Skip to main content
U.S. flag

An official website of the United States government

Dot gov

The .gov means it’s official.
Federal government websites often end in .gov or .mil. Before sharing sensitive information, make sure you’re on a federal government site.

Https

The site is secure.
The https:// ensures that you are connecting to the official website and that any information you provide is encrypted and transmitted securely.

The 1996 National Survey of Homeless Assistance Providers and Clients: A Comparison of Faith-Based and Secular Non-Profit Programs

Publication Date

Introduction

One of the most dramatic findings to emerge from the 1996 National Survey of Homeless Assistance Providers and Clients (NSHAPC) is the tremendous growth in the number and variety of homeless assistance programs during the late 1980s and early 1990s.  There is now a virtual “industry” of homeless assistance programs, and initial analyses of NSHAPC data provide a first glimpse at this system of programs: in February 1996, about 40,000 programs across the country received an estimated 3 million service contacts in 21,000 service locations.  Housing programs are the most common type of program (40 percent of the total), followed by food programs (33 percent), and health programs (7 percent).  Other types of programs account for the remaining 20 percent of homeless assistance programs.  About one-half of homeless assistance programs are located in central cities, another one-third are in rural areas, and the remaining 19 percent are in suburban/urban fringe communities.

Much of the recent growth in homeless assistance has been fueled by new investments of public funds.  Beginning in 1987 with the passage of the Stewart B. McKinney Homeless Assistance Act (P.L. 100-77), federal government investments in homeless assistance programs grew from $350 million to almost $1.5 billion in fiscal year 1995.  There is now “a two billion dollar a year infrastructure designed to deal with the problem” of homelessness (National Alliance to End Homelessness 2000), and the Bush administration recently announced more than $1 billion in grants to HUD’s Continuum of Care and Emergency Shelter Grant programs — an amount it describes as “the largest amount of homeless assistance in history” (U.S. Department of Housing and Urban Development 2001).

Faith-based and other secular non-profit organizations have a long history of helping homeless people.

Homeless service providers have not always benefited from such public support.  Faith-based and other secular non-profit organizations have a long history of helping homeless people and others in desperate need, and prior to the early 1980s almost all formal services for homeless people were administered by such organizations with little or no government support.  Many religious non-profits continue to operate with little or no government funding and yet they play a critical role in helping homeless people:  preliminary analyses of NSHAPC reveal that they administer over one-half of all food programs and about a quarter of housing and other (non-health) programs.

A more detailed examination of religious or faith-based non-profits within the homeless assistance service delivery system is clearly needed.  It is of particular interest given President Bush’s desire to improve funding opportunities for faith-based and other community organizations.1  The White House recently issued a report (2001) summarizing initial findings from five federal Departments, including the Department of Health and Human Services and the Department of Housing and Urban Development, on barriers to the participation of faith-based and other community organizations in the delivery of social services in collaboration with the federal government.  The report describes a bias against faith- and community-based organizations among federal social service programs by restricting certain organizations from applying for funding, and overburdening small organizations with cumbersome regulations and requirements.  It also notes that small faith-based and secular non-profit service providers receive very little government support compared to the services they provide.  To address some of these problems, President Bush has established Centers of Faith-Based and Community Initiatives within the Department of Health and Human Services (DHHS) and other federal agencies.  The centers have been charged with developing and coordinating outreach and other department-wide efforts to better inform faith-based and other community organizations on initiatives and funding opportunities; proposing programs that would increase the participation of faith-based and other community organizations in federal, state, and local initiatives; and reviewing and promoting compliance of relevant programs with “charitable choice” provisions2 (U.S. Department of Health and Human Services 2001).  In his budget for 2002, the President also established the Compassion Capital Fund to “provide grants to charitable organizations to emulate model social service programs and to encourage research on the best practices of social service organizations…”  This fund has been appropriated $30 million for 2002 and these will be used to match private giving with Federal dollars.

Non-profits are a very diverse group of organizations — so diverse that some people have even questioned the usefulness of grouping them together within a single sector.  They are distinctive in that they “engage people in collective purposes outside of the market and the state and are independently organized and self governing” but are extremely varied in their origins, sizes, finances, the types of activities they undertake, the people they serve, and the means they use to reach their goals (Boris 1998).  In general, much of what we know about organizations in the non-profit sector has been limited to public-serving organizations that are eligible to receive tax-deductible donations (501[c][3] organizations), that are large enough to be required to file annual reports to the IRS (over $25,000 in revenues), and that are required to do so.  Relatively little is known about faith-based non-profits because they are not required to report to the IRS (Boris 1998).3

This study examines new data from the National Survey of Homeless Assistance Providers and Clients (NSHAPC) with an explicit focus on comparing homeless assistance programs administered by faith-based organizations and secular non-profit service providers.  It provides a basic but comprehensive picture of the numbers and characteristics of faith-based versus secular service providers, including how the two groups compare in terms of specific types of programs; numbers of program contacts;  types of clients and special population focus (battered women, families, runaway youth, people with alcohol, drug or mental health problems, etc.); program administrators’ assessments of homeless clients’ needs; how clients are referred to the program; where clients of housing programs go after leaving the program; and how much government funding they receive.  In addition, when national findings vary in important ways by urban-rural location or region of the country, these too are reported.


1.  President Bush’s Faith-Based and Community Initiative has raised a number of legal objections which for some people remain unresolved.  A Senate bill introducing the Charity Aid, Recovery, and Empowerment (CARE) Act has received bipartisan agreement and aims to “harness the enormous potential of charitable organizations to help the Federal Government solve pressing social problems” (Lieberman 2002).  The Act would provide for (1) tax incentives to spur more private charitable giving, (2) innovative programs to promote savings and economic self-sufficiency among low-income families, (3) technical assistance to help smaller social service providers do more good works, (4) narrowly-targeted efforts to remove unfair barriers facing faith-based groups in competing fairly for federal aid, and (5) additional federal funding for essential social service programs.

2.  Charitable Choice is a legislative provision designed to allow religious organizations to compete for federal funding on the same basis as other social service providers, without impairing the religious character of such organizations and without diminishing the religious freedom of beneficiaries of assistance.  Currently, Charitable Choice applies to three block grant programs at DHHS: the Temporary Assistance to Needy Families (TANF) program, the Community Services Block Grant (CSBG) program, and the Substance Abuse Prevention and Treatment Block Grant and discretionary grants.  The provision also covers the PATH (Projects for Assistance in Transition from Homelessness) program.

3.  While the term “faith-based organization” is not defined in the law, it is generally used to refer to religious organizations or religiously affiliated not-for-profit entities.  Those providing social services fall into two categories: (1) sectarian (pervasively religious organizations such as churches, synagogues, mosques, etc.) and (2) non-sectarian (separate, secular organizations created by religious organizations to provide social services) (U.S. Government Accounting Office 2002).

The National Survey of Homeless Assistance Providers and Clients

The 1996 National Survey of Homeless Assistance Providers and Clients (NSHAPC) is a comprehensive nationally representative survey of programs providing homeless assistance services and the clients of these programs.  The survey was conceived, developed, and funded by 12 federal agencies under the auspices of the Interagency Council on the Homeless.4  The Census Bureau carried out the data collection on behalf of the sponsoring agencies.

NSHAPC data are drawn from a nationally representative sample of homeless assistance programs in 76 primary sampling areas including (1) the 28 largest metropolitan statistical areas in the United States; (2) 24 small and medium-sized metropolitan statistical areas selected at random to be representative of the nation’s four main geographical regions (the northeast, south, midwest, and west) and size; and finally, (3) 24 rural areas (groups of counties or, in New England, groups of smaller administrative units (MCDs) within counties), selected at random from a sampling frame defined as the catchment areas of Community Action Agencies, and representative of geographical regions.5

NSHAPC defines a program as a set of services offered to the same group of people at a single location.

Within each primary sampling area, all homeless assistance programs offering a direct service were identified.6  For the purposes of NSHAPC, a program — defined as a set of services offered to the same group of people at a single location — had to meet the following specific criteria: (1) be managed or administered by the agency (i.e., the agency provides the staff and funding); (2) be designed to accomplish a particular mission or goal; (3) be offered on an ongoing basis; (4) be focused on homeless persons as an intended population (although not always the only population); and (5) not be limited to referrals or administrative functions.  In rural areas, which often lack homeless-specific services (Aron and Fitchen 1996), the definition was expanded to include agencies serving some homeless people even if this was not a focus of the agency.  About one-fourth of the rural programs in NSHAPC were included as a result of this expanded definition and they are retained for this analysis.  Sixteen types of homeless assistance programs were defined: emergency shelters, transitional housing programs, permanent housing programs, voucher distribution programs for emergency housing, programs accepting vouchers for emergency housing, food pantries, soup kitchens/meal distribution programs, mobile food programs, physical health care programs, mental health programs, alcohol/drug programs, HIV/AIDS programs, outreach programs, drop-in centers, migrant housing programs, and other types of programs.7

The study collected information on homeless assistance programs in two ways.  An initial computer-assisted telephone interview (CATI) survey was conducted with representatives of 6,307 service locations offering 11,983 homeless assistance programs.8  The CATI data yielded basic descriptive information about homeless assistance programs.  This was followed by a more detailed mail-survey of 5,694 programs.  The mail survey was completed by a staff person who knew the program and its clients well, and covered a variety of topics including client needs, the extent to which these needs were met, and whether services9 to meet these needs were available through their own program or other programs in the community.10  The findings reported here are drawn from both sources of program data: the CATI and the mail surveys.  In addition to these, interviews with a sample of 4,207 clients of these programs were also part of NSHAPC.11  These client data were not analyzed for this study.  The NSHAPC figures reported here have all been weighted to be nationally representative of homeless assistance programs and service locations on an average day in February 1996.12]


4.  The 12 federal sponsoring agencies are the U.S. Departments of Housing and Urban Development, Health and Human Services, Veterans Affairs, Agriculture, Commerce, Education, Energy, Justice, Labor, and Transportation, and the Social Security Administration, and the Federal Emergency Management Agency.

5.  Central cities are the main or primary cities of metropolitan statistical areas (MSAs).  Suburban and urban fringe areas are defined as what is left of MSAs after taking out the central cities, and may include smaller cities, suburbs, towns, and even open land if it is in the counties making up the MSA.  Rural areas are defined as all areas outside of MSAs, and may include small cities (under 50,000), towns, villages, and open land.

6.  Developing a list of all types of homeless assistance providers in each PSU involved multiple steps including developing an initial unduplicated list of potential providers, screening potential providers to determine if they offered programs that met the survey’s definition, and updating the provider list.  For more details about these procedures see Tourkin and Hubble (1997).

7.  As it happens, no migrant housing programs were identified by NSHAPC and one type of “other” program — financial/housing assistance — was encountered frequently enough to warrant a category of its own.  For a more detailed description of each of these programs, see Appendix A.

8.  A “service location” is the physical location at which one or more programs operate.

9.  NSHAPC defined a “service” as a good or activity offered to people using a program, but not qualifying on its own as a program.

10.  After extensive mail and telephone follow-up, the Census Bureau ultimately reached a response rate of about 70 percent for programs that received a mail survey, were active at the time they received the survey, and were in fact the program thought to have been identified through the CATI.  Response rates, however, varied by both program type and the number of programs at the service location.  Food programs and certain types of other programs were the least likely to complete a mail survey, while health programs unattached to shelter/housing or soup kitchen programs were the most likely to complete a mail survey.  Also, programs co-located with four or more other homeless assistance programs were less likely to return the mail survey.

11.  A “client” is anyone who uses a program whether or not (s)he is homeless.  Interviews were conducted with clients of any age as long as they were not accompanied by a parent or guardian.  For more information on the client survey and NSHAPC in general, see Burt et al. (1999) and Burt, Aron, and Lee (2001).

12.  All statements comparing two or more percentage figures have been tested for statistical significance, meaning a statistical test was used to determine if differences between the percentages are “significant” in a statistical sense.  A 90 percent criterion was used for the tests.  Thus, all comparisons discussed in the text are statistically significant at p = 0.10 or better, meaning that there is only a 10 percent chance that the difference is not a true difference.

Numbers and Types of Homeless Assistance Programs

NSHAPC documented just under 40,000 homeless assistance programs operating on an average day in February 1996 (see Table 1 and Figure 1).  Secular non-profits run almost one-half (47 percent) of these, followed by faith-based non-profits (32 percent), and government agencies (13 percent).  Another 0.6 percent of all programs are run by for-profit entities, typically hotels/motels accepting vouchers in exchange for providing a homeless person with one or more nights of accommodation, and for 7 percent of programs, it is not clear what type of organization or agency administers the program.13

Table 1:
NSHAPC Programs by Type of Agency Operating Programs

  Total Number
of Programs
Faith-Based Non-Profit Secular Non-Profit Government For-Profit Unidentified
All Program Types 39,664 (100%) 31.8 47.3 13.4 0.6 6.9
  Housing 15,879 (100%) 23.8 54.6 12.1 0.8 8.7
  Emergency Shelter 5,687 (100%) 26.7 61.2 5.6 0.2 6.3
Transitional Shelter 4,395 (100%) 26.9 57.7 9.9 1.4 4.2
Permanent Housing 1,918 (100%) 10.7 51.1 27.9 0.9 9.4
Distribute Vouchers 3,080 (100%) 24.1 44.2 17.0 0.3 14.4
Housing For Vouchers 799 (100%) 16.7 38.5 14.2 2.5 28.1
  Food 13,003 (100%) 53.1 37.4 5.0 0.4 4.2
    Soup Kitchen/Meal Distribution 3,484 (100%) 61.2 30.3 2.7 0.1 5.7
Food Pantry 9,028 (100%) 51.3 39.4 6.1 0.5 2.8
Mobile Food 491 (100%) 30.1 49.1 0.2 1.0 19.5
Health 2,739 (100%) 4.8 37.7 45.3 0.6 11.6
  Physical Health Care 715 (100%) 5.9 30.1 58.7 0.0 5.3
Mental Health 801 (100%) 1.2 31.2 64.8 0.6 2.3
Alcohol or Drug 778 (100%) 7.9 46.7 23.5 0.7 21.3
HIV/AIDS 446 (100%) 4.0 46.1 26.8 1.6 21.5
  Other 8,043 (100%) 22.1 52.2 18.8 0.7 6.2
  Outreach 3,307 (100%) 15.3 58.1 17.5 0.6 8.5
Drop-In Center 1,790 (100%) 25.1 60.5 7.2 0.3 6.9
Financial/Housing Assist. 1,378 (100%) 20.1 32.8 45.3 0.1 1.7
Other 1,568 (100%) 34.8 47.1 11.6 2.0 4.5

Source:  Urban Institute analysis of NSHAPC program data.  Data represent "an average day in February 1996."


Figure 1:
Distribution and Number of Programs Run by Each Sponsoring, for All Programs, for Housing Programs, and Food Programs.

Figure 1: Distribution and Number of Programs Run by Each Sponsoring, for All Programs, for Housing Programs, and Food Programs.

Faith-based organizations administer over half of all food programs and about one-quarter of all housing and drop-in center programs in the country.

In general, faith-based organizations account for over one-half (53 percent) of all food programs while secular non-profits provide most (55 percent) of the housing programs.  The single largest administrators of health programs are government agencies (which run 45 percent of all homeless assistance health programs).  Data on detailed program types shed some additional light on this general pattern.  Faith-based non-profits run most soup kitchens (61 percent), food pantries (51 percent), and many types of non-food programs.  They are responsible for about one-quarter of all emergency shelter, transitional housing, and voucher distribution programs.  Faith-based non-profits also administer over one-quarter of all drop-in centers across the country.

Secular non-profits are also clearly involved in providing many different kinds of homeless assistance services.  In addition to running the majority of shelters and permanent housing programs, secular non-profits administer over one-third of all food programs, including close to 40 percent of all food pantry programs and one-half of all mobile food programs.  They also oversee almost 40 percent of all health programs, including almost one-half of all alcohol and drug programs and HIV/AIDS programs.  Finally, around 60 percent of all outreach and drop-in center programs are run by secular non-profits.

Although government agencies administer less than 15 percent of all homeless assistance programs in the country, they are responsible for almost two-thirds (65 percent) of all mental health care programs and well over one-half (59 percent) of all physical health programs.  They also run about one-quarter of alcohol/drug and HIV/AIDS homeless assistance programs.  In the area of housing, government entities account for more than one-quarter (28 percent) of all permanent housing programs and close to one-half (45 percent) of all financial/housing assistance programs.

Faith-based programs play a much larger role in urban than in rural areas.

Analyses of program type and sponsoring agency by urban-rural location (Table 1A) and region of the country (Table 1B) reveal some interesting deviations from the national picture.  Table 1A makes clear that faith-based programs play a much larger role in urban areas (both central city and suburban) than they do in rural areas.  This is true for all types of programs taken together as well as for specific types of programs.  For food programs, for example, faith-based agencies account for 63 percent of all central city food programs but only 38 percent of all rural food programs.  Government agencies, by contrast, play a critical role in delivering health services in rural areas and a much smaller role in central cities and suburban areas.14  Over two-thirds of homeless assistance rural health programs are administered by government agencies.  In all other places, secular non-profits run the majority of these health programs.  With the exceptions of food programs (which are dominated by faith-based providers in urban areas) and health programs (which are dominated by government agencies in rural areas), secular non-profits tend to be much more consistent in terms of the types of programs they provide in urban versus rural areas.  They administer over one-half of all housing programs and just over one-half of all “other” types of programs in both urban and rural areas.

Table 1a:
NSHAPC Programs by Urban/Rural Status

  Total Number
of Programs
Faith-Based Non-Profit Secular Non-Profit Government For-Profit Unidentified
All Program Types 39,664 (100%) 31.8 47.3 13.4 0.6 6.9
Central Cities
All 19,388 (100%) 36.8 45.9 9.9 0.7 6.7
Housing 7,894 (100%) 28.7 53.8 9.6 0.8 7.2
Food 6,018 (100%) 63.4 28.3 2.6 0.2 5.5
Health 1,379 (100%) 7.5 56.8 29.1 0.7 5.8
Other 4,097 (100%) 23.5 53.0 14.6 1.2 7.7
Suburbs
All 7,694 (100%) 35.1 48.0 7.4 1.1 8.3
Housing 3,230 (100%) 24.2 53.6 8.7 1.8 11.8
Food 3,020 (100%) 53.0 40.0 2.6 0.4 4.0
Health 251 (100%) 2.9 51.0 32.0 2.6 11.5
Other 1,192 (100%) 26.2 52.9 11.0 0.4 9.6
Rural Areas
All 12,583 (100%) 21.9 48.9 22.6 0.2 6.4
Housing 4,754 (100%) 15.5 56.6 18.6 0.0 9.3
Food 3,965 (100%) 37.6 49.1 10.3 0.7 2.3
Health 1,110 (100%) 1.8 11.1 68.4 0.0 18.8
Other 2,754 (100%) 18.3 50.7 28.6 0.0 2.5

Source:  Urban Institute analysis of NSHAPC program data.  Data represent "an average day in February 1996."


Table 1b:
NSHAPC Programs by Region of the Country

  Total Number of Programs Faith-Based Non-Profit Secular Non-Profit Government For-Profit Unidentified
All Program Types 39,664 (100%) 31.8 47.3 13.4 0.6 6.9
Northeast
All 7,097 (100%) 28.6 53.6 10.1 0.6 7.0
Housing 2,870 (100%) 16.4 61.3 12.9 0.6 8.8
Food 2,401 (100%) 53.1 37.2 3.6 0.5 5.7
Health 306 (100%) 6.6 69.1 14.1 0.7 9.4
Other 1,521 (100%) 17.4 62.1 14.5 0.7 5.3
South
All 11,101 (100%) 39.0 40.7 13.6 0.5 6.2
Housing 4,309 (100%) 30.0 50.3 10.3 1.1 8.3
Food 4,113 (100%) 58.1 32.2 6.1 0.0 3.5
Health 863 (100%) 4.7 26.9 57.0 0.1 11.2
Other 1,817 (100%) 33.5 43.5 17.9 0.1 5.0
Midwest
All 11,853 (100%) 31.6 43.7 16.2 0.5 8.1
Housing 4,678 (100%) 24.5 47.6 16.9 0.4 10.6
Food 3,945 (100%) 54.6 34.3 6.7 0.8 3.6
Health 736 (100%) 2.8 39.7 35.5 0.0 22.0
Other 2,494 (100%) 16.8 52.6 24.0 0.4 6.3
West
All 9,333 (100%) 25.8 54.6 12.4 1.0 6.2
Housing 3,892 (100%) 21.2 62.9 8.0 1.0 6.9
Food 2,478 (100%) 42.4 51.0 1.7 0.2 4.7
Health 816 (100%) 6.0 34.7 53.8 1.7 3.6
Other 2,147 (100%) 22.3 51.3 17.2 1.6 7.6

Source: Urban Institute analysis of NSHAPC program data. Data represent "an average day in February 1996."


There are some interesting regional variations too.  Faith-based organizations appear to be responsible for a greater share of programs in the south (where they administer 39 percent of all homeless assistance programs) than in the west or northeast (where they administer just over one-quarter of programs).  The west is the only region of the country where faith-based organizations do not run the majority of all food programs (secular non-profits do).  Although they continue to play a major role in sponsoring food programs, the greater importance of faith-based service providers in the south is largely due to their increased involvement in housing and “other” types of homeless assistance programs.  Faith-based providers run 30 percent of housing programs in the south (compared to only 16 percent of those in the northeast) and 33 percent of all “other” types of homeless assistance programs (compared to only 17 percent in both the mid-west and northeast).  Interestingly, the south and west, where government agencies run the majority of homeless assistance health programs, are responsible for the predominance of government-run health programs nationally.  In the northeast, government agencies are responsible for less than 15 percent of all health programs (almost 70 percent of these programs are in fact run by secular non-profits).


13.  Programs with unidentified sponsoring agencies were eliminated from further analysis.  Those administered by government agencies are retained for some of the initial tables.

14.  Recall that the definition of what constituted a program for the purposes of NSHAPC had to be  expanded in rural locations to include agencies serving some homeless people even if this was not a focus of the agency.  The expansion brought in 8.5 percent of l secular non-profit programs, but only 4.3 percent of faith-based rural programs.

Program Contacts

In addition to documenting numbers (and types) of programs, the NSHAPC survey asked program administrators about how many people they expected to serve through their program on an average day in February 1996.  Summing these estimates across all programs (or across programs of a given type) yields an estimate of the total number of “program contacts” on an average day in February 1996.  Nationally, this figure is about 3 million contacts on a given day (see Table 2), with food programs accounting for slightly more than one-half of these contacts (1.6 million), housing programs for about 600,000, health programs for about 140,000, and other programs for about 700,000.

Table 2:
Comparison of NSHAPC Programs and Program Contacts
by Type of Agency Operating Programs
  Total Faith-Based Non-Profit Secular Non-Profit Government
Programs
All Types 36,674 (100%) 34.4 51.1 14.5
Housing 14,371 (100%) 26.3 60.3 13.4
Food 12,410 (100%) 55.7 39.1 5.2
Health 2,405 (100%) 5.4 43.0 51.6
Other 7,488 (100%) 23.7 56.0 20.2
Programs Contacts
All Types 3,022,492 (100%) 37.8 45.8 16.4
Housing 600,422 (100%) 29.9 54.8 15.4
Food 1,586,978 (100%) 53.1 44.9 2.0
Health 140,665 (100%) 4.8 41.5 53.7
Other 694,427 (100%) 16.1 41.0 42.9

Source:  Urban Institute analysis of NSHAPC program data.  Data represent an "average day in February 1996."


It is important to understand that estimates of “program contacts” are not the same as numbers of people or even service units.  One individual may use both an emergency shelter and a soup kitchen on a given day, while a second person eats breakfast at a drop-in center, lunch at a soup kitchen, and is then contacted by a mobile food program at night.  Both the emergency shelter and the soup kitchen would report the first person as a “person served,” and the drop-in center, soup kitchen, and mobile food program would each report the second person as a “person served.”  The sum of these reports far exceeds the two people actually served on that day and is not, therefore, an estimate of the number of people served.  Nor are “program contacts” the same as “service units.”  A person can often access a variety of services at a single location.  For example, a person staying at an emergency shelter might receive food, healthcare, and a place to sleep.  Each service received by that person would be counted as a “service unit.”  The emergency shelter made one “program contact” with the person staying at the shelter, but delivered multiple “service units” (a night’s accommodation, a meal, and a health care visit).

Comparisons of numbers of programs and program contacts by general program type and administering agency are presented in Table 2 and Figure 2.  Table 2 shows that the two sets of distributions are very similar, but it is interesting to note that faith-based non-profits are responsible for a somewhat higher share of all program contacts (38 percent) than programs (34 percent).  This is probably because food programs (the majority of which are run by faith-based organizations) are more likely than other types of programs to see large numbers of people on a given day.  There are no major differences in these findings by urban-rural status or region.

Figure 2:
A Comparison of Programs and Program Contacts:

Distribution of Programs

Figure 2a: A Comparison of Programs and Program Contacts: Distribution of Programs

Figure 2b: A Comparison of Programs and Program Contacts: Distribution of Programs

Clients of Homeless Assistance Programs

NSHAPC asked program administrators a number of questions about the clients of their programs, including whether “on an average day in February 1996 that the [emergency shelter or other type of] program operated, did the program serve (1) families with children (follow-up questions ask what share were single- versus two-parent, and among the single parent, what share were female headed), (2) two-parent families with children, (3) adults without children, and (4) unaccompanied youth or children, 17 years of age or under?  It is important to note that the answers to these questions reflect the types of clients actually served by homeless assistance programs, and not the types of clients they are willing or able to serve.  It should also be noted that these clients may or may not be homeless.  Clients of housing programs such as emergency and transitional shelters are clearly homeless, but clients of soup kitchen or health programs need not be homeless.15  Tables 3and 4 and Figure 3 present the share of faith-based and secular non-profits serving each of the following types of clients: men by themselves, women by themselves, female-headed households with children, other households with children, and youth.  The data do not reveal what share of clients are from these groups, but simply whether or not the program serves any such clients.

Figure 3:
Percentage of Faith-Based and Secular Nonprofit Programs Serving Various Population Groups

Figure 3a: Percentage of Faith-Based and Secular Nonprofit Programs Serving Various Population GroupsFigure 3b: Percentage of Faith-Based and Secular Nonprofit Programs Serving Various Population GroupsFigure 3c: Percentage of Faith-Based and Secular Nonprofit Programs Serving Various Population Groups

Faith-based programs are more likely than secular non-profits to serve different types of clients, especially single men.

As the first rows of Tables 3 and 4 indicate, faith-based programs are generally more likely than secular non-profits to serve each of the five client groups covered.  Almost 90 percent of faith-based programs serve single men, and almost 85 percent serve women by themselves and single women with children.  The group most often served by secular non-profits is single women (80 percent of them report serving such clients), followed by females with children (served by 77 percent of secular non-profits), and single men (72 percent).  Not surprisingly given their unique needs and circumstances, the group least likely to be served by either type of program is youth.  Youth are served by only 36 percent of faith-based programs and 31 percent of secular non-profits. 

Table 3:
Proportion of Faith-Based Non-profit Programs Serving Each Population Group
  Total # of programs Men by themselves Women by themselves Female-headed with children Other households with children Youth
All Program Types 12,599 88.4 84.9 84.3 77.9 35.7
Housing 3,783 75.4 67.6 67.9 53.1 19.0
  Emergency Shelter 1,520 80.2 69.7 67.5 52.9 22.9
  Transitional Shelter 1,181 58.9 49.7 57.8 33.4 10.3
  Permanent Housing 205 77.0 58.4 53.6 49.6 17.8
  Distribute Vouchers 743 93.0 92.0 91.6 87.4 27.8
  Housing For Vouchers 134 66.5 81.8 52.5 43.5 5.1
Food 6,907 94.9 93.7 95.8 92.8 45.3
  Soup Kitchen/Meal Distribution 2,131 97.5 96.7 90.3 86.0 60.3
  Food Pantry 4,628 93.7 92.4 98.9 96.4 38.2
  Mobile Food 148 97.1 94.0 76.5 75.9 50.6
Health 131 96.0 82.2 47.0 46.2 23.5
  Physical Health Care . . . . . .
  Mental Health . . . . . .
  Alcohol or Drug . . . . . .
  HIV/AIDS . . . . . .
Other 1,778 90.4 87.4 77.3 75.0 35.2
  Outreach 505 90.3 87.5 78.4 74.4 49.6
  Drop-In Center 450 85.2 68.5 55.3 52.1 34.8
  Financial/Housing Assistance 277 89.2 98.5 86.3 85.2 14.8
  Other 546 95.4 97.4 90.0 89.4 32.4

Source:  Urban Institute analysis of NSHAPC program data.  Data represent "an average day in February 1996."


Table 4
Proportion of Secular Non-profit Programs Serving Each Population Group
  Total # of programs Men by themselves Women by themselves Female-headed with children Other households with children Youth
All Program Types 18,751 71.5 80.4 76.7 59.8 30.6
Housing 8,664 56.4 72.1 70.8 41.9 18.5
  Emergency Shelter 3,480 36.9 74.7 74.6 35.9 25.6
  Transitional Shelter 2,535 53.6 60.2 59.1 32.7 11.9
  Permanent Housing 980 89.6 81.9 47.9 39.0 7.0
  Distribute Vouchers 1,361 85.3 82.6 98.4 75.2 23.1
  Housing For Vouchers 307 65.4 62.4 76.2 48.0 8.4
Food 4,858 86.8 92.7 86.2 80.8 42.3
  Soup Kitchen/Meal Distribution 1,057 90.8 92.9 63.1 58.2 48.6
  Food Pantry 3,560 86.7 92.3 97.1 91.3 40.9
  Mobile Food 241 72.4 98.3 25.5 24.2 35.5
Health 1,034 90.0 90.5 62.2 58.4 29.9
  Physical Health Care 215 92.5 92.5 75.9 72.2 58.0
  Mental Health 250 93.4 93.1 52.1 49.0 17.9
  Alcohol or Drug 363 86.4 85.1 55.7 52.5 16.0
  HIV/AIDS 206 89.5 94.9 71.5 65.6 39.5
Other 4,195 80.5 80.9 81.6 72.7 42.2
  Outreach 1,922 77.6 84.9 80.0 73.8 43.5
  Drop-In Center 1,083 81.1 88.0 75.2 65.1 43.3
  Financial/Housing Assistance 452 96.8 50.5 96.2 65.5 11.6
  Other 738 77.4 78.5 86.2 85.0 55.7

Source:  Urban Institute analysis of NSHAPC program data.  Data represent "an average day in February 1996."


A closer look at specific types of programs reveals that differences in the client groups served by faith-based and secular non-profits depend on both the population group and the type of program.  Beginning with single men, Tables 3 and 4 show that the vast majority (88 percent) of faith-based non-profits serve single men, while a smaller but still substantial share (72 percent) of secular non-profits serve this group.  For food programs, almost 95 percent of faith-based programs serve single men compared to 87 percent of secular programs.  The share of programs serving single men is lowest among housing programs (75 percent of faith-based programs and 56 percent of secular programs), but differences between faith-based and secular non-profits are found among all types of homeless assistance programs.

The patterns are less clear for single women.  While 85 percent of all faith-based programs serve single women compared to 80 percent of secular non-profits, among specific types of programs — such as health or housing programs — a larger share of secular non-profits than faith-based programs serve single women.  This is true of most types of housing programs and several types of health programs.  Almost all food programs run by either type of sponsoring agency serve single women, but a larger share of faith-based agencies running “other” types of programs serve single women than do secular non-profits.

Patterns of service for female-headed households with children are similar to those for single women.  Faith-based programs as a group are generally more likely to serve female-headed families than are secular programs, but secular health, housing, and “other” programs are all more likely to have such clients than are faith-based programs.  Thus, only for food programs do a much larger percentage of faith-based agencies (96 percent) serve women with children than do secular non-profits (86 percent).

Other families (i.e., those headed by two parents, one male parent, or no parents) are less frequently served by both secular and faith-based non-profits, with 78 percent of faith-based and 60 percent of secular non-profit programs serving these families.  Still, 93 percent of faith-based food programs serve this population as do 81 percent of secular non-profits.  The share of programs serving these family types drops dramatically when programs other than food programs are examined.  Fifty-three and 42 percent of faith-based and secular housing programs serve this group of clients, and 46 and 58 percent of health programs do.

The last population group is unaccompanied youth, who are served by just 36 percent of all faith-based programs and 31 percent of secular programs.  This is the one group that is not served by most faith-based food programs:  45 percent of these programs report having unaccompanied youth among their clients.  Compared to similar programs run by faith-based agencies, youth are served by a larger share of secular health and “other” programs, but the percentages that serve youth are small among all types of programs. 

In general, faith-based programs of all types are much more likely than secular programs to serve single men.  This is the only client group for which this is true.  For other groups of clients, faith-based providers are more likely to serve them if one looks at all types of programs combined, but this finding is driven almost entirely by food programs, which account for the majority of faith-based programs.  When one looks at other types of programs such as housing or health programs, secular non-profits are often more likely to serve some types of clients than are faith-based non-profits (see Figure 3).

Although the number of programs decreases dramatically when these findings are broken down by urban-rural status and region, one broad pattern does emerge:  both faith-based and secular programs in rural areas are more likely to serve clients from each population group than do the same programs in suburban areas and especially in urban areas.  This may reflect the inclusion of more generic social service programs in the sampling plan used for NSHAPC in rural areas.  The relative scarcity of homeless programs and homeless clients in rural areas may mean that rural area programs, even homeless-specific ones, may not have enough volume to specialize and may have to be more flexible in serving anyone in need of assistance.


15.  Certain questions in the mail survey are limited to the needs of clients “who are homeless.”  Later sections of this report indicate when the findings pertain to homeless clients specifically.

Special Focus of Homeless Assistance Programs

In addition to asking what types of clients they serve, NSHAPC asked program administrators whether their programs focused on one or more special population groups, such as children, veterans, or HIV patients.16  Table 5 lists the share of faith-based non-profits, secular non-profits, and government run programs that focus on each specific population group.  Figure 4 also displays the same information graphically.

Table 5:
Do NSHAPC Programs Have a Special Focus?

(Share of programs with any type of special focus)

  All Programs Faith-Based Non-Profit Secular Non-Profit Government
All Program Types 15,350 of 36,493 2,753 of 12,526 9,998 of 18,643 2,598 of 5,323
42.1% 22.0% 53.6% 48.8%
  Housing 53.5 40.7 62.4 38.3
    Emergency Shelter 59.5 39.0 69.4 50.1
    Transitional Shelter 63.7 60.4 68.0 47.8
    Permanent Housing 54.7 44.4 61.9 45.5
    Distribute Vouchers 26.0 11.0 37.8 16.8
    Housing For Vouchers 45.4 47.0 49.0 33.9
  Food 19.9 11.6 31.6 21.4
    Soup Kitchen/Meal Distribution 28.4 18.9 48.1 23.3
    Food Pantry 15.4 7.8 24.6 20.9
    Mobile Food 49.3 25.8 63.5 100.0
  Health 74.6 54.9 77.7 74.2
    Physical Health Care 48.7 7.5 60.1 47.0
    Mental Health 83.5 . 80.9 85.1
    Alcohol or Drug 87.0 80.3 84.1 95.1
    HIV/AIDS 83.6 . 80.7 90.0
  Other 45.3 19.1 53.7 53.0
    Outreach 52.0 25.2 53.0 72.1
    Drop-In Center 46.8 28.0 49.1 92.2
    Financial/Housing Assist. 39.9 2.3 75.6 30.8
    Other 35.0 14.5 48.6 41.0
 

Source:  Urban Institute analysis of NSHAPC program data.  Data represent "an average day in February 1996." 


Figure 4:
Percentage of Programs with a Special Focus

Figure 4: Percentage of Programs with a Special FocusFaith-based programs are much less likely than secular non-profits or government programs to have a special focus.

From Figure 4 it is clear that faith-based programs are significantly less likely to have a special focus than are secular non-profits or government programs.  Only 22 percent of all faith-based programs report having a special focus, compared to 54 percent of secular non-profits and 49 percent of government programs.  The percentage of faith-based programs with a special focus is especially low among food programs: only 12 percent of these programs have any special focus compared to 32 percent of secular non-profit and 21 percent of government-run food programs.  Faith-based food programs located in central cities are slightly more likely to specialize, but the percentage of programs that do so (15 percent) is still much smaller than that of secular programs (35 percent) and government programs (34 percent).  These results reflect the low rates of specialization among food programs, the type of program most often run by faith-based agencies.

Faith-based housing programs located in central cities are much more likely than those in other areas to have a special focus.

Housing and health programs are much more likely to specialize, no matter what type of agency sponsors them.  Forty-one percent of faith-based housing programs have a special focus, including 60 percent of transitional housing programs.  These percentages are still lower than those for secular non-profit housing programs (62 percent of all housing and 68 percent of transitional housing programs have a special focus), but they are higher than the corresponding percentage of government housing programs with a special focus (38 percent of all housing and 48 percent of transitional programs).  Faith-based housing programs located in central cities are significantly more likely to specialize than those in suburban or rural areas.  Among health programs, about 55 percent of faith-based programs have a special focus, compared to 78 percent of secular non-profit and 74 percent of government health programs.  There are too few faith-based health programs to examine these results by urban status or region of the country.  Among programs in the “other” category, only 19 percent of faith-based programs have a special focus, compared to 54 percent of secular non-profits and 53 percent of government programs.  These differences remain when comparing programs in central cities, suburban, and rural areas, and in all regions of the country.

In general, these results indicate that faith-based programs of all types are less likely to have a special focus than those run by secular non-profits.  This is especially true of faith-based food programs, which very rarely report having a special focus.  These findings fit our characterization of faith-based programs as programs that provide a basic set of services to a diverse group of clients.  They are less likely to specialize on one or more groups of clients (families, youth, etc.) or on one or more special needs (victims of domestic violence, HIV positive, etc.).  Instead, faith-based programs seem to be more general providers of basic assistance to many types of people, including many poor people who may not be literally homeless.

Another question of interest is: What special focus do homeless assistance programs tend to have?  The answer to this question was gleaned from two sources: reports about the primary mission of the program as well as any special populations the program had as a focus.  This information was only developed for “core” homeless assistance programs with relatively large sample sizes, namely emergency shelters, transitional and permanent housing programs, and soup kitchens.  Table 6 lists the shares of faith-based, secular non-profit, and government organizations that report an overall program focus.

Table 6:
What Special Focus Do NSHAPC Programs Have?

  Number of Programs Faith-Based Non-Profit Secular Non-Profit Government
Emergency Shelter with 5,320 (100%) 1,520 (100%) 3,480 (100%) 320 (100%)
  No specialization 40.6 63.2 30.4 44.6
  Mental health (MH) focus 3.7 2.5 4.1 5.2
  Chemical dependency (CD) focus 8.6 15.5 5.3 12.6
  MH/CD focus 1.4 2.7 0.9 1.0
  HIV/AIDS focus 1.4 1.8 1.3 0.3
  Domestic violence (DV) focus 30.3 5.2 42.1 20.1
  Youth focus 8.3 1.7 11.3 6.8
  Family focus 5.6 7.4 4.5 9.3
Transitional Shelter with 4,149 (100%) 1,181 (100%) 2,535 (100%) 433 (100%)
  No specialization 43.4 54.8 35.6 57.6
  Mental health (MH) focus 8.3 3.5 9.6 14.2
  Chemical dependency (CD) focus 14.4 16.6 15.2 4.2
  MH/CD focus 5.2 2.9 6.3 5.2
  HIV/AIDS focus 3.1 1.2 4.2 1.7
  Domestic violence (DV) focus 14.0 7.7 18.2 6.6
  Youth focus 4.4 5.6 4.6 0.2
  Family focus 7.1 7.6 6.3 10.2
Permanent Housing with 1,719 (100%) 205 (100%) 980 (100%) 534 (100%)
  No specialization 63.6 61.6 52.8 84.2
  Mental health (MH) focus 15.7 8.8 22.1 6.6
  Chemical dependency (CD) focus 5.2 11.0 5.2 2.9
  MH/CD focus 5.8 5.6 7.8 2.2
  HIV/AIDS focus 9.8 13.0 12.1 4.2
Soup Kitchen with 3,284 (100%) 2,131 (100%) 1,057 (100%) .
  No specialization 83.2 84.9 79.4 .
  Mental health (MH) focus 6.1 4.4 9.8 .
  Chemical dependency (CD) focus 6.7 7.6 5.2 .
  Family focus 2.4 2.9 1.6 .
  HIV/AIDS focus 1.5 0.2 4.0 .

Source:  Urban Institute analysis of NSHAPC program data.  Data represent "an average day in February 1996."


The most common focus among emergency shelters is domestic violence.

The most common focus of all emergency shelter programs is domestic violence, with 30 percent of all programs reporting victims of domestic violence as a primary focus.  Only 5 percent of faith-based programs have this focus, however, compared to 42 percent of secular non-profit and 20 percent of government emergency shelters.  The most common focus among faith-based emergency shelters is chemical dependency, reported by 16 percent of such programs.  Outside of central cities, faith-based emergency shelters focus on serving families more often than do similar programs within central cities.  These figures indicate a major difference in the focus of emergency shelters operated by secular non-profits and faith-based non-profits.  Whereas a substantial proportion of secular and government programs run shelters specifically for victims of domestic violence, faith-based shelters are much less likely to have any special focus, and only a small percentage are focused on domestic violence.

Among transitional housing programs, the most common focus of faith-based non-profits is again chemical dependency (17 percent report this), but an almost equal share (15 percent) of secular non-profits have this as a focus of their transitional housing programs.  Again, in suburban and urban fringe areas, faith-based transitional housing programs are more likely to focus on families or youth than do similar central city programs.17 Domestic violence is the most common focus among secular non-profit transitional housing programs (15 percent), while mental health is most common among government programs (14 percent).  For permanent housing programs, the most common focus of faith-based programs is HIV/AIDS followed by chemical dependency, while secular non-profits focus are more likely to focus on mental health (22 percent have this focus).

The vast majority of soup kitchens have no special focus.

Not surprisingly, the vast majority of soup kitchens have no special focus.  Among those that do, the most common focus is mental health or chemical dependency.  Eight percent of faith-based soup kitchens focus on chemical dependency and 4 percent focus on mental health, compared to 5 percent of secular non-profits that focus on chemical dependency and 10 percent on mental health.  There are not enough government-run soup kitchens to include in the analysis.

These figures indicate that there are few common missions or population groups targeted by a large percentage of transitional and permanent housing programs.  Those that do specialize tend to focus on domestic violence, chemical dependency, or mental health.  Among emergency shelter programs, however, a substantial proportion of secular non-profits, which run the majority of these programs, focus on domestic violence victims.  This focus is not shared by faith-based emergency shelters, which are much less likely to have any special focus.


16.  A program’s special focus was determined based on responses to questions about its primary population focus (30 groups were asked about) and the service location’s primary mission.  If answers to either of these indicated a focus, the program was classified according to that focus.  Classification rules included: any combination that included domestic violence was classified as having a domestic violence focus; any combination that included HIV/AIDS was classified as having an HIV/AIDS focus; and any combination that included youth was classified as having a youth focus.

17.  In rural areas, there are too few faith-based transitional and permanent housing programs to be included in the analysis.

Needs of Homeless Assistance Clients

A major focus of the NSHAPC mail survey was administrators’ perceptions of homeless clients’ needs.18  The survey asked about 59 distinct service needs ranging from food, clothing, and housing to employment and health care (see Table 7 for a complete listing).  The basic question asked of all respondents to the mail survey was:  Please consider your current clients in (your program) who are currently homeless.  How many of these clients need assistance from you or others in obtaining (e.g. health assessment) services?  Based on the responses to these questions, one can determine the share of programs with high levels of need for the various services identified in the survey. 

It is very important to keep in mind that these findings are based on administrators’ perceptions, and are not derived from in-depth needs assessments or even self-reports by homeless clients themselves.19  This distinction is important because the answers given by program staff will reflect only what they think they know about their clients, and this can be quite inaccurate.  The administrator of a soup kitchen, for example, may know about clients’ food needs, but he/she may not know what share of them need mental health services or even whether or not they are homeless.

Table 7:
Clients' Service Needs

(% of programs with most or all clients in need of service)
Service Need Faith-Based Non-profit Secular Non-profit Government   Service Need Faith-Based Non-profit Secular Non-profit Government
Food 75.1 68.2 53.7   General Health Care/HIV
Clothing 51.1 36.9 29.4   Health assmt 35.2 50.9 48.5
Life Skills Services   Primary care 32.5 46.8 45.7
Money mgmt 55.1 53.9 50.4   Acute care 7.1 13.3 7.8
Household skills 32.7 35.2 32.5   Prenatal care 6.6 7.4 8.5
Conflict resolution 41.6 51.1 37.6   Immunizations 14.1 16.3 17.0
Parenting training 28.1 29.4 21.1   HIV treatment 3.1 7.8 4.0
Case Management Services   HIV prevention 25.8 35.9 26.6
Needs assmt 58.4 77.2 64.8   TB test 23.3 27.8 25.9
Ind. Goals/Srv Plans 54.3 75.5 60.5   TB treatment 4.8 5.9 6.7
Referral 39.8 62.7 56.8   Dental care 34.8 44.1 35.2
Follow-up 35.3 50.3 43.8   Hospice 2.4 3.0 1.2
Housing Services   Health education 34.2 43.7 35.8
Locate housing 49.2 64.3 53.5   Substance Abuse Services
Apply for housing 31.0 46.2 37.5   Alcohol/drug testing 21.4 19.8 20.0
Financial Assistance 42.1 57.7 55.7   Clin alcohol/drug assmt 20.6 26.2 22.8
Landlord relat 16.1 27.9 19.4   Detoxification 6.8 9.0 8.2
Education Services   Outpatient treatment 8.5 14.4 21.9
School liaison 9.9 15.5 10.9   Residual treatment 6.4 9.1 5.4
Head start 5.7 9.7 12.8   AA 17.3 17.3 16.8
Other child ed 7.7 11.9 11.9   Substance abuse ed 27.6 32.5 31.2
Tutoring 8.5 9.7 10.5   Mental Health Services
ESL 2.0 2.8 3.6   MH assmt 16.2 32.1 30.5
GED 11.0 16.3 10.3   Medical admin. 7.5 16.8 18.8
Family literacy 5.5 6.0 4.0   Crisis intervention 10.4 27.4 22.3
Basic literacy 6.1 6.8 4.9   Outpatient therapy 11.2 26.2 23.4
Adult education 17.2 18.0 16.4   Inpatient/res treatment 4.7 10.1 8.3
Employment Services   Peer group 14.5 24.3 15.7
Job skills assmt 42.4 49.7 51.6   MH education 15.7 25.3 29.2
Job search skills 39.3 41.9 40.6   Other Services
Job referral 43.1 47.0 47.1   Child care 17.7 26.5 11.6
Sp. Job training 25.7 28.5 25.8   DV counseling 8.3 21.9 9.8
Vocational rehab 17.0 18.5 18.1   Legal assistance 7.8 20.3 11.9
Vol job placement 9.7 11.1 6.2   Veterans services 3.6 4.2 10.0

Source:  Urban Institute analysis of NSHAPC program data.  Data represent "an average day in February 1996."


18.  Questions and findings from this section of the survey are limited to homeless clients.

19.  The NSHAPC client survey did ask respondents to identify their top three service needs, but they were not asked about detailed needs separately.  Also, the NSHAPC public-use client data cannot be analyzed by agency type (faith-based, secular non-profit, etc.).

All Client Needs

When asked about each of the 59 different service needs, administrators were asked if “all,” “most,” “some,” or “none” of their clients had this need.  Table 7 lists all 59 service needs and reports what share of programs run by faith-based non-profits, secular non-profits, and government organizations have an administrator who thinks that “most” or “all” of their homeless clients need the given service.

Secular non-profits tend to report higher levels of client need than do faith-based non-profits.

In general, administrators of secular non-profits tend to report higher levels of client need than do faith-based non-profits.  Faith-based service providers report greater shares of clients needing food and clothing, but the observed difference in the shares of clients needing food is not statistically significant.  For many services, greater shares of clients of secular non-profit agencies are thought to need the service, but there are many services — especially in the areas of education, employment and training, and substance abuse services — where the levels of client need are surprisingly similar among the two types of agencies.

In assessing differences in client needs it is important once again to remember that some of these differences are likely to stem from the types of programs the two groups of agencies tend to sponsor.  Most faith-based homeless assistance programs are food programs (see Table 1), and it is natural that administrators of these programs would be more aware of basic needs such as food and clothing.  Even among food programs alone, however, a smaller percentage of faith-based non-profits report high levels of need than do secular non-profits (the only statistically significant exception to this is for clothing).

While it is difficult to know exactly why administrators of secular non-profits report that their clients have greater levels of need for many services, there are several possible explanations.  First, the clients of the two types of agencies may simply differ.  Prior analyses showed that faith-based programs, especially faith-based food programs, are more likely than secular programs to serve single men.  While it is unlikely that single men have fewer needs than other types of clients, the diversity of the client population of faith-based food programs serve may result in a more diffuse set of needs among all clients.  That is, if a program serves many types of clients — single mothers with children, single men, youth, etc. — it may be that there are few individual services that are needed by “all” or even “most” of these clients.  Thus, the heterogeneity of clients of faith-based programs may result in fewer needs that are common to all or most clients.

Other differences in the client groups may also explain the observed differences in client needs.  Compared to faith-based agencies, secular non-profits may have more clients who are currently homeless.  Although NSHAPC specifically asks about service needs for all currently homeless clients, comparisons of NSHAPC program and client data have shown that administrators overestimate the share of clients who are homeless.  This is especially true of food programs — administrators of these programs estimated that about 35 percent of their clients are currently homeless while the NSHAPC client data indicate that only 27 percent of food program clients are currently homeless.(20)  It is easy to see how program administrators could misidentify clients of food programs as being homeless, when in fact they may be formerly homeless clients or else they may simply be very poor and precariously housed.  In soup kitchens and other programs providing basic services other than housing, many clients are not screened.  This is especially true of food programs, which are known to serve many people who may not be literally homeless at the time of service.  Thus, compared to secular non-profits, faith-based programs may have fewer clients who are literally homeless and may therefore, have fewer immediate service needs.

Finally, the clients of faith-based non-profits may have fewer needs because these programs are less likely than other programs to have a special focus.  Administrators of faith-based assistance programs may also be less aware of the full range of specific client needs compared to secular non-profits.  Instead, faith-based programs may tend to concentrate on the basic needs of clients and provide services that address these needs.

Although the same general patterns hold when service needs are examined by urban-rural status and region of the country, there are a few interesting differences.  First, as might be expected, needs are generally perceived to be greater among clients of central city programs.  Second, faith-based programs in the northeast and in the west tend to report higher levels of need for many types of services relative to secular non-profits in these regions and relative to faith-based programs nationally.


20.  See Table 15.5, Estimates of Proportion Homeless Based on Program and Client Data on p. 15-21 of the NSHAPC Technical Report (Burt et al. 1999).

Primary Client Needs, Availability of Services, and Location of Services

The above analysis provides a broad sketch of homeless client needs for a wide range of services.  The remainder of this section examines a few of these needs in more detail, namely the twelve service needs that have been identified by over half of all program administrators as being needed by all or most clients (see Table 8).  In addition to level of need, we examine how often the needs are being met and where (on-site by the same program, on-site by another program, or off-site somewhere else in the community).

Table 8:
Major Needs, Availability of Services, and Location of Services
  % of programs reporting how many clients experience a specific need % of programs reporting how often clients' needs are being met % of programs reporting where services are available
Major Need Program All Most Some None Missing/ Don’t Know Mean Always Usually Sometimes Seldom Missing/ Don’t Know Mean on-site off-site not available
Food Faith-Based Non-profit 49.0 26.1 17.4 3.6 3.8 3.17 54.3 27.0 8.9 1.4 8.5 3.17 85.2 37.3 0.1
Secular Non-profit 39.3 29.0 21.0 6.9 3.9 2.97 53.1 23.6 10.0 1.8 11.4 3.05 75.4 44.0 0.0
Clothing Faith-Based Non-profit 13.1 38.0 34.3 6.8 7.8 2.49 33.5 32.0 14.5 4.1 15.8 2.63 67.8 41.4 0.4
Secular Non-profit 9.6 27.4 49.9 6.3 6.8 2.33 26.6 33.7 22.5 2.5 14.8 2.55 61.0 54.2 0.0
Money mgmt Faith-Based Non-profit 23.2 31.9 15.6 8.8 20.5 2.49 13.6 15.3 18.0 22.9 30.2 1.59 29.8 40.6 10.5
Secular Non-profit 23.2 30.7 22.8 5.6 17.8 2.54 20.2 22.6 20.7 12.9 23.6 2.03 52.8 39.4 7.2
Conflict resolution Faith-Based Non-profit 13.6 28.0 29.1 6.5 22.8 2.26 11.6 19.9 23.8 12.2 32.4 1.66 36.8 43.9 4.7
Secular Non-profit 20.1 31.0 22.8 5.3 20.8 2.45 23.8 25.0 18.0 6.1 27.1 2.12 51.9 46.0 2.6
Needs assmt Faith-Based Non-profit 30.4 28.0 17.8 8.1 15.8 2.65 20.4 18.8 23.6 10.8 26.4 1.96 40.6 52.2 2.7
Secular Non-profit 50.6 26.6 12.7 4.1 6.0 3.18 45.9 24.9 12.4 5.9 10.8 2.89 74.0 39.4 2.0
Ind. Goals Faith-Based Non-profit 26.0 28.3 17.0 9.6 19.2 2.52 16.1 16.7 21.0 14.1 32.1 1.70 33.3 47.5 5.4
Secular Non-profit 47.8 27.7 11.5 5.0 8.1 3.10 41.6 21.0 15.1 8.9 13.4 2.68 69.5 34.1 5.3
Referral Faith-Based Non-profit 12.5 27.3 32.8 6.3 21.1 2.25 15.0 26.1 22.4 7.5 29.2 1.90 40.3 50.5 1.7
Secular Non-profit 26.1 36.7 25.4 5.0 6.9 2.77 38.6 31.5 14.2 2.2 13.4 2.80 72.1 48.1 0.4
Follow-up Faith-Based Non-profit 13.2 22.1 28.3 15.1 21.2 2.12 6.9 14.2 22.4 19.1 37.4 1.34 34.1 34.5 8.8
Secular Non-profit 24.1 26.2 29.1 11.5 9.1 2.54 22.3 23.1 22.9 9.8 21.9 2.14 60.4 33.4 4.2
Locate housing Faith-Based Non-profit 20.3 29.0 30.5 6.8 13.5 2.49 9.9 22.5 32.4 13.0 22.2 1.85 38.9 60.0 4.5
Secular Non-profit 31.4 33.0 25.1 5.5 5.1 2.85 18.2 31.2 31.4 8.4 10.9 2.37 64.5 57.8 3.3
Financial assistance Faith-Based Non-profit 10.2 31.9 33.9 8.4 15.7 2.28 5.4 21.1 32.4 14.8 26.3 1.64 37.2 59.2 3.5
Secular Non-profit 23.9 33.7 27.8 8.5 6.0 2.67 13.2 29.6 30.3 11.0 16.0 2.13 45.9 60.8 3.3
Job skills assmt Faith-Based Non-profit 10.3 32.1 29.1 9.3 19.2 2.24 7.0 14.6 30.9 17.8 29.8 1.51 19.7 58.5 5.7
Secular Non-profit 14.6 35.2 32.9 10.0 7.4 2.47 13.2 26.5 30.2 12.1 18.1 2.05 33.5 64.7 2.6
Health assmt Faith-Based Non-profit 12.2 23.0 35.4 9.1 20.4 2.18 9.5 16.5 30.9 11.8 31.3 1.61 18.2 57.8 3.8
Secular Non-profit 23.3 27.6 28.8 9.7 10.7 2.54 20.7 22.9 23.6 12.0 20.9 2.11 28.8 61.1 4.8

Source:  Urban Institute analysis of NSHAPC program data.  Data represent "an average day in February 1996."


Administrators of faith-based programs do not know the service needs of many of their clients.

Looking at all twelve service needs, it should first be noted that program administrators do not always know about the needs of their clients.  Responses in the “missing/don’t know” category range from about 4 percent for basic needs such as food and clothing to more than 20 percent for needs like “conflict resolution” and “referral” among clients of faith-based agencies.  Note also that faith-based administrators are much more likely than secular non-profit administrators to report not knowing about specific service needs of their clients.

Only for food and clothing, are faith-based programs meeting needs more often than secular non-profit programs.

Once again, secular non-profits generally report higher levels of need for most services than do faith-based non-profits.  The findings on how often client needs are met mirror those for service needs:  only when it comes to food and clothing are faith-based programs meeting needs more often than secular non-profit programs.  For most other services, secular non-profits usually or always meet client needs much more often than do faith-based programs.  Again, it is important to keep in mind that large shares of faith-based administrators simply do not know if client needs are being met.

In general, these results suggest that the administrators of secular programs are more likely to know about their clients, their clients are more likely to have high levels of need for most services, and their clients are more likely to have these needs met.  With the exceptions of food and clothing, services are available on-site more frequently through secular non-profit programs rather than through faith-based programs.  About 85 percent and 68 percent of faith-based programs report that food and clothing, respectively, are available on-site, compared to 75 percent and 61 percent of secular non-profits.  No other service, however, is offered on-site by more than 40 percent of faith-based programs.  Secular programs, by contrast, seem to offer a wide variety of services on-site.  For example, the share of secular versus faith-based non-profits offering various services is as follows:  needs assessments (74 versus 41 percent), referrals (72 versus 40 percent), and help with locating housing (65 versus 39 percent).  For off-site services, similar shares of each type of program report specific services being available, with few significant differences.

Unlike secular non-profits, faith-based agencies are more likely to provide basic services such as food and clothing, and less likely to provide other types of services.  Faith-based programs are less likely to report a high level of need for all other services, are less likely to meet most clients’ needs, and are less likely to offer these services on-site.  However, the vast majority of faith-based programs report high levels of need for food and clothing, most programs are consistently able to meet these two needs, and most also offer services to meet these needs on-site.

Clients of central city programs have greater needs than those of suburban and rural programs.

The same patterns of client needs, needs being met, and availability of services are present when programs are broken down by urban-rural status and by region of the country.  Generally, administrators of central city programs report higher levels of client need and those of suburban and rural programs report slightly lower levels of need.  Needs are being met at about the same rates for most services, although a few individual services, such as financial assistance, seem to be emphasized in rural areas more so than in central cities.  Looking at regions of the country, there is quite a bit of variation in administrators’ perceptions of client needs and availability of services, but there is no discernible pattern to these differences.  Among programs in the northeast, client needs are slightly lower than the national figures, and faith-based administrators report that these needs are met less often than is the case for all programs nationally, while secular non-profits report that needs are met slightly more often.  Also, both faith-based and secular programs in the northeast are less likely to offer basic services such as food, clothing, and financial assistance on-site, but are more likely to offer other specific services such as conflict resolution, individual goals, and health assessments. 

Programs in the south seem to meet client needs slightly more often than nationally, and they also offer basic services (food, clothing, financial assistance) on-site more frequently.  In the mid-west, there are no discernible differences, except that programs also tend to offer services on-site more frequently than the national figures.  Finally, programs in the west tend to emphasize different client needs and are able to meet different needs as well, although there is no pattern as to which needs are more prevalent and which are being met more frequently.  Programs in the west also tend to offer services on-site less frequently than nationally.

Referrals to and from Programs

The CATI survey asked respondents where most of their clients came from — referrals from other programs, self-referral, outreach by program staff, or other.21 Respondents were also allowed to report that there was no single source from which most of their clients came.  Responses to this question are reported in Table 9 for faith-based programs and Table 10 for secular non-profits, and in Figure 5 for both types of sponsoring agencies.  Over one-third of non-profits (both faith-based and secular) did not identify a specific client source, either because there was no single source from which most of their clients came, because there was some “other” source, or because they simply did not answer the question.

Table 9:
Client Referrals Among Faith-Based Non-profit Programs
  Total Number of programs From other program Self-referred Program staff outreach Other source or missing
All Program Types 12,599 (100%) 26.0 34.3 4.2 35.4
  Housing 3,783 (100%) 40.4 27.2 2.1 30.3
    Emergency Shelter 1,520 (100%) 28.4 34.9 0.9 35.7
    Transitional Shelter 1,181 (100%) 58.7 10.5 3.6 27.2
    Permanent Housing 205 (100%) 49.0 17.1 10.9 23.0
    Distribute Vouchers 743 (100%) 24.2 45.6 0.0 30.2
    Housing For Vouchers 134 (100%) 91.2 2.3 0.0 6.5
  Food 6,907 (100%) 20.5 39.1 3.9 36.6
    Soup Kitchen/Meal Distribution 2,131 (100%) 10.4 54.9 3.9 30.7
    Food Pantry 4,628 (100%) 25.3 32.2 2.5 40.0
    Mobile Food 148 (100%) 12.1 25.7 47.6 14.6
  Health 131 (100%) 22.6 14.2 11.0 52.1
    Physical Health Care .   . . . .
    Mental Health .   . . . .
    Alcohol or Drug .   . . . .
    HIV/AIDS .   . . . .
  Other 1,778 (100%) 17.3 32.5 9.7 40.4
    Outreach 505 (100%) 21.9 16.1 28.5 33.5
    Drop-In Center 450 (100%) 15.1 40.8 2.5 41.7
    Financial/Housing Assist. 277 (100%) 12.7 42.6 0.0 44.7
    Other 546 (100%) 17.3 35.9 3.3 43.5

Source:  Urban Institute analysis of NSHAPC program data.  Data represent "an average day in February 1996."


Table 10:
Client Referrals Among Secular Non-profit Programs

  Total Number of programs From other program Self-referred Program staff outreach Other source or missing
All Program Types 18,751 (100%) 35.2 23.0 7.7 34.1
  Housing 8,664 (100%) 46.3 17.4 3.9 32.5
    Emergency Shelter 3,480 (100%) 39.4 19.4 2.3 38.9
    Transitional Shelter 2,535 (100%) 57.3 12.5 4.6 25.6
    Permanent Housing 980 (100%) 44.1 9.5 6.8 39.6
    Distribute Vouchers 1,361 (100%) 41.1 27.5 3.5 27.9
    Housing For Vouchers 307 (100%) 63.1 13.9 8.5 14.4
  Food 4,858 (100%) 28.6 31.2 6.8 33.4
    Soup Kitchen/Meal Distribution 1,057 (100%) 17.0 46.8 5.5 30.7
    Food Pantry 3,560 (100%) 31.9 27.9 6.1 34.0
    Mobile Food 241 (100%) 31.0 10.1 22.7 36.1
  Health 1,034 (100%) 29.6 21.9 13.3 35.2
    Physical Health Care 215 (100%) 14.7 46.9 11.7 26.8
    Mental Health 250 (100%) 43.5 18.6 15.4 22.5
    Alcohol or Drug 363 (100%) 28.3 17.1 10.8 43.8
    HIV/AIDS 206 (100%) 30.4 8.3 17.0 44.3
  Other 4,195 (100%) 21.3 25.7 15.3 37.8
    Outreach 1,922 (100%) 22.0 19.2 26.9 32.0
    Drop-In Center 1,083 (100%) 17.8 27.6 6.8 47.7
    Financial/Housing Assist. 452 (100%) 12.6 40.5 2.9 44.0
    Other 738 (100%) 29.7 30.7 5.1 34.5

Source:  Urban Institute analysis of NSHAPC program data.  Data represent "an average day in February 1996."


Figure 5:
Referral Sources Among Faith-Based and Secular Non-Profits

Figure 5a: Referral Sources Among Faith-Based and Secular Non-Profits

Figure 5b: Referral Sources Among Faith-Based and Secular Non-Profits

The largest source of clients for faith-based programs is self-referrals.

Some interesting patterns emerge from the remaining respondents who did identify how clients came to their programs.  For programs of all types, the data show that the largest source of clients for faith-based programs is self-referral (34 percent identify this as their largest source) but for secular non-profits the most common source is through another program (35 percent report this as the largest source).  There are also some interesting differences among specific types of programs.  In general, for both faith-based and secular non-profits, clients of housing programs are more likely to come to the program through referral from another program while clients of food programs are more likely to come to the program on their own (self-refer).  The third main route to programs and services — outreach by program staff — seems to be a significant source of clients of mobile food and (not surprisingly) outreach programs.  While this is true of both faith-based and secular non-profits, mobile food programs run by faith-based organizations are much more likely to receive clients through outreach and self-referral (48 and 26 percent, respectively) than are mobile food programs run by secular non-profits (23 and 10 percent, respectively).

Clients of homeless assistance programs in rural areas are much more likely to self-refer.

Looking at client sources for faith-based and secular non-profits by urban-rural location and region of the country shows very similar patterns.  There are some clear differences by geography, with clients of rural programs being much more likely than those of central city and especially suburban programs to self-refer, but the differences between faith-based and secular providers are fairly consistent across these geographical patterns.  The same is generally true of regional differences.  It is interesting to note, however, that differences in self-referrals between faith-based and secular providers are especially large in the northeast and south.  Only 18 percent of administrators of secular non-profit programs in the south identify self-referrals as their main source of clients, compared to 33 percent of administrators of faith-based programs.

In addition to sources of clients, administrators of housing programs were asked where their clients went after leaving the program.  More specifically, the NSHAPC mail survey asked administrators of housing programs the following question: “we are interested in where your families go when they are no longer served by your [emergency shelter] program.  For what percentage of your families do you know this information?  For these families, please estimate the percent that went to the following destinations on departure from your [emergency shelter] program last year: the streets or other outside locations, other emergency shelter, transitional housing, family or friend’s housing, private unsubsidized housing, government subsidized housing (e.g., Section 8, Public or Rural Rental Housing), special permanent housing for the disabled (mentally ill, developmentally disadvantaged, HIV), other group home, hospital, jail or prison, and other (specify).”  Separate but similarly worded questions were asked for unaccompanied clients.22  The answers to these questions are reported in Table 11 for all types of homeless assistance housing programs combined, Table 12 for emergency shelters alone, and Table 13 for transitional housing programs alone.  Administrators of housing programs tend to know where their clients go after leaving their program.  On average, administrators of faith-based housing programs know the destinations of about 75 percent of their unaccompanied individual clients and 90 percent of their family clients.  The corresponding figures for secular non-profit housing programs are 86 percent of unaccompanied individuals and 90 percent of families.  The top three destinations are the same irrespective of whether the client is alone or with children and whether the program is run by a faith-based or secular non-profit agency.  These are (1) a family or friend’s housing, (2) private unsubsidized housing, and (3) government subsidized housing.  In general, clients of secular non-profits, especially family clients, are more likely than other housing clients go to one of these places.  For example, 75 percent of families in secular non-profit housing programs (with known destinations) go to one of these destinations, compared to 66 percent of family clients of faith-based housing programs, and only 57 percent of individual clients of secular non-profit housing programs.  Over one-third of family clients of secular non-profits leave the program for government subsidized housing (the largest single destination of these clients), but a similarly large share (29 percent) of family clients of faith-based housing go here as well.  The second most common destination of families for both faith-based and secular providers is private unsubsidized housing, with about 20 to 22 percent of families going here.  Not surprisingly given the priority given to families, individual clients of both faith-based and secular housing programs are much less likely than families to go into government subsidized housing (13-14 percent versus 29-34 percent).

Table 11:
Client Referrals from Faith-Based and Secular Non-profit Programs
  Percentage of Clients Moving on to…
Programs Serving Individuals Programs Serving Families
Faith-Based Nonprofit Secular Nonprofit Faith-Based Nonprofit Secular Nonprofit
median percentage known 75 86 90 90
The streets or other outside locations 12.7 8.7 8.0 4.3
Other emergency shelter 11.6 8.9 9.6 6.8
Transitional housing 9.7 7.3 10.1 6.6
Family or friend's housing 17.6 25.1 17.1 19.7
Private unsubsidized housing 21.6 17.7 20.0 22.4
Government subsidized housing 13.0 14.3 29.0 33.6
Special permanent housing for disabled 3.4 3.7 2.1 1.8
Other group home 2.5 3.0 1.8 0.7
Hospital 1.4 2.0 0.5 1.0
Jail or prison 3.0 2.0 1.0 0.7
Other 3.5 8.1 1.2 5.2

Source:  Urban Institute analysis of NSHAPC program data.  Data represent "an average day in February 1996."

Note:  Because some programs reported referrals for more than 100% of clients, columns may not sum to 100%.


Table 12:
Client Referrals from Faith-Based and Secular Non-profit Emergency Shelter Programs
  Percentage of Clients Moving on to…
Programs Serving Individuals Programs Serving Families
Faith-Based Nonprofit Secular Nonprofit Faith-Based Nonprofit Secular Nonprofit
median percentage known 74 80 80 84
The streets or other outside locations 15.7 7.8 12.0 5.0
Other emergency shelter 13.8 12.1 15.1 10.1
Transitional housing 11.2 6.4 14.1 8.4
Family or friend's housing 15.5 30.7 16.2 21.7
Private unsubsidized housing 17.6 17.0 13.2 20.1
Government subsidized housing 11.5 10.3 23.5 30.1
Special permanent housing for disabled 3.2 1.5 1.7 0.9
Other group home 2.6 2.7 2.0 0.9
Hospital 1.2 1.8 0.7 0.8
Jail or prison 1.9 2.0 1.1 0.9
Other 3.3 8.0 1.1 5.2

Source:  Urban Institute analysis of NSHAPC program data.  Data represent "an average day in February 1996."

Note:  Because some programs reported referrals for more than 100% of clients, columns may not sum to 100%.


Table 13:
Client Referrals from Faith-Based and Secular Non-profit Transitional Housing Programs
  Percentage of Clients Moving on to…
Programs Serving Individuals Programs Serving Families
Faith-Based Nonprofit Secular Nonprofit Faith-Based Nonprofit Secular Nonprofit
median percentage known 80 90 90 97.5
The streets or other outside locations 8.5 10.8 2.6 3.6
Other emergency shelter 9.0 4.9 1.7 2.0
Transitional housing 7.4 9.6 4.7 4.5
Family or friend's housing 22.4 20.1 18.9 16.1
Private unsubsidized housing 29.3 20.2 29.9 27.1
Government subsidized housing 14.3 18.0 35.7 40.9
Special permanent housing for disabled 2.8 5.8 2.5 2.5
Other group home 1.8 3.7 1.4 0.3
Hospital 1.0 1.8 0.3 0.5
Jail or prison 4.8 1.0 0.8 0.1
Other 1.8 4.7 1.5 2.8

Source:  Urban Institute analysis of NSHAPC program data.  Data represent "an average day in February 1996."

Note:  Because some programs reported referrals for more than 100% of clients, columns may not sum to 100%.


Individual clients of faith-based emergency shelters are more likely to leave the shelter for the streets and other outside locations.

A closer examination of emergency and transitional shelters alone reveals that the destinations of emergency shelter clients are much more varied than those of transitional shelter clients.  Even among emergency shelter programs, clients of faith-based programs are more likely to go to many different destinations than are clients of secular non-profit emergency shelters (see Table 12).  Individuals leaving faith-based emergency shelter programs are more likely than those in secular programs to go into transitional housing and to the streets or other outside locations, and they are less likely to go to a family or friend’s house.  Compared to families served by secular non-profits, family clients of faith-based emergency shelter programs are more likely to go to another emergency or transitional shelter or an outside location, and they are less likely to go to private or government housing or even the home of a friend or family member.  Other notable findings from Table 12 are that faith-based and secular emergency shelters report almost equal shares of individual clients leaving their programs for private unsubsidized housing (about 17 percent), and for families the single most common destination for clients is the same: government subsidized housing (the destination of 30 percent of families in secular emergency shelters and 24 percent of families in faith-based emergency shelters).

The destinations of clients of transitional housing programs are somewhat less varied than those of clients of emergency shelters (see Table 13).  This may be due to the greater stability (economically and otherwise) of transitional housing program clients compared to their emergency shelter counterparts.  For both faith-based and secular transitional housing programs, the top destination of individual clients is private unsubsidized housing (29 percent of faith-based and 20 percent of secular clients) and for family clients, the top destination is government subsidized housing (36 percent of faith-based and 41 percent of secular non-profit clients).  The second and third most common destinations are also the same for both types of agencies, for individual clients they are family/friend’s housing and government subsidized housing, and for family clients they are private unsubsidized housing and family/friend’s housing.  In general, the most striking feature of Table 13 is how similar faith-based and secular non-profits are in terms of where the clients of their transitional housing programs go after leaving the program.


21.  These questions cover all clients of homeless assistance programs, not just currently homeless ones.

22.  Questions in this section of the survey pertain to all clients of the given homeless assistance program, as opposed to only currently homeless clients.

Government Funding of Homeless Assistance Programs

Revenue streams in non-profit organizations are “a complex mix of private and public dollars raised through grants, contracts, fees for services, sales, donations, investment income, special events income, and income from commercial ventures” (Boris 1998) and NSHAPC collected only very limited funding information from service providers.  The CATI portion of NSHAPC asked respondents two basic questions about funding: “what percentage of your current funding for the [emergency shelter] program comes from private funding such as individual contributions, foundation or corporate grants, United Way, funding from religious organizations or churches or other private sources?” and “what percentage of [your] current funding for the [emergency shelter] program comes from federal, state, or local government?”  The two responses had to sum to 100 percent.23  A final funding question, with an open-ended response, asked the respondent “from what agency does this [federal, state, or local government] funding come?” but responses to this last question were very limited and, not surprisingly, many local providers are unable to distinguish between government funds coming from state versus federal agencies, or between different agencies within the federal government.

The majority of faith-based programs receive no government funding at all.

The share of program funds coming from government sources of any type are reported in Tables 14 and 15 and Figure 6.  There are significant differences between faith-based and secular non-profit homeless assistance programs in levels of government funding.  The majority of faith-based programs (62 percent) receive no government funding at all and close to 90 percent receive less that one-half of their funding from government sources.  Among secular non-profits, less than one-quarter (23 percent) receive no government funds and only 40 percent receive less than one-half of their funds from government agencies.  Similarly large differences are found when one looks at the share of programs that are fully funded by the government.  Twenty-two percent of secular non-profits rely on government funds exclusively, compared to less than 3 percent of faith-based programs.  To some extent, differences between faith-based and secular non-profits in levels of government funding are affected by the types of programs the two groups of providers tend to run.  Faith-based groups run the majority of food programs and these are much less likely than housing programs (the majority of which are run by secular non-profits) to receive government funding.

Table 14:
Percent of Funding from Government Sources Among Faith-Based Non-profits
  Number of Programs Percent of Funding
0 1-24 25-49 50-74 75-99 100
All Program Types 11,902 (100%) 61.5 17.9 8.6 5.3 4.1 2.7
  Housing 3,587 (100%) 55.7 17.3 10.0 6.4 6.1 4.5
    Emergency Shelter 1,436 (100%) 54.1 21.8 9.9 7.5 4.7 2.0
    Transitional Shelter 1,116 (100%) 57.2 15.6 9.4 5.5 9.7 2.6
    Permanent Housing 190 (100%) 53.1 5.7 4.6 16.4 8.5 11.6
    Distribute Vouchers 714 (100%) 61.5 9.9 12.6 3.1 3.1 9.8
    Housing For Vouchers 131 (100%) 32.6 39.9 10.5 4.4 2.3 10.4
  Food 6,495 (100%) 60.7 21.2 8.8 5.2 2.7 1.4
    Soup Kitchen/Meal Distribution 2,056 (100%) 60.4 22.2 8.1 3.7 4.0 1.5
    Food Pantry 4,294 (100%) 60.3 20.9 9.4 6.0 2.0 1.4
    Mobile Food 145 (100%) 76.9 13.3 3.2 5.1 1.5 0.0
  Health 126 (100%) 78.4 1.7 2.5 3.5 10.6 3.3
    Physical Health Care .   . . . . . .
    Mental Health .   . . . . . .
    Alcohol or Drug .   . . . . . .
    HIV/AIDS .   . . . . . .
  Other 1,693 (100%) 75.8 7.5 5.0 3.3 4.7 3.6
    Outreach 494 (100%) 65.8 13.4 3.7 3.3 5.0 8.9
    Drop-In Center 425 (100%) 75.7 7.0 7.2 1.2 5.4 3.5
    Financial/Housing Assist. 256 (100%) 64.9 1.2 13.2 10.9 9.0 0.8
    Other 518 (100%) 91.0 5.5 0.4 1.3 1.6 0.2

Source:  Urban Institute analysis of NSHAPC program data.  Data represent "an average day in February 1996."


Table 15:
Percent of Funding from Government Sources Among Secular Non-profit Programs

  Number of Programs Percent of Funding
0 1-24 25-49 50-74 75-99 100
All Program Types 17,271 (100%) 23.2 7.6 9.0 15.1 23.2 21.9
  Housing 8,016 (100%) 13.0 6.3 9.9 19.6 28.6 22.4
    Emergency Shelter 3,365 (100%) 8.3 6.5 11.9 27.6 36.4 9.3
    Transitional Shelter 2,321 (100%) 11.2 8.0 11.6 18.5 30.4 20.2
    Permanent Housing 851 (100%) 15.4 4.4 6.0 11.1 22.5 40.5
    Distribute Vouchers 1,192 (100%) 27.8 4.7 4.6 5.3 11.1 46.4
    Housing For Vouchers 287 (100%) 13.3 3.4 6.9 20.3 13.9 42.2
  Food 4,430 (100%) 40.8 14.7 10.1 14.4 9.2 10.7
    Soup Kitchen/Meal Distribution 979 (100%) 25.2 11.1 11.8 17.5 9.0 25.4
    Food Pantry 3,277 (100%) 46.4 16.0 10.1 12.1 9.6 5.9
    Mobile Food 174 (100%) 23.7 12.3 1.8 40.0 2.4 19.8
  Health 914 (100%) 17.3 1.4 3.4 10.9 31.5 35.6
    Physical Health Care 198 (100%) 26.9 2.4 3.4 22.6 29.9 14.8
    Mental Health 225 (100%) 7.9 1.4 6.1 5.1 33.5 46.0
    Alcohol or Drug 320 (100%) 20.3 0.7 1.2 4.6 30.1 43.1
    HIV/AIDS 170 (100%) 12.9 1.8 3.6 16.7 33.3 31.8
  Other 3,911 (100%) 25.6 3.5 7.1 7.6 25.8 30.3
    Outreach 1,722 (100%) 14.6 3.0 6.2 8.5 30.2 37.5
    Drop-In Center 1,067 (100%) 33.7 3.2 8.9 7.6 21.2 25.3
    Financial/Housing Assist. 403 (100%) 43.4 1.6 2.2 2.0 27.2 23.6
    Other 719 (100%) 30.2 6.5 9.6 8.4 21.3 24.0

Source:  Urban Institute analysis of NSHAPC program data.  Data represent "an average day in February 1996."


Figure 6:
Percentage of Funding From Government Sources

Figure 6: Percentage of Funding From Government Sources


23.  Note that NSHAPC did not ask about in-kind contributions, such as the free use of space or buildings, donated food, volunteers’ time, etc.  In-kind contributions (which can come from both government and private sources) can account for a significant share of a program’s resources.

Conclusions

Faith-based and other community-based non-profit organizations have a long history of helping people in need.  The NSHAPC data analyzed here provide yet more evidence of the continuing importance of faith-based organizations in serving people who are homeless or on the brink of homelessness.  Faith-based non-profits run the majority of homeless assistance food programs in this country, including over 60 percent of soup kitchens and over one-half of all food pantries.  But their contributions are not limited to food assistance.  Faith-based providers oversee about one-quarter of all emergency and transitional shelters and voucher distribution programs and more than one-quarter of all drop-in centers across the country.  Secular non-profits are also a critical component of the nation’s homeless service delivery system.  In addition to running the majority of shelters and permanent housing programs, they administer over one-third of all food programs, including close to 40 percent of all soup kitchens and one-half of all mobile food programs.  These providers also oversee almost 40 percent of all health programs, including almost one-half of all alcohol and drug programs and HIV/AIDS programs.  Finally, around 60 percent of outreach and drop-in center programs are run by secular non-profits.

Faith-based programs play a much larger role in urban areas (both central city and suburban) than they do in rural areas.  This is true for all types of programs taken together as well as for specific types of programs.  Government agencies, by contrast, play a critical role in delivering health services in rural areas and a much smaller role in central cities and suburban areas.  With the exceptions of food programs (which are dominated by faith-based providers in urban areas) and health programs (which are dominated by government agencies in rural areas), secular non-profits tend to be much more consistent in terms of the types of programs they administer in urban versus rural areas.  Looking at these data by region reveals that compared to the west or northeast, faith-based organizations appear to be responsible for a greater share of programs in the south (largely due to their increased involvement in housing and “other” types of homeless assistance programs).

In general, compared to programs run by secular non-profits, faith-based service providers of all types serve a wider variety of clients (they are especially more likely to serve single men) and are less likely to have some type of special focus (on a group of clients such as families or youth, or on a group with special needs such as victims of domestic violence or people with HIV/AIDS).  They also report lower levels of specific types of needs among their clients.  This last finding may in fact be the result of the first two.  If the clients of faith-based agencies are more diverse, then their needs are also likely to be more diverse (and there will be lower levels of any one need).  Faith-based agencies may also be serving more clients who are not literally homeless.  Finally, faith-based non-profits are much less likely than their secular counterparts to receive government funding.  The majority receive no government funds at all, while less than one-quarter of secular non-profits do.  Almost a quarter of secular non-profits rely on government funds exclusively, compared to less than 3 percent of faith-based programs.

In reviewing these findings it is important to understand that while there may be true differences in the characteristics of faith-based and secular non-profit service providers, many of the observed overall differences are likely to be driven by the types of programs the two groups of agencies tend to run.  Faith-based service providers oversee over half of all food programs, while secular non-profits run a similarly large share of homeless housing programs.  Food programs by their very nature, and especially as compared to housing programs, are likely to serve larger numbers and types of clients, not have a special focus, and receive less government funding.  As this analysis has shown, however, there are some interesting differences between faith-based and secular non-profits even within a single type of program, like food or housing programs.  One should also remember that the NSHAPC data reflect the homeless assistance system of 1996 and do not, therefore, reflect events such as the terrorist attacks of September 11, 2001 that may have resulted in increasing demands on charitable organizations.  Future research on homeless assistance programs should examine whether the clients of faith-based, secular non-profit, and government-run agencies differ in any fundamental ways and whether the relationships between agencies and their clients vary by type of program or administering agency.  More work is also needed on how different types of agencies choose their focus (in response to current funding streams, the agency’s basic mission, assessments of needs within the community, etc.), and on the effectiveness of social service programs run by faith-based organizations.  Finally, it would also be useful to know if clients are even aware of the faith-based versus secular status of non-profit agencies, and if so, whether (and why) they prefer one over the other.

The American public is supportive of the general idea of government funding of social services provided by faith-based groups, but this support diminishes when they are given specific details about how this might work (Morin 2001).  There are still many ways that government can support the social service components of faith-based organizations (see Appendix B), and only a few of these raise unresolved legal objections.  To the extent that faith-based agencies are already eligible for government grants and other forms of support and are wanting additional government funds, any unnecessary barriers need to be identified and removed.  These might include improving their knowledge and understanding of government funding opportunities, and assisting them in developing their financial, administrative, and managerial capacities. 

Finally, many observers believe that the country has done an adequate job of building up an emergency response system for homeless people and must now go beyond this by focusing on prevention and longer-lasting housing and support services (National Alliance to End Homelessness 2000, Burt 2001).  Faith-based providers have a very important role to play in this.  They have missions that extend beyond homeless assistance and may have much longer-term (possibly lifelong) relationships with members of their communities.  Secular non-profits may be more narrowly focused on homeless assistance.  Adequate and affordable housing, a living wage, and critical support services such as childcare and substance abuse treatment, are key to reducing homelessness.  But more basic support services are likely to remain a key ingredient in helping prevent poor people from becoming homeless and ensuring that those who do become homeless do so only once and for a short period of time.  The NSHAPC data make clear that many faith-based organizations are already providing very high levels of such assistance.

Appendices

Appendix A: Types of Programs Included in NSHAPC

NSHAPC covered 16 different types of homeless assistance programs, defined as follows:

  1. Emergency shelter programs provide short-term housing on a first-come first-served basis where people must leave in the morning and have no guaranteed bed for the next night OR provide beds for a specified period of time, regardless of whether or not people leave the building.  Facilities which provide temporary shelter during extremely cold weather (such as churches) and emergency shelters or host homes for runaway or neglected children and youth, and victims of domestic violence were also included.
  2. Transitional housing programs have a maximum stay for clients of two years and offer support services to promote self-sufficiency and to help them obtain permanent housing. They may target any homeless sub-population such as persons with mental illnesses, persons with AIDS, runaway youths, victims of domestic violence, homeless veterans, etc.
  3. Permanent housing programs for homeless people provide long-term housing assistance with support services for which homelessness is a primary requirement for program eligibility.  Examples include the Shelter Plus Care Program, the Section 8 Moderate Rehabilitation Program for Single-Room Occupancy (SRO) Dwellings, and the Permanent Housing for the Handicapped Homeless Program administered by the Department of Housing and Urban Development (HUD).  These programs also include specific set-asides  of assisted housing units or housing vouchers for homeless persons by public housing agencies or others as a matter of policy, or in connection with a specific program (e.g., the HUD-VA Supported Housing Program, "HUD-VASH").  A permanent housing program for homeless people does NOT include public housing, Section 8, or federal, state, or local housing assistance programs for low-income persons that do not include a specific set-aside for homeless persons, or for which homelessness is not a basic eligibility requirement.
  4. Voucher distribution programs provide homeless persons with a voucher, certificate, or coupon that can be redeemed to pay for a specific amount of time in a hotel, motel, or other similar facility.
  5. Programs that accept vouchers for temporary accommodation provide homeless persons with accommodation, usually in a hotel, motel, board and care, or other for-profit facility, in exchange for a voucher, certificate, or coupon offered by a homeless assistance program.
  6. Food pantry programs are programs which distribute uncooked food in boxes or bags directly to low income people, including homeless people.
  7. Soup kitchen programs include soup kitchens, food lines, and programs distributing prepared breakfasts, lunches, or dinners.  These programs may be organized as food service lines, bag or box lunches, or tables where people are seated, then served by program personnel.  These programs may or may not have a place to sit and eat the meal.
  8. Mobile food programs are programs which visit designated street locations for the primary purpose of providing food to homeless people.
  9. Physical health care programs provide health care to homeless persons, including health screenings, immunizations, treatment for acute health problems, and other services that address physical health issues.  Services are often provided in shelters, soup kitchens, or other programs frequented by homeless people.
  10. Mental health care programs provide services for homeless persons to improve their  mental or psychological health or their ability to function well on a day-to-day basis.  Specific services may include case management, assertive community treatment, intervention or hospitalization during a moment of crisis, counseling, psychotherapy, psychiatric services, and psychiatric medication monitoring.
  11. Alcohol/drug programs provide services to assist a homeless individual to reduce his/her level of alcohol or other drug addiction, or to prevent substance abuse among homeless persons.  This may include services such as detoxification services, sobering facilities, rehabilitation programs, counseling, treatment, and prevention and education services.
  12. HIV/AIDS programs provide services for homeless persons where the services provided specifically respond to the fact that clients have HIV/AIDS, or are at risk of getting HIV/AIDS.  Services may include health assessment, adult day care, nutritional services, medications, intensive medical care when required, health, mental health, and substance abuse services, referral to other benefits and services, and HIV/AIDS prevention and education services.
  13. Drop-in center programs provide daytime services primarily for homeless persons such as television, laundry facilities, showers, support groups, and service referrals, but do not provide overnight accommodations.
  14. Outreach programs contact homeless persons in settings such as on the streets, in subways, under bridges, and in parks to offer food, blankets, or other necessities; to assess needs and attempt to engage them in services; to offer medical, mental health, and/or substance abuse services; and/or to offer other assistance on a regular basis (at least once a week) for the purpose of improving their health, mental health, or social functioning, or increasing their  use of human services and resources.  Services may be provided during the day or at night.
  15. Migrant housing is housing that is seasonally occupied by migrating farm workers.  During off-season periods it may be vacant and available for use by homeless persons.
  16. Other programs: providers could describe other programs they offered, as long as the programs met the basic NSHAPC definition of a homeless assistance program.  Types of programs actually identified through the survey include housing/financial assistance (e.g., from Community Action, county welfare, or housing agencies); Emergency Food and Shelter Program agencies; job training for the homeless, clothing distribution, and other programs.
Appendix Table A1:
Number of NSHAPC Programs by Type of Agency Operating Programs
  Total Number of Programs Faith-Based Non-Profit Secular Non-Profit Government For-Profit Unidentified
All Program Types 39,664 12,599 18,751 5,324 243 2,747
  Housing 15,879 3,783 8,664 1,924 120 1,388
    Emergency Shelter 5,687 1,520 3,480 320 11 356
    Transitional Shelter 4,395 1,181 2,535 433 62 184
    Permanent Housing 1,918 205 980 534 17 181
    Distribute Vouchers 3,080 743 1,361 523 10 443
    Housing For Vouchers 799 134 307 114 20 224
  Food 13,003 6,907 4,858 645 50 542
    Soup Kitchen/Meal Distribution 3,484 2,131 1,057 95 3 197
    Food Pantry 9,028 4,628 3,560 548 42 249
    Mobile Food 491 148 241 1 5 96
  Health 2,739 131 1,034 1,241 17 317
    Physical Health Care 715 42 215 420 0 38
    Mental Health 801 9 250 519 4 18
    Alcohol or Drug 778 61 363 183 5 165
    HIV/AIDS 446 18 206 119 7 96
  Other 8,043 1,778 4,195 1,515 56 499
    Outreach 3,307 505 1,922 579 19 281
    Drop-In Center 1,790 450 1,083 129 5 123
    Financial/Housing Assist. 1,378 277 452 625 1 24
    Other 1,568 546 738 182 31 71
Source:  Urban Institute analysis of NSHAPC program data.  Data represent "an average day in February 1996."
Appendix Table A1a:
Number of NSHAPC Programs by Urban/Rural Status
  Total Number of Programs Faith-Based Non-Profit Secular Non-Profit Government For-Profit Unidentified
All Program Types 39,664 12,599 18,751 5,324 243 2,747
Central Cities
All 19,388 7,143 8,902 1,915 133 1,294
Housing 7,894 2,263 4,244 759 62 565
Food 6,018 3,815 1,704 157 10 331
Health 1,379 104 783 402 10 80
Other 4,097 961 2,170 598 50 317
Suburbs
All 7,694 2,702 3,696 571 82 642
Housing 3,230 781 1,731 281 58 380
Food 3,020 1,601 1,208 80 12 119
Health 251 7 128 80 6 29
Other 1,192 312 630 131 5 114
Rural Areas
All 12,583 2,754 6,153 2,838 28 811
Housing 4,754 738 2,689 884 0 443
Food 3,965 1,491 1,946 408 28 92
Health 1,110 20 123 759 0 208
Other 2,754 505 1,395 787 0 68
Source:  Urban Institute analysis of NSHAPC program data.  Data represent "an average day in February 1996."
Appendix Table A1b:
Number of NSHAPC Programs by Region of the Country
  Total Number of Programs Faith-Based Non-Profit Secular Non-Profit Government For-Profit Unidentified
All Program Types 39,664 12,599 18,751 5,324 243 2,747
Northeast
All 7,097 2,032 3,807 718 40 500
Housing 2,870 472 1,759 369 17 254
Food 2,401 1,275 893 85 11 137
Health 306 20 211 43 2 29
Other 1,521 265 944 220 10 81
South
All 11,101 4,333 4,516 1,515 53 686
Housing 4,309 1,293 2,166 446 49 356
Food 4,113 2,391 1,326 252 1 142
Health 863 40 232 492 1 97
Other 1,817 608 791 325 2 91
Midwest
All 11,853 3,741 5,184 1,915 57 956
Housing 4,678 1,146 2,229 790 17 496
Food 3,945 2,155 1,352 266 31 142
Health 736 20 292 261 0 162
Other 2,494 419 1,311 598 9 157
West
All 9,333 2,404 5,097 1,162 92 580
Housing 3,892 825 2,447 312 38 270
Food 2,478 1,050 1,264 41 6 117
Health 816 49 283 439 14 30
Other 2,147 480 1,101 369 34 163
Source:  Urban Institute analysis of NSHAPC program data.  Data represent "an average day in February 1996."

Appendix B: Ways That Government Financially Supports Faith-Based Social Services

Luis Lugo, Director of the Religion Program at the Pew Charitable Trusts, has developed a nine-point scale in describing the variety of ways that the government can support faith-based agencies in organizations in their delivery of social welfare services.  The nine types of government support — ordered from high to low in terms of general public approval — are:(24)

  1. The president and other leaders trumpet the success of innovative and effective faith-based programs, encouraging citizens, corporations and foundations to increase their support for these efforts.
  2. Citizens and corporations directly support their favorite charities, including religiously affiliated nonprofits (i.e., 501(c)(3)s) and congregations, and receive a tax deduction.  A new study by Price-WaterhouseCoopers estimates that President Bush’s proposal to extend the charitable deduction to the 85 million taxpayers who do not itemize their taxes could stimulate an additional $14.6 billion a year in charitable giving, with the lion’s share going to religious organizations.
  3. Citizens and corporations directly support their favorite charities, including religious nonprofits and congregations, and receive a tax credit.  For example, President Bush is encouraging states to provide a tax credit (up to 50 percent of the first $500 for individuals and $1,000 for married couples) against state income or other taxes for donations to charities — whether secular or religious — that are battling poverty and its effects.  (Note: This proposal could become quite controversial if federal welfare dollars were to be used to offset the cost of these credits.)
  4. State and local government job training and juvenile delinquency programs have recruited volunteers from churches as mentors.  Conversely, a federal volunteer program, AmeriCorps, placed nearly 6,000 of the total 40,000 positions in 2000 in religious nonprofits such as the Catholic Network for Volunteer Service and the National Jewish Coalition for Literacy.
  5. Religiously affiliated nonprofits such as Lutheran Social Services and Catholic Charities USA have received billions of public dollars to run a variety of social service programs, including Head Start, emergency shelters, adoption services and refugee resettlement.
  6. Government provides both in-kind, non-cash assistance and formula grant support to religiously affiliated nonprofits.  In-kind assistance often is provided informally, for example, by allowing a welfare-to-work program to use a desk in the county welfare office and copying program brochures.  Formula grants designate money for specific resources, for instance, computers for qualified low-income housing projects, according to objective, non-discretionary criteria (usually, the number of clients served).  While these grants are made to 501(c)(3)s), both religious and secular, these organizations often redistribute funds to on-the-ground programs, including church-based social services.
  7. A large secular nonprofit such as Goodwill Industries with the administrative capacity and experience to work with the government signs a contract to provide social services, and in turn subcontracts some of the services to other organizations, including church-based ministries.
  8. Government provides clients with certificates or vouchers, and they in turn select the provider of their choice, including church-based social service ministries.  Vouchers are a restricted subsidy that falls between cash and direct government provision of services, and are currently used in higher education, child care, job training, housing and health care.
  9. The newest, and most controversial, option is made possible by the charitable choice provision of the 1996 Welfare Reform Act.  Charitable choice permits churches, synagogues and mosques as well as other pervasively religious organizations to compete for government contracts on the same basis as secular, non-governmental service providers, but prohibits the use of public funds for religious worship or proselytizing as well as discrimination among clients on the basis of religious belief.  However, congregations may continue to use religion as a criterion for personnel decisions, as under current law.
Appendix Table B1:
Unweighted Number of NSHAPC Programs by Type of Agency Operating Programs
  Total Number of Programs Faith-Based Non-Profit Secular Non-Profit Government For-Profit Unidentified
All Program Types 11,983 3,880 5,888 1,162 103 950
  Housing 5,035 1,236 2,811 497 64 427
    Emergency Shelter 1,692 524 974 91 12 91
    Transitional Shelter 1,728 425 1,056 141 20 86
    Permanent Housing 751 105 452 110 12 72
    Distribute Vouchers 572 147 229 91 3 102
    Housing For Vouchers 292 35 100 64 17 76
  Food 3,860 2,065 1,418 128 11 238
    Soup Kitchen/Meal Distribution 1,278 723 430 37 3 85
    Food Pantry 2,414 1,272 922 90 7 123
    Mobile Food 168 70 66 1 1 30
  Health 769 55 429 205 10 70
    Physical Health Care 168 11 93 48 0 16
    Mental Health 214 9 115 75 2 13
    Alcohol or Drug 210 26 119 42 5 18
    HIV/AIDS 177 9 102 40 3 23
  Other 2,319 524 1,230 332 18 215
    Outreach 1,113 210 615 174 8 106
    Drop-In Center 584 156 316 42 2 68
    Financial/Housing Assist. 151 42 54 46 1 8
    Other 471 116 245 70 7 33
Source:  Urban Institute analysis of NSHAPC program data.  Data represent "an average day in February 1996."
Appendix Table B1a:
Unweighted Number of NSHAPC Programs by Urban/Rural Status
  Total Number of Programs Faith-Based Non-Profit Secular Non-Profit Government For-Profit Unidentified
All Program Types 11,983 3,880 5,888 1,162 103 950
Central Cities
All 7,763 2,590 3,769 721 70 613
Housing 3,235 843 1,795 290 45 262
Food 2,385 1,348 798 75 6 158
Health 556 47 320 144 6 39
Other 1,587 352 856 212 13 154
Suburbs
All 3,778 1,173 1,912 358 32 303
Housing 1,608 358 913 171 19 147
Food 1,342 655 565 43 4 75
Health 186 7 103 47 4 25
Other 642 153 331 97 5 56
Rural Areas
All 441 117 207 83 1 33
Housing 191 35 103 36 0 17
Food 133 62 55 10 1 5
Health 27 1 6 14 0 6
Other 90 19 43 23 0 5
Source:  Urban Institute analysis of NSHAPC program data.  Data represent "an average day in February 1996."
Appendix Table B1b:
Unweighted Number of NSHAPC Programs by Region of the Country
  Total Number of Programs Faith-Based Non-Profit Secular Non-Profit Government For-Profit Unidentified
All Program Types 11,983 3,880 5,888 1,162 103 950
Northeast
All 3,090 908 1,640 294 19 229
Housing 1,308 248 793 153 12 102
Food 1,073 559 401 44 3 66
Health 183 10 123 30 2 18
Other 526 91 323 67 2 43
South
All 2,155 859 922 221 12 141
Housing 906 287 460 91 8 60
Food 704 433 214 15 1 41
Health 126 12 65 38 1 10
Other 419 127 183 77 2 30
Midwest
All 2,876 1,093 1,263 274 23 223
Housing 1,130 320 586 109 17 98
Food 1,056 619 344 35 4 54
Health 147 13 69 48 0 17
Other 543 141 264 82 2 54
West
All 3,660 962 1,957 359 48 334
Housing 1,592 350 920 138 27 157
Food 982 433 442 34 2 71
Health 300 20 164 84 7 25
Other 786 159 431 103 12 81
Source:  Urban Institute analysis of NSHAPC program data.  Data represent "an average day in February 1996."

Endnotes

24.  This list is drawn directly from a recent publication by the Pew Charitable Trusts (Lugo 2001).

Populations
Older Adults