Introduction:
With the 2010 Olympic Games nearing, much attention has been given to Metro Vancouver’s homelessness problem. This issue, however, is not a new one and in the past few years the problem has been spreading into the suburbs. The homeless population in Metro Vancouver has grown from 1121 in 2002 to 2660 in 2008 (SPARC, 2008). Two of the communities expressing the most rapid growth of homeless populations include Burnaby and the Tri-Cities both with 100% or greater growth from 2005 to 2008. Homeless populations in these regions are largely confined to green spaces such as parks and forests. By identifying the needs of homeless people and thus identifying the environments which meet these needs we can develop a series of factors and constraints on which to base a mutli-criteria analysis of Metro Vancouver. This analysis can then be used to identify areas in which homeless people may be found, also known as homeless refuge sites.
The introduction of spatial analysis will allow us to better enumerate the homeless population by allowing us to focus on areas with high incidence of homelessness. Furthermore, this will eventually allow us to take the homelessness studies one step further by incorporating estimations of hidden homeless as defined by Fiedler, R., N. Schuurman, and J. Hyndman (2006). Another issue this will address is the nomadic nature of homeless populations. With the increasing growth of the Metro Vancouver region, many homeless people are being forced to relocate. Policy implications such as where to place a shelter or offer a hot meal program become increasingly difficult when the population displays high levels of mobility. For example, a “meals on wheels” type program may be more effective in some regions than a meal program at a fixed location as the homeless population may be spread out over a large area. Although spatial analysis is a great tool to help find a solution to the homelessness issue nothing can be done without a proper understanding of truly how big the problem is.
Enumerating homeless people is a challenging task. Regardless of which enumeration method is used there are faults. Much like the Bogard (2001) study, many enumeration techniques focus on identifying homeless populations by sleeping locations. In most cases these locations are in isolated areas where access at night is simply not safe. Moreover, indirect counts, which are a result of obtaining data from service providers geared towards the homeless, have an inherent bias in that they do not properly capture the portion of the homeless population who do not use these services (Berry, 2007). Moreover, these estimates are conducive to double counting in that people who use a soup kitchen, for example, may also use a shelter that day resulting in a double count. A common problem with all of these enumeration methods is that they habitually undercount the true homeless population.
To further understand the difficulty of enumerating homeless populations it is imperative to bring forth the semantic confusion when it comes to defining homelessness. The terms hidden homeless, street/service homeless, couch surfers and sheltered homeless are some of the terms used in different studies to describe homeless people of varying degrees. Many of these terms are use-driven, that is they are defined based on the ideas, opinions and assumptions of the people or groups running the study. This leads to very little consistency between different counts and in some cases ultimately leads to results which are not comparable with others (Burt, 1995).
Although the methods used to enumerate homeless people are imperfect, if used in combination with each other they would provide a better picture of homelessness within a given region (Berry, 2007). In Metro Vancouver, the Social Planning and Research Council of BC, or SPARC, has conducted a homeless count every three years since 2002. The basic features of this count include:
- Performed in a 24-hour period
- Two components to count both sheltered and non-sheltered (“street/service”) homeless
- Worked with community-based organizations and benefited from their local knowledge of homelessness by allowing them to operate the counts in their communities
- Separate questionnaires for night time and daytime components
- Had over 800 volunteers in the 2008 count
- Collection of simple demographic details for each of the homeless
- Collected locational information of each interview
This count, albeit thorough, still self admittedly undercounts the homeless population in Metro Vancouver. Undercounting in this case largely stems from the issues in defining homelessness. For example, couch surfers were not included in this survey unless a volunteer had frequent contact with them. Furthermore, hidden homeless (Fiedler, R., N. Schuurman, and J. Hyndman, 2006) are all but excluded in this survey. All weaknesses considered, the Metro Vancouver homeless population count is a good one and could be used as a model in developing homelessness counts in other regions.
Two features of the Metro Vancouver count that make it hard to replicate easily are the support of community based organizations and the large number of volunteers. Homelessness has been a hotbed issue in Metro Vancouver for many years and as such has received a lot of attention from all levels of government. This has led to funding for homeless support programs and the establishment of community based organizations like Hope for Freedom Society. Not all regions have access to these kinds of resources and therefore efforts must be organized to produce these counts in an efficient yet thorough manner. By utilizing the power of geographical information systems, researchers can help identify potential areas which may be inhabited by homeless people thus allowing them to streamline their resources.