ArcNews Online
 

Spring 2010
 

E-mail to a Friend

Mapping Urban Inequalities with GIS

By Linda Loubert, Economics Department, Morgan State University, Baltimore, Maryland

Highlights

  • ArcGIS is used to geocode 911 calls and crime data to socioeconomic and demographic data to determine a focus/study area.
  • Esri Business Analyst mapped all businesses around a proposed emergency shelter site.
  • GIS is important to homelessness prevention.

Mapping urban areas can help cities target policies that are most efficient and effective for their communities, particularly for those who are less fortunate. However, finding a solution to a problem such as homelessness entails understanding the associated issues. GIS has become fundamental to that process.

Homelessness prevention, of course, should be the first priority. But when that has not taken place, it becomes necessary to have a structure ready to supply fundamental care and services. Finding a location for shelters gets to be a tricky situation for local governments because businesses find it undesirable to have homeless people close by and, therefore, resist their accommodation, hoping shelters will not be near their businesses or, as the slogan goes, Not in My Back Yard (NIMBY).

Social scientists at Morgan State University, Baltimore, Maryland, studied the impact of locating a permanent homeless shelter for the City of Baltimore with the intent of uncovering all perspectives of building a new structure. Their findings could be applicable to any city. Beginning with some statistics on homeless people, the study found a clear indication of the critical need for some type of permanent structure because

  • More than 800,000 people may be homeless on any given day; 200,000 of them may be children (Burt, M. R., 2001. What Will It Take to End Homelessness? Washington, D.C.: The Urban Institute).
  • During a typical year, 900,000 to 1.4 million children are homeless.
  • Ten percent of all poor people may be homeless, even if only for a short while.
  • Seventy-five percent of homeless individuals access services in central cities (The Annual Homeless Assessment Report to Congress, 2007).

When more than 50 percent of their income has to go for housing, this tends to push low-income people into homelessness even faster. Also contributing to the problem is that U.S. health care policies have removed institutional support for people with severe mental illness, along with a drastic reduction in long-term hospitalization for the mentally ill; this has pushed these individuals out into the streets.

  see enlargement
This shows the population density for defined neighborhood boundaries and locations of current service providers within a 1.5-mile radius. Within this radius are at least 60 percent of the providers of services to the homeless.

The "visible" homeless people are generally overrepresented in central cities of large urban areas. In Baltimore, as in other cities, homelessness is a serious social and public health problem, so the city believed building a new emergency shelter for more than 200 people would help alleviate some of the problems for homeless individuals. Building the shelter, called the Housing Resource Center, is a strategy to address the City of Baltimore's 10-year plan to end homelessness. This project reflects various aspects of best practices to the extent that it integrates a 24/7 emergency shelter with an array of supportive services (health, counseling, and employment).

The study involved key stakeholders to understand the impact of this shelter as it related to homeless people, businesses, service providers, and neighborhoods located less than one mile from the proposed site. The study also included the developers of buildings for homeless people who could contribute design ideas that would incorporate safety measures for the shelter residents and the residents of the surrounding community, as well as appropriate architectural designs for the area.

ArcGIS was used by the Institute for Urban Research at Morgan State through an Esri university site license. Using its overlays and tools, the institute's researchers incorporated ArcGIS in this study beginning with community mapping; they collected information from the city, local businesses, and neighbors of the proposed site. They captured mobility patterns of homeless individuals using GPS. ArcGIS provided the tools to geocode 911 calls and crime data to U.S. census block groups, and socioeconomic and demographic data from the U.S. census was added to paint a picture of the focused area for analysis. The researchers took population density into account for defined neighborhood boundaries and the location of current service providers within a 1.5-mile radius (showing at least 60 percent of the providers of services to the homeless).

It should be noted that the City of Baltimore has only used temporary emergency shelters, scattered throughout the city, not a permanent one. Even though the neighborhood is densely populated, the study showed that the proposed location of the site would be in an unpopulated area of the neighborhood, under the viaduct of an interstate highway.

With Esri Business Analyst, all businesses around the proposed site were identified. Businesses in the neighborhoods surrounding the proposed emergency shelter represent 13 percent of all businesses in the city. The area consists of the downtown district.

From this kind of study, the question naturally arises: Will the shelter bring more crime and/or disturbances? To answer this question, researchers geocoded emergency medical services (EMS) calls and other crime data to U.S. census block groups for 2004 and 2008. Since a private sponsor opened a multipurpose soup kitchen in 2007, within 1,000 feet of the proposed facility, the homeless traffic was assumed to have increased during that time; this gave good reason to use years 2004 and 2008 for analysis of crime and EMS data. Based on standard deviations, the results indicate that the proposed site would not increase crime with an influx of more homeless people.

The study concluded that businesses and neighboring communities possessed a rather negative view of having a permanent shelter in their area. Homeless people were seen as loiterers and panhandlers who sleep in public spaces and relieve themselves on private property and who should not be concentrated in one area of the city. Service providers and developers perceived homelessness as a societal health illness, with the need for compassion and effective policy to relieve the symptoms. The homeless individuals who spoke during the focus group study indicated that their desire for help was only for private residency, not group residency, as the proposed structure would provide.

Using ArcGIS Desktop and Esri Business Analyst, the study concluded that the site would be in a sparsely populated area of a few blocks within a densely populated neighborhood that included some businesses. The crime and EMS data showed that no increase in crime would occur because of the site when standard deviations were examined.

Using GIS along with qualitative analysis, such as the focus group of stakeholders, cities can better understand the needs of the homeless population.

About the Author

Linda Loubert, Ph.D., is an assistant professor in the Economics Department at Morgan State University, Baltimore, Maryland, and an affiliate researcher in the Institute for Urban Research at Morgan State.

More Information

For more information, contact Linda Loubert, Ph.D., assistant professor, Economics Department (e-mail: linda.loubert@morgan.edu), or visit the Institute for Urban Research at iur.morgan.edu. Other key personnel for this study from Morgan State University were Mary Anne Akers, Ph.D., School of Architecture and Planning; Jonathan VanGeest, Ph.D., School of Community Health & Policy; Sidney Wong, Ph.D., School of Architecture and Planning; Azza Kamal, Ph.D., School of Architecture and Planning; and Marvin Perry, Office of Sponsored Programs.

Contact Us | Privacy | Legal | Site Map