distribution of their programs, to create visual displays of resource allocation for political and educational purposes, and to advocate programmatic directions. Internally HUD’s regional offices use GIS in a variety of ways, including: ascertaining the eligibility of localities for place-based HUD funding, documenting HUD’s investments, investigating fair housing, and responding to disasters. Most of the GIS efforts in HUD’s field offices do not go beyond point and thematic mapping, because of limited understanding of spatial analysis, comparative spatial statistics, and housing indicator development (Michael Martin, U.S. HUD Milwaukee field office, personal communication, 2002). Typically, the use of GIS in HUD-related advocacy has revolved around education and communication with traditional HUD intra-agency groups created to help build support for HUD projects and activities. In particular, HUD has used GIS with organizations that promote fair and affordable housing. Specific projects include planning and disseminating information. This includes: answers to the question, “What does HUD fund in my community?”; building a community consensus on affordable housing needs; promoting better understanding of local real estate investment; and multimedia GIS to visualize, and to help non-experts visualize planning project alternatives. Used in these ways, GIS can be an effective tool for encouraging community engagement in decision making and planning.

OCR for page 51
Visualization, Communication, and Power “If information is power and if community is built through dialogue, then informatics permits both to emerge for those who would otherwise have no voice and no space for collective action” (Pickles, 1995, 10). Local organizations can become powerful players using GIS. Although local people may not think about their neighborhood in terms of GIS data layers, they may consider community issues comprehensively, in an integrated way in which “the quality of life in [the] family and [the] neighborhood revolves around housing, and work, and safety, and education, and goods and services, and neighbors and social networks and… many other things” (Michael Barndt, Nonprofit Center, Milwaukee, personal communication, 2001). Crime, transportation, housing, health care, families, and many other issues are in the minds of community actors as they think about their neighborhoods and attempt to develop solutions to meet neighborhood needs. GIS is a tool that can help integrate and analyze complex information and display it with the visual clarity of a map. Currently, at the local level, GIS is used largely for mapping or visualization to identify problems, show them spatially, and use the maps to advocate for public policy changes. Mapping serves several functions. First, a map can transmit new information to community residents in a way that is visual and easy to understand. For example, seeing and showing that crime is concentrated in one area of a neighborhood sends a powerful message. Residents can take a map to city hall or the city council to argue for additional resources for their communities. Community-level affordable housing advocate, Stella Adams, stresses the importance of spatial data to empower people to advocate for themselves (Stella Adams, North Carolina Affordable Housing Center, personal communication, 2002). Using HUD housing data, Resident Advisory Councils can go online and see what is going on in their community. Ideally, these HUD datasets would incorporate local knowledge from these groups within the framework of a larger dataset. Adams says, “Having access to GIS empowered me to be able to do things I couldn’t. [People] need the proof to show to elected officials. A map gives them legitimacy. Then you’re paid attention to.” Many of the decisions regarding housing and urban development are ultimately made by local municipalities and by states although HUD provides resources and develops the national housing and urban agenda. HUD has long sought a strategy that would support local efforts and provide tools for communities, local governments, and other interested partners to

OCR for page 51
use GIS and spatial data for local planning. Community 20/20 software was designed to enhance access to geographic information, to put HUD in the forefront of using spatially enabled technology, and to facilitate local planning processes and community efforts to prepare their consolidated plans (Box 2.2). HUD’s follow-up program, Enterprise GIS, continues this effort (Box 2.5). Understanding Local Conditions and Needs Informal settlements along the U.S.–Mexico border (colonias) are a good example of the need for understanding local conditions, for attention to processes at work at the regional level, and for putting relevant geographic information in the hands of local decision makers. HUD could use GIS in the colonias to evaluate the question the appropriate scale for the analysis of urban problems, assess the regional labor markers, identify housing and rental prices as a determinant of colonias development, monitor transnational processes between the United States and Mexico, and analyze the changing economic practices of unregulated urban settlements (see Box 3.1). Aerial photographs make effective visual tools because customers can “see” the area in question and better understand the other data that is overlaid on the map. The partnership between HUD and the U.S. Geological Survey in the colonias has made good use of these photos (Figure 3.2). Although visualization is a powerful tool for community groups and policy makers alike, the complexity of some urban problems requires advanced analytical techniques including statistics and modeling. Spatial analysis can inform our understanding of social problems and suggest public policy response. For example, understanding private housing markets is critical to housing voucher use. Housing vouchers work well in urban areas where there is a surplus of housing and landlords are happy to have the opportunity to rent their apartments through HUD’s voucher program. In tight housing markets, landlords have little incentive to accept housing vouchers. When demand for housing exceeds supply, prospective tenants bid up the price of rental units. In these situations, landlords often prefer to let the market work, since rising demand leads to higher rental prices. Individuals who have vouchers in these situations may find that they cannot secure a place to live. GIS could provide a framework for keeping track of trends in the private housing market including rental prices, and lead to the development of additional means to provide affordable housing in urban areas. GIS can be used to analyze the availability and spatial distribution of housing for people with low income. Nationally, the availability of affordable housing is lower in the suburbs than in the cities. In the 1990s, housing costs

OCR for page 51
rose faster than family income (Lichter and Crowley, 2002). Although housing economists at HUD currently track these trends at the national level, tracking at the city and neighborhood levels will provide additional information. Availability and dissemination of local data are key issues in integrating the fine-grained detail of housing market issues within the larger context of national trends. Analysis that integrates data at the local, regional, and national scales is required for the development of urban public policy. Confounding the integration of analysis at various scales is the lack of digital data and GIS expertise at the local and neighborhood levels. The role that HUD can play in promoting the inclusion of local data in national databases is discussed in Chapter 5. A Spectrum of User Needs Users of HUD data vary in terms of technical ability and access to resources. Regardless, most people want timely, accurate, and accessible information. People want information about their neighborhood, such as the availability of homes to rent or buy, and the location of social services and transportation routes. More advanced HUD data users want this information in the form of spatially enabled data with accessible metadata to show, find, and explain interesting trends and patterns. BOX 3.1 Colonias: U.S.–Mexico Community-based GIS for Economic Development HUD defines the colonias as “rural communities and neighborhoods located within 150 miles of the U.S.–Mexican border that lack adequate infrastructure and frequently also lack other basic services.1 Colonias have emerged in rural areas but they are predominantly residential areas for workers and families working in nearby urban centers or in agricultural occupations. Colonias vary in age, size, and composition, but because of their informal nature and recent origin, little is known about the trajectory of their growth. Typically they are unplanned, unregulated settlements with high rates of poverty, which is a factor that compounds the difficulty of developing infrastructure such as roads, water and sewer systems, improved housing, street lighting, and other services. 1   Definitions of a colonia vary among agencies and groups (<www.hud.gov/whatcol.cfm>).

OCR for page 51
In the 1990s, Congress required states along the Mexican border to set aside a portion of their HUD-allocated Community Development Block Grant (CDBG) funds to alleviate poverty and improve housing in the colonias. Other federal agencies, including the EPA and Department of Agriculture, also have projects that deal with the colonias. FIGURE 3.1 House in the colonias. SOURCE: Alina Simone, Texas Low-Income Housing Information Service. HUD’s goal in its colonias project is twofold: understand the conditions within the settlements to identify emerging issues and challenges, and to inform PD&R and HUD’s responsibilities in the colonias, and to carry out this research without doing extensive on-the-ground fieldwork as a demonstration of the potential usefulness of GIS approaches to urban problems. Monitoring the colonias along the U.S.–Mexico border is being undertaken in conjunction with the U.S. Geological Survey (USGS), through an interagency agreement. This agreement seeks to develop a

OCR for page 51
joint HUD–USGS cooperative GIS-enabled web site. The demonstration site for the project is the city of Eagle Pass, Texas. Three other pilot communities, one in each of the remaining Mexican border states, will be developed based on the experience gained in Eagle Pass. FIGURE 3.2 Aerial photograph of colonias in Berino, New Mexico. Note the multiple boundary lines pictured (black line, dotted line, white line) because of the lack of a consistent geographic definition of a colonia. SOURCE: Robert Czerniak, Department of Geography, New Mexico State University. PD&R is in the process of overlaying USGS aerial photos and data on housing and water facilities and integrating these layers into HUD’s enterprise GIS platform. Data from other agencies, such as the Department of Health and Human Services, EPA, and the Census Bureau will subsequently be integrated.

OCR for page 51
Technically sophisticated users may want substantial flexibility in querying and downloading HUD data; access to confidential data; and the ability to integrate multiple data sources, integrate data from different time periods, and use the data for multivariate spatial analysis. Less technically sophisticated users and users with few resources may have different demands such as easy-to-use tools to access HUD data for making maps or charts and tables that address their needs. Such users will need an Internet-delivered or web-based interface so that they can access and use HUD data at their local library or community center using basic hardware and software. The level of expertise and the needs of users are important considerations for the design of web-based GIS. HUD’s EGIS is setting out to address these user needs. Conclusion: HUD is faced with a broad a spectrum of user needs from the basic to the more technologically advanced. The full range of users must be considered. A web-based interface is important for some of HUD’s clients and for certain applications but the quality and usability of data are essential for all clients and all applications. In addition to simple, web-based interfaces, flexible querying is required to support more sophisticated applications. Recommendation: HUD should continue to develop a spectrum of tools to meet user needs. For users with limited financial or technical resources, HUD should provide web-based mapping of HUD data and related information. For more advanced applications, HUD should develop tools for flexible querying, extracting and downloading data, including standard file formats for exchanging data. DATA DISSEMINATION HUD disseminates data through multiple avenues including the Internet, HUD USER, and academically affiliated and unaffiliated researchers. Datasets and dissemination strategies are not centralized. Information disseminated to HUD’s local and regional offices is not always adequate or easy to use. HUD’s current methods for disseminating data are discussed below.

OCR for page 51
HUD’s Enterprise GIS (EGIS) The EGIS is a portal to access HUD’s data and to coordinate relevant data from other private and public agencies either through direct access or links. In its current state, it provides limited use to more advanced researchers, but the EGIS is useful for visualization purposes. Users can map locations and add layers for political and program boundaries, and add point data from HUD or other programs. More sophisticated users can download data to their own personal computers but EGIS does not support analysis of urban and housing issues at different scales or geographies (see Box 2.5). Using GIS effectively is a complex goal involving HUD's information and communication technology (ICT) infrastructure, personnel skill and training, organizational structure, partnerships, and technology choices. HUD’s R-MAPS R-MAPS, PD&R’s packaging of 13 spatially enabled datasets provides a wealth of HUD-related housing data. The Guide to PD&R Data Sets1 includes basic information about each of the datasets, but to get a full list of variables, variable definitions, and other specifics, users must download data directories and other supporting documents, which makes the use of the datasets cumbersome for sophisticated users and impossible for less sophisticated users (Box 2.4). Custom Data Tools To disseminate data, HUD has invested in customized tools like Community 20/20. Unfortunately, customized tools are limited in capability compared with the commercial and public domain or open-source tools. HUD would do well to avoid further custom tool development for data dissemination. Online data tools that support a broad range of users will promote the use of geographic data. Research Clearinghouse HUD can build relationships with research communities to facilitate use of spatially enabled data to examine housing and urban issues. The agency 1   Available at <http://www.hud.gov>.

OCR for page 51
has taken steps to initiate similar information sharing resources, for example, the Regulatory Barriers Clearinghouse.2 Along with the creation of a USDI, described in Chapter 2, HUD can support the development of a node for urban and housing research in the National Geospatial Data Clearinghouse of the NSDI. A good model is the USGS node for information about geospatial or spatially referenced data.3 Strategies for Developing a Research Clearinghouse Discussed below are several strategies to develop a research clearinghouse for housing and urban issues. These include better documentation, support for user conferences and online data user groups, a HUD USER newsletter focused on spatial analysis, and training to develop spatial analytical capabilities both in-house and for local data users. HUD could use the research clearinghouse to encourage researchers to examine urban and housing issues at different geographic scales and work closely with communities to develop research questions and create research products that are useable by communities, non-profits, and other local groups. As HUD makes additional housing-related data available via the Internet, the agency could further facilitate researchers’ access to and use of these datasets by creating better documentation that includes accurate and complete metadata. HUD could also support conferences on urban and housing themes or support particular tracts of panels within existing conferences to promote the understanding of the importance of data standards and to encourage the advancement of spatial methods to examine urban questions and problems. The Internet permits the formation of online data user groups. The American Housing Survey offers an example.4 HUD could encourage the formation of online user groups around the use of HUD data sets and discussion of broader urban and housing issues. The HUD USER newsletter is another means of disseminating information about current urban and housing research and the availability of relevant data and tools. HUD could create a new newsletter devoted to spatially enabled urban and housing research, methodologies, and issues about datasets. A newsletter could provide a source of information on HUD funded projects and other related projects including descriptions, implementing agencies, datasets employed, and status reports. 2   See <http://www.huduser.org/regbarriers/>. 3   See <http://nsdi.usgs.gov>. 4   See <http://www.huduser.org/datasets/ahs.html>.

OCR for page 51
Finally, HUD’s public includes highly sophisticated users with significant resources and users with far fewer resources. Additional training is necessary to bring resources to less sophisticated and resource-poor users. Many of the institutions, groups, and individuals who should be part of discussions about housing and urban issues lack access to technology and training to fully utilize GIS. Community organizations, neighborhood residents, and even HUD field office staff require both technical training and products that are easy to access and use. HUD’s Community Outreach Partnership Center (COPC) Program Many of the tasks involved in disseminating and using data to analyze urban conditions are best done locally. Regional or metropolitan centers can work collaboratively with communities to inform research questions and agendas and to collect and present data in ways that are useful to the community. These centers and others, such as State Census Data Centers, can serve as regional or metropolitan points to gather data across time periods, negotiate partnerships with other data producers, clean data, and disseminate it. A number of Community Outreach Partnership Centers (COPCs) have begun this work. Taking advantage of Environmental Systems Research Institute’s (ESRI’s) Arc Internet Map Server (ArcIMS), a product that allows for the development of interactive-mapping web sites, a number of COPCs have developed web sites for the community and city in which they work. Communities produce maps to meet neighborhood needs. Researchers post commonly used data such as U.S. Census data along with a variety of boundaries such as political jurisdictions or census geography, and users can create their own maps. Neighborhood Knowledge Los Angeles (NKLA), developed by graduate students at University of California at Los Angeles’s COPC,5 is one of the more sophisticated sites. The site allows users to post their own data and projects, and use data posted by the university. Users can also post the results and descriptions of the projects for which they have used data, in the process building a library of examples to generate future ideas. These COPC web sites are what are known as “thin” GIS sites. They provide data for mapping or visualization directly on the site. They do not necessarily encourage more sophisticated analysis that would include analysis of problems at multiple scales or downloading data for multivariate analysis. HUD’s COPC program is structured to emphasize outreach rather than 5   <http://copc.sppsr.ucla.edu/>.

OCR for page 51
research. Intermediaries would be needed to develop research capacity in the community. The potential for cooperative centers can be seen in the example of the Nonprofit Center of Milwaukee, a coalition with more than 200 local members (including the local HUD field office) that created the Community Information System (Box 3.2). Most of the COPC projects rely on interactive mapping to distribute data. Providing access to data over the Internet reduces the level of technical sophistication necessary to create basic maps; however, users who lack Internet access or have slow Internet connections are at a disadvantage. Furthermore, users are limited to available data. Finally, map-making is part art, part science, and a lack of expertise can result in maps that distort the data they are intended to present. BOX 3.2 Milwaukee’s Community Information System The Community Information System in Milwaukee, Wisconsin, facilitates access to data and builds the capacity for community and non-profit organizations’ data use. The center provides access to data and training in the use of data; and works closely with its partners to build a sustainable neighborhood data clearinghouse, offer data and GIS services on demand to neighborhood organizations, build the capacity of local organizations to organize and interpret data, and use technology to create tools to lower the costs of accessing and analyzing data. Rather that log on to a web site, community users work closely with the expert staff of the center to identify research questions, negotiate partnerships to share data, develop maps to present the data, and build capacity to use the data and maps to influence public policy and address urban problems and issues. The Role of PD&R in Data Dissemination As outlined earlier in this chapter, PD&R plays an active role in the interagency work that HUD is doing in the colonias, in relationships with other HUD clients and partners (such as urban researchers and community groups), and in dissemination of spatially enabled and housing-related data with R-MAPS. In the past, some GIS efforts at HUD such as Community 20/20 were hampered by a lack of technical input. Updating and maintaining data for Community 20/20 may have proved difficult in part because no program office at HUD had clear responsibility or ownership of the initiative.

OCR for page 51
PD&R’s current active role in providing technical expertise, research, and analysis for the development of HUD’s enterprise GIS is necessary and appropriate. Conclusion: A user friendly, web-based GIS is an efficient means for providing information about local housing conditions and making basic data available to the public. The utility of the information depends on the accuracy and the relevance of the basic data and the methods by which the information was derived. The development of a well-designed web-based GIS is a long-term process. User input is critical to this process. Recommendation: To improve dissemination and promote the use of spatial data, HUD should: Involve users in design of the web-based GIS; Sponsor conferences and workshops for clients and partners about using spatial data; Support online groups for HUD spatial database users; and Produce an Internet newsletter devoted to spatial data and analysis. PD&R is well-positioned to: Work with HUD clients and data users to derive the most appropriate GIS designs and to identify needed data and functions. Manage data confidentiality. For some sensitive data, PD&R will need to develop a policy on releasing confidential data as well as algorithms to suppress sensitive data to protect privacy. Take a lead in establishing a node for housing and related economic and demographic data in the NSDI’s National Geospatial Data Clearinghouse. Support the functions of an agency-wide enterprise GIS across all relevant HUD units. HUD GIS SUPPORT TOOLS GIS Applications and Needed Technological Support HUD’s mission has a strong spatial component so there is tremendous potential and need for spatially referenced data and geographic analysis for policy development. For example, one can use GIS to determine mortgage distribution, patterns of segregation, and neighborhood change. GIS has

OCR for page 51
multiple capabilities and each is valuable in its own way, for basic mapping, data handling, spatial analysis, etc. Local governments and community groups are beginning to take advantage of GIS and are developing capacity to do more sophisticated spatial analysis. GIS technology encompasses three integrated components at different levels of technological sophistication: database management technology, spatial analysis technology, and visualization technology. Database management technology is needed to store, retrieve, and convert large volumes of geospatial data. Spatial analysis technology is needed to make explicit a myriad of geospatial relationships among people, housing, and places. Last but not least, map visualization technology is used to view stored data. These capabilities can be made available in various ways, some of which can be combined to provide application support. HUD can assemble synergistic combinations of the GIS capabilities to increase its GIS potential by use of the following: Browser (thin client), which provides access to display only; and Browser with plug-in (thick client), which allows access to display and manipulation (Jankowski and Nyerges, 2001). Three levels of spatial analysis each add more explicit information about geospatial relationships. First, map display enables visual analysis. Second, geometric analysis of the points, lines, and polygons can be combined with their attribute (i.e., non-spatial character) qualifiers, such as population, ethnic background, housing stock. Third, spatial-temporal analysis takes into consideration the surrounding social characteristics, such as housing stock maturity and crime levels of an area regarding their effects on public housing. Mapping capabilities also come in a variety of levels, each level adding to the potential for visual insight into the spatial analysis performed and the database management undertaken. Map types of various forms (e.g., data magnitude by area in a choropleth6 map or spatially dispersed data observations as in a dot map) can be used to portray complex social and economic relationships in a spatial manner. Linking maps to scatter plots or box diagrams can promote exploration of more interactive spatial relationships, and, in a sense, help to develop deeper knowledge about a housing and urban development topic. Database technology comes in various levels of sophistication, each level adding more flexibility for retrieving and making available the stored 6   Choropleth maps are divided into parts corresponding to the physical extents of the enumeration areas and these parts are shaded according to the value of a variable for that area. See < http://www.mimas.ac.uk/argus/Tutorials/CartoViz/PopViz/Choro1.html#Density> for details.

OCR for page 51
geospatial data. Three of these data management levels include the following approaches: Standalone workstation approach to data management encourages duplication of data and duplicative efforts in processing. Local area network approach although useful within a building to reduce duplication close by, does not foster broad-based sharing of information. Enterprise network approach to data storage fosters data sharing, but is more expensive because it requires more coordination of data production and use. More sophisticated GIS capabilities can be promoted for in-house HUD use and for the use of HUD’s technically advanced partners. Efforts to devolve spatial analytical abilities to local users involving training and introduction of support tools are also needed. Potential in-house and advanced capabilities that HUD could develop include: Data aggregation for wider data distribution, Data aggregation functions to provide high-resolution data to users while protecting confidentiality, A software tool to perform data format conversion for major datasets, Robust GIS software platforms with multiple levels of capability to support different levels of user ability, and Spatial statistical modeling software. HUD can provide web services that select, cross-reference, aggregate, document, and feed HUD-specific data into local systems. Ensuring that these web service components are interoperable, accessible, and protective of privacy are major challenges. To support web services that allow online map production, HUD would do well to track the development of the “geographic markup language”(GML)7 and related GIS and web-service standards efforts of the Open GIS Consortium and the World Wide Web Consortium. GIS support tools can respond to the wide range of needs and capabilities that characterize HUD’s user community. Advanced tools and capabilities such as models are useful for analyzing urban development patterns and for developing housing policy. 7   GML is an encoding system for geographic features that is intended to support both data storage and data transport. See the OpenGIS Consortium for details (<http://www.opengis.org/info/techno/specs/00-029/GML.html#GMLOverview>).

OCR for page 51
Simulation Models of Urban Development and Housing Interventions in one neighborhood may affect the residents not only of that neighborhood but of other neighborhoods that are linked through the metropolitan housing and labor market. Research is aimed at developing simulation models suitable for use in analyzing these complex urban effects. The Urban Institute Model (de Leeuw and Struyk, 1975) and the NBER HUDS8 model (Kain and Apgar, 1985) illustrate early attempts to provide simulation models for policy analysis of metropolitan housing markets. Both models attempted to simulate the effects of housing policies such as the Housing Allowance Program using a micro-simulation of the housing market. The usefulness of these models in practical policy applications was limited by the high cost of developing the models, extensive data and computational requirements, limited theoretical and statistical methods, lack of geographic detail, and limited validation of the models. Modeling efforts have continued since the 1980s but have not been directly tied to housing policy. For example, micro-simulation models use a large sample of households to assess the effects of financial policies, often over long intervals that require “aging” persons and households in the sample over time. These models have been developed and used widely for policy analysis since Guy Orcutt developed the approach in the 1950s (Hanushek and Citro, 1991; Orcutt, 1957, 1960). Other examples of microsimulation models include the Dynamic Simulation of Income Model (DYNASIM), used by the Urban Institute to evaluate effects of alternative social security rules on different types of families; the Cornell Simulation Model (CORSIM), and the Micro-Analysis of Transfers to Households (MATH) Model developed by Mathematica, Inc. These and other microsimulation models for policy applications provide considerable detail about persons and households, but, as they are intended for national policy applications, they lack spatial detail that would make them useful for analysis of housing and community development policies. Models have also been developed for metropolitan-scale land use and transportation planning, and have come into widespread use since the 1980s. The Clean Air Act Amendments of 1991, and ISTEA and TEA219 designate Metropolitan Planning Organizations as the principal agents in developing coordinated regional transportation plans, and require coordination of land 8   National Bureau of Economic Research Harvard Urban Development Simulation (NBER HUDS). 9   Respectively, Intermodal Surface Transportation Efficiency Act of 1991 and Transportation Equity Act for the 21st Century.

OCR for page 51
use and transportation planning processes. The DRAM/EMPAL10 model (Putman, 1983) has been widely used for metropolitan-scale land use modeling, but lacks any representation of the housing market, and has only a modest degree of spatial detail. The MEPLAN (Echenique, 1994) and TRANUS (de la Barra, 1995) models11 are larger-scale, metropolitan models of land use and transportation that share an approach based on spatial input-output modeling, and do include some representation of markets for real estate, though with relatively little spatial detail. The UrbanSim model has been developed recently to evaluate metropolitan transportation and growth management policies. This model uses a micro-simulation approach that makes extensive use of GIS and parcel-level data, and simulates processes of household location, business location, and real estate development and prices (Waddell, 2002). Urban simulation models can be used to analyze the dynamics of neighborhoods and metropolitan areas and to evaluate the effects of housing and community development policies. Efforts in these areas would benefit from the lessons of early housing policy simulation efforts. Advances from ongoing development of micro-simulation and metropolitan land use and transportation models could be integrated into housing policy analysis. Conclusion: At present, local data users including local governments and advocacy groups use GIS mostly for visualization. This allows users to view only a few variables at a time, but does not promote the application of spatial data to complex urban and community issues. While capability varies among users, the development of spatial analytical ability is important for both professional researchers and for local data users. Recommendation: To help community groups and local governments develop spatial analysis capabilities, HUD should support the development of tools for spatial analysis. PD&R should support the development of on-line/downloadable analytical tools that incorporate multivariate techniques. SUMMARY HUD serves a broad spectrum of data users and stakeholders in urban and community development including some of the most disadvantaged and 10   The Disaggregated Residential Allocation Model (DRAM) and the Employment Allocation Model (EMPAL). 11   See <http://tmip.fhwa.dot.gov/clearinghouse/docs/landuse/compendium/dvrpc_toc.stm> for a review of these models.

OCR for page 51
underrepresented groups in the nation. GIS and relevant spatial data can provide HUD’s users from technically sophisticated urban researchers to neighborhood advocacy groups with the ability to collect, analyze, and present data in powerful ways. HUD uses a variety of tools to disseminate and promote the analysis of these data. More progress is required in building relationships for collecting data, disseminating spatial information and know-how to HUD’s clients, and devolving spatial analytical capabilities to communities.