Neighborhood benchmarking—a GIS approach

William Langely
GIS Supervisor
City of Garland, Texas

Tracking neighborhood vitality is an important issue for local governments. Declining neighborhoods will adversely affect the city's tax base, increase crime rates and drive away desirable homeowners creating a spiral of decline. This becomes not only a financial issue but also a quality of life issue for a city's residents. Some measurement of these trends must be used in order to intervene. But how can the quality and vitality of any particular neighborhood be measured?

The City of Garland, Texas is taking a proactive approach to measuring the vitality of its neighborhoods with a combination of data collection, analysis, and long-range planning tools. At the center of Garland's efforts is the Neighborhood Benchmarking Program (NBP), which takes basic planning concepts and marries these with performance indicators, GIS technologies, and administrative strategies. The City's Geographic Information Systems Department supported these efforts by developing a Neighborhood Information System (NIS) in cooperation with the City's Organizational Development Team (ODT), an internal consulting group.

Building an information system such as the NIS presented some unique challenges. The NIS data warehouse contains data compiled from various sources, such as the City of Garland's tax system, Code Compliance System, Pavement Management System, Utility Billing System, and information from other sources such as the local council of governments and realtor-oriented databases. Compiling these data required researching data sources, extraction from their native systems, conversion into Garland's GIS database format, geo-coding by address and other geo-based attributes, and in some cases creating new graphics and assigning new geo-codes.

To serve as an objective indicator of neighborhood conditions, the NIS compiles housing and nuisance code violations, street condition ratings, property values, litter index ratings, and other data from multiple city databases into one location for analysis. In addition, housing turnover rates and economic indicators are derived from external sources. ODT also adds survey results and field-collected neighborhood appearance ratings to the NIS.

Coordination of all these data into one location—as well as its analysis at the neighborhood level—enables staff to construct neighborhood profiles, monitor changes in neighborhood indicators and conditions, and develop a coordinated approach to addressing neighborhood issues. Bringing this information together in a GIS environment will help to identify spatial patterns and citywide neighborhood trends.

Project requirements
The first step in this project, as any other, was to define the ultimate objectives, such as what information would be needed out of the system and what types of analysis would be performed against the data. This is, of course, necessary in order to decide what data needed to go into the system. ODT expressed the following goals for the NIS:

  1. Assess and monitor the condition and health of Garland neighborhoods.
  2. Identify problems and challenges that are unique to particular neighborhoods, as well as those that face all Garland neighborhoods.
  3. Highlight at-risk neighborhoods and others showing signs of loss in neighborhood appeal or vitality.
  4. Tailor neighborhood efforts and services, instead of using a one-size-fits-all approach.
  5. Evaluate the effectiveness of code enforcement, grant monies, neighborhood associations, and other efforts used in maintaining or improving neighborhood conditions.
  6. Provide information that will enable policymakers to design programs and to target resources to produce the greatest return on investment.
  7. Furnish data that will help to determine cause-and-effect relationships regarding neighborhood health or decline.
Compiling data from multiple sources can require different methods for each source, but the basic method is as follows:
  1. Extract from native system
  2. Convert to GIS database format
  3. Scrub addresses
  4. Load into GIS
Creating neighborhood polygons
The above process, while joining the necessary information to the GIS, does not permit analysis of these data as a neighborhood unit. In order to do this, a geographic entity would need to be created to associate the data now joined to the parcel base to each neighborhood. Once neighborhood polygons were defined, a spatial association between each parcel and the neighborhood that contains it can be created and the parcel record populated with a neighborhood ID. This ID will permit a neighborhood-by-neighborhood analysis of the data now associated with each parcel via the situs address geocode.

First, the geographic boundaries of a neighborhood would have to be defined and agreed upon. A neighborhood can be defined in many ways. While valid opinions about what constitutes a neighborhood range across the spectrum from physical to cultural, a decision had to be made as to what geographic unit would constitute a neighborhood for the purposes of building a GIS dataset. The agreed-upon solution was to use subdivision phase as a neighborhood unit, because the houses of a common subdivision phase would be largely of the same age, price range, and quality of construction. The subdivision phase solution would also permit neighborhoods to be redefined by using different collections of subdivision phases.

Subdivision phases had been digitized by Garland's Engineering Department but were created to a CADD and not a GIS standard, and therefore were not topologically clean or suitable for the purposes of GIS analysis. Cleanup was performed to create enclosed polygons, and conflation to the existing parcel base. This provided the geographic unit to then reference all of the necessary data.

Data sources
With the neighborhoods defined, and the graphical elements produced and cleaned to a GIS standard, compiling data from various systems both within and without the city to meet the above stated goals could begin. Garland uses a variety of applications both developed in-house and purchased off the shelf to meet the needs of city operations. Client-server systems as well as mainframe systems were involved in this project. The following is a list of the data sources utilized:

  • Code Compliance System
  • Realtor-oriented databases
  • County Appraisal data
  • Garland Tax Data
  • U.S. Census Data
  • Pavement Management System
  • Garland Utility Billing System
  • Field Data Collection
  • Resident Surveys
Future enhancements
Some future steps in this initiative include plotting changes in crime statistics alongside instances of housing and environmental code violations, plotting changes in property values alongside changes in crime or code violations, and identifying spatial patterns or relationships between the presence of neighborhood associations and the number of code violations or crimes or the amount of property values. Also, a comprehensive neighborhood needs assessment coupled with a strategic plan to address neighborhood issues will be developed.

William Langely can be reached at (972) 205-2214 or at