San Bernardino County2009 Point-In-Time

Homeless Count & Survey Report

Research conducted by:

CLA & Associates2070 North Tustin AvenueSanta Ana, CA 92705(714) 606-2000www.cla-associates.net

County of San Bernardino Human ServicesOffice of Homeless Services

San Bernardino County2009 Point-In-Time

Homeless Count & Survey Report

San

Bern

ardi

no C

ount

y Po

int i

n Ti

me

2009

page

3

TABLE OF CONTENTS

I. Introduction .................................................................................... 8A. San Bernardino County Continuum of Care ....................................... 8B. 2009 Point-In-Time Homeless Count & Survey ..................................... 9

II. Executive Summary ..........................................................................10

A. Number of homeless in San Bernardino Continuum of Care ............ 10B. General Profile of Homelessness in the County ................................. 12C. Conclusions ........................................................................................... 16

III. Homeless Count Results ..................................................................17A. Homeless Definition .............................................................................. 17B. Numbers and Characteristics ............................................................... 18

IV. Homeless Survey Results .................................................................19A. Demographics of the Homeless Population in San Bernardino County B. Location, Frequency and Length of Homelessness ............................. 20 C. Usual Sleeping Place ............................................................................. 27D. Service Use; Assistance; Income; and Factors that Prevent Sheltered Respondents from Obtaining and Holding a Job ............... 30E. Factors that Prevent Obtaining Permanent Housing (Sheltered Only) 32 F. Mental and Physical Illnesses and Disabilities .................................... 39G. Main Reason for Current Homelessness ............................................. 43

V. Homeless Count & Survey Methodology .......................................46A. Overview of Planning Process ............................................................. 46B. Street Count Method............................................................................ 47C. Shelter Count Method .......................................................................... 52D. Homeless Survey Method ..................................................................... 53E. Differences Between the 2007 and 2009 Homeless Counts ............... 56

VI. Homeless Count Challenges and Assumptions ..............................59A. Challenges to the Street and Shelter Counts .................................... 59B. Challenges to the Survey Process ........................................................ 60

VII. Conclusions .....................................................................................62

IX. Appendices

Appendix A: Terms & Definitions ..............................................................64

Appendix B: Enumeration Form – Unsheltered .......................................68

Appendix C: Enumeration Form – Sheltered ............................................69

Appendix D: Unsheltered Homeless Survey Instrument .........................70

Appendix E: Sheltered Homeless Survey Instrument ..............................76

Appendix F: Differences Between Unsheltered Homeless Persons Surveyed on the “Night of Count” and Over Subsequent Weeks ...........82

Appendix G: CoC Subpopulation Computations, Unsheltered Homeless Population ..................................................................................................89

Appendix H: CoC Subpopulation Computations, Sheltered Homeless Population ..................................................................................................99

Appendix I: List of Census Block Groups Canvassed ...............................110

Appendix J: Acknowledgements ..............................................................116

Tables & Figures

Table 1. Distribution of the Homeless Population by Type, Gender and Age Group ..............................................................................12

Table 2: Count and Percentage of Homeless Persons in San Bernardino County by Type, Gender and Age ....................................18

Table 3: Enumerated and Surveyed Homeless Persons in San Bernardino County by Type (Unsheltered/Sheltered), Gender and Age Group ..........................................................................................................20

Table 4: Racial/Ethnic Distributions of the Homeless Populations .......22

Table 5: Educational Attainment in the Sheltered Homeless Population ..................................................................................................23

Table 6: Educational Institutions Attended by Two or More Homeless Students in the San Bernardino County .................................24

Table 7: Location Most Recently Became Homeless and Length of Time in San Bernardino County Before Becoming Homeless Most Recently .....28

Table 8: Other Non-Employment Related Sources of Income by Sheltered/Unsheltered ..............................................................................37

Table 9: Other Factors Keeping Sheltered Respondents From Getting or Holding on to a Job ..............................................................................39

Table 10: Other Main Reasons Now Homeless by Sheltered/Unsheltered ..............................................................................45

Table 11: Distribution of Surveys Administered to Unsheltered Homeless Persons Collected on the Night of the County by Deployment Center .54

Table 12: Distribution of Surveys Administered to Sheltered Homeless Persons by Shelter Name ...........................................................................55

Figure 1: Racial/Ethnic Distribution of the Homeless Population in San Bernardino County .............................................................................21

Figure 2: Household Composition in the Sheltered and Unsheltered Homeless Populations ...............................................................................26

Figure 3: Household Composition by Gender .........................................27

Figure 4: Lifetime Days Homeless by Sheltered/Unsheltered ..................30

Figure 5: Places Usually Slept in Last 30 Days by Sheltered/Unsheltered .........................................................................31

Figure 6: Service Use in the Unsheltered Homeless Population .............32

Figure 7: Service Use in the Sheltered Homeless Population ..................33

Figure 8: Needed but Unable to Obtain Services (Sheltered Only)..........34

Figure 9: Government Assistance Programs in the Unsheltered Homeless Population .................................................................................35

Figure 10: Government Assistance Programs in the Sheltered Homeless Population .................................................................................35

Figure 11: Top Three Other Non-employment Related Sources of Income by sheltered/Unsheltered..........................................................................37

Figure 12: Sheltered Only: Factors Keeping Respondents from Getting or Holding on to a Job ..............................................................................38

Figure 13: Factors that Prevent Obtaining Permanent Housing (Sheltered Only) .........................................................................................40

Figure 14: Serious Physical Illness or Disability by Sheltered/Unsheltered and Gender .........................................................42

Figure 15: Conditions and Forms of Disability and Disease Currently Experienced by Sheltered and Unsheltered Homeless Persons ................................................................42

Figure 16: Top “Main Reasons now Homeless” by Sheltered/Unsheltered .........................................................................44

San

Bern

ardi

no C

ount

y Po

int i

n Ti

me

2009

page

8

I. Introduction

A. The San Bernardino County Continuum of CareThe Continuum of Care (CoC) in San Bernardino County is an example of the community model of homeless care promoted by the United States Department of Housing and Urban Development (HUD) since 1994. A CoC system is designed to address the critical problem of homelessness through a coordinated, community-based process of system-building to address identified needs. The approach is predicated on the understanding that homelessness is not caused merely by a lack of shelter, but involves a variety of underlying, unmet needs - physical, economic, and social. These underlying causes can be related both to the individual and larger systemic factors, such as lack of affordable housing. San Bernardino County is one of approximately 400 CoCs nationally. HUD is the primary source for homeless services funding in most CoC areas. The CoC in San Bernardino County is supported by the Office of Homeless Services, a subdivision of the Human Services Department.

The Office of Homeless Service (OHS) was created in September 2007 by the San Bernardino County Board of Supervisors. OHS works to develop a countywide public and private partnership that coordinates services designed to reduce and prevent homelessness. By providing comprehensive services and resources for homeless persons, and increasing permanent supportive housing opportunities for very low income and long-term homeless

San

Bern

ardi

no C

ount

y Po

int i

n Ti

me

2009

page

9

persons, OHS strives to end homelessness in San Bernardino County. The Office of Homeless Services carries out this directive by developing and implementing a countywide 10-Year Strategy to End Homelessness through advocacy, policy review, technical assistance, priority setting, linkages and coordination, provision of financial and other resources, and articulation of the vision of the San Bernardino County Homeless Partnership.

B. 2009 Point-In-Time Homeless Count and Survey

During the evening of Thursday, January 22, 2009 and the early morning hours of Friday, January 23, 2009, OHS and Human Services Department staff, in conjunction with CLA & Associates (CLA), and over 500 community volunteers, conducted the 2009 San Bernardino County Point-In-Time Homeless Count and Survey. This enumeration of homeless persons in San Bernardino County is the second such effort since 2005, when a count became a prerequisite to applying for CoC grant funding from HUD. HUD provides the methodological and procedural guidelines for enumerating the homeless in a document entitled, A Guide to Counting Unsheltered Homeless People (Second Revision, January, 2008). A detailed description of the logistics and planning process is provided in Section V, Homeless Count and Survey Method. Different levels of scientific rigor and adherence to HUD guidelines in the 2007 and 2009 Point-In-Time processes produced counts that are not directly comparable. These issues are also discussed in Section V of this document.

San

Bern

ardi

no C

ount

y Po

int i

n Ti

me

2009

page

10

II. Executive Summary

A. Number of homeless in San Bernardino Continuum of Care

The definition of homelessness provided by the United States Department of Housing and Urban Development (HUD) determined the persons to be enumerated in this study as: “people who for various reasons have found it necessary to live in emergency shelters or transitional housing for some period of time,” and “… unsheltered homeless people who sleep in places not meant for human habitation (for example, streets, parks, abandoned buildings, and subway tunnels) and who may also use shelters on an intermittent basis.”

The 2009 Point-In-Time Homeless Count resulted in a total of 1,736 persons. Volunteers and shelter staff tallied those observed in shelters, out-of-doors, and in motels/hotels with vouchers issued by a social service provider, on the evening of January 22, 2009 and the early morning hours of January 23, 2009. This number was calculated based upon direct observations and recognized statistical projection methods which HUD deems appropriate for this research process. The count consists of three subtotals:

• The number of projected unsheltered homeless is 968 persons;• The number of sheltered homeless (those in emergency shelter or transitional

housing) is 747 persons; and• The number of hotel/motel vouchers issued to homeless persons or families and

used on the night of the count is 21.

San

Bern

ardi

no C

ount

y Po

int i

n Ti

me

2009

page

11

Those included in the street count (also known as unsheltered homeless) were living in a place not meant for human habitation, such as a car, sidewalk, or abandoned building. Those included in the shelter count were living in an emergency shelter or in transitional housing for homeless persons, who originally came from the street or an emergency shelter.

The 2009 Point-in-Time Count employed a rigorous methodology that produced a conservative count. It does not, nor was it intended to, convey the total number of persons experiencing homelessness in San Bernardino County throughout the year in 2009. The rigorous method employed in this study is likely to have resulted in an undercount. Homeless persons not physically observed were not counted. Based upon the number of children observed and the household composition data resulting from the surveys of unsheltered homeless persons, the sole quantitative estimate (See Appendix G) suggests that the Point-In-Time process resulted in an undercount of a minimum of 290 adults.

For various reasons, drawing comparisons between the 2007 and 2009 homeless counts is problematic. These reasons, described in detail in Section V. E., include: (1) use of an unapproved multiplier in 2007, (2) possible duplication of observed homeless in 2007, and (3) questions regarding use of a canvassing or sampling method during the 2007 process. For these reasons, it is recommended that the 2009 process serve as the baseline for all future Point-In-Time Homeless Counts in San Bernardino County, with respect to both methodology and trend analysis.

San

Bern

ardi

no C

ount

y Po

int i

n Ti

me

2009

page

12

B. General Profile of Homelessness in the County

Many of the attributes, needs and patterns of service utilization are significantly different between the unsheltered and sheltered homeless populations. In these cases, significant differences are reported by population type (unsheltered or sheltered) to facilitate comparison. The distribution of the observed homeless population in San Bernardino County by type, gender and age group is depicted in Table 1.

Demographics and Family Composition

Table 1. Distribution of the Homeless Population by Type, Gender and Age Group.

*This total omits 21 vouchers because recipient demographics are unknown and is one more than the total reported because of rounding error resulting from extrapolation.

• More than one fifth (21.3%) of the homeless population consists of children and youth. The largest proportion (n= 595, 61.5%) of the unsheltered homeless population is males, estimated by the volunteer enumerators to be 18 to 59 years of age. By a narrow margin, the largest proportion (n= 226, 30.3%) of the sheltered homeless population consists of females, 18 to 59 years of age.

• In descending order of frequency, homeless persons in San Bernardino County are White (39.4%), Latino/ Hispanic (24.4%), Black or African-American (23.0), two or more races (7.2%), Asian/ Pacific Islander (3.1%) and American Indian (2.9%). Compared to the county’s general population, Latino/Hispanics (especially) and Asian/ Pacific Islanders are underrepresented in the homeless population. Conversely, Black or African American persons, persons of two or more races, and American Indians are overrepresented in the homeless population compared to their proportions in the total county population.

San

Bern

ardi

no C

ount

y Po

int i

n Ti

me

2009

page

13

• The majority of both the unsheltered (53.5%) and sheltered (51.8%) homeless populations live alone, including 66.4% of all males and 35.4% of all females. About 6.5% of the “households” or family groups of unsheltered homeless persons include dependent children, compared to 34% of households in the sheltered population. 32.7% of all females live in family groups including dependent children compared to 5.7% of all homeless males.

Veteran and Student Status

• Two hundred fifty-five, or 17.8% of all unsheltered homeless adults are veterans, compared to 55 (11.4%) sheltered homeless adults. The largest proportions of homeless veterans served in the conflict in Vietnam (40.7%) and during peacetime (42.6%).

• Overall, (15.6%) of homeless persons in San Bernardino County are currently students. Significantly more of these are sheltered homeless (20.9%) than unsheltered homeless persons (10.4%). Just less than one third (32.3%) of all homeless students completing surveys report attending San Bernardino Valley College.

Where They Became Homeless

• Over three quarters (77.0%) of the surveyed homeless report living in San Bernardino County when they most recently became homeless. Comparatively small proportions came to San Bernardino from Los Angeles County (7.8%), out of state (5.8%), Riverside County (5.4%), and from other counties in California (4.0%).

Frequency and Duration of Homelessness

• The length of time since homeless persons last lived in a permanent housing situation such as an owned or rented house or apartment ranges from one day to 42 years, with an average (mean) of three years. Sheltered homeless report significantly less time since last permanent housing (just less than two years) than homeless persons on the street (unsheltered); about four years and three months.

• Most homeless survey respondents (89.5%) report one episode of homelessness in the last twelve months, but responses range up to 12. The average number of episodes of homelessness in the last twelve months is 1.2. Similarly, the majority (69.8%) report one episode of homelessness in the last three years, and responses range up to 18 (homeless once every 45 days), with a mean of 1.8 times homeless in three years.

• Adding together all the different times, lifetime total homelessness ranges from one day to 42 years (reported by a 43 year old male). The mean, or average, cumulative length of homelessness is about 3.5 years, the most frequently reported length of time is one year, and the median length of time (the point above which and below which half the time periods of fall) is two years.

San

Bern

ardi

no C

ount

y Po

int i

n Ti

me

2009

page

14

Where the Homeless Usually Sleep

• Transitional Housing (38.4%), Emergency Shelters (36.4%) and Friend’s home (9.7%) (the top three locations in which sheltered respondents usually slept in the last 30 days), account for over 80% of the locations named in that population. The top four locations in which unsheltered homeless usually slept—Outdoor Encampment/ Street (46.4%), Car, Van, Truck or RV (14.7%), Abandoned Building (14.7%) and Friend’s Home (9.7%)—similarly account for over 80% of the locations named by unsheltered persons.

Victimization

• About one third of all homeless persons in San Bernardino County have been victims of a crime while they were homeless. About one quarter (27.2% of all victims) experienced non-violent crimes only, 37.7% were victims of violent crime, and 35.2% were victims of both violent and non-violent crime. Significantly more “street” or unsheltered homeless (38.1%) than sheltered homeless (29.8%) reported victimization. Type of crime is unrelated to shelter vs. street (unsheltered), but significantly more (82.4%) victimized females than males (64.2%) had experienced violent crime.

Chronic Homelessness and Disabling Conditions

• The Department of Housing and Urban Development identifies chronically homeless persons as those who 1) have a mental, emotional or developmental disability or a diagnosable substance abuse problem that is expected to be of long duration and substantially limits their ability to work or to perform one or more activities of daily living, 2) are unaccompanied (not in “households” or family units) and 3) have been continuously homeless for a year or more OR had at least four episodes of homelessness in the past three years. By this definition, 27.1% of unsheltered adults are chronically homeless, compared to 12.9% of sheltered adults.

• 23.6% of the unsheltered and 9.1% of the sheltered homeless population is severely mentally ill.

• 22.6% of the unsheltered and 2.5% of the sheltered homeless are chronic substance abusers.

• 2.1% of the unsheltered and 1.2% of the sheltered population consists of homeless persons with HIV/ AIDS.

Domestic Violence

• 8.2% of the unsheltered and 12.0% of the sheltered homeless population was currently experiencing domestic violence.

San

Bern

ardi

no C

ount

y Po

int i

n Ti

me

2009

page

15

Employment and Sources of Income

• Significantly more (24.4%) sheltered homeless persons are employed than unsheltered (10.0%). Proportionally more homeless females (22.6%) than males (14.0%) are employed. In the sheltered population, more than 20% report the following factors that keep them from getting, or holding on to a job: “There are no jobs” (33.0%), “Need more Education” (30.0%), “No Transportation” (27.4%), “Need more training” (25.1%), and “No Permanent Address” (21.8%).

• Significantly more unsheltered homeless persons (47.3%) than those in shelters (29.6%) report receiving no government assistance of any kind. Among the 52.7% of unsheltered homeless that do receive some form of government assistance, 52.9% receive Food Stamps, 26.9% Supplemental Security Income (SSI) and 21.8% Medi-Cal. Among the 70.4% of the sheltered homeless that receive some sort of government assistance, 73.5% get Food Stamps, 50.8% are beneficiaries of Medi-Cal, 34.3% receive Cash Aid, and 22.7% receive aid to Women, Infants and Children (WIC).

• Other top sources of non-employment related income include recyclables: unsheltered (48.9%), sheltered (24.1%); money from family or friends: unsheltered (29.5%), sheltered (36.7%); and panhandling: unsheltered (28.2%), sheltered (12.7%).

Service Use

• Significantly more unsheltered (29.7%) than sheltered homeless persons (7.0%) do not use services of any kind. Among the 70.3% of unsheltered homeless that do use services, the most frequently reported are free meals (76.5%) and food bank/ pantry (41.6%). The proportion of unsheltered homeless persons using services falls off after that, to 16.3% each reporting shelter day services and emergency shelter. 15.7% of the unsheltered homeless report utilizing legal assistance and 9.6% report using health care services. Among the 93% of sheltered homeless that do use services, 49.6% report using emergency shelters, 40.1% report using transitional housing, and 36.9% report using a food bank/ pantry. 10.3% of sheltered service users report obtaining legal assistance, and 17.1% report receiving health care. 21.1% of sheltered homeless persons report that health care is a needed service they have been unable to obtain.

Main Reasons Now Homeless

• The most frequently reported reasons for current homelessness are, “Lost job” 25.5% unsheltered, 34.3% sheltered; “Alcohol or drug problem” 9.4% unsheltered, 14.2% sheltered; “Domestic violence” 3.6%% unsheltered, 17.2% sheltered; “Parents household fell apart” 11.5% unsheltered, 9.6% sheltered; “Divorce or separation” 10.9% unsheltered, 9.6% sheltered; and “Argument, asked to leave last housing” 8.9% unsheltered, 10.5% sheltered.

San

Bern

ardi

no C

ount

y Po

int i

n Ti

me

2009

page

16

C. Conclusions

Because substantially different methodologies were employed, the 2009 count should be understood as a new baseline, rather than as a trend in relation to the 2007 Point-in-Time Count. The sampling strategy employed was designed to produce one count for the entire county. It does not support precise estimates of the distribution of homeless persons within municipal, supervisorial, or other jurisdictional boundaries.

An investment in infrastructure to implement the sampling plan for the 2009 Point-in-Time Count was made this year in the form of maps geo-coded to identify “outdoor congregating areas” where unsheltered homeless persons had been observed. This critical Geographic Information Systems work was performed by Mr. Brent Rolf, San Bernardino County Information Services Department GIMS Team. The cost of future counts could be reduced and the accuracy improved by updating this work periodically before the 2011 Point-in-Time Count. The Office of Homeless Services and the county’s mental health outreach workers can both contribute to, and profit from, gathering this information and mapping these locations.

The outcomes of the 2009 Count have both federal and local public policy implications. All Point-In-Time data are submitted to HUD and are analyzed to produce a portrait of national homelessness. Federal legislators are regularly informed of HUD’s policy priorities through the prism of the research data that come from local Continuums of Care. The survey data in particular provide a snapshot of the various factors leading to homelessness. Information about factors leading to or prolonging homelessness can be incorporated at the local level through processes like the San Bernardino County 10-Year Strategy to End Homelessness as a means of identifying gaps in services, and addressing new services that may need to be added to the existing service continuum.

Another potentially significant federal policy implication is the increasing focus on family homelessness. New federal funding that focuses on homelessness prevention and permanent housing is beginning to work its way into local communities, which is a noticeable shift from the policy priority in past years on reducing the population of individual chronic homeless persons. While current national data suggest that the chronic homeless use a disproportionately large share of services, there has long been anecdotal information suggesting that families – particularly those with minor children – were increasingly seeking assistance. The current economic recession seems to have accelerated the burden on local service providers to assist families, yet most of these agencies have limited capacity and resources, especially with the dire financial situation in the State of California. The 2009 Point-In-Time data are one source that local leaders can look to in beginning to align local policies with the new federal funding priorities on prevention and permanent housing. However, future research within San Bernardino County is needed to study how the differences in service provision impact both homeless families and homeless individuals over time, and the overall effectiveness of this policy shift toward prevention and permanent housing in reducing, and perhaps eliminating, homelessness.

San

Bern

ardi

no C

ount

y Po

int i

n Ti

me

2009

page

17

III. Homeless Count Results

A. Homeless Definition

Noting that it has been controversial since the 1980s, HUD frames the definition of homelessness in terms of a distinction between the “literally homeless,” and the “precariously housed.” The literally homeless are, “people who for various reasons have found it necessary to live in emergency shelters or transitional housing for some period of time. This category also includes unsheltered homeless people who sleep in places not meant for human habitation (for example, streets, parks, abandoned buildings, and subway tunnels) and who may also use shelters on an intermittent basis.”

The literally homeless are distinguished from the precariously housed, who are at high risk of becoming literally homeless, but who for now, “may be doubled up with friends and relatives or paying extremely high proportions of their resources for rent.” The McKinney-Vento Act, which governs HUD’s homeless assistance programs, specifically targets persons living in shelters or in places not meant for human habitation, but not people in precarious housing situations. Consequently, for the Point-In-Time Count homelessness is defined to include only persons living in places not meant for human habitation, in shelters, and in transitional housing.

San

Bern

ardi

no C

ount

y Po

int i

n Ti

me

2009

page

18

B. Numbers and Characteristics

Six hundred sixteen persons were physically observed on the night of the street count in places not meant for human habitation. Projections calculated for eight high-density stratum block groups selected into the sample that were not canvassed due to resource constraints increase this total by 31.6 to 647.6. The low-density stratum projections add 320.8 persons, building to an estimate of 968.4 unsheltered homeless persons. Adding the 747 enumerated in shelters and the 21 hotel/motel vouchers brings the total estimate to 1,736 homeless persons observed in San Bernardino County on the night of the Point-In-Time Count. The enumeration forms (see Appendices B and C) utilized to count both sheltered and unsheltered homeless persons included tallies in eight demographic categories. Table 1 presents the gender and age-group detail associated with the sheltered, unsheltered and total homeless counts.

Note that more than three of five (n= 595, 61.5%) unsheltered homeless persons are males, estimated by the volunteer enumerators to be 18 to 59 years of age. Unsheltered females in this age group are substantially fewer (n= 225, 23.2%). In contrast, the numbers of male and female sheltered homeless persons 18 to 59 years of age are nearly identical (males, n= 222, 29.7%; females, n= 226, 30.3%). Just less than five percent (4.7%) of the total estimated homeless population consists of children from birth to five years of age, and 6.6% of youth from six to 17 years of age.

Table 2. Count and Percent of Homeless Persons in San Bernardino County by Type, Gender and Age Group

* This total is one lower than the sum of the row. This is due to rounding error.** Includes 21 vouchers issued on the night of the Point-In-Time Count. Because demographics of voucher users are unknown, percentages in this row are computed based upon a total of 1,715.

Male Children (0 to 5)

Male Youth (6 to 17)

Male Adults (18 to 59)

Male Adults (60 & >)

Female Children (0

to 5)Female Youth

(6 to 17)Female Adults

(18 to 59) Female Adults TotalCount Count Count Count Count Count Count (60 & >) Count

(%) (%) (%) (%) (%) (%) (%) Count (%)(%)

2 49 595 55 16 13 225 14 968*(0.21) (5.1) (61.5) (5.7) (1.7) (1.3) (23.2) (1.5) (100)

79 64 222 12 83 60 226 1 747(10.6) (8.6) (29.7) (1.6) (11.1) (8) (30.3) (0.13) (100)

81 113 817 67 99 73 451 15 1,736**

(4.7) (6.6) (47.6) (3.9) (5.8) (4.3) (26.3) (0.87) (100)

San

Bern

ardi

no C

ount

y Po

int i

n Ti

me

2009

page

19

IV. Homeless Survey Results

Separating the enumeration process from administration of the surveys in 2009 resulted in a more manageable process. Conducting interviews over subsequent weeks reduced time on the ground on the night of the Point-In-Time Count (few surveys were administered), facilitating the reach and mobility of the volunteer teams. In the weeks following the count, OHS staff continued administering surveys to unsheltered homeless persons encountered in outdoor locations (e.g. parks, on the street) but also to persons that were attracted to day services or free meal events following the Point-in-Time Count. While simplifying the process on the night of the count, this method of survey data collection simultaneously complicates the process of projecting results to the unsheltered homeless population that was actually observed on the night of the count. The issue is confidence in the extent to which the survey data accurately represent the observed unsheltered homeless population.

To check this, the survey data were analyzed to determine whether the 57 unsheltered persons surveyed during the night of the count differ from the 179 unsheltered persons interviewed in the weeks after the count. These analyses show that proportionally fewer night-of-count interviewees were homeless for the first time, more reported military service, more report panhandling as a source of income, a larger proportion reports no service use at all, and proportionally fewer are employed or are current students. Proportionally more unsheltered persons surveyed on the night of the count report a serious physical illness or disability, fewer report experiencing depression, but more report alcohol or drug abuse. (See detail in Appendix F). One might conclude from this summary that the unsheltered population interviewed on the night of the count consists of a higher proportion of the “service resistant” and/or chronic homeless, as chronic homelessness is defined by HUD.

San

Bern

ardi

no C

ount

y Po

int i

n Ti

me

2009

page

20

However, the proportion interviewed on the night of the count (26.3%) is not significantly different than the proportion among the unsheltered homeless interviewed during the weeks following the count (25.7%). It cannot be stated with certainty, however, how the homeless persons who qualified to be interviewed at outdoor locations or during free meal events may have differed from the unsheltered homeless persons enumerated on the streets at night during the Point-In-Time Count. It is possible, for example, that fewer asocial and/or mentally ill homeless persons joined those congregating at the free meal events.

Finally, in spite of face-to-face administration of the survey on the night of the count and efforts to “group proctor” survey administration following the count, in many cases branching sequences in the questionnaire were not followed and/ or items were left missing. All reported percentages are based upon the valid responses; a number frequently lower than the total number of surveys.

A total of 236 surveys with reasonably useful data were completed with unsheltered homeless persons and 271 with sheltered homeless persons. About 35 of these surveys were conducted in Spanish. The English versions of the sheltered and unsheltered questionnaires are reproduced in Appendices D and E.

A. Demographics of the Homeless Population in San Bernardino County

Table 3 depicts the gender and age group distributions of homeless persons enumerated and surveyed during the 2009 Point-In-Time Count.

Table 3. Enumerated and Surveyed Homeless Persons in San Bernardino County by Type (Unsheltered/ Sheltered), Gender and Age Group

San

Bern

ardi

no C

ount

y Po

int i

n Ti

me

2009

page

21

The largest proportion of the unsheltered population is males, 18 to 59 years of age (n= 595, 61.5%), and by a narrow margin, females 18 to 59 years of age (n= 226, 29.4%) in the sheltered homeless population. Note that 5.3% (n= 51) of the unsheltered population consists of children and youth 17 years of age or younger. Point-In-Time methods typically undercount unsheltered youth. Those living on their own are difficult to locate at night, and because they fear intervention by Child Protective Services, homeless parents frequently sequester their children in vehicles, motel rooms or with alternating family members or friends. Children and youth comprise a much higher proportion (n= 142, 18.6%) of the sheltered population.

Race/ Ethnicity

Figure 1 illustrates the proportions of homeless persons in San Bernardino County by race/ ethnicity. In descending order of frequency, the largest three racial/ ethnic groups are White (39.4%), Latino/ Hispanic (24.4%) and Black or African American (23.0%).

Throughout this section, attributes of the homeless population are presented collectively when differences between the surveyed sheltered and unsheltered populations are not statistically significant, and separately when they are.

San

Bern

ardi

no C

ount

y Po

int i

n Ti

me

2009

page

22

Table 4. Racial/Ethnic Distribution of the San Bernardino Total County and Homeless Populations

Racial/ Ethnic Category Total County Population1

Count(%)

Homeless Population

Count(%)

Latino/ Hispanic 970,484(47.6)

119(24.4)

White 716,438(35.1)

192(39.4)

Black 170,805(8.4)

112(23.0)

Asian, Pacific Islander 120,467(5.9)

15(3.1)

Two or More Races 47,996(2.4)

35(7.2)

American Indian 12,156(0.6)

14(2.9)

Total 2,038,346(100.0)

487(100.0)

1

Table 4 compares racial/ ethnic proportions in the total county population and the surveyed homeless population. The first two data rows show that Latino/ Hispanic persons are significantly underrepresented in the homeless population, and that White persons are slightly overrepresented.

Black or African American persons, persons of two or more races, and American Indians are significantly overrepresented in the homeless population compared to their proportions in the total county population. Conversely, like Latinos, Asian and Pacific Islanders are significantly underrepresented in the homeless population in San Bernardino County. Twenty of the 507 homeless interviewees declined to report their race/ ethnicity.

1 California State Department of Finance Demographics Unit data file “E-3 Race/ Ethnic Population Estimates with Age and Sex Detail, 2007 downloaded from: http://www.dof.ca.gov/research/demographic/data/e-3/

To facilitate data collection, the unsheltered questionnaire includes fewer items than the sheltered survey questionnaire. When a survey item was not administered to unsheltered respondents, “(Sheltered Only)” appears in the section heading.

San

Bern

ardi

no C

ount

y Po

int i

n Ti

me

2009

page

23

Education (Sheltered Only) and Student Status

Over 63% of the sheltered homeless have at least a high school education; only 8% have a Bachelor’s degree or higher.

Table 5. Educational Attainment in the Sheltered Homeless Population

Education Count(%)

Less than 6th Grade 2(0.8)

Less than High School Diploma 70(27.9)

High School Diploma or GED 92(36.7)

Some College, No Degree 58(23.1)

AA Degree/ Technical Certificate 21(8.4)

Bachelor’s Degree 5(2.0)

Graduate or Professional Degree 3(1.2)

Total 251(100.0)

Of the 494 survey respondents who answered the question, 79 (15.6%) are currently students. Significantly more sheltered homeless (n= 55, 20.9%) are students than unsheltered homeless persons (n= 24, 10.4%); χ2 (1, 494) = 10.137, p < .002.

San

Bern

ardi

no C

ount

y Po

int i

n Ti

me

2009

page

24

Table 6. Educational Institutions Attended by Two or More Homeless Students in San Bernardino County

School, College or University Count(%)

San Bernardino Valley College 21(32.3)

Victor Valley College 10(15.4)

San Bernardino Adult School 5(7.7)

Barstow Community College 4(6.2)

Barstow Adult School 2(3.1)

Chaffey College 2(3.1)

Everest (formerly Bryman) 2(3.1)

Hillside Adult School 2(3.1)

Summit Career College 2(3.1)

Upland Adult School 2(3.1)

Other Institutions 13(20.0)

Total 65(100.0)

Sixty-five of the 79 current students identified the school they attend. Table 6 lists all schools, colleges and universities reported by two or more homeless students. These account for 52 of the 65 (80.0%) homeless students that identified a school. The remaining thirteen students (20.0%) each listed a unique institution, including “Adult Education” with no further description, Crafton Hills College, CSU San Bernardino, “Home school,” Intercoast College, Mojave High School, Parole Office GED Program, Platt College-Ontario, “ROP” with no further description, University of Phoenix, Willow Park High School, and Zupanic High School.

San

Bern

ardi

no C

ount

y Po

int i

n Ti

me

2009

page

25

Employment

Significantly more sheltered homeless persons report employment (n= 64, 24.4%) than do unsheltered homeless survey respondents (n= 23, 10.0%); χ2 (1, 492) = 17.515, p < .001. More females (n= 51, 22.6%) than males (n= 36, 14.0%) are employed; χ2 (1, 483) = 5.965, p < .02. Seventy-nine of the 87 homeless persons reporting employment indicated whether they held a permanent or a temporary job. The majority of employed sheltered homeless (n= 42, 72.4%) report permanent jobs, compared to 16 (27.6%) in temporary jobs. Conversely, the majority of employed unsheltered homeless (n= 13, 61.9%) are in temporary jobs, compared to eight (38.1%) in permanent jobs. The difference in permanent/ temporary status between employed sheltered and unsheltered homeless persons is significant; χ2 (1, 79) = 7.816, p < .01.

Foster Care

Just less than one in five homeless survey respondents (n= 93, 19.1%) was in foster care with a family, in a group home or some other type of placement. This proportion does not differ significantly between sheltered/ unsheltered, or by gender.

Veteran Status

A veteran is defined as a person who served in the active military, naval or air service, and who was discharged or released under conditions other than dishonorable. Sixty-seven (13.6%) of the 494 homeless respondents who answered the question, “Have you ever served on active duty in the Armed Services of the United States?” responded affirmatively. The survey did not inquire about discharge status. Fifty-eight of the 67 indicated whether they served in the “Regular Military” (n= 54, 93.1%), National Guard (n= 1, 1.7%), or Reserves (n= 3, 5.2%).

Significantly more unsheltered homeless persons (n= 40, 17.5%) than sheltered (n= 27, 10.2%) report military service; χ2 (1,494) = 5.551, p < .02, as do males (n= 57, 22.3% of all males) compared to females (n= 7, 3.1% of all females); χ2 (1,483) = 38.513, p < .001.

Fifty-five survey respondents reported their era of service: One served in World War II, one in the Korean conflict, and five (3.3%) each during the first Gulf war and during the War on Terror in Iraq and Afghanistan. One of these reports serving during the War on Terror and during peacetime. Twenty-two (40.7%) served in Vietnam and 23 (42.6%) during peacetime. Although the relationship is not statistically significant, it is noteworthy nonetheless that twelve of 23 (52.2%) sheltered veterans served during peacetime, compared to eleven of 38 (28.9%) homeless veterans encountered on the street.

San

Bern

ardi

no C

ount

y Po

int i

n Ti

me

2009

page

26

Victimization and Criminal History

Collectively, just over one third (n= 167, 33.7%) of all homeless survey respondents reported that they had been victims of a crime while they were homeless. Forty-four (27.2% of all victims) had experienced non-violent crimes only, 61 (37.7%) are victims of violent crime, and 57 (35.2%) of both violent and non-violent crime. Significantly more “street” homeless (n= 88, 38.1%) than sheltered homeless (n= 79, 29.8%) reported victimization; χ2 (1, 496) = 3.792, p < .04.

Type of crime is unrelated to shelter vs. street (unsheltered), but significantly fewer females who reported victimization (n= 13, 17.6%) compared to males (n= 29, 35.8%) had experienced only non-violent crime, and significantly more (n= 61, 82.4% of all victimized females) than males (n= 52, 64.2% of all victimized males) had experienced violent crime.Collectively, 140 (28.8%) homeless survey respondents reported a felony conviction. Felony conviction is unrelated to type (sheltered vs. unsheltered), but is highly related to gender. Twice the proportion of males (n= 95, 37.7%) than females (n= 43, 18.8%) reported a felony conviction; χ2 (1, 481) = 20.994, p < .001.

Household Composition

A detailed description of the computations performed to estimate the count and proportions of households with dependent children in the unsheltered and sheltered homeless populations is detailed in Appendices G and H, respectively. Figure 2 provides a general overview of household composition.

51.8%53.5%

4.4%

18.7%

6.0%

20.0%

32.7%

5.7%

2.0% 1.3%3.2%

0.9%

0.0%

10.0%

20.0%

30.0%

40.0%

50.0%

60.0%

Lives alone Withnonfamily only

With familyonly, nochildren

With familyonly including

children

With family &others, nochildren

With family &other,

includingchildren

Figure 2. Household Composition in the Sheltered and Unsheltered Homeless Populations

Sheltered

Unsheltered

San

Bern

ardi

no C

ount

y Po

int i

n Ti

me

2009

page

27

The majority of both the sheltered (light green bars) and unsheltered homeless populations (blue bars) live alone. Significantly larger proportions of the unsheltered population live with nonfamily members only (18.7%, compared to 4.4% of the sheltered homeless) and with family members only in households that do not include children (20.0%, compared to 6.0% of the sheltered homeless. Conversely, nearly one third (32.7%) of the sheltered population lives with family members only, including children, compared to 5.7% of the unsheltered population; χ2 (1, 481) = 88.378, p < .001.

Lives alone

With

nonfamily

only

With

fam

ily only, n

o children

With

fam

ily only in

cluding childre

n

With

fam

ily &

others, n

o children

With

fam

ily &

other, i

ncluding chi.. .

FemaleMale

66.4%

11.7% 12.6%

5.7% 1.2% 0.8%

35.4%

11.2% 13.5%

32.7%

2.2% 3.6%0.0%

10.0%

20.0%

30.0%

40.0%

50.0%

60.0%

70.0%

Figure 3 . Household Composition by Gender

Female

Male

Figure 3 illustrates significant differences in the proportions of males and females in various household types. The proportions living with non-family members only and with family members in households that do not include children are equivalent, but proportionally fewer females (35.4%) than males (66.4%) live alone, and more females (32.7%) than males (5.7%) live with family members in households that do include children; χ2 (5, 470) = 68.887, p < .001. This pattern is observed in both sheltered and unsheltered homeless populations.

B. Location, Frequency and Length of Homelessness

Over three quarters (n= 384, 77.0%) of the surveyed homeless reported living in San

San

Bern

ardi

no C

ount

y Po

int i

n Ti

me

2009

page

28

Bernardino County when they most recently became homeless. Comparatively small proportions came to San Bernardino from Los Angeles County (n= 39, 7.8%), out of state (n= 29, 5.8%), Riverside County (n= 27, 5.4%), and from other counties in California (n= 20, 4.0%).

The instruction, “If you came to San Bernardino as a homeless person, the answer to (the next) question…is “0,” preceded the question, “How long had you been living in San Bernardino County before becoming homeless most recently?” So, survey respondents who answered “0” to the latter question presumably came to the county while homeless. The length of time in San Bernardino County before becoming homeless most recently (rightmost column of Table 6, data rows 2-5) reported by persons who indicated that they were living elsewhere when they most recently became homeless is difficult to interpret. If they were living elsewhere when they most recently became homeless, logic dictates that they came to San Bernardino County as homeless persons.

Table 7. Location Most Recently Became Homeless and Length of Time in San Bernardino County Before Becoming Homeless Most Recently

* Excludes one outlier of 10,950 days (30 years)** Excludes two outliers of 13,505 and 14,600 days (37 and 40 years, respectively)

Similarly, although it’s possible, it seems unlikely that the 28 persons who became homeless while living in San Bernardino County (second cell from the left in the first data row) did so on the day of the Point-In-Time Count. This would explain reporting that they had lived in the county zero days before most recently becoming homeless, however. Note that the mean length of time in San Bernardino County among persons who lived here when they became homeless is 15 years (5,476.5 days), but the median and mode length of time is

San

Bern

ardi

no C

ount

y Po

int i

n Ti

me

2009

page

29

ten years.

First Time Homeless, Length of Homelessness This Time and Length of Time Since Last Permanent Housing

The majority (n=283, 57.2%) of survey respondents report that this is the first time they had been homeless. Among persons homeless for the first time during the Point-In-Time Count, answers to the question, “How long have you been homeless this time?” range from one to 8,030 days (22 years), with a mean of 796.5 days (a little more than two years and two months) and a median (the point above which and below which half the scores fall) of one year.

Among those reporting this is not the first time they had been homeless, the length of time homeless on this occasion ranges from one to 5,475 days (15 years) with a mean of 737.1 days (two years and one week) and a median of one year.

Four hundred seventy-eight survey respondents reported the length of time since they last lived in a permanent housing situation such as a house or apartment that they owned or rented. Collectively, the length of time since permanent housing ranges from one to 15,330 days (42 years), with a mean of 1,095.7 days (three years). Sheltered homeless survey respondents report significantly less time since last permanent housing (M= 696.6 days—just less than two years) than homeless persons on the street (M= 1,552 days—about four years and three months); F (1, 476) = 29.08, p< .001.

Episodes and Lifetime Length of Homelessness

Most survey respondents (n= 427, 89.5%) report one episode of homelessness in the last twelve months, but responses range up to 12. The average number of episodes of homelessness in twelve months is 1.2. Similarly, the majority (n= 330, 69.8%) report one episode of homelessness in the last three years, and responses range up to 18 (homeless once every 45 days), with a mean of 1.8 times homeless in three years.

Adding together all the different times, survey respondents report the total time they had been homeless over their entire lives, ranging from one day to 42 years (reported by a 43 year old male). The mean cumulative length of homelessness is about 3.5 years, the most frequently reported length of time is one year, and the median length of time is two years.

San

Bern

ardi

no C

ount

y Po

int i

n Ti

me

2009

page

30

9.8%

2.3%

9.8%

6.5%

13.0%

6.9%

20.1%

12.0%

15.4%

17.5%

11.4% 11.1%11.8%

15.7%

7.1%

14.7%

1.6%

13.4%

0.00%

5.00%

10.00%

15.00%

20.00%

25.00%

1 month orless

> 1 monthto 3 months

> 3 monthsto 6 months

> 6 monthsto 1 year

> 1 year to2 years

> 2 years to3 years

> 3 years to5 years

> 5 years to10 years

> 10 years

Figure 4. Lifetime Days Homeless by Sheltered/ Unsheltered

Sheltered Unsheltered

Figure 4 illustrates the fact unsheltered survey respondents report significantly longer periods of cumulative homelessness (blue bars signifying longer periods of time are higher on the right side of the figure), compared to sheltered respondents (light green bars signifying shorter periods of time tend to be higher on the left side of the figure); χ2 (1,471) = 52.313, p < .001.

C. Usual Sleeping Place

Survey respondents were asked to check one location from a list of eleven options to indicate where they usually slept in the 30 days preceding the point at which they completed the survey questionnaire. Of the 482 persons who answered, 95% (n= 458) endorsed one option, but the other 5% (n= 24) identified between two and eight places. Consequently, the total number of options endorsed exceeds the number of respondents. Figure 5 illustrates the proportions of the unsheltered and sheltered homeless populations that usually sleep in the six locations that account for over 80% of those reported by all survey respondents. The percentage in blue font by the location label on the left side of the graphic is the overall proportion, and the percentages near each bar in the graphic are the sheltered and unsheltered proportions. The between-type differences are statistically significant for each location except “Friend’s Home.”

San

Bern

ardi

no C

ount

y Po

int i

n Ti

me

2009

page

31

0.8%

14.7%

3.5%

14.7%

9.7%11.2%

36.4%

2.2%

38.4%

5.8%

3.9%

46.4%

0 0.05 0.1 0.15 0.2 0.25 0.3 0.35 0.4 0.45 0.5

Abandoned Building

Car, Van, Truck, RV

Friend's Home

Emergency Shelter

Transitional Housing

Outdoor Encampment,Street

Figure 5. Places Usually Slept in Last 30 Days by Sheltered/ Unsheltered

Unsheltered

Sheltered

23.7%

23.2%

20.5%

10.4%

8.7%

7.3%

The top three locations reported by sheltered respondents, Transitional Housing (38.4%), Emergency Shelters (36.4%) and Friend’s home (9.7%), account for over 80% of the locations named in that population. The top four locations in which unsheltered homeless usually sleep, Outdoor Encampment/ Street (46.4%), Car, Van, Truck or RV (14.7%), Abandoned Building (14.7%) and Friend’s Home (11.2%) similarly account for over 80% of the locations named.

With regard to the five options not listed on the graph, roughly equal proportions of sheltered (4.3%) and unsheltered (4.0%) report usually sleeping in a hotel or motel during the 30 days preceding the survey. Significantly more unsheltered (4.7%) than sheltered survey respondents (1.3%) identified a church, a family member’s home (4.3% sheltered, compared to 0.9% unsheltered) and a garage (2.7% sheltered, compared to 0.9% unsheltered). Conversely, proportionally more unsheltered survey respondents (4.9%) usually sleep in a park than sheltered homeless persons (0.8%).

San

Bern

ardi

no C

ount

y Po

int i

n Ti

me

2009

page

32

D. Service Use; Assistance; Income; and Factors that Prevent Sheltered Respondents from Obtaining and Holding a Job

Service Use

Survey respondents identified which, if any, of twelve services or forms of assistance they currently use. Replies included between zero and eleven services. Significantly more unsheltered (n= 70, 29.7%) than sheltered homeless (n= 19, 7.0%) indicate that they are not using any services; χ2 (1,507) = 44.715, p < .001. Moreover, among those who do use services, the average number of services or forms of assistance reported by unsheltered persons (n= 166, M= 2.2) is slightly but significantly lower than the average number of services reported by sheltered homeless persons (n= 252, M= 2.6), F (1, 416) = 5.918, p< .02. In the sheltered population, the proportions of females (n= 145, 96.0%) and males (n= 102, 92.7%) that do use services is approximately equivalent, but in the unsheltered population, proportionally more males (n= 112, 75.7%) than females (n= 51, 63.8%) report using one or more services; χ2 (1,228) = 3.624, p < .05.

Restricting the analysis to those who do use services or assistance of some kind, Figures 6 and 7 illustrate the proportions in each homeless population that report use of each service.

1.8%

4.8%

6.0%

7.2%

9.6%

9.6%

13.9%

15.7%

16.3%

16.3%

41.6%76.5%

0.0% 10.0% 20.0% 30.0% 40.0% 50.0% 60.0% 70.0% 80.0%

T rans itio nal H o us ing

H elp W A pplic atio ns

J o b T raining

M ental H ealth C are

A lc / D rug C o uns eling

H ealth C are

B us P as s es

Legal A s s is tanc e

E mergenc y S helter

S helter D ay S vc s

F o o d B ank/ P antry

F ree M eals

Figure 6. Service Use in the Unsheltered Homeless Population

Note that because of the high proportion (76.5%) of unsheltered respondents using Free Meals, the axis at the bottom of Figure 6 extends from 0.0% to 80.0%. In contrast, the axis in Figure 7 extends from 0.0% to 50.0%. Free Meals and Food Bank/ Pantry are the first and second most frequently reported services used in the unsheltered population, and the third and fourth in the sheltered population. Legal assistance is the fifth most used service among unsheltered persons, but the eleventh among those in shelters.

San

Bern

ardi

no C

ount

y Po

int i

n Ti

me

2009

page

33

6.7%

10.3%

11.1%

12.7%

15.1%

15.1%

17.1%

19.4%

24.6%

36.9%

40.1% 49.6%

0.0% 5.0% 10.0% 15.0% 20.0% 25.0% 30.0% 35.0% 40.0% 45.0% 50.0%

Help W Applications

Legal Assistance

Job Training

Alc/ Drug Counseling

Mental Health Care

Shelter Day Svcs

Health Care

Bus Passes

Food Bank/ Pantry

Free Meals

Transitional Housing

Emergency Shelter

Figure 7. Service Use in the Sheltered Homeless Population

Considering just those who do use services, the rates of service use between sheltered and unsheltered populations are significantly different for Emergency Shelter and Transitional Housing, Health Care, and Mental Health Care (more sheltered); and Food Bank/ Pantry and Free Meals, (more unsheltered). The differences between homeless populations in use rates for the other listed services are not statistically significant.

Services Needed but not Obtained (Sheltered Only)

Unsheltered survey respondents only were asked whether they had needed any of ten listed services (and given the opportunity to describe a service not listed), but had been unable to obtain them. One hundred eighty respondents endorsed between one and all ten services, with a mean of 3.0 needed services that were not obtained. Three sheltered respondents went on to describe an additional service they needed but had been unable to obtain, bringing the response rate for this series to 183/ 271= 67.5%.

San

Bern

ardi

no C

ount

y Po

int i

n Ti

me

2009

page

34

9.4%

14.4%

20.0%

21.1%

21.1%

22.2%

25.0%

40.0%

40.6%

54.4%

0.0% 10.0% 20.0% 30.0% 40.0% 50.0% 60.0%

Counseling

Job Training

Educational

Medical

Food

Clothing

Temporary Shelter

Financial Assistance

Transportation

Permanent Housing

Figure 8. Needed but Unable to Obtain Services (Sheltered Only)

The additional services needed but which sheltered survey respondents were unable to obtain include child care (n= 2), “only a job,” “optometry, dermatology, dentist,” and “self esteem.” One respondent remarked, “I couldn”t get any help until my fiancé and I had a baby.” A significantly higher proportion of males (33.3%) than females (17.0%) endorsed, “Temporary Shelter,” χ2 (1,178) = 6.381, p < .02. Conversely, proportionally more females (13.0%) than males (3.8%) reported that counseling was a needed service they were unable to obtain, χ2 (1,178) = 4.488, p < .03. Likewise, a higher proportion of females (66.0%) than males (41.0%) checked “Permanent Housing,” χ2 (1,178) = 11.046, p < .002.

Government Assistance

All survey respondents were asked which, if any, of 11 types of government assistance they currently receive. Like service use, significantly more unsheltered survey respondents (n= 107, 47.3%) than those in shelters (n= 76, 29.6%) report receiving no government assistance of any kind; χ2 (1,483) = 16.142, p < .001. As was the case with service use, restricting the analysis to just those who do receive at least one form of government assistance, the mean number of types of government assistance reported by unsheltered respondents (n= 119, M= 1.4) is significantly lower than the average reported by those in shelters (n= 181, M= 2.2 types of assistance), F (1, 298) = 42.226, p < .001.

San

Bern

ardi

no C

ount

y Po

int i

n Ti

me

2009

page

35

0.8%

2.5%

3.4%

4.2%

5.0%

7.6%

8.4%

8.4%

21.8%

26.9%

52.9%

0.0% 10.0% 20.0% 30.0% 40.0% 50.0% 60.0%

WIC

Veteran's Disability C omp

Other Govt Asst

C ash Aid

Veteran's B enefits

S ocial S ecurity

Med Indigent Asst

General R elief

Medi-C al

S S I

F ood S tamps

Figure 9. Government Assistance Programs in the Unsheltered Homeless Population

1.1%

2.2%

3.9%

5.0%

5.0%

8.3%

12.2%

22.7%

34.3%

50.8%

73.5%

0.0% 10.0% 20.0% 30.0% 40.0% 50.0% 60.0% 70.0% 80.0%

V eteran's D is ability C o mp

V eteran's B enefits

S o c ial S ec urity

Other G o vt A s s t

G eneral R elief

M ed Indigent A s s t

S S I

WIC

C as h A id

M edi-C al

F o o d S tamps

Figure 10. Government Assistance Programs in the S heltered Homeless Population

The number of respondents who do receive at least one type of government assistance is utilized as the denominator to calculate proportions, so the percentages illustrated by Figures 9 and 10 are the proportions in the sheltered and unsheltered homeless populations receiving each type of aid among those that receive at least one type of government assistance. Note that because of the high proportion (73.5%) of sheltered survey respondents receiving Food Stamps, the axis in Figure 10 extends to 80%, whereas the maximum value on the axis in Figure 9 is 60%. Supplemental Security Income (SSI) is the second most frequently reported form of government assistance received in the unsheltered population among those who receive any kind of assistance, but the fifth in the sheltered population. Conversely, over one third (34.3%) of the sheltered homeless who get some type of government assistance report receiving Cash Aid, compared to just 4.2% of unsheltered survey respondents who receive some type of assistance.

The proportions of sheltered survey respondents receiving Food Stamps, Medi-Cal, Cash

San

Bern

ardi

no C

ount

y Po

int i

n Ti

me

2009

page

36

Aid and Women, Infant and Children’s Aid (WIC) are significantly larger than among unsheltered respondents. Significantly more unsheltered respondents report receiving Supplemental Security Income than do those in shelters. The proportions receiving General Relief, Medical Indigent Assistance, Veteran’s Benefits, Social Security, Veteran’s Disability Compensation and Other Government Assistance are not significantly different.

In the sheltered population, proportionally more females (n= 122, 80.8%) than males (n= 57, 51.8%) report receiving some type of government assistance, χ2 (1,261) = 24.799, p < .001, but on the street, the proportions of male (n= 78, 52.7%) and female (n= 39, 48.8%) unsheltered survey respondents reporting government assistance are not significantly different.

Sheltered survey respondents only were asked to report their total monthly income from all government sources (county, state, federal). Three persons who receive Food Stamps or Medi-Cal, and in one case WIC, reported zero monthly income from these sources. Because these forms of aid have some non-zero value, these respondents are omitted from the distribution. One hundred thirty-two sheltered homeless persons reported monthly income from government assistance programs ranging from $75.00 to $1,800, with an average of $565.22 and a median of $555.00. The average reported by sheltered persons with dependent children in their households (M= $593.45) is not significantly different than the average reported by those without children (M= $526.67).

Other Forms of Income

Presented with a list of twelve options and an opportunity to describe other forms of income than government assistance that are unrelated to employment, all survey respondents were asked, “How else do you get money?” Two hundred seventy-one (53.5%) homeless respondents reported one or more sources of income unrelated to employment. The proportion reporting income from one or more of these sources is significantly higher (n= 160, 67.8%) among unsheltered than among sheltered (n= 111, 41.0%) survey respondents; χ2 (1,507) = 36.516, p < .001. Recalling that just less than one quarter of the sheltered homeless compared to one tenth of the unsheltered homeless are employed, a significantly larger proportion of those who are not employed (n= 242, 59.8%) report one or more of these sources of income compared to those who are employed (n= 26, 29.9%); χ2 (1,492) = 25.761, p < .001.

Figure 11 illustrates the proportions of the sheltered and unsheltered survey respondents that obtain income from the top three non-employment related sources. The proportions who report receiving money from family or friends are not significantly different, but the differences between populations who collect recyclables and who panhandle are statistically significant.

San

Bern

ardi

no C

ount

y Po

int i

n Ti

me

2009

page

37

12.7%

28.2%

36.7%

29.5%

24.1%

48.9%

0.0% 5.0% 10.0% 15.0% 20.0% 25.0% 30.0% 35.0% 40.0% 45.0% 50.0%

Panhandling

Money from Family orFriends

Recylables

Figure 11. Top Three Other Non-employment Related Sources of Income by Sheltered/ Unsheltered

Unsheltered

Sheltered

Table 7 presents counts and proportions in the sheltered and unsheltered homeless populations reporting other, non-employment related sources of income. These are listed in descending order of the collective proportion, which ranges from 4.1% (selling blood/ plasma) to 1.5% (retirement pension). Only the proportions of sheltered and unsheltered survey respondents reporting child support are significantly different.

Table 8. Other Non-Employment Related Sources of Income By Sheltered/ Unsheltered

Source of Income ShelteredCount

(%)

Unsheltered Count

(%)Selling Blood/ Plasma 7

(4.2)7

(4.0)Selling Other Found Items 4

(2.4)9

(5.1)Sex Work/ Prostitution 4

(2.4)8

(4.5)Street Performance 2

(1.2)7

(4.0)Child Support 9

(5.4)0

(0.0)Gambling 2

(1.2)5

(2.8)Robbery/ Theft 3

(1.8)1

(0.6)Selling Drugs 0

(0.0)4

(2.3)Retirement Pension 1

(0.6)4

(2.3)

San

Bern

ardi

no C

ount

y Po

int i

n Ti

me

2009

page

38

Since many employed survey respondents listed their jobs as an “other” source of income, only the responses from those indicating that they are not currently employed are considered here. Thirty-one (73.8%) of the 42 “other” sources of income are identified as “side jobs,” “odd jobs,” or “temporary work.” Descriptions of these include “clean house,” “construction work contracts,” “day labor,” “mow lawns” or “yard work,” “painting,” “repair bikes,” “work as tattoo artist” and “wash cars.” Other forms of income include “church,” “going to school,” “compensated work therapy,” “income from property,” “income tax return,” “inheritance,” “sold wedding ring to pawn shop,” and “wife’s SSI.”

Factors that Prevent Obtaining or Holding a Job (Sheltered Only)

Two hundred seven sheltered survey respondents that did not report current employment were asked, “What is keeping you from getting or holding on to a job?” One hundred eighty of these endorsed between one and twelve of eighteen options provided on the questionnaire. The average number of reasons endorsed is 2.61. Figure 12 illustrates the reasons selected by 10.0% or more of the survey respondents in descending order of frequency. Seven additional reasons are presented in Table 9.

11.7%

15.1%

15.6%

15.7%

17.9%

19.4%

21.8%

25.1%

27.4%

30.0%

33.0%

0.0% 5.0% 10.0% 15.0% 20.0% 25.0% 30.0% 35.0%

Health Problems

Do Not Have Phone

No Child Care

Disability

Need Clothes

Criminal R ecord

No Permanent Address

Need More Training

No Transportation

Need More Education

There Are No Jobs

Figure 12. Sheltered Only: Factors Keeping Respondents from Getting or Holding on to a Job

San

Bern

ardi

no C

ount

y Po

int i

n Ti

me

2009

page

39

Table 9. Other Factors Keeping Sheltered Respondents From Getting or Holding on to a Job

Factor ShelteredCount

(%)Need more tools 12

(6.7)Do not have Social Security card/ Work permit

11(6.1)

Alcohol or drug problem 10(5.6)

Do not have photo ID 9(5.0)

Do not want to work 4(2.3)

Do not have shower/ bath 4(2.3)

I am retired 3(1.7)

Significantly different proportions of males and females endorsed four of the 18 factors that prevent sheltered survey respondents from obtaining or keeping a job. These include, “Need more education” (Males 20.0%, Females 38.0%); “No child care” (Males 2.7%, Females 25.0%); “Alcohol or drug problem” (Males 10.8%, Females 2.0%); and “Need more tools” (Males 12.2%, Females 2.0%).

E. Factors that Prevent Obtaining Permanent Housing (Sheltered Only)

Sheltered survey respondents were presented nine response options and an opportunity to write in another answer to the question, “What is keeping you from getting permanent housing?” Two hundred fifty-three of 271 sheltered respondents endorsed between one and nine of these reasons, with a mean of 2.5 reasons. Figure 13 illustrates the distribution of responses in descending order of frequency. Note that the full text of the third most frequently endorsed factor preventing permanent housing is “Can’t put together moving costs (security deposit, first/ last month’s rent)”.

San

Bern

ardi

no C

ount

y Po

int i

n Ti

me

2009

page

40

2.4%

10.3%

12.3%

15.8%

17.4%

32.4%

37.2%

51.8%

66.4%

0.0% 10.0% 20.0% 30.0% 40.0% 50.0% 60.0% 70.0%

Do not want to

No housing available

Criminal record

No transportation

Eviction record

Bad credit

Can't get moving costs

No job/ No income

Can't afford rent

Figure 13. Factors that Prevent Obtaining Permanent Housing (Sheltered Only)

A significantly higher proportion of sheltered males (63.6%) than females (42.9%) selected, “No job/ No income” as the factor preventing them from obtaining permanent housing;” χ2 (1,247) = 10.403, p < .002. Conversely, higher proportions of females endorsed, “Bad Credit” (39.3% of females, 24.3% of males), χ2 (1,247) = 6.181, p < .01, “Eviction Record” (25.0% of females, 7.5% of males), χ2 (1,247) = 12.952, p < .001, and “Can’t put together moving costs” (47.1% of females, 23.4% of males), χ2 (1,247) = 14.737, p < .001.

F. Mental and Physical Illnesses and Disabilities

The computation of Continuum of Care Subpopulation Categories including the chronic homeless, severely mentally ill, and chronic substance abuse is detailed in Appendices G and H. This section provides an overview of the survey responses utilized as components of these calculations.

San

Bern

ardi

no C

ount

y Po

int i

n Ti

me

2009

page

41

Female

Male

0.0%

5.0%

10.0%

15.0%

20.0%

25.0%

30.0%

35.0%

40.0%

45.0%

50.0%

18.7%

38.0%30.9%

47.6%

Figure 14. Serious Physical Illness or Disability by Sheltered/ Unsheltered and Gender

Female

Male

ShelteredUnsheltered

Figure 14 illustrates two facts: first, that a significantly larger proportion of unsheltered (street) homeless survey respondents report a serious physical illness or disability (44.8% across both genders—averaging the pale yellow and green bars on the right side of the graphic), than do sheltered respondents (24.4% across both genders—averaging the bars on the left side of the graphic); χ2 (1,496) = 22.798, p < .001. Second, proportionally more males report awareness of a serious physical illness or disability than females in both shelters and on the street, but the difference is statistically significant only among sheltered survey respondents; χ2 (1,260) = 5.238, p < .02.

Considering the 496 survey respondents that answered both questions (“Have you ever been told by a health care provider that you have a mental illness?” and “Do you believe that you have a mental illness?”), 358 (72.2%) report that they have never been told by a health care provider that they have a mental illness and which the respondent refutes. Seventy-nine (15.9%) survey respondents have been told by a health care provider that they have a mental illness, and they also believe that they do. Thirty-one (6.2%) have been told that they have a mental illness but don’t believe it, and 28 (5.6%) have not been told by a health care provider that they have a mental illness, but think that they do. Relying upon self-report of a diagnosis delivered by a health care provider, a significantly higher proportion (n= 60, 26.0%) of unsheltered homeless survey respondents have a mental illness than those completing questionnaires in shelters (n= 50, 18.7%); χ2 (1,498) = 3.780, p < .04. The difference between males (24.5%) and females (19.5%) reporting a diagnosis of mental illness is not statistically significant.

San

Bern

ardi

no C

ount

y Po

int i

n Ti

me

2009

page

42

All survey respondents indicated whether they were currently experiencing nine conditions, forms of disability, and disease. Two hundred sixteen (42.6%) endorsed none of these, and 291 (57.4%) indicated that they were currently experiencing between one and seven of the listed conditions. A significantly larger proportion of unsheltered (n= 151, 64.0%) than sheltered survey respondents (n= 140, 51.7%) report one or more of these conditions, disabilities or diseases; χ2 (1,507) = 7.833, p < .005. Among these 291 respondents, the average number of conditions reported is 1.86, but this average is significantly higher among unsheltered (M= 2.1) than sheltered (M= 1.6 conditions) survey respondents, F (1, 289) = 11.922, p < .002.

1.3%1.9%2.6%

13.5%3.0%

16.8%8.6%

13.5%

12.9%8.4%

9.4%21.3%

3.4%34.8%

17.2%36.8%

39.5%56.8%

0.0% 10.0% 20.0% 30.0% 40.0% 50.0% 60.0%

AIDS/ HIV-Related Illness

Developmental Disability

Drug Abuse

PTSD

Domestic Violence

Chronic Health Problem

Alcohol Abuse

Physical Disability

Depression

Figure 15. Conditions and forms of Disability and Disease Currently Experienced by S heltered and Unsheltered Homeless Persons

Unsheltered

Sheltered

Figure 15 illustrates the fact that a higher proportion of unsheltered survey respondents report each condition except domestic violence, which is higher in the sheltered population. Except for AIDS/ HIV-related Illness, PTSD, and domestic violence, the differences between unsheltered (street) and sheltered populations are statistically significant.

The best method of producing CoC population and subpopulation counts involves computing the ratios within each sheltered and unsheltered gender from the survey data, and multiplying these by the counts by gender produced during the Point-In-Time Count.

Understanding the needs and characteristics of homeless people who do not use shelters, many of whom are chronically homeless, is critical to the 10-Year Strategy to End Homelessness in San Bernardino County. HUD’s definition of chronic homelessness is: “An unaccompanied homeless individual with a disabling condition who has either been continuously homeless for a year or more OR has had at least four (4) episodes of homelessness in the past three (3) years. To be considered chronically homeless, persons must have been sleeping in a place not meant for human habitation (e.g., living on the streets) and/or in emergency shelter during that time. Note that HUD’s definition of

San

Bern

ardi

no C

ount

y Po

int i

n Ti

me

2009

page

43

chronic homelessness does not include families. In addition, to be identified as chronically homeless, an individual must have a disabling condition, defined as follows: A diagnosable substance use disorder, serious mental illness, developmental disability, or chronic physical illness or disability, including the co-occurrence of two or more of these conditions. A disabling condition limits an individual’s ability to work or perform one or more activities of daily living.”2

Using an estimation process described in Appendices G and H, the number of chronically homeless persons in the unsheltered homeless population in San Bernardino County is 241; 27.1% of unsheltered adults, and 24.9% of the total unsheltered population. Among the sheltered population enumerated during the Point-In-Time Count, 62 are chronically homeless (12.9% of sheltered adults and 8.1% of the total sheltered population).

G. Main Reason for Current Homelessness

Four hundred thirty-one (85.0%) survey respondents indicated from a list of nineteen options the main reason for their current homelessness. Although the questionnaire guided respondents to “Please check only one,” about 15% of those who answered endorsed between two and six reasons, so the total number of reasons exceeds the number of respondents. Proportionally more sheltered respondents (88.2%) than unsheltered respondents (81.4%) answered this question. Six of the nineteen response options garnered an endorsement from at least ten percent of either the sheltered or unsheltered population, or from both. These reasons are illustrated by Figure 16.

2 “Guide to Counting Unsheltered Homeless People: Second Revision,” Office of Community Planning and Development, U.S. Department of Housing and Urban Develop-ment, January 15. 2008, pg. 9.

San

Bern

ardi

no C

ount

y Po

int i

n Ti

me

2009

page

44

0.0% 5.0% 10.0% 15.0% 20.0% 25.0% 30.0% 35.0%

Argument, Asked to Leave

Divorce or Separation

Parents' Household Fell Apart

Domestic Violence

Alcohol or Drug Problem

Lost Job

10.5%

9.6%

9.6%

17.2%

14.2%

34.3%

8.9%

10.9%

11.5%

3.6%

9.4%

25.5%

Figure 16. Top "Main Reasons Now Homeless" by Sheltered/ Unsheltered

Unsheltered

S heltered

The proportions of street (25.5%) and sheltered (34.3%) survey respondents that cite “lost job” as a main reason for their current homelessness are significantly different; χ2 (1,431) = 3.887, p < .04. Likewise, the rate of unsheltered survey respondents (3.6%) attributing their current homelessness to “domestic violence” is significantly lower than among sheltered respondents (17.2%); χ2 (1,431) = 19.633, p < .001. Differences between sheltered and unsheltered respondents on the other four “main reasons” are not statistically significant.

San

Bern

ardi

no C

ount

y Po

int i

n Ti

me

2009

page

45

Table 10. Other “Main Reasons Now Homeless” by Sheltered/Unsheltered

Reason ShelteredCount

(%)

UnshelteredCount

(%)Mental health issue 16

(6.7)11

(5.7)Illness or medical problem 14

(5.9)10

(5.2)Just out of jail 16

(6.7)3

(1.6)Just out of hospital/ treatment program

3(1.3)

16(8.3)

Foreclosure 6(2.5)

9(4.7)

Lost government benefits/ assistance

5(2.1)

9(4.7)

Landlord raised rent 3(4.2)

0(0.0)

Landlord stopped renting because of foreclosure

5(2.1)

5(2.6)

Just out of alcohol/ drug treatment program

10(4.2)

0(0.0)

Too old for foster care 3(1.3)

5(2.6)

For being lesbian, bisexual, gay or transgender

0(0.0)

8(4.2)

Natural disaster (fire or flood) 2(0.8)

2(1.0)

Gambling problem 0(0.0)

1(0.5)

San

Bern

ardi

no C

ount

y Po

int i

n Ti

me

2009

page

46

V. Homeless Count and Survey Methodology

A. Overview of Planning Process

Community outreach and volunteer recruitmentThe first step in any homeless count planning effort is to establish contact with shelter and other homeless service providers. Although by no means the only stakeholder groups with input into the planning process these providers have daily contact with the population to be studied. In addition, many service providers and their staff are the first to begin outreach to the homeless to convey the importance of the Point-In-Time process. Contact with shelter and homeless service providers in San Bernardino County began in the fall of 2008. In addition to extending an opportunity to provide input into the planning process and encouraging them to inform their clients when the homeless count would be conducted, providers were also asked to assist with identifying “hot spots”, or outdoor locations where homeless persons were known to congregate. As described later in this report, identification of these hot spot locations was critical to the selection of the geographic areas where field teams of volunteer enumerators would be sent.

Cities within the county boundaries were the second focus of outreach efforts. Through their mayors and city managers, all jurisdictions within the county were informed regarding the intent of the Point-In-Time process and the timeline of activity. Their assistance with both hot spot identification and volunteer recruitment was solicited. City staff from many different departments, ranging from police and fire departments to parks and recreation, often assists the homeless in some fashion on a daily basis. These staff members are most likely to have both valuable information about outdoor congregating locations and an interest in making the overall process successful. Some cities actually assigned dedicated staff members to assist the County staff with the planning efforts.

Recruiting volunteer enumerators and surveyors was the third focus of our community outreach efforts. Through a variety of means, dedicated County staff began outreach efforts to many different types of organizations in the fall of 2008. The volunteer coordination team communicated with and conducted on-site presentations with: homeless service

San

Bern

ardi

no C

ount

y Po

int i

n Ti

me

2009

page

47

providers, high school and college service learning groups, churches and other faith based organizations, and volunteer networks such as 2-1-1. Given the ambitious goal of canvassing approximately 200 geographic areas across the entire county, known as Census block groups, within a five hour time frame on one evening, the planning team predicted that several hundred volunteers would be needed.

Approximately 500 volunteers were ultimately dispatched in three or four-person field teams during the Point-In-Time Count on the evening of January 22, 2009 and the early morning hours of January 23, 2009. Depending upon the geographic areas to which teams were assigned, some teams only canvassed one area (typically in dense urban areas), while others volunteered to canvass two or three contiguous areas (typically those in larger desert or rural areas). Volunteers were required to attend a one and one half hour training in the weeks prior to the count that reviewed the enumeration and survey forms, appropriate procedures for ensuring their safety and the privacy and dignity of those being enumerated and surveyed, and other logistical information.

It is important to note that the Point-In-Time Count & Survey process is an unfunded federal mandate. However, because it is a prerequisite to applying for homeless service funding, most Continuums of Care conduct some type of homeless count consistent with HUD guidelines. In addition to the funds budgeted for the planning and implementation of the overall process, the County also budgeted funds to purchase hygiene and food items to provide to homeless individuals that agreed to participate in the survey process.

B. Street Count Method

San Bernardino County is 215 miles wide and 150 miles from North to South with a total land area of 20,052 square miles. The relative size of the county is frequently described by observing that two states the size of Rhode Island, three the size of Delaware, and the states of Connecticut and Hawaii would all fit within the San Bernardino County boundaries with open space remaining.3 Even excluding much of its desert and mountainous terrain, canvassing the entire county with a limited pool of volunteers on one night is not practical. Consequently, rather than a census approach (canvassing the entire county) a sampling strategy was devised for the 2009 Point-In-Time Count. The strategy employed in 2009 was based upon a methodology used in New York City’s 2003 Manhattan Point-In-Time Count, revised in 2004.4

In this model featured by HUD, the New York City Department of Homeless Services (NYDHS) divided the borough into 899 surface study areas composed of clusters of census blocks and 143 underground subway stations/platforms. NYDHS then assigned a density classification of high, medium, or low to each surface study area and subway station and selected a stratified sample. After completing the count in 2003, NYDHS found that the average densities in low and medium areas were very similar and they were combined. In the 2004 count, a high-density area was defined as including at least one homeless person.

3 http://www.usacitiesonline.com/casanbernardinocounty.htm4 “Guide to Counting Unsheltered Homeless People: Second Revision,” Office of Commu-nity Planning and Development, U.S. Department of Housing and Urban Development, January 15. 2008, pgs. 113-115.

San

Bern

ardi

no C

ount

y Po

int i

n Ti

me

2009

page

48

Each high-density study area or subway station and a random sample of low density locations were canvassed. The number of low-density locations included in the sample count was determined by a confidence interval criterion of 95 percent confidence that the weighted sample count was within 10 percent of the actual number of unsheltered homeless persons in these areas.

To implement this strategy locally, personnel working with the San Bernardino County Office of Homeless Services (OHS) identified 298 individual “hotspots” or outdoor congregating areas where homeless persons had been observed in the months preceding the count. These locations were geocoded by the Information Services Department, Geographic Information Systems unit (ISD-GIS) and found to be contained in 158 (14.4%) of the county’s 1,094 eligible block groups. Five block groups were excluded from the 1,099 U.S. Census-defined Block groups in San Bernardino County because they encompass military installations or desert tracts with zero population.5 Fifty-three (33.5%) block groups of the 158 contained between two and fourteen “hotspots” and 105 block groups (66.5%)contained one. OHS obtained estimated counts of homeless for 220 (73.8%) of the 298“hotspots.” These estimates ranged from one to 200 homeless persons with a mean of 6.55 and a median (the point above which and below which half the values fall) of 3.5. The estimated counts obtained summed to 1,441 homeless persons.

In the context of developing a sampling design before the final density strata were defined, the 78 hotspots lacking estimated counts of homeless persons were assigned “imputed” counts. Several approaches might have been taken here. To be conservative, a count of “1” might have been arbitrarily assigned to these hotspots, but because some of them were described as multiple encampments, significantly more homeless persons than that may have been found in these spots. If the decision was made to classify block groups with hotspots into multiple “density” levels, the within-stratum variance would have been high. This would not have biased sample selection, but may have made it less statistically efficient by elevating sampling variance.

Consequently, contextual information such as the estimated counts in proximate locations and/or similar settings was utilized to replace missing values for estimated counts in 78 cases. These “imputed estimates” improved the likelihood that the respective block groups would be classified into the proper density strata. The imputed estimates would not bias sample selection and most likely would have made it more efficient by reducing within-stratum variance. Accordingly, the actual Point-in-Time Count, which resulted from teams carefully canvassing the block groups selected into the final sample, would have been more precise.

With the 78 estimated-count missing values replaced by imputed estimates, the sum of homeless persons increased to 1,757 distributed across 158 block groups in San Bernardino County. Accumulating the “hotspot” estimated counts, the average number of homeless persons per block group was 10.46, the median was 5.5, and the mode (most frequent

5 A U.S. Census block group is a geographic unit that typically encompasses a population of between 600 to 3,000 residents, with an optimum of 1,500 persons.

San

Bern

ardi

no C

ount

y Po

int i

n Ti

me

2009

page

49

value) was 2.0. Emulating the revised New York sampling design, the 158 block groups with estimates of homeless greater than one (1.0) were ultimately assigned to a High Density stratum. The remainder, (1,094 – 158= 936) including block groups where no hotspots were identified, were assigned to the Low Density stratum.

1. Sample selection

All block groups in the High Density stratum were selected with certainty. Before sampling in the Low Density stratum, some pairs of block groups were consolidated to form single sampling units to help even the workload of field teams on the night of the count, thereby increasing field productivity. Pairs of block groups were linked as a single sampling unit if they were contiguous and in the same census tract, and if their combined area was less than 1,000 acres. Following these criteria, the number of sampling units in the Low Density stratum was reduced from 936 (1,094 block groups in the first-stage sampling frame minus 158 block groups selected with certainty = 936) to 631. Next, sampling units in the Low Density stratum were sorted by acreage, low to high, and a random selection of 50 sampling units was obtained using Systematic Random Sampling (sometimes called “interval sampling”). Thirty of the selections were single block groups, and twenty were linked pairs of block groups.

A confidence interval of plus or minus ten percent is the target suggested by the Manhattan sampling strategy approved by HUD. This refers to the level of precision required of estimates of an entire study area that are based upon sample data. When a confidence interval is specified we can be 95% confident that the true population parameter lies within an interval extending that percentage above and below any proportion derived from sample data. A population parameter is the result one would obtain if every sampling unit in the Low Density stratum was canvassed.

Sampling error, as indexed by the confidence interval around reported proportions, varies in relation to sample size and to the variability of the results (among other factors). In general, as the sample grows larger, the confidence interval grows narrower. That is, inferences about population parameters based upon larger samples are more precise (are associated with narrower confidence intervals) than inferences based upon smaller samples.

Also, as the proportion of some attribute in the sample (e.g. sampling units in which homeless persons are observed) approaches a fifty/ fifty split, sampling error increases, resulting in a wider confidence interval. Conversely, sampling error decreases as the proportion of a given attribute approaches a five/ ninety-five split, (e.g. homeless persons are observed in 5% of the sampling units and not in 95%) resulting in a more precise estimate and a narrower confidence interval. Homeless persons were observed in seven of the 49 canvassed Low Density stratum sampling units, fixing this proportion at .143/ .857. The sample of 49 selected from a population of 631 results in a confidence interval of plus or minus 9.61%.

San

Bern

ardi

no C

ount

y Po

int i

n Ti

me

2009

page

50

Sampling from both strata combined resulted in 208 selections, encompassing 228 individual block groups. Due to resource constraints, eight block groups of the 158 selected with certainty, and one of the 50 sampling units in the Low Density stratum were not canvassed on the night of the count. This diminished the total of block groups canvassed from 208 to 199. This sampling strategy produced a weighted, estimated count with a formula:

Count = Ni*Ki

where Ni is the Count for the ith Block group; and Ki=(631/49) if the ith Block group is from the Low Density stratum, and Ki=1 otherwise.

Using equivalent mathematics, but expressing the procedure slightly differently, the count of unsheltered homeless persons in San Bernardino County in the Low Density strata is computed by multiplying the mean number of persons observed (not the total, but the average number in each of the eight demographic categories) in the canvassed Low Density sampling units by the total number of Low Density units (n= 631). So, for example, volunteer enumerators observed a total of 17 males estimated to be 18 to 59 years of age in the 49 Low Density sampling units. 17/49= .347; .347 * 631= a projected count of 218.92 males in the Low Density sampling units. So the final (street) count of unsheltered homeless persons in this demographic category in San Bernardino County is computed by adding 218.9 to the subtotal from the High Density block groups (375.8), producing a total of 594.7 unsheltered males, 18 to 59 years of age. Repeating this process in each age by gender category results in an extrapolation of 347.7 homeless persons in Low Density sampling units. At a confidence interval of +/- 9.61%, we are 95% confident that the true number lies between 314.3 and 381.1. Consequently, the total unsheltered population lies somewhere between 314.3 + 620.6= 935 and 381.1 + 620.6= 1,002.

2. Description of Enumeration Method

Street Count

A two-stage sample, with selection of geographic units in the first stage, and homeless persons to be surveyed within those selected geographic units in the second stage, was designed to collect data for the 2009 San Bernardino County Point-In-Time Homeless Count.

First Stage. The first-stage sampling frame consists of 1,094 of the 1,099 U.S. Census-defined block groups in San Bernardino County. Five block groups were excluded from the sample frame because they encompass military installations or desert tracts with zero population.

Available volunteer resources permitted 199 block groups to be thoroughly canvassed on the night of the count. To enhance statistical efficiency, block groups where homeless persons are most likely to be found were oversampled. To implement this strategy, the sampling frame was stratified on the basis of rough estimates of homeless persons to be found in each block group.

San

Bern

ardi

no C

ount

y Po

int i

n Ti

me

2009

page

51

Personnel working with the San Bernardino County Office of Homeless Services (OHS) identified 298 individual “hotspots” or outdoor congregating areas where homeless persons had been observed in the months preceding the count. OHS also provided estimated counts of homeless for 220 of the 298 hotspots. For the 78 hotspots lacking count information, estimates were “imputed” from contextual information such as the estimated counts in proximate locations and/or similar settings. The total of OHS and imputed estimates summed to 1,757 homeless persons in outdoor locations.

The hotspot locations were geocoded and assigned to their respective block groups by the Information Services Department, Geographic Information Management Systems unit (ISD-GIMS). All hotspots were found to be contained in 158 of the county’s 1,099 block groups, with a median block group value of 5.5 homeless persons.

Two strata were created for the sampling frame: High Density and Low Density. The 158 block groups with estimates of homeless greater than one (1.0) were assigned to the High Density stratum. The remainder, including block groups where no hotspots were identified, was assigned to the Low Density stratum. The High Density block groups were selected with certainty (all 158 were included in the sample), and the remaining 936 block groups in the county were combined using specific criteria to form 631 sampling units. From these, a random sample of 50 was selected.

A total of 616 persons in eight demographic categories were physically observed on the night of the street count in places not meant for human habitation; 589 in High Density block groups and 27 in Low Density sampling units. Projections calculated for eight High Density stratum block groups selected into the sample that were not canvassed due to resource constraints increase the High Density total of 589 by 31.6 to 620.6. Projections based upon the Low Density stratum count of 27 in 49 sampling units to the total of 631 sampling units add 347.7 persons, building to an estimate of 968 unsheltered homeless persons.

Second Stage. On the night of the Point-In-Time Count, all homeless persons physically observed in the selected block groups were counted. In each block group, a number of homeless persons (depending on the size of the respective block group count) were sub-sampled and approached for interview. Fifty-seven interviews with unsheltered homeless persons were completed on the night of the Point-In-Time Count.

In addition to the tallies of persons observed in eight age-by-gender classifications, volunteers provided hand-written notes on the enumeration forms. Volunteers were trained to preserve the privacy of the homeless by refraining from knocking on vehicle windows or doors to verify that persons were indeed sleeping inside. However, many of the notes on enumeration forms indicated the presence of recreational vehicles or campers; sport/ utility vehicles or vans; and passenger cars in which volunteer enumeration teams suspected that persons were sleeping. Similarly, notes indicating the presence of tents or

San

Bern

ardi

no C

ount

y Po

int i

n Ti

me

2009

page

52

campsites that volunteers suspected were occupied were recorded. Again, though, the “do not disturb” policy, protected the privacy of homeless persons and safety of field team volunteers, constrained volunteers from entering makeshift shelters to determine whether or not they were occupied and may have lowered the count.

A total from the volunteer enumerators’ notes would not be useful because they were not instructed to record vehicles or campsites, and such reporting was not consistently done by all field teams. HUD explicitly prohibits using, “unscientific ‘adjustment factors’ to derive…counts of the unsheltered population…to account for people not seen during the point-in-time count6.” Consequently, this information has not been utilized to augment the San Bernardino Point-In-Time Count.

C. Shelter Count Method

Enumerating sheltered homeless persons is a relatively straightforward activity. HUD defines sheltered homeless persons as adults, children, and unaccompanied youth who, on the night of the count, are living in shelters for the homeless, including:

• Emergency shelters;• Transitional housing;• Domestic violence shelters;• Residential programs for runaway/ homeless youth; and• Any hotel, motel, or apartment that accepts voucher arrangements paid by a

public or private agency because the person or family is homeless. This does not include those living in permanent housing supported by a HUD Housing Choice Voucher (also known as Section 8 vouchers).

HUD further advises that the following types of people should not be counted as part of the sheltered population:

• Persons living doubled up in conventional housing;• Formerly homeless persons living in Section 8 SRO, Shelter Plus Care, Supportive

Housing Program permanent housing or other permanent housing units;• Children or youth, who because of their own or a parent’s homelessness or

abandonment now reside temporarily or for a short anticipated duration in hospitals, residential treatment facilities, emergency foster care, or detention facilities;

• Adults in mental health facilities, chemical dependency facilities, or criminal justice facilities7.

6 “Guide to Counting Unsheltered Homeless People: Second Revision,” Office of Commu-nity Planning and Development, U.S. Department of Housing and Urban Development, January 15. 2008, pgs. 12-13.7 “A Guide to Counting Sheltered Homeless People” U.S. Department of Housing and Ur-ban Development, Office of Community Planning and Development; October 2, 2006.

The process of determining where to find sheltered homeless in San Bernardino County was made relatively simple by the use of existing contact lists of known emergency and transitional shelter programs provided by the County OHS. The shelter staff on duty the night of the Point-In-Time Count were provided with a copy of the shelter enumeration form and briefed in the days prior to the count on how to appropriately complete the form. The enumeration form used by the shelter staff is substantially similar to the form used by field teams counting the unsheltered homeless. The only difference is the addition of a box that asks for the number of hotel/motel vouchers provided that night, if any.

Although there was a desire by the Point-In-Time Count planning committee to include homeless being discharged from institutions, no jails, hospitals or similar institutions were included in the process this year because limited staff resources during the planning processprecluded establishing relationships with discharge staff in the months prior to the count. Further, on the evening of the count itself, limited volunteer availability did not permit assignment of enumerators to these institutions.

D. Homeless Survey Method

Separate survey instruments for the unsheltered or street homeless and for the sheltered homeless were used in the 2009 Point-In-Time process. Almost all of the questions were identical in both surveys, however because those in sheltered environments were typically in settings more conducive to survey administration and could take more time to fill out the questionnaires, several additional questions were added to the sheltered instrument.

Proctoring surveys with the unsheltered homeless is a much more difficult proposition, particularly at night. Keeping the total number of questions to a minimum was imperative to produce accurate data efficiently. To achieve both of these objectives, with the help of The Salvation Army of San Bernardino, the research consultant pilot-tested the unsheltered homeless survey questionnaire. The working draft included approximately 40 questions that were deemed necessary to provide requested data to HUD, or to provide guidance to OHS staff for future homeless service planning efforts. In a focused group discussion, participants were asked to judge the clarity of the wording and overall ordering of the questions. The resulting feedback was informative regarding final revisions of the unsheltered survey instrument.

Of the 236 surveys completed by unsheltered homeless persons, 57 were administered on the night of the Point-In-Time Count by volunteer enumerators. Due to the limited timeframe in which to both conduct the Count and attempt to administer some number of surveys to the unsheltered homeless, volunteers were advised to select potential survey respondents based upon the size of the group in which they were found. For groups of up to three homeless persons, volunteers were to attempt to survey all members of the group; in groups of four to eight persons, volunteers were to attempt to survey every other person; and in groups of nine or more persons, volunteers were to attempt to survey every third person. The remaining 179 surveys of unsheltered homeless were gathered by the

San

Bern

ardi

no C

ount

y Po

int i

n Ti

me

2009

page

54

staff of the San Bernardino County OHS in several outdoor service and congregating areas during the months of February and March, 2009.

Administration of surveys to sheltered homeless was coordinated through the County OHS. Some service providers, particularly those serving domestic violence victims, had their own staff proctor the survey with their clients. The remaining providers allowed OHS staff to personally proctor the survey with their clients. The process of gathering survey data from sheltered homeless persons also took place between the first week of February, 2009 and was completed at the end of April, 2009. Some shelter provider staff members, as well as the staff of the County OHS, were trained on the appropriate administration procedure for the sheltered homeless survey. The distribution of unsheltered surveys conducted on the night of the Point- In-Time Count is presented in Table 11 by deployment center, and the distribution of sheltered surveys by shelter name in Table 12.

Table 11. Distribution of Surveys Administered to Unsheltered Homeless Persons Collected on the Night of the Count by Deployment Center

Deployment Center Count(%)

HS Lena 15(26.3)

Fontana 12(21.1)

Barstow 10(17.5)

Del Rosa 5(8.8)

Yucca Valley 5(8.8)

Ontario 4(7.0)

Rancho Cucamonga 4(7.0)

Redlands 2(3.5)57

(100.0)

One hundred seventy-nine surveys were administered to unsheltered homeless persons over the weeks following the count by personnel of the San Bernardino County Human Services Office of Homeless Services. Forty (22.3%) of these were administered at the Frazee Community Center, 32 (17.9%) at the City of Ontario homeless encampment, and 107 at other locations throughout San Bernardino County.

San

Bern

ardi

no C

ount

y Po

int i

n Ti

me

2009

page

55

Table 12. Distribution of Surveys Administered to Sheltered Homeless Persons by Shelter Name

Shelter Count(%)

Central City Lutheran Mission

42(15.5)

High Desert Homeless Services

28(10.3)

Salvation Army 27(10.0)

Victor Valley Domestic Violence

20(7.4)

Mercy House 19(7.0)

Desert Manna Ministries 16(5.9)

Veronica’s House of Mercy

11(4.1)

Time for Change Foundation

10(3.7)

High Desert Domestic Violence Program

8(3.0)

New Hope Village, Inc. 6(2.2)

Pacific Lifeline LMWS, Inc.

6(2.2)

Mimi’s House 4(1.1)

Desert Sanctuary (DV) 3(1.1)

Option House (DV) 3(1.1)

DOVES of Big Bear Valley, Inc.

2(0.7)

Shelter Unidentified 66(24.4)

Total 271(100.0)

San

Bern

ardi

no C

ount

y Po

int i

n Ti

me

2009

page

56

E. Differences Between the 2007 and 2009 Homeless Counts

Differences in methodology and implementation of the 2007 and 2009 homeless count processes make comparison of the results between years extremely difficult. The County OHS expects the methodology employed in 2009 to be used in 2011, which will produce data amenable to comparison. Significant differences in methodology, as well as other issues not addressed in the 2007 report, however, make it difficult to make scientifically supportable comparisons between 2007 and 2009. The primary differences between years fall into three categories:

1. Use of an unapproved “multiplier” - page 16 of the 2007 report describes that the method by which the number of homeless persons found in cars, vans/RV’s and encampments was calculated using a multiplier. The 2007 report states that, “Out of concern for the privacy of the

homeless community and the safety of the enumerators, census workers were instructed not to approach vehicles or to enter encampment areas. Instead, when they could not clearly count the number of occupants, enumerators were asked to just note on their census tally sheet how many vehicles or encampments they discovered.”

As a result of not actually observing a determinant number of homeless persons in each of these types of locations, a multiplier was created, “by asking survey respondents who reported staying in cars, vans, RVs, or encampments how many people typically stayed there, producing an average number of people for each of these sleeping locations. The multiplier for cars was 1.2424, for vans/RVs it was 1.7692, and for encampments it was 2.6842. These multipliers were applied to every car, van/RV, or encampment indentified…”

However, HUD’s A Guide to Counting Unsheltered Homeless People specifically notes that, “In the past, many CoCs used unscientific ‘adjustment factors’ to derive their counts of the unsheltered population…using an adjustment factor to account for people not seen during the point-in-time count. HUD no longer allows CoCs to use such adjustment factors.”8

The multiplier used to calculate the number of homeless persons believed to be in cars, vans, RVs and encampments appears to be an adjustment factor not allowed under the HUD guidelines. Because the field workers noted only the number of vehicles or encampments, but not the number of persons observed in these locations, the final estimates of homeless persons created using these multipliers are

8 “Guide to Counting Unsheltered Homeless People: Second Revision, “Office of Commu-nity Planning and Development, U.S. Department of Housing and Urban Development, January 15, 2008, pg. 12-13.

San

Bern

ardi

no C

ount

y Po

int i

n Ti

me

2009

page

57

not data derived from HUD approved methods. The charts on page 21-23 of the 2007 report lists the numbers

of sheltered and unsheltered homeless persons estimated in 30 jurisdictions derived using these unapproved multipliers.

The 2007 enumeration form, as shown on page 67 of the 2007 report, asks for reporting of vehicles and encampments, rather than actually observed homeless in the following locations: • “# of cars with sleeping occupants”• “# of Vans or RVs with electrical or water connections”• “# of encampments with people”• “# of Persons reported by Park Rangers”

If the contribution to the 2007 homeless census created by using these multipliers (3,017—41.2%) is removed from the total count of homeless persons reported that year (7,331), the final 2007 homeless count is reduced to 4,314 persons.

2. Possible duplication of observed homeless - There was no explanation in the 2007 report of how, per the HUD guidelines, potential duplication of observed homeless was avoided under certain conditions. The primary intent of a point-in-time process is to produce an unduplicated count of the homeless in a particular area during the specified point-in-time. On page 29 it is noted that law enforcement helicopters were employed

to search hard-to-reach locations. However, there is no mention of which areas the helicopter teams were assigned to or what efforts were made to ensure that their counts were not duplicative of what the ground teams were observing.

As noted above, the last box on the 2007 enumeration form asks for field teams to note the number of homeless persons reported by park rangers. This is also problematic because there is no discussion in the report of how the final analysis determined that park ranger observations were not duplicative of information submitted by field teams in the same areas.

3. Was the entire county canvassed or a selected sample of areas? - Page 1 of the 2007 report states that field volunteers “canvassed” all 244 census tracts in the county. However, it is unclear if the final data reported are based solely upon actually observed homeless in all 244 census tracts, or if the numbers were in fact extrapolations or projections. The term “canvassed” suggests an attempt at fully observing a designated geographic area. Given the vast size of San Bernardino County, as noted earlier in this report, it is unclear whether the approximately 300 volunteers and homeless workers in 2007 fully canvassed all of the County’s 244 census tracts. In the last paragraph on page 29 of the 2007 report it is noted that “selected

tracts were traveled…” This suggests two issues not consistent with the basic methodology and canvassing process noted earlier in the report: (1) that only certain “selected” tracts were in fact visited by volunteers, rather than all of the county’s census tracts, and (2) that the field volunteers “traveled” around their assigned tracts, rather than actually “canvassing”

San

Bern

ardi

no C

ount

y Po

int i

n Ti

me

2009

page

58

them. A “drive-by” count appears to have been conducted in 2007. The numbers of homeless persons actually observed would vary widely

depending upon which strategy was employed – a full canvass of each census tract, or a selective observation of certain locales within each census tract, based on specific intelligence.

San

Bern

ardi

no C

ount

y Po

int i

n Ti

me

2009

page

59

VI. Homeless Count Challenges and Assumptions

A. Challenges to the Street and Shelter Counts

Research targeting a homeless population encounters a number of challenges. Any homeless population – whether studied in its entirety, or in smaller subpopulation categories – is by its very nature highly mobile. Additionally, most homeless persons are disposed to hide themselves from public scrutiny through a variety of means and for many different reasons. The tendencies of this population to either intentionally hide from enumerators or to simply be difficult to find as they try to stay “under the radar” is a significant hindrance to a homeless count process. Some of the reasons the homeless can be hard to enumerate include:

• Parents who hide their children for fear of having to turn their children over to Child Protective Services;

• Chronically homeless persons who may not seek services or government benefits;• Persons who live in vehicles and must move every few days;• Persons who overcrowd into homes, apartments and motel rooms above the

acceptable occupancy levels;• Homeless youth who tend to be less visible than homeless adults;• And, homeless individuals and groups that camp in locations that are not easily

accessible, such as riverbeds, under bridges, or in abandoned buildings.

In San Bernardino County in particular, another significant challenge to an accurate homeless count and survey process is the geographic size of the county. Although the county’s total population in 2007 was 2.05 million, that population was spread across 20,052 square miles, much of which is largely uninhabited desert, farm or park land. This geographic

San

Bern

ardi

no C

ount

y Po

int i

n Ti

me

2009

page

60

spread presents a particular difficulty when trying to determine whether recreational and other large vehicles are housing homeless persons or leisure campers. These vast expanses of desert and park land attract many leisure campers throughout the year, and it can be difficult to distinguish them from homeless living in vehicles or encampments, unless they self-identify themselves to enumerators. Further, this has made providing homeless services across the county in a regionally coordinated way difficult to achieve, as many areas are simply too far from population centers to create an effective and economically viable service provision system. Unlike more densely populated urban counties, there are few service locations that see a large concentration of homeless, or are generally known by the local community to be a central service location for homeless persons.

The definition of homelessness established by HUD limits the Point-In-Time Count to a potentially small subset of the total homeless population. This definition of homelessness makes the final count numbers necessarily limited to a potentially small subset of the total homeless population. Most service providers agree that the definition used by HUD for this process is quite conservative, primarily focusing on those that are chronically homeless or those that have sought assistance with a shelter provider. This narrow focus leaves out other at-risk or precariously housed populations, such as:

• Those that are long-term residents of motels/hotels;• Those that are “doubled-up” in informal living arrangements with friends or family

members; and• Those that are temporarily sheltered in hospitals, jails, or rehabilitation facilities.

In addition to these specific populations that could not be included in this process, because of their attempts to hide from others in efforts to protect themselves, enumerators undoubtedly failed to observe homeless persons in highly inaccessible areas, such as forested and desert areas, abandoned buildings, and other areas that were perceived to be inaccessible or unsafe. In spite of these practical limitations, every effort was made to count all observable homeless, and not to count any homeless person or persons more than once.

B. Challenges to the Survey Process

Although any survey process is subject to difficulty, surveying any homeless population can be an especially challenging undertaking. The results of such a process are dependent upon the willingness of the participant to self-report accurately, which is a variable factor with any group of respondents. However, a further complication with surveying a homeless population in particular is the fact that some questions must ask about sensitive topics in order to acquire the information required by, in this case, the federal mandates of the process. While many respondents were very open about their circumstances, explaining that they saw it as a means of helping others, some respondents were understandably uncomfortable with providing in-depth information about their experiences.

Another challenge to the Point-In-Time survey process is the usually difficult circumstances

San

Bern

ardi

no C

ount

y Po

int i

n Ti

me

2009

page

61

under which the unsheltered respondents must participate. This is particularly true for those that were observed the night of the count and were willing to complete a survey – most of those respondents participated in inclement weather, out-of-doors.

San

Bern

ardi

no C

ount

y Po

int i

n Ti

me

2009

page

62

VII. Conclusions

The HUD directed Point-In-Time process first became a requirement for communities seeking homeless services funding in 2005, and biennially thereafter. However, the impetus for requiring a standardized homeless census of some sort began many years earlier. The significance of that timeline is that it highlights one of the primary policy objectives of HUD at that time, which was to reduce the population of chronically homeless. In particular, the theory that reducing the number of the most service-intensive homeless clients would not only reduce the overall homeless population, but also make available more services for those who were not as consistently accessing social service resources. With examples like “Million-Dollar Murray”9coming to light during a time of economic prosperity, this goal seemed even more sensible.

However, social service providers have long held that there has always been a consistent population of family homeless who needed a different level and type of service than the chronic homeless. Proof of this, for the most part, has been largely anecdotal in nature. But, the current economic crisis seems to have exacerbated these reports, as some community data submitted to HUD seems to suggest. As The 2008 Annual Homeless Assessment Report to Congress states, “The number of homeless individuals was fairly stable, while homelessness among persons in families increased by about 43,000 or 9

9 Gladwell, M. (2006, Feburary 13). Million-Dollar Murray. The New Yorker website: www.newyorker.com. Retrieved January 2, 2009, from Malcolm Gladwell’s website: http://www.gladwell.com/2006/2006_02_13_a_murray.html

San

Bern

ardi

no C

ount

y Po

int i

n Ti

me

2009

page

63

percent. Accordingly, the share of family households among all sheltered households also increased, by nearly 3 percentage points.”10

As a result of both the current economic situation, and many years of support by homeless advocates for policy change to address these two subpopulations, federal homeless policy seems to be shifting toward a multi-tiered definition of chronic homelessness, which includes families as well as individuals. The recently reauthorized McKinney-Vento Act, which is a primary funding source for homeless services through HUD, expanded the definition of chronic homelessness to include families. Further, the newest federal funding source for homeless services, known as the Homelessness Prevention and Rapid Re-Housing Program (HPRP), which is a $1.5 billion component of the American Recovery and Reinvestment Act of 2009, is intended to prevent homelessness and to encourage the resettlement of homeless families and individuals in permanent, rather than long-term shelter housing. However, the HPRP funds and the change in definitions will not begin to noticeably affect the provision of services for many months, if not several years. It is in this arena that continued research into the benefits and impacts of these policy changes on local populations and service provision are greatly needed.

Even with the change in federal funding policies, there are many policies on which local leaders can have vast influence. There are numerous reasons that individuals and families become homeless, some of which involve individual issues (for example, losing a job, domestic violence or mental illness) and others of which involve larger regional issues (gaps in services and shortage of permanent housing options). Both types of issues can be greatly influenced by local policies and the amount of resources invested in services. Research into the existing regional barriers to the homeless in finding permanent housing can help inform policy directives, such as the 10-Year Strategy to End Homelessness. However, the outcomes of such policy decisions can only be known with far more local research, in addition to that which is currently required by HUD.

10 U.S. Department of Housing and Urban Development. (2009). The 2008 Annual Homeless Assessment Report to Congress. Washington D.C. Retrieved on July 14, 2009 from HUD website: http://www.hudhre.info/documents/4thHomelessAssessmentReport.pdf

San

Bern

ardi

no C

ount

y Po

int i

n Ti

me

2009

page

64

Appendix ATerms and Definitions

Census Block GroupA census block group (BG) is a cluster of census blocks having the same first digit of their four-digit identifying numbers within a census tract. (See Census Track below.) Block Groups (or BGs) generally represent a population of between 600 and 3,000 people, with an optimum size of 1,500 people.

Census TractA relatively permanent statistical subdivision of a county used for the purpose of presenting data. Census tract boundaries normally follow visible features, but may follow governmental unit boundaries and other non-visible features in some instances. The spatial size of census tracts varies widely depending on the density of settlement, ranging from a population of 2,000 to 8,000 inhabitants (an average of about 4,000 inhabitants is preferred).

Chronically Homeless PersonDefined by HUD as an unaccompanied homeless individual with a disabling condition who has either been continuously homeless for a year or more or has had at least four (4) episodes of homelessness in the past three (3) years. In defining chronically homeless, the term “homeless” means “a person sleeping in a place not meant for human habitation (e.g. living on the streets) or in an emergency shelter”. HUD defines “disabling condition” as “a diagnosable substance use disorder, serious mental illness, developmental disability, or chronic physical illness or disability, including the co-occurrence of two or more of these conditions.”

Continuum of Care (CoC) The Continuum of Care refers to a community plan to organize and deliver housing and services to meet the specific needs of people who are homeless as they move to stable housing and maximum self-sufficiency. It includes action steps to end homelessness and prevent a return to homelessness. It also refers to the system of services that help people move from homelessness to housing.

San

Bern

ardi

no C

ount

y Po

int i

n Ti

me

2009

page

65

Deployment CenterThe designated sites throughout San Bernardino County where volunteers reported to participate in the Unsheltered Street Count. Volunteers worked in assigned teams to canvass a specific area to tally observed homeless persons.

Domestic Violence ProgramAny residential program whose primary mission is to serve victims of domestic violence. Services are facilitated through education, counseling, legal advocacy, and support groups to improve and enhance the level of confidence and self-esteem of the participants empowering them to take control of their own lives.

Emergency Shelter ProgramA temporary shelter with services that are designed to facilitate homeless individuals and families transition from sleeping in places not meant for human habitation to appropriate housing. Emergency Shelter is provided free of charge for a maximum of ninety-days. On a case-by-case basis, clients may remain for longer than ninety days if they require a longer period to accomplish a specific goal.

EncampmentTemporary, makeshift housing created by homeless persons, such as tents and boxes on sidewalks, under bridges, or other outdoor locations not typically meant for human habitation.

EnumeratorWorked in field teams of three or more and tallied the number of observed homeless persons in a defined geographic area (census block group/s) during the unsheltered street count on the night of January 22, 2009.

ExtrapolationA technique for estimating the total number of homeless persons in a population or category based upon a) the number of unsheltered and sheltered homeless persons observed during the Homeless Count (to obtain a complete count) and b) proportions derived from survey data to estimate subpopulation numbers, i.e. chronic homelessness, etc.

FamilyFamily is defined as any of the following: Minor parents with child(ren); one or more adults with legal custody of minor child(ren); a couple in which one person is pregnant; grandparents or others who are legal guardians with child(ren) present; multi-generational families with grandparents, parents (adult child) and minor child(ren).

Foster YouthA term describing young people who are wards of the foster care system up to age 18. Those youth discharged from the child welfare system when reaching the majority age of 18, judicially relieving the care, custody, and control of the young adult are frequently referred to as an “emancipated foster youth”.

GeocodingThe process of assigning geographic identifiers (e.g., codes or geographic coordinates expressed as latitude-

longitude) to map features and other data records, such as street addresses.

Homeless VeteransAn eligible Veteran is defined as one who: (1) served on active duty in the US armed forces for more than 160 days and was discharged with other than a dishonorable discharge; (2) was discharged or released from active duty because of a service connected disability; or (3) served on active duty during a period of war, or in a campaign or expedition to which a campaign badge is authorized.

San

Bern

ardi

no C

ount

y Po

int i

n Ti

me

2009

page

66

Homeless (HUD McKinney-Vento Act definition)An individual who lacks a fixed, regular, and adequate nighttime residence. An individual who has a primary nighttime residence that is: a supervised publicly or privately operated shelter designed to provide temporary living accommodations (including welfare hotels, congregate shelters, and transitional housing for the mentally ill), or an institution that provides a temporary residence for individuals intended to be institutionalized, or a public or private place not designed for, or ordinarily used as, a regular sleeping accommodation for human beings.

Homeless CountFor the 2009 San Bernardino County Homeless Count, this process (also known as an enumeration, census, or “complete count”) was performed by using U.S. Department of Housing and Urban Development-recommended practices for counting homeless people and estimating the number of homeless people at any given time. This comprehensive study includes a (1) Street Count, (2) Shelter Count, and (3) Homeless Demographic Survey. The census was conducted on the night of January 22, 2009.

HUD - United States Department of Housing and Urban DevelopmentA federal department created in 1965 to increase homeownership, support community development and housing free from discrimination. Since 1987 HUD has been responsible for funding homeless programs, which today comprise the Continuum of Care.

IndividualA person over age 18, not accompanied by minor child(ren).

Individuals with DisabilitiesA person has a disability if she or he has at least one of the following: (1) has a physical or mental impairment that substantially limits a major life activity, (2) has a record of such impairment, or (3) is regarded as having such an impairment.

Institutions For the purposes of the 2009 San Bernardino County Homeless Count, jails (including juvenile detention centers), emergency rooms, hospitals, residential facilities, and drug and alcohol rehabilitation facilities that shelter persons who will become homeless upon discharge.

McKinney-Vento ActThe primary federal legislation used to (1) establish an Interagency Council on the Homeless; (2) use public resources and programs in a more coordinated manner to meet the critically urgent needs of the homeless of the Nation; and (3) provide funds for programs to assist the homeless, with special emphasis on elderly persons, handicapped persons, families with children, Native Americans, and veterans.

Permanent Housing or Permanent Supportive HousingLong-term housing that is commonly community-based with supportive services designed to enable homeless persons with disabilities to live as independently as possible in a permanent setting. Permanent housing can be provided in one structure or several structures at one site or in multiple structures at scattered sites.

Point-In-Time Count A one-day, statistically reliable, unduplicated count of sheltered and unsheltered homeless individuals and families in San Bernardino County. It is required that communities using HUD Continuum of Care Homeless Assistance Funding to serve the homeless conduct a count every two years, beginning in 2005, during the last ten days of January.

Section 8 The common name for the federal housing subsidy program that is administered locally by housing authorities.

San

Bern

ardi

no C

ount

y Po

int i

n Ti

me

2009

page

67

The “Section 8” program is formally known as the Housing Choice Voucher program. The subsidy program is both tenant and project-based. The Section 8 voucher program provides assistance in order for the voucher recipient to pay no more than 30% of their gross monthly income on rent in a unit that complies with the rent guidelines. Housing authorities may spend a portion of their Section 8 certificate program funds to specific projects and thus subsidizing the unit.

Sheltered HomelessSomeone who is homeless and is being temporarily housed in a homeless shelter (emergency and transitional) or motels/hotels that accept vouchers.

Shelters See Emergency Shelter Program and Transitional Housing Program

Single Room Occupancy (SRO) Private rooms that contain either food preparation or sanitary facilities, or both, that are designed for occupancy by a single individual.

Substance Abuse, Individuals Individuals who have acknowledged addiction problems related to alcohol and drug use and who seek services or housing to support their sobriety.

Survey ProctorPresented survey questions and recorded responses of homeless persons throughout San Bernardino County.

Transitional Housing ProgramA type of housing that facilitates the movement of homeless individuals and families to permanent housing, typically within twenty-four (24) months. Services provided in transitional housing must promote residential stability and increased skill level or income in order to prepare homeless persons to live more independently.

Unaccompanied YouthIndividuals who are under 18 years of age who have experienced homelessness on their own, without parent or guardian. A minor with children of his or her own is considered a member of a family unit and not an unaccompanied youth.

Unsheltered HomelessSomeone who is either living on the streets, or in a vehicle, encampment, abandoned building, garage, or any other place not normally used or meant for human habitation.

Young Adults (18-24)Programs serving unaccompanied persons age 18 to 24. This group faces particular challenges because they have reached adulthood in legal terms but still require supportive services and housing. Included in this group are youth, who, because they have reached 18 years, no longer can be served by certain child dependency and delinquency systems but nonetheless require assistance.

San

Bern

ardi

no C

ount

y Po

int i

n Ti

me

2009

page

68

Appendix BEnumeration Form – Unsheltered

Point-in-Time Enumeration Form

Tally Sheet for Block Group Number:

Other Location/ Shelter Name:

Team Code & Team Leader’s Name:

# of

MALE

CHILDREN

(0 to 5

Years of age)

# of

MALE

YOUTH

(6 to 17

Years of age)

# of

ADULT

MALES

(18 to 59 Years of age)

# of

ADULT

MALES

(60 Years of age & older)

Total: Total: Total: Total:

# of

FEMALE

CHILDREN

(0 to 5

Years of age)

# of

FEMALE

YOUTH

(6 to 17

Years of age)

# of

ADULT

FEMALES

(18 to 59 Years of age)

# of

ADULT

FEMALES

(60 Years of age & older)

Total: Total: Total: Total:

Notes:

Use Reverse If Needed

Count

Each

Person

Only

Once!

San

Bern

ardi

no C

ount

y Po

int i

n Ti

me

2009

page

69

Appendix CEnumeration Form - Sheltered

Point-in- T ime E n ume rat ion F o rm for Shelte rs

L ocation/ Shelter N ame: ____________________________________________________

Res pons ible A dm inis trator ’s Name: __________________________________________

Res pons ible A dm inis trator ’s P hone N um ber : ___________ _____________________

# of M A L E

C H I L D R E N(0 to 5

Years of age)

# of M A L E

Y O U T H(6 to 17

Years of age)

# of A D U L T

M E N(18 to 59

Years of age)

# of A D U L T

M E N(60 Years of age & older)

Total : Total : Total : Total :

# of F E M A L E

C H I L D R E N(0 to 5

Years of age)

# of F E M A L E

Y O U T H(6 to 17

Years of age)

# of A D U L T

F E M A L E S(18 to 59

Years of age)

# of A D U L T

F E M A L E S(60 Years of age & older)

Total : Total : Total : Total :

# of M A L E

C H I L D R E N(0 to 5

Years of age)

# of M A L E

Y O U T H(6 to 17

Years of age)

# of A D U L T

M E N(18 to 59

Years of age)

# of A D U L T

M E N(60 Years of age & older)

Total : Total : Total : Total:

Notes :

Use Reverse I f Needed

Count each persononly once!

Please E nter the number of vouchers issued to be used on January 22, 2009:

Please fax the completed form to (909) 200-3805 no later than 10:00 p.m. the night of the count, January 22, 2009.

San

Bern

ardi

no C

ount

y Po

int i

n Ti

me

2009

page

70

Appendix DUnsheltered Homeless Survey Instrument

Survey for UNSHELTERED Homeless Persons

1. Have you been staying in a room, an apartment, or a house?

Yes

No [SKIP TO 3]

2. That’s all the questions I have right now. Thank you very much for your participation. Take

care. [TERMINATE INTERVIEW]

3. Have you been staying in a . . .

Garage

Abandoned building

Outdoor encampment/ Streets (Not in car)

Car, Van, Truck or RV

Emergency Shelter or Transitional Living*

Motel

Some Other Place

(Please Describe)

DON’T KNOW [BACK TO 2—TERMINATE INTERVIEW]

REFUSED [BACK TO 2—TERMINATE INTERVIEW]

Room —Not Including a Motel

An Apartment

A House

1a. Is that a permanent or a temporary situation?

Permanent [CONTINUE TO Q2]

Temporary [SKIP TO Q3]

OR

1

San

Bern

ardi

no C

ount

y Po

int i

n Ti

me

2009

page

71

4. In the last 30 days, where have you usually slept? Have you usually slept in a . . . [CHECK ONE]

¨ Outdoor Encampment/ Streets (Not in car) ¨ Abandoned building ¨ Park ¨ Church¨ Car Van Truck or RV ¨ Friend’s home ¨ Family member’s home.¨ Hotel/ Motel ¨ Emergency shelter ¨ Transitional Housing ¨ Garage ¨ OTHER (Please Describe)

5. Where were you living when you most recently became homeless?¨ San Bernardino County ¨ Riverside County ¨ Out of state¨ Los Angeles County ¨ Other county in California

[IF CAME TO SAN BERNARDINO AS A HOMELESS PERSON, ANSWER TO Q6=0]6. How long had you lived in San Bernardino County before becoming homeless most

recently? (Please use the most appropriate interval) ______ Days ______ Weeks ______ Months ______Years

7. How long has it been since you last lived in a permanent housing situation such as a house or apartment that you owned or rented? (Please use the most appropriate interval) ______ Days ______ Weeks ______ Months ______Years

8. Isthisthefirsttimeyouhavebeenhomeless? ¨ No ¨ Yes

[IF FIRST TIME HOMELESS, ANSWER TO Q9 = ANSWER TO Q7]9. How long have you been homeless this time? (Please use the most appropriate interval)

______ Days ______ Weeks ______ Months ______Years

[IF ANSWER TO Q7 IS 12 MONTHS (I YEAR) OR MORE, ANSWER TO Q10 IS “1 TIME.”]10. In the last 12 months, how many different times have you been homeless including this

present time? ______ Times

11. In the last 3 years, how many different times have you been homeless including this present time? ______ Times

12. Adding together all the different times, how much time have you been homeless in your entire life? (Please use the most appropriate interval) ______ Days ______ Weeks ______ Months ______Years

2Survey for UNSHELTERED Homeless Persons

San

Bern

ardi

no C

ount

y Po

int i

n Ti

me

2009

page

72

3Survey for UNSHELTERED Homeless Persons

13. Have you ever been the victim of a crime while you were homeless? ¨ NO [SKIP TO Q15]

¨ YES [CONTINUE]

14. Were you the victim of one or more: ¨ Violent crimes ¨ Non-violent crimes ¨ Both violent and non-violent crimes?

15. Have you ever served on active duty in the Armed Forces of the United States? ¨NO [SKIP TO Q17]

¨ YES ¨ Regular Military ¨ National Guard ¨ Reserves 16. Did you serve in . . .

¨ World War II ¨ Korea ¨ Vietnam ¨ First Gulf War ¨ Iraq, Afghanistan, or Other Place in Global War on Terror ¨ Peace Time ¨ OTHER

17. Do you live alone, without family, partner, or friends? ¨ Yes, Lives alone [SKIP TO Q18]

¨ No, Lives with: # Male(s) #Female(s)

¨ Spouse or partner ______ ______¨ Other family member(s) ______ ______¨ Friend(s) ______ ______¨ Child/ Children ______ ______ ¨ Parent or legal guardian ______ ______ ¨ Other ______ ______Please Describe:

18. Are you currently using any of the following kinds of services or assistance? (Check all that apply)

¨ Emergency Shelter ¨ Transitional Housing ¨ Free Meals ¨ Food Bank or Pantry ¨ Bus Passes ¨ Job Training ̈ Alcohol or Drug Counseling ¨ Shelter Day Services ¨ Legal Assistance ¨ Health Care ¨ Mental Health Care¨ Help Filling Out Application(s)¨ OTHER SERVICE(S) ¨ Not Using Any Services

San

Bern

ardi

no C

ount

y Po

int i

n Ti

me

2009

page

73

4Survey for UNSHELTERED Homeless Persons

19. Do you currently get: (Check all that apply) ¨ General Relief (GR) ¨ Food Stamps ¨ Medi-Cal/ Medi-Care

¨ MIA (Medical Indigent Services/ Assistance) ¨Veteran’sBenefits ¨ WIC ¨ Social Security

¨ SSI (Supplemental Security Income) or SSD ¨ Cash Aid/ CalWORKS ¨ Veteran’s Disability Compensation

¨OtherGovernmentalAssistance(StateDisabilityBenefits,Worker’sCompensation, Unemployment) ¨ Not Receiving Any Form of Governmental Assistance

20. How else do you get money? ¨ From Family or Friends ¨ Retirement Pension ¨ Child Support ¨ Panhandling ¨ Selling Blood/ Plasma ¨ Robbery or Theft ¨ Selling Recyclables ¨ Selling Other Found Items

¨ Gambling ¨ Sex Work/ Prostitution ¨ Street Performance (Music, Singing, etc.)

¨ Selling Drugs (Prescription or Illegal) ¨ OTHER

21. Are you currently employed? ¨ No ¨ Yes ¨ Permanent Job ¨ Temporary Job

22. Are you currently a student? ¨ No ¨ Yes What school do you attend?

23. Are you aware of any serious physical illness or disability that you have? ¨ No ¨ Yes

24. Have you ever been told by a health care provider that you have a mental illness? ¨ No ¨ Yes

25. Do you believe that you have a mental illness? ¨ No ¨ Yes

San

Bern

ardi

no C

ount

y Po

int i

n Ti

me

2009

page

74

5Survey for UNSHELTERED Homeless Persons

26. Are you currently experiencing any of the following: ¨ Physical Disability ¨ Domestic/ partner violence or abuse ¨ Depression ¨ Alcohol Abuse ¨ Drug Abuse ¨ Post Traumatic Stress Disorder (PTSD) ¨ AIDS/HIV-Related Illness ¨ Chronic (Long Lasting) Health Problem ¨DevelopmentalDisability(Achronic—longlasting—conditionthatsignificantlylimitsaperson’s ability to speak, hear, see, walk, learn, or to perform basic tasks) ¨ DON’T KNOW ¨ REFUSED

27. Does this condition keep you from working? ¨ No ¨ Yes

28. Were you ever in foster care (with a family, in a group home or other placement)? ¨ No ¨ Yes

29. What do you think is the one main reason that you are now homeless? (Check only one) ¨ Lost Job ¨ Parent household fell apart

¨ Foreclosure ¨ Landlord raised rent ¨ Landlord stopped renting because of foreclosure

¨ Gambling problem ¨ Alcohol or drug problem ¨ Mental health issue ¨ Illness or medical problem ¨ Domestic violence ¨ Argument—asked to leave last housing ¨ Just out of jail ¨ Just out of hospital/ treatment program ¨ Divorce or separation ¨ Just out of alcohol/ drug treatment program ¨ Too old for foster care ¨Lostgovernmentbenefits/assistance ¨ For being lesbian, bisexual, gay or transgender ¨Naturaldisaster(fireorflood) ¨ OTHER (Please Describe)

30. Do you have a felony conviction? ¨ No ¨ Yes

San

Bern

ardi

no C

ount

y Po

int i

n Ti

me

2009

page

75

6Survey for UNSHELTERED Homeless Persons

31. How do you identify yourself? [CHECK ONE] ¨ White ¨ Black ¨ American Indian or Alaskan Native

¨ Latino/ Hispanic ¨ Two or more races ¨Asian/PacificIslander[INCLUDESPERSONSFROMINDIA]

32. ¨ Male ¨ Female ¨ Transgender

33. What is your date of birth? ___ ___ / ___ ___ / ___ ___ Your Initials? ___ ___ ___

THANK YOU

San

Bern

ardi

no C

ount

y Po

int i

n Ti

me

2009

page

76

Appendix ESheltered Homeless Survey Instrument

January 2009 Point-in-Time Survey

for SHELTERED Homeless Persons

1. In the last 30 days, where have you usually slept? Have you usually slept in a . . .

[Please Check One] Outdoor Encampment/ Streets (Not in car)

Abandoned building Park Church

Car Van Truck or RV Friend’s home Family member’s home

Hotel/ Motel Emergency shelter

Transitional Housing Garage

OTHER (Please Describe)

2. Where were you living when you most recently became homeless?

San Bernardino County Riverside County Out of state

Los Angeles County Other county in California

[If You Came To San Bernardino as a Homeless Person, The Answer To Question 3 Is “0”]

3. How long had you lived in San Bernardino County before becoming homeless most recently?

(Please use the most appropriate interval—Days Weeks, Months, and/ or Years) ______ Days ______ Weeks ______ Months ______Years

4. How long has it been since you last lived in a permanent housing situation such as a house

or apartment that you owned or rented?

(Please use the most appropriate interval—Days, Weeks, Months, and/ or Years)

______ Days ______ Weeks ______ Months ______Years

5. Is this the first time you have been homeless? No Yes

[If this is the FIRST TIME you have been homeless, the answer to Question 6 will be the same as your answer to Question 4]6. How long have you been homeless this time?

(Please use the most appropriate interval—Days, Weeks, Months, and/ or Years) ______ Days ______ Weeks ______ Months ______Years

1

San

Bern

ardi

no C

ount

y Po

int i

n Ti

me

2009

page

77

2Survey for SHELTERED Homeless Persons

[f your answer to Question 4 is 12 months (I year) or more, the answer to Question 7 is“1 time.”]7. In the last 12 months, how many different times have you been homeless including this

present time? ______ Times

8. In the last 3 years, how many different times have you been homeless including this pres-ent time? ______ Times

9. Adding together all the different times, how much time have you been homeless in your entire life? (Please use the most appropriate interval—Days, Weeks, Months, and/ or Years) ______ Days ______ Weeks ______ Months ______Years

10. Have you ever been the victim of a crime while you were homeless? ¨ NO [SKIP TO Question 12] ¨ YES [PLEASE CONTINUE TO Question 11]

11. Were you the victim of one or more: ¨ Violent crimes ¨ Non-violent crimes,

¨ Both violent and non-violent crimes?

12. Have you ever served on active duty in the Armed Forces of the United States? ¨NO ¨YES ¨ Regular Military ¨ National Guard ¨ Reserves

[PLEASE SKIP TO Question 14]

13. Did you serve in . . . ¨ World War II ¨ Korea ¨ Vietnam ¨ First Gulf War ¨ Iraq, Afghanistan, or Other Place in Global War on Terror ¨ Peace Time ¨ OTHER

San

Bern

ardi

no C

ount

y Po

int i

n Ti

me

2009

page

78

14. Do you live alone, without family, partner, or friends? ¨ No, I Live with: ¨ Yes, I Live alone [PLEASE SKIP TO Question 16]

# Male(s) #Female(s)¨ Spouse or partner ______ ______¨ Other family member(s) ______ ______¨ Friend(s) ______ ______¨ Child/ Children ______ ______ ¨ Parent or legal guardian ______ ______ ¨ Other ______ ______Please Describe:

15. Are any of these persons completing a survey like this one?¨ No. ¨ Yes. If so, what are that person’s initials? ____ ____ ____

16. Are you currently using any of the following kinds of services or assistance? (Please check all that apply) ¨ Emergency Shelter ¨ Transitional Housing ¨ Free Meals ¨ Food Bank or Pantry ¨ Bus Passes ¨ Job Training¨ Alcohol or Drug Counseling ¨ Shelter Day Services¨ Legal Assistance ¨ Health Care ¨ Mental Health Care¨ Help Filling Out Application(s)¨ OTHER SERVICE(S)

¨ Not Using Any Services

17. Do you currently get: (Please check all that apply) ¨ General Relief (GR) ¨ Food Stamps ¨ Medi-Cal/ Medi-Care ¨ MIA (Medical Indigent Services/ Assistance) ¨Veteran’sBenefits ¨ WIC ¨ Social Security ¨ SSI (Supplemental Security Income) or SSD ¨ Cash Aid/ CalWORKS ¨ Veteran’s Disability Compensation ¨OtherGovernmentalAssistance(StateDisabilityBenefits,Worker’sCompensation,Unemployment) ¨ Not Receiving Any Form of Governmental Assistance

18. What is your total monthly income from all government sources (County, State, Federal) $____________ per month

3Survey for SHELTERED Homeless Persons

San

Bern

ardi

no C

ount

y Po

int i

n Ti

me

2009

page

79

19. How else do you get money? ¨ From Family or Friends ¨ Retirement Pension ¨ Child Support ¨ Panhandling ¨ Selling Blood/ Plasma ¨ Robbery or Theft ¨ Selling Recyclables ¨ Selling Other Found Items ¨ Gambling ¨ Sex Work/ Prostitution ¨ Street Performance

(Music, Singing, etc.) ¨ Selling Drugs (Prescription or Illegal) ¨ OTHER

20. What is your total monthly income from all of the sources listed in Question 19 above? $____________ per month

21. Are you currently employed? ¨ No ¨ Yes ¨ Permanent Job ¨ Temporary Job

21a. What is keeping you from getting or holding on to a job? (Please check all that apply) ¨ Need more education ¨ Do not have phone ¨ Criminal Record ¨ Need more training ¨ Health problems ¨ No child care ¨ Need clothes ¨ Disability ¨ No permanent address ¨ Do not have shower/ bath ¨ Alcohol/ drug problem ¨ No transportation ¨ Do not have tools ¨ Do not have Social Security card/ work permit ¨ Do not have photo ID ¨ There are no jobs ¨ Do not want to work

¨ I am retired ¨ OTHER

22. Are you currently a student? ¨ No ¨ Yes What school do you attend?

22a. What is keeping you from getting permanent housing? (Please check all that apply) ¨ Can’t afford rent ¨ No transportation ¨ Criminal Record ¨ No job/ no income ¨ Bad Credit ¨ Do not want to ¨ No housing available ¨ Eviction Record ¨Can’tputtogethermovingcosts(securitydeposit,first/lastmonth’srent) ¨ OTHER

4Survey for SHELTERED Homeless Persons

San

Bern

ardi

no C

ount

y Po

int i

n Ti

me

2009

page

80

5Survey for SHELTERED Homeless Persons

23. Are you aware of any serious physical illness or disability that you have? ¨ No ¨ Yes

24. Have you ever been told by a health care provider that you have a mental illness? ¨ No ¨ Yes

25. Do you believe that you have a mental illness? ¨ No ¨ Yes

26. Are you currently experiencing any of the following: ¨ Physical Disability ¨ Domestic/ partner violence or abuse

¨ Depression ¨ Alcohol Abuse ¨ Drug Abuse ¨ Post Traumatic Stress Disorder (PTSD) ¨ AIDS/HIV-Related Illness ¨ Chronic (Long Lasting) Health Problem ¨DevelopmentalDisability(Achronic—longlasting—conditionthatsignificantlylimitsa

person’s ability to speak, hear, see, walk, learn, or to perform basic tasks) ¨ DON’T KNOW ¨ REFUSED

27. Does this condition keep you from working? ¨ No ¨ Yes

28. Were you ever in foster care (with a family, in a group home or other placement)? ¨ No ¨ Yes

29. What do you think is the one main reason that you are now homeless? (Please check only one)

¨ Lost Job ¨ Parent household fell apart ¨ Foreclosure ¨ Landlord raised rent ¨ Landlord stopped renting because of foreclosure

¨ Gambling problem ¨ Alcohol or drug problem ¨ Mental health issue ¨ Illness or medical problem ¨ Domestic violence ¨ Argument—asked to leave last housing ¨ Just out of jail ¨ Just out of hospital/ treatment program ¨ Divorce or separation ¨ Just out of alcohol/ drug treatment program ¨ Too old for foster care ¨Lostgovernmentbenefits/assistance ¨ For being lesbian, bisexual, gay or transgender ¨Naturaldisaster(fireorflood) ¨ OTHER (Please Describe)

San

Bern

ardi

no C

ount

y Po

int i

n Ti

me

2009

page

81

6Survey for SHELTERED Homeless Persons

29a. Have you needed any of the following services, and been UNABLE to obtain them? (Please check all that apply) ¨ Temporary Shelter ¨ Clothing ¨ Food

¨ Financial Assistance ¨ Medical ¨ Counseling ¨ Educational ¨ Job Training ¨ Permanent Housing ¨ Transportation ¨ OTHER (Please Describe)

29b. What is your age? _______ years old.

29c. What is the highest level of education you have completed? ¨ Less than 6th Grade ¨AAdegree/technicalcertificate ¨ Less than High School Diploma ¨ Bachelor’s degree ¨ High School Diploma or GED ¨ Graduate or Professional degree ¨ Some College, no degree

30. Do you have a felony conviction? ¨ No ¨ Yes

31. How do you identify yourself? [PLEASE CHECK ONE] ¨ White ¨ Black ¨ American Indian or Alaskan Native

¨ Latino/ Hispanic ¨ Two or more races ¨Asian/PacificIslander[INCLUDESPERSONSFROMINDIA]

32. ¨ Male ¨ Female ¨ Transgender

33. What is your date of birth? ___ ___ / ___ ___ / ___ ___ Your Initials? ___ ___ ___

THANK YOU

San

Bern

ardi

no C

ount

y Po

int i

n Ti

me

2009

page

82

Appendix FDifferences between Unsheltered Homeless Persons Surveyed on the “Night of Count” and Over Subsequent Weeks

Fifty-seven (24.2%) of the 236 unsheltered homeless persons completing surveys were interviewed during the Point-In-Time Count, and 179 (75.8%) were surveyed after the count. It is important to know whether the homeless persons surveyed afterward (three quarters of the unsheltered sample) are equivalent to those interviewed during the count (one quarter of the unsheltered sample) because the survey results will be generalized to the entire estimated population of unsheltered homeless. To determine whether the unsheltered homeless persons who were interviewed during the day in the weeks following the Point-In-Time count have the same characteristics as those interviewed at night when they were counted, a variety of comparisons were performed.

Demographics

Night-of-count and “after-count” unsheltered homeless interviewees are not significantly different by gender or by race/ ethnicity. Persons interviewed on the night of the count are just less than four and one half years older on average (n= 50, M= 45.5 years of age) than those interviewed over subsequent weeks (n= 177, M= 41.1 years); F (1, 225) = 4.003, p < .05.

San

Bern

ardi

no C

ount

y Po

int i

n Ti

me

2009

page

83

Where Have You Usually Slept?

From a list of eleven places, unsheltered survey respondents reported sleeping in one to eight “usual” places in the last 30 days. The reported frequency is not significantly different between groups for “Outdoor encampment/ Streets (Not in car),” “Car, van, truck or RV,” “Transitional housing,” “Park,” “Friend’s home,” “Emergency shelter,” “Garage,” and “Family member’s home.” A significantly higher proportion of unsheltered survey respondents interviewed on the night of the count (n= 13 of 52, 22.7%) report usually sleeping in an abandoned building compared to those interviewed in the weeks after the count (n= 20 of 177, 11.3%); χ2 (1, 229)= 6.117, p < .02. Likewise, more persons interviewed during the street count (n= 5 of 52, 9.6%) reported usually sleeping a “Hotel/ motel” than those interviewed later (n= 4 of 177, 2.3%); χ2 (1, 229) = 5.759, p < .035. Finally, three of 52 (5.8%) unsheltered persons interviewed the night of the count usually sleep in a church, compared to none of the 177 interviewed later; χ2 (1, 229) = 10.347, p < .015.

Where Were you Living When you Most Recently Became Homeless?

Proportionally fewer (n= 37, 66.1%) unsheltered homeless persons interviewed the night of the count became homeless while living in San Bernardino County than among those interviewed after the count (n= 147, 84.0%); χ2 (1, 231) = 22.771, p < .001.

How Long in San Bernardino Since Last Lived in Permanent Housing?

The 184 unsheltered persons who were living in San Bernardino when they became homeless were asked how long they had lived in the county before becoming homeless most recently. Twenty-nine night-of-count interviewees provided answers ranging from seven days to 46 years, with an average of 12.5 years. Answers from unsheltered persons interviewed in the weeks after the count ranged from one month to 61 years, with an average of 20 years, one month and six days. These averages are significantly different, F (1, 153) = 5.373, p < .025.

The length of time since night-of-count unsheltered survey respondents had “lived in a permanent housing situation such as a house or apartment that you owned or rented” ranged from seven days to 16 years, with an average of three years, eight months and 17 days. Answers from those interviewed after the count ranged from one month to 42 years with an average of four years, four months and 28 days. This difference is not statistically significant.

San

Bern

ardi

no C

ount

y Po

int i

n Ti

me

2009

page

84

Length of Time/Number of Occasions Homeless

A significantly lower proportion (n=23, 41.8%) of night-of-count interviewees report that this is the first time they have been homeless than unsheltered persons surveyed after the count (n= 104, 59.4%); χ2 (1, 230) = 5.248, p < .02. Length of the current episode of homelessness ranges from three days to 42 years, and there is no significant difference between group averages. The number of different times unsheltered respondents report they have been homeless in the last 12 months (including the present time) ranges from one to 12, however 200 (90.9%) of the 220 who answered this question say they’ve been homeless one time in the past 12 months. Although the mean number of episodes of homelessness in the last year is slightly higher among night-of-count respondents, the difference with those interviewed after the count is not statistically significant.

The number of occasions respondents have been homeless in the last three years ranges from one to 18. Two unsheltered survey respondents reported more episodes of homelessness in the last 12 months than in the last three years. In both cases, the number of episodes within three years was set to the higher “12 month” report. One hundred sixty-two (74.3%) of the unsheltered survey respondents report one episode of homelessness in the past three years, and again, although the mean number of episodes is higher among night-of-count respondents (M= 2.33), the difference with those interviewed after the count (M= 1.70 episodes) is not statistically significant.

Lifetime total length of homelessness ranges from seven days to 42 years, with an overall unsheltered respondent mean of five years and 16 days. The small difference between night-of-count and “after count” survey respondents is not statistically significant.

Victimization While Homeless

More than one third of unsheltered survey respondents (n= 88, 38.1%) report that they have been the victim of a crime while homeless. Eighty-five of these 88 respondents indicated whether they had been victims of violent crime (n= 28, 32.9%), non-violent crime (n= 26, 30.6%) or both violent and non-violent crimes (n= 31, 36.5%) while homeless. There are no significant differences by point of interview in the rate or type of victimization while homeless.

Military Service

Forty (17.5%) of the 229 unsheltered survey respondents who answered the question report military service. Of these, 35 served in active duty, one as a reserve and four did not indicate the type of service. A significantly higher proportion (n= 17, 30.9%) of homeless persons interviewed on the night of the count report military service than among unsheltered homeless persons interviewed after the count (n= 23, 13.2%); χ2 (1, 229) = 9.072, p < .005.

San

Bern

ardi

no C

ount

y Po

int i

n Ti

me

2009

page

85

Thirty-three unsheltered homeless survey respondents identified the era of their military service: 14 (42.4%) served in Vietnam, 11 (33.3%) during peace time, four (12.1%) during the first Gulf War, three (9.1%) during the War on Terror in Iraq or Afghanistan, and one each in Grenada and Korea. Primarily because of the small number of cases, neither relationship is statistically significant, but more than half (eight of 15, 53.3%) of those with military service interviewed on the night of the count are Vietnam-era veterans, compared to (n= 6, 26.1%) of those interviewed later, and just two of 15 (13.3%) night-of-count interviewees served during peace time compared to nine of 23 (39.1%) interviewed later.

Lives Alone or With Others

Among the unsheltered homeless persons interviewed on the night of the count, 31 of 54 (57.4%) live alone, compared to 92 of 176 (52.3%) persons interviewed in the weeks after the Point-In-Time count. This difference is not statistically significant, and neither is the difference in the average number of persons with whom the respondent lives (1.0 for night-of-count respondents and .91 for those interviewed later).

Current Use of Services or Assistance

More than half (n= 31, 54.4%) of the unsheltered homeless persons interviewed during the enumeration process on the night of the count report no service use at all, compared to about one fifth (n= 39, 21.8%) of those interviewed over the following weeks; χ2 (1, 236) = 22.020, p < .001. Among the unsheltered respondents who do use services/ assistance, the mean number of services used by night-of-count and later interviewees is nearly identical (M= 2.15 and M= 2.20 services, respectively).

Sources of Support and Income

Just more than half (n= 28, 52.8%) of those interviewed on the night of the count and of those interviewed over subsequent weeks (n= 91, 52.6%) report receiving some type of government assistance. Among those who do receive some sort of government assistance, the mean number of types of aid (about 1.4) is the same in both groups. Equivalent proportions of night-of-count (n= 18, 31.6%) and later interviewees (n= 58, 32.4%) report no non-governmental source of income (e.g. from family or friends, selling recyclables).

A significantly larger proportion (n= 18, 41.9%) of night-of-count unsheltered homeless persons report panhandling as a source of income than among those interviewed later (n= 32, 23.9%); χ2 (1, 177) = 5.192, p < .025. Likewise, the proportion interviewed on the night of the count who sell “other found items” (n= 6, 14.0%) is significantly greater than among those interviewed over the following weeks (n= 3, 2.3%); χ2 (1, 176) = 9.164, p < .01.

Just one of 53 (1.9%) unsheltered homeless persons interviewed on the night of the count is employed (a temporary job) compared to 22 of 177 (12.4%) of those interviewed later (nine in permanent, 13 in temporary jobs), a statistically significant difference; χ2 (1, 230) = 5.037, p < .02.

San

Bern

ardi

no C

ount

y Po

int i

n Ti

me

2009

page

86

Student Status

Very similar to the pattern of employment, just one of 53 (1.9%) unsheltered homeless persons interviewed on the night of the count is a student compared to 23 of 178 (12.9%) of those interviewed later, a statistically significant difference; χ2 (1, 231) = 5.341, p < .015.

Physical and Mental Illness

In response to the question, “Are you aware of any serious physical illness or disability that you have?” 32 of 53 (60.4%) of the night-of-count interviewees replied in the affirmative, compared to 71 of 177 (40.1%) unsheltered homeless persons interviewed later; χ2 (1, 230) = 6.773, p < .01. Very similar proportions, however, replied affirmatively to the question, “Have you ever been told by a health care provider that you have a mental illness?” (n= 15, 28.3% among those interviewed on the night of the count and n= 45, 25.3% among those interviewed later). Also with roughly equivalent proportions, 14 (26.4%) night-of-count interviewees and 52 (29.5%) of the unsheltered homeless persons interviewed later replied “yes” to the question, “Do you believe that you have a mental illness?”

The group interviewed the same night they were counted and the homeless persons interviewed over the following weeks do not differ significantly on the proportions reporting physical disability (overall, 36.8%), domestic/ partner violence or abuse (overall, 8.4%), post traumatic stress disorder (overall, 13.5%), AIDS/ HIV-related illness (overall, 1.9%), a chronic health problem (overall, 21.3%), or a developmental disability (overall, 13.5%). The groups differ significantly, though, on depression and drug and alcohol use. Unsheltered homeless persons interviewed in the weeks after the Point-In-Time count report a significantly higher rate of depression (n= 72, 62.6%) than those interviewed during the count (n= 16, 40%); χ2 (1, 155) = 6.181, p < .015. Conversely, the rate of self-reported alcohol abuse is higher among night-of-count interviewees (n= 19, 47.5%) than among the homeless persons interviewed during the day over the following weeks (n= 35, 30.4%); χ2 (1, 155) = 3.807, p < .05. The unsheltered homeless persons interviewed during the night of the count also report a significantly higher rate of drug abuse (n= 12, 30.0%) than those interviewed later (n= 14, 12.2%); χ2 (1, 155) = 6.755, p < .015.

Foster Care

The proportion of night-of-count interviewees (n= 11, 22.4%) ever in foster care is somewhat higher than among unsheltered homeless persons interviewed later (n= 31, 17.8%), but this difference isn’t statistically significant.

Felony Conviction

Similarly, a somewhat but not significantly higher proportion (n= 16, 32%) of the night-of-count unsheltered interviewees report that they have a felony conviction, compared to

San

Bern

ardi

no C

ount

y Po

int i

n Ti

me

2009

page

87

those interviewed over subsequent weeks (n= 49, 27.5%).

Main Reason Now Homeless

From a list of nineteen potential reasons, unsheltered survey respondents were asked, “What do you think is the one main reason that you are now homeless?” One hundred ninety-two of the 236 unsheltered homeless persons who were interviewed endorsed one or more of these reasons. Although five of 41 (11.1%) unsheltered homeless persons interviewed on the night of the count compared to six of 147 (4.1%) of those interviewed later indicated that a “mental health issue” was the main reason they were now homeless, this difference is not statistically significant. The one significant between-group difference in this series is observed on, “lost government benefits/ assistance.” Here, the rate is significantly higher among night of count interviewees (n= 5, 11.1%) than among unsheltered homeless persons interviewed over the subsequent weeks (n= 4, 2.7%); χ2 (1, 192) = 5.428, p < .035.

Summary of Appendix F

Night of count interviewees are just less than four and one half years older on average (M= 45.5 years of age) than those interviewed over the subsequent weeks (M= 41.1 years).

A higher proportion of unsheltered survey respondents interviewed on the night of the count (22.7%) usually sleeps in an abandoned building compared to those interviewed in the weeks after the count (11.3%). Similarly, more persons interviewed during the street count (9.6%) usually sleep in a “Hotel/ motel” than those interviewed later (2.3%). A higher proportion of unsheltered persons interviewed the night of the count (5.8%) usually sleep in a church, compared to none of those interviewed later.

Proportionally fewer (66.1%) unsheltered homeless persons interviewed the night of the count became homeless while living in San Bernardino County compared to those interviewed after the count (84%). Of those living in San Bernardino when they became homeless, night-of-count interviewees had lived in the county less time (an average of 12.5 years before becoming homeless most recently) compared to unsheltered persons interviewed in the weeks after the count who had lived in San Bernardino County an average of 20 years before becoming homeless.

Fewer night-of-count interviewees (41.8%) than unsheltered persons surveyed after the count (59.4%) report that this is the first time they have been homeless.

A higher proportion interviewed on the night of the count (30.9%) than unsheltered homeless persons interviewed after the count (13.2%) report military service.

More than half (53.3%) of those with military service interviewed on the night of the count are Vietnam-era veterans, compared to (26.1%) of those interviewed later, and just 13.3% night-of-count interviewees served during peace time compared to 39.1% of those

San

Bern

ardi

no C

ount

y Po

int i

n Ti

me

2009

page

88

interviewed later.

Over half (54.4%) of the unsheltered homeless persons interviewed during the enumeration process on the night of the count report no service use at all, compared to about one fifth (21.8%) of those interviewed over the following weeks.

A higher proportion (41.9%) of night-of-count unsheltered homeless persons report panhandling as a source of income than among those interviewed later (23.9%) and a higher proportion (14.0%) interviewed on the night of the count sells “other found items” than among those interviewed over the following weeks (2.3%).

Just one (1.9%) of the unsheltered homeless persons interviewed on the night of the count is employed compared to (12.4%) of those interviewed later. Similarly, (1.9%) of the night-of-count interviewees is a student compared to (12.9%) of those interviewed later.

The respondents reporting awareness of a serious physical illness or disability, were 60.4% of the night-of-the-count respondents, compared to 40.1% of the unsheltered homeless persons interviewed after the night of the count. Unsheltered homeless persons interviewed in the weeks after the Point-In-Time Count report a significantly higher rate of depression (62.6%) than those interviewed during the count (40%). Conversely, the rate of self-reported alcohol abuse is higher among night-of-count interviewees (47.5%) than among the homeless persons interviewed later (30.4%) and night-of-count interviewees also report a higher rate of drug abuse (30.0%) than those interviewed later (12.2%).

A higher proportion of night of count interviewees (11.1%) than unsheltered homeless persons interviewed over the subsequent weeks (2.7%) say the primary reason they are homeless is, “lost government benefits/ assistance.”

San

Bern

ardi

no C

ount

y Po

int i

n Ti

me

2009

page

89

Appendix GCoC Subpopulation Computations, Unsheltered Homeless Population

To facilitate comparisons by gender and age category, Table A presents counts and proportions of the sheltered and unsheltered homeless persons enumerated and surveyed during the 2009 Point-In-Time Count.

Table A. Comparison of Enumerated and Surveyed Homeless Persons in San Bernardino County by Type (Unsheltered/ Sheltered), Gender and Age Group

Male Children (0 to 5)

Male Youth (6 to 17)

Male Adults (18 to 59)

Male Adults (60 & >)

Female Children (0

to 5)Female Youth

(6 to 17)Female Adults

(18 to 59)

Age and/or Gender Unknown Total

Count Count Count Count Count Count CountFemale Adults

Count Count(%) (%) (%) (%) (%) (%) (%) (60 & >) (%) (%)

Count(%)

2 49 595 55 16 13 225 14 0 968(0.21) (5.1) (61.5) (5.7) (1.7) (1.3) (23.2) (1.5) 0 (100)

0 1 132 13 0 1 73 5 11 2360 (0.42) (55.9) (5.5) 0 (0.42) (30.9) (2.1) (4.7) (100)

79 64 222 12 83 60 226 1 21 768(10.3) (8.3) (28.9) (1.6) (10.8) (7.8) (29.4) (0.13) (2.7) (100)

0 0 102 8 0 1 142 3 15 2710 0 (37.6) (3) 0 (0.37) (52.4) (1.1) (5.5) (100)

San

Bern

ardi

no C

ount

y Po

int i

n Ti

me

2009

page

90

Based upon unsheltered survey responses, Table B indicates the counts and ratios needed to calculate estimates of the unsheltered homeless persons in San Bernardino County in CoC household categories.

Table B. Household Composition Counts Based on Unsheltered Survey Data

Part 1: Homeless Population N=236 Unsheltered

Survey Respondents*

1) Number of Unsheltered Survey Respondents with Dependent Children* n= 230

15

Unsheltered Households With Dependents

2a) Total number of children in these households (15 persons in row 1 reported a total of 27 children)

27

2b) Average number of children per household (27/15)

M = 1.80

2c) Total number of adults (including respondent) in these households

29

2d) Average number of adults per household (29/15) M = 1.93

2e) Total number of persons in these households (Sum of 2a and 2c)

56

2g) Average number of persons per household (56/ 15)

M = 3.73

Unsheltered Households Without Dependents

3a) Number of Survey Respondents Without Dependent Children

215

-93.50%3b) Total number of adults (including respondent) in these households

390

3c) Average number of adults per household (390/ 215)

M = 1.81

All Unsheltered Households

4) Total number of persons in these households (Sum of 2e and 3b)

446

4a) Average number of persons in these households 446/230

1.94

San

Bern

ardi

no C

ount

y Po

int i

n Ti

me

2009

page

91

* Based upon the direct report from 230 of 236 unsheltered survey respondents who indicated whether they were living with any male or female children or other adults.

Table A shows that the extrapolated final count indicates 968 unsheltered homeless persons on the street in San Bernardino County on a given January night. Of these, 80 (8.3%) are homeless and unsheltered children and 888 (91.7%) are adults.

To estimate the number of households represented among these unsheltered persons, the number of children (n= 80) is divided by the average number of children per household (1.80 from row 2b in Table 2), producing an estimate of 45 households with dependent children (45 X 1.80 = 81, just one fewer child than actually counted). Multiplying 45 households times the average number of adults in households with dependent children (1.93) results in an estimate of 87 adults. Subtracting these 81 children + 87 adults from the estimate of 968 unsheltered homeless leaves 800 adults. Dividing 800 by the average number of adults in households without dependent children (1.81) produces an estimate of 442 households without dependent children. Adding the 45 households with dependent children to the 442 without children produces a total estimate of 487 households among the 968 unsheltered homeless persons in San Bernardino County.

As indicated by Table B, however, based upon survey data, 6.5% of the unsheltered survey respondents live in households with children, and 93.5% live in households without children. Forty-five is 9.2% of 487. To achieve the 6.5% ratio, the number of households must increase to 647. Multiplying 602 (647- 45) times the average number of adults in households without dependent children (1.81) produces a total of 1,090 unsheltered homeless adults; 290 fewer than the 800 remaining adults in the final count. This calculation quantifies the minimum undercount on January 22nd. Based upon the number of children observed and the household composition data resulting from the surveys of unsheltered homeless persons, it appears very likely that the Point-In-Time process resulted in an undercount of a minimum of 290 adults.

This is suggested as a minimum because the number of unsheltered homeless children enumerated during Point-In-Time counts is typically lower than the actual number. During the evening and nighttime hours of the count, unsheltered homeless children are likely to have been tucked away in locations that their parents may perceive as safer (e.g. in a car, with housed relatives, or in a motel room). If this is the case, the number of unsheltered homeless children might increase, as well as the corresponding number of adults in households needed to match the 6.5/ 93.5 ratio. Any undercount of children can not be reliably estimated, however, and the calculated minimum adult undercount is the only empirically supported figure.

San

Bern

ardi

no C

ount

y Po

int i

n Ti

me

2009

page

92

CoC Subpopulation Category Computations: Unsheltered Homeless

Because the proportion of homeless persons in each of the CoC subpopulation categories depicted in the first column of Table C is likely to vary by age and gender, respondents in the survey sample are classified into four demographic categories: adult males between the ages of 18 and 59, males over the age of 60, adult females between the ages of 18 and 59, and females over the age of 60. Persons who did not specify their age or gender on the unsheltered survey are classified as unknown. These categories parallel those on the enumeration forms. Table C presents counts and proportions from the surveyed unsheltered homeless population only. Extrapolations to the enumerated unsheltered homeless population are calculated in subsequent tables.

As indicated by Table C, the largest unsheltered homeless subpopulation (n= 61) among survey respondents consists of the chronically homeless; however these proportions vary by gender. Proportionally more males 60 years of age and older (61.5%) are chronically homeless, followed by females in this age group (40.0%) and males 18 to 59 years of age (29.3%).

Table C. Subpopulation Counts and Proportions within Age by Gender Categories in the Surveyed Unsheltered Homeless Population

Part 2: Homeless Subpopulations

Adult Male Senior Adult Female

Senior Female Persons of Unknown Age

and Gender

Total Count & % in the

Survey SampleMale

a) Chronically Homeless*

39 8 9 2 3 61

-29.3 -61.5 -12.2 -40 -27.3 -25.8b) Severely Mentally Ill**

35 2 15 0 0 52

-26.3 -15.4 -20.3 0 0 -22c) Chronic Substance Abuse***

31 3 15 1 1 51

-23.3 -23.1 -20.3 -20 -9.1 -21.6e) Veterans 25 8 4 0 3 40

-18.8 -61.5 -5.4 0 -27.3 -16.9f) Persons with HIV/ AIDS****

2 0 1 0 0 3

-2.3 0 -2.1 0 0 -1.9g) Victims of Domestic Violence****

6 0 6 1 0 13

-7 -15.8 -12.5 -20 0 -8.4

San

Bern

ardi

no C

ount

y Po

int i

n Ti

me

2009

page

93

* To be classified as chronically homeless, respondents must report 1) a mental, emotional or developmental disability, or a diagnosable substance abuse problem that is expected to be of long duration and substantially limits their ability to work or to perform one or more activities of daily living, 2) that they are unaccompanied and 3) that they have been continuously homeless for a year or more OR had at least four episodes of homelessness in the past three years.

** To be classified as severely mentally ill, respondents must report that they had been told by a health care provider that they have a mental illness and that the problem keeps them from working/ or (if missing) that they are currently unemployed.

*** To be classified as having a chronic substance abuse problem, respondents must have reported that they had a problem with alcohol OR drugs and reported that the problem keeps them from working/ or (if missing) that they are currently unemployed. Only 155 of 236 unsheltered homeless survey respondents replied yes or no to questions about alcohol abuse and drug abuse.

**** Only 155 of 236 unsheltered homeless survey respondents replied yes or no to a question about AIDS/ HIV-related illness and to a question about domestic violence.

To estimate the number of unsheltered homeless persons in each of the CoC subpopulations, the extrapolated total number of persons in each demographic category is multiplied by the proportions derived from the survey data falling into each of the subpopulation categories. For example, based upon extrapolation methods reported earlier, it can be estimated that there are 595 homeless males between the ages of 18 and 59 on the streets in San Bernardino County on any given January night (See Table A). Based on the survey results (Table C), 29.3% of males in this age category are chronically homeless. The product of 595 X .293 is 174 chronically homeless unsheltered adult males in this age group.

As depicted by Table D, using this procedure successively in each of the four demographic categories produces a total of 241 unsheltered chronically homeless adults in San Bernardino County. Note that children are omitted from Table D, because persons under the age of 18 were not surveyed. This procedure utilized to calculate estimates of the number of chronic homeless persons in the unsheltered population is repeated to estimate the count and percentages of persons in each subpopulation presented in Tables D through I.

San

Bern

ardi

no C

ount

y Po

int i

n Ti

me

2009

page

94

Table D. Estimate of Chronically Homeless Adults in the Unsheltered Population Based on Survey Data

• Number of unsheltered chronically homeless persons = 241• Percentage of unsheltered adults chronically homeless: 241/889=27.1%• Percentage of chronically homeless in total unsheltered population= 241/968=

24.9%

San

Bern

ardi

no C

ount

y Po

int i

n Ti

me

2009

page

95

Table E. Estimate of Severely Mentally Ill in the Unsheltered Homeless Population Based on Survey Data

Demographic Category Extrapolated Final Count of Adult

Unsheltered Homeless

SubpopulationSurvey Count

(%)

Total Count

Adult Male 595 35(26.3)

156

Senior Male 55 2(15.4)

8

Adult Female 225 15(20.3)

46

Senior Female 14 0(0.0)

0

Total 889(100.0)

52(23.1)

210

• Number of unsheltered severely mentally ill homeless persons = 210• Percentage of unsheltered adults severely mentally ill: 210/889=23.6%• Percentage of severely mentally ill in total unsheltered population: 210/968=

21.7%

Table F. Estimate of Chronic Substance Abusers in the Unsheltered Homeless Population Based on Survey Data

Demographic Category

Extrapolated Final Count of

Adult Unsheltered Homeless

SubpopulationSurvey Count

(%)

Total Count

Adult Male 595 31(23.3)

139

Senior Male 55 3(23.1)

13

Adult Female 225 15(20.3)

46

Senior Female 14 1(20.0)

3

Total 889(100.0)

50(22.2)

201

• Number of unsheltered substance abusing homeless persons = 201• Percentage of unsheltered chronic substance abusing adults: 201/889=22.6%• Percentage of chronic substance abusers in total unsheltered population:

201/968= 20.8%

San

Bern

ardi

no C

ount

y Po

int i

n Ti

me

2009

page

96

A veteran is defined as a person who served in the active military, naval or air service, and who was discharged or released under conditions other than dishonorable. Forty unsheltered homeless respondents responded affirmatively to the question, “Have you ever served on active duty in the Armed Services of the United States?” The survey did not inquire about discharge status.

Table G. Estimate of the Number of Veterans in the Unsheltered Homeless Population Based on Survey Data

Demographic Category

Extrapolated Final Count of

Adult Unsheltered Homeless

SubpopulationSurvey Count

(%)

Total Count

Adult Male 595 25(18.8)

112

Senior Male 55 8(61.5)

34

Adult Female 225 4(5.4)

12

Senior Female 14 0(0.0)

0

Total 889(100.0)

39(17.3)

158

• Number of unsheltered homeless persons who are veterans = 255• Percentage of veterans among adult unsheltered homeless: 158/889=17.8%• Percentage of veterans in total unsheltered population: 158/ 968= 16.3%

San

Bern

ardi

no C

ount

y Po

int i

n Ti

me

2009

page

97

Table H. Estimate of the Number of Persons with HIV/ AIDS in the Unsheltered Population Based on Survey Data

Demographic Category

Extrapolated Final Count of

Adult Unsheltered Homeless

SubpopulationSurvey Count(%)*

Total Count

Adult Male 595 2(2.3)

14

Senior Male 55 0(0.0)

0

Adult Female 225 1(2.1)

5

Senior Female 14 0(0.0)

0

Total 889(100.0)

3(1.9)

19

*Proportions computed on the basis of 155 valid responses.

• Number of unsheltered homeless persons infected with HIV/ AIDS = 19• Percentage of adult homeless persons with HIV/ AIDS: 19/889=2.1%• Percentage of all unsheltered homeless persons with HIV/ AIDS: 19/ 968= 1.96%

Table I. Estimate of the Number of Persons Experiencing Domestic Violence in the Unsheltered Population Based on Survey Data

Demographic Category

Extrapolated Final Count of

Adult Unsheltered Homeless

SubpopulationSurvey Count

(%)

Total Count

Adult Male 595 6(7.0)

42

Senior Male 55 0(0.0)

0

Adult Female 225 6(12.5)

28

Senior Female 14 1(20.0)

3

Total 889(100.0)

13(8.4)

73

• Number of unsheltered homeless persons who have experienced domestic violence = 73

San

Bern

ardi

no C

ount

y Po

int i

n Ti

me

2009

page

98

• Percentage of adult unsheltered homeless persons who have experienced domestic violence 73/889=8.2%

• Percentage of all unsheltered homeless persons who have experienced domestic violence 73/965=7.6%

San

Bern

ardi

no C

ount

y Po

int i

n Ti

me

2009

page

99

Appendix HCoC Subpopulation Computations, Sheltered Homeless Population

To facilitate comparisons by gender and age category, Table J presents counts and proportions of the sheltered and unsheltered homeless persons enumerated and surveyed during the 2009 Point-In-Time Count.

Table J. Comparison of Enumerated and Surveyed Homeless Persons in San Bernardino County by Type (Unsheltered/ Sheltered), Gender and Age Group

San

Bern

ardi

no C

ount

y Po

int i

n Ti

me

2009

page

100

Based upon sheltered survey responses, Table K indicates the counts and ratios needed to calculate estimates of the sheltered homeless persons in San Bernardino County in CoC household categories.

Table K. Household Composition Counts Based on Sheltered Survey Data

Part 1: Homeless Population N= 271 Sheltered Survey Respondents*

1) Number of Sheltered Survey Respondents with Dependent Children* n= 251

83(34.0%)

Sheltered Households With Dependents

2a) Total number of children in these households (83 persons in row 1 reported a total of 183 children)

183

2b) Average number of children per household (183/83)

M = 2.21

2c) Total number of adults (including respondent) in these households

144

2d) Average number of adults per household (144/83)**

M = 1.73

2e) Total number of persons in these households (Sum of 2a and 2c)

327

2g) Average number of persons per household (327/ 83)

M = 3.94

Sheltered Households Without Dependents

3a) Number of Survey Respondents Without Dependent Children

161(66.0%)

3b) Total number of adults (including respondent) in these households

235

3c) Average number of adults per household (235/ 161)

M = 1.46

All Sheltered Households

4) Total number of persons in these households (Sum of 2e and 3b)

562

4a) Average number of persons in these households 562/244

2.30

San

Bern

ardi

no C

ount

y Po

int i

n Ti

me

2009

page

101

* Based upon direct reports from 251 of 271 sheltered survey respondents who indicated whether they were living with any male or female children, minus duplicated reports from seven linked pairs.

Table J shows that the extrapolated final count indicates 768 sheltered homeless persons in San Bernardino County on a given January night. Of these, 286 (37.2%) are homeless sheltered children and 482 (62.8%) are adults.

To estimate the number of households represented among these sheltered persons, the number of children (n= 286) is divided by the average number of children per household as reported by sheltered survey respondents (2.21 from row 2b in Table K), producing an estimate of 129 households with dependent children (129 X 2.21 = 285, just one fewer child than actually counted). Multiplying 129 households times the average number of adults in households with dependent children (1.73) results in an estimate of 223 adults. Subtracting these 285 children + 223 adults from the count of 768 sheltered homeless persons leaves 260 adults. Dividing 260 by the average number of adults in households without dependent children (1.46) produces an estimate of 178 households without dependent children. Adding the 129 households with dependent children to the 178 without children produces a total estimate of 307 households among the 768 sheltered homeless persons in San Bernardino County.

As indicated by Table K, based upon survey data, 34.0% of the sheltered survey respondents live in households with children, and 66.0% live in households without children. One hundred twenty-nine is 42.0% of 307. To achieve the 34% ratio computed on the basis of sheltered survey respondents’ reports, the number of households would have to increase to 379. Multiplying 250 (379- 129) times the average number of adults in households without dependent children (1.46) produces a total of 365 sheltered homeless adults; 105 fewer than the 260 remaining adults in the final shelter count. Based upon the number of children observed and the household composition data resulting from the surveys of unsheltered homeless persons, it appears likely that youth were enumerated in “youth only” shelters where they were unaccompanied by an adult, or that some shelters did not report a tally for the Point-In-Time Count. Either of these circumstances might account for the discrepancy between the expected and actual counts of sheltered adults.

CoC Subpopulation Category Computations: Sheltered Homeless

Because the proportion of homeless persons in each of the CoC subpopulation categories depicted in the first column of Table L is likely to vary by age and gender, respondents in the survey sample are classified into five demographic categories: adult males between the ages of 18 and 59, males over the age of 60, adult females between the ages of 18 and 59, females over the age of 60, and persons who did not specify their age or gender on the

San

Bern

ardi

no C

ount

y Po

int i

n Ti

me

2009

page

102

sheltered survey and are classified as unknown. These categories parallel those on the enumeration forms. The 21 persons of unknown age and gender in the shelter enumeration process were issued housing vouchers on the night of the count. Table L presents counts and proportions from the surveyed sheltered homeless population only. Extrapolations to the enumerated sheltered homeless population are calculated in subsequent tables.

As indicated by Table L, proportionally, the largest sheltered homeless subpopulation (n= 30, 12.9%) among survey respondents consists of victims of domestic violence; however these proportions vary by gender. Proportionally, the second largest subpopulation is the chronically homeless (n= 33, 12.2%). Proportionally more males 60 years of age and older (61.5%) are chronically homeless, followed by females in this age group (40.0%) and males 18 to 59 years of age (29.3%).

Table L. Subpopulation Counts and Proportions within Age by Gender Categories in the Surveyed Sheltered Homeless Population

* To be classified as chronically homeless, respondents must report 1) a mental, emotional or developmental disability, or a diagnosable substance abuse problem that is expected to be of long duration and substantially limits their ability to work or to perform one or more activities of daily living, 2) that they are unaccompanied and 3) that they have been continuously homeless for a year or more OR had at least four episodes of homelessness in the past three years.

** To be classified as severely mentally ill, respondents must report that they had been told by a health care provider that they have a mental illness and that the problem keeps them from working/ or (if missing) that they are currently unemployed.

*** To be classified as having a chronic substance abuse problem, respondents must

San

Bern

ardi

no C

ount

y Po

int i

n Ti

me

2009

page

103

have reported that they had a problem with alcohol OR drugs and reported that the problem keeps them from working/ or (if missing) that they are currently unemployed. 233 of 271 sheltered homeless survey respondents replied yes or no to questions about alcohol abuse and drug abuse.

**** 265 of 271 sheltered homeless survey respondents replied yes or no to a question about serving on active duty in the armed forces.

***** 233 of 271 sheltered homeless survey respondents replied yes or no to a question about AIDS/ HIV-related illness and to a question about domestic violence.

To estimate the number of sheltered homeless persons in each of the CoC subpopulations, the extrapolated total number of persons in each demographic category is multiplied by the proportions derived from the survey data falling into each of the subpopulation categories. For example, based upon a shelter census, there are 222 homeless males between the ages of 18 and 59 in shelters in San Bernardino County on any given January night (See Table J). Based on the survey results (Table L), 18.6% of males in this age category are chronically homeless. The product of 222 X .186 is 41 chronically homeless sheltered adult males in this age group.

As depicted by Table M, using this procedure successively in each of the five demographic categories produces a total of 62 sheltered chronically homeless adults in San Bernardino County. Note that children are omitted from Table M, because persons under the age of 18 were not surveyed. This procedure utilized to calculate estimates of the number of chronic homeless persons in the sheltered population is repeated to estimate the count and percentages of persons in each subpopulation presented in Tables M through R.

San

Bern

ardi

no C

ount

y Po

int i

n Ti

me

2009

page

104

Table M. Estimate of Chronically Homeless Adults in the Sheltered Population Based on Survey Data

Demographic Category

Final Count of Adult Sheltered

Homeless

SubpopulationSurvey Count

(%)

Total Count

Adult Male 222 19(18.6)

41

Senior Male 12 1(12.5)

2

Adult Female 226 9(6.3)

14

Senior Female 1 1(33.3)

1

Persons of unknownAge or Gender

21 3(20.0)

4

Total 482(100.0)

33(12.2)

62

• Number of sheltered chronically homeless persons = 62• Percentage of sheltered adults chronically homeless: 62/482= 12.9%• Percentage of chronically homeless in total sheltered population= 62/768= 8.1%

San

Bern

ardi

no C

ount

y Po

int i

n Ti

me

2009

page

105

Table N. Estimate of Severely Mentally Ill in the Sheltered Homeless Population Based on Survey Data

Demographic Category

Final Count of Adult Sheltered

Homeless

SubpopulationSurvey Count

(%)

Total Count

Adult Male 222 12(11.8)

26

Senior Male 12 1(12.5)

2

Adult Female 226 8(5.6)

13

Senior Female 1 0(0.0)

0

Persons of unknownAge or Gender

21 2(13.3)

3

Total 482(100.0)

23(8.5)

44

• Number of sheltered severely mentally ill homeless persons = 44• Percentage of sheltered adults severely mentally ill: 44/482=9.1%• Percentage of severely mentally ill in total sheltered population: 44/768= 5.7%

San

Bern

ardi

no C

ount

y Po

int i

n Ti

me

2009

page

106

Table O. Estimate of Chronic Substance Abusers in the Sheltered Homeless Population Based on Survey Data

Demographic Category

Final Count of Adult Sheltered

Homeless

SubpopulationSurvey Count

(%)

Total Count

Adult Male 222 3(3.4)

8

Senior Male 12 1(16.7)

2

Adult Female 226 1(0.8)

2

Senior Female 1 0(0.0)

0

Persons of unknownAge or Gender

21 0(0.0)

0

Total 482(100.0)

5(2.1)

12

• Number of sheltered chronic substance abusing homeless persons = 12• Percentage of chronic substance abusing sheltered adults: 12/482=2.5%• Percentage of chronic substance abusers in total sheltered population: 12/768=

1.6%

A veteran is defined as a person who served in the active military, naval or air service, and who was discharged or released under conditions other than dishonorable. Twenty-seven sheltered homeless respondents responded affirmatively to the question, “Have you ever served on active duty in the Armed Services of the United States?” The survey did not inquire about discharge status.

San

Bern

ardi

no C

ount

y Po

int i

n Ti

me

2009

page

107

Table P. Estimate of the Number of Veterans in the Sheltered Homeless Population Based on Survey Data

Demographic Category Final Count of Adult Sheltered

Homeless

SubpopulationSurvey Count

(%)

Total Count

Adult Male 222 20(19.6)

44

Senior Male 12 4(50.0)

6

Adult Female 226 3(2.1)

5

Senior Female 1 0(0.0)

0

Persons of unknownAge or Gender

21 0(0.0)

0

Total 482(100.0)

27(10.2)

55

• Number of veteran sheltered homeless persons = 55• Percentage of veteran sheltered adults: 55/482=11.4%• Percentage of veterans in total sheltered population: 55/768= 7.2%

San

Bern

ardi

no C

ount

y Po

int i

n Ti

me

2009

page

108

Table Q. Estimate of the Number of Persons with HIV/ AIDS in the Sheltered Population Based on Survey Data

Demographic Category

Final Count of Adult Sheltered Homeless

SubpopulationSurvey Count(%)*

Total Count

Adult Male 222 1(1.1)

2

Senior Male 12 1(16.7)

2

Adult Female 226 1(0.8)

2

Senior Female 1 0(0.0)

0

Persons of unknownAge or Gender

21 0(0.0)

0

Total 482(100.0)

3(1.3)

6

*Proportions computed on the basis of 155 valid responses.

• Number of sheltered homeless persons currently experiencing AIDS/ HIV-related illness = 6

• Percentage of sheltered adults currently experiencing AIDS/ HIV-related illness: 6/482=1.2%

• Percentage of persons currently experiencing AIDS/ HIV-related illness in total sheltered population: 6/768= 0.8%

San

Bern

ardi

no C

ount

y Po

int i

n Ti

me

2009

page

109

Table R. Estimate of the Number of Persons Experiencing Domestic Violence in the Sheltered Population Based on Survey Data

Demographic Category

Final Count of Adult Sheltered

Homeless

SubpopulationSurvey Count

(%)

Total Count

Adult Male 222 3(3.4)

8

Senior Male 12 0(0.0)

0

Adult Female 226 25(19.7)

45

Senior Female 1 0(0.0)

0

Persons of unknownAge or Gender

21 2(22.2)

5

Total 482(100.0)

30(12.9)

58

• Number of sheltered homeless persons currently experiencing domestic violence= 58

• Percentage of sheltered currently experiencing domestic violence: 58/482=12.0%

• Percentage currently experiencing domestic violence in total sheltered population: 58/768= 7.6%

San

Bern

ardi

no C

ount

y Po

int i

n Ti

me

2009

page

110

Appendix IList of Census Block Groups Canvassed

San

Bern

ardi

no C

ount

y Po

int i

n Ti

me

2009

page

111

BKGPIDFP-1 BKGPIDFP-2 Region City/Community Name

Second City/Community Name

060710091048 VICTOR VALLEY ADELANTO AREA060710103005 DESERT BAKER060710093001 BARSTOW BARSTOW AREA060710094001 BARSTOW BARSTOW AREA060710094003 BARSTOW BARSTOW AREA060710116003 DESERT BARSTOW AREA060710118003 DESERT BARSTOW AREA060710119001 DESERT BARSTOW AREA060710120001 BARSTOW BARSTOW AREA060710036024 VALLEY BLOOMINGTON060710036025 VALLEY BLOOMINGTON60710119001 DESERT CHINA LAKE060710091045 VICTOR VALLEY CITY OF ADELANTO060710095004 BARSTOW CITY OF BARSTOW060710095006 BARSTOW CITY OF BARSTOW060710120002 BARSTOW CITY OF BARSTOW060710004027 VALLEY CITY OF CHINO060710036018 VALLEY CITY OF COLTON060710025015 VALLEY CITY OF FONTANA060710026013 VALLEY CITY OF FONTANA060710026015 VALLEY CITY OF FONTANA060710032002 VALLEY CITY OF FONTANA060710033008 VALLEY CITY OF FONTANA060710040002 VALLEY CITY OF FONTANA060710040009 VALLEY CITY OF GRAND TERRACE060710100102 VICTOR VALLEY CITY OF HESPERIA060710100112 VICTOR VALLEY CITY OF HESPERIA060710100134 VICTOR VALLEY CITY OF HESPERIA060710100143 VICTOR VALLEY CITY OF HESPERIA060710100201 VICTOR VALLEY CITY OF HESPERIA060710074065 VALLEY CITY OF HIGHLAND060710074074 VALLEY CITY OF HIGHLAND060710076021 VALLEY CITY OF HIGHLAND060710076022 VALLEY CITY OF HIGHLAND060710076024 VALLEY CITY OF HIGHLAND060710079011 VALLEY CITY OF HIGHLAND060710079021 VALLEY CITY OF HIGHLAND060710072006 VALLEY CITY OF LOMA LINDA060710072007 VALLEY CITY OF LOMA LINDA060710078003 VALLEY CITY OF LOMA LINDA060710003015 VALLEY CITY OF MONTCLAIR060710106001 DESERT CITY OF NEEDLES060710107001 DESERT CITY OF NEEDLES060710107002 DESERT CITY OF NEEDLES060710013031 VALLEY CITY OF ONTARIO060710013045 VALLEY CITY OF ONTARIO060710016006 VALLEY CITY OF ONTARIO060710022031 VALLEY CITY OF ONTARIO

060710020075 VALLEYCITY OF RANCHO CUCAMONGA

060710021002 VALLEYCITY OF RANCHO CUCAMONGA

San

Bern

ardi

no C

ount

y Po

int i

n Ti

me

2009

page

112

060710021007 VALLEYCITY OF RANCHO CUCAMONGA

060710022043 VALLEYCITY OF RANCHO CUCAMONGA

060710080026 VALLEY CITY OF REDLANDS060710081001 VALLEY CITY OF REDLANDS060710081002 VALLEY CITY OF REDLANDS060710083014 VALLEY CITY OF REDLANDS060710023021 VALLEY CITY OF RIALTO060710035012 VALLEY CITY OF RIALTO060710035014 VALLEY CITY OF RIALTO060710035015 VALLEY CITY OF RIALTO060710035016 VALLEY CITY OF RIALTO060710036012 VALLEY CITY OF RIALTO060710036013 VALLEY CITY OF RIALTO060710036016 VALLEY CITY OF RIALTO060710036021 VALLEY CITY OF RIALTO060710037002 VALLEY CITY OF RIALTO060710038002 VALLEY CITY OF RIALTO060710038004 VALLEY CITY OF RIALTO060710040001 VALLEY CITY OF RIALTO060710041009 VALLEY CITY OF SAN BERNARDINO060710042023 VALLEY CITY OF SAN BERNARDINO060710046015 VALLEY CITY OF SAN BERNARDINO060710047002 VALLEY CITY OF SAN BERNARDINO060710048003 VALLEY CITY OF SAN BERNARDINO060710049001 VALLEY CITY OF SAN BERNARDINO060710050001 VALLEY CITY OF SAN BERNARDINO060710055004 VALLEY CITY OF SAN BERNARDINO060710055005 VALLEY CITY OF SAN BERNARDINO060710056003 VALLEY CITY OF SAN BERNARDINO060710057001 VALLEY CITY OF SAN BERNARDINO060710057002 VALLEY CITY OF SAN BERNARDINO060710058001 VALLEY CITY OF SAN BERNARDINO060710058003 VALLEY CITY OF SAN BERNARDINO060710059001 VALLEY CITY OF SAN BERNARDINO060710059002 VALLEY CITY OF SAN BERNARDINO060710060001 VALLEY CITY OF SAN BERNARDINO060710060009 VALLEY CITY OF SAN BERNARDINO060710063013 VALLEY CITY OF SAN BERNARDINO060710065004 VALLEY CITY OF SAN BERNARDINO060710071026 VALLEY CITY OF SAN BERNARDINO060710074044 VALLEY CITY OF SAN BERNARDINO060710074061 VALLEY CITY OF SAN BERNARDINO060710074082 VALLEY CITY OF SAN BERNARDINO060710078001 VALLEY CITY OF SAN BERNARDINO060710079014 VALLEY CITY OF SAN BERNARDINO

060710104142 TWENTYNINE PALMCITY OF TWENTYNINE PALMS

060710104153 TWENTYNINE PALMCITY OF TWENTYNINE PALMS

060710104163 TWENTYNINE PALMCITY OF TWENTYNINE PALMS

060710008091 VALLEY CITY OF UPLAND060710008103 VALLEY CITY OF UPLAND

San

Bern

ardi

no C

ount

y Po

int i

n Ti

me

2009

page

113

060710008112 VALLEY CITY OF UPLAND060710008114 VALLEY CITY OF UPLAND060710009001 VALLEY CITY OF UPLAND060710009003 VALLEY CITY OF UPLAND060710098002 VICTOR VALLEY CITY OF VICTORVILLE060710099011 VICTOR VALLEY CITY OF VICTORVILLE060710099031 VICTOR VALLEY CITY OF VICTORVILLE060710099032 VICTOR VALLEY CITY OF VICTORVILLE060710099034 VICTOR VALLEY CITY OF VICTORVILLE060710100032 VICTOR VALLEY CITY OF VICTORVILLE060710100044 VICTOR VALLEY CITY OF VICTORVILLE060710087044 VALLEY CITY OF YUCAIPA060710087054 VALLEY CITY OF YUCAIPA060710087062 VALLEY CITY OF YUCAIPA060710087083 VALLEY CITY OF YUCAIPA060710088002 VALLEY CITY OF YUCAIPA060710022044 VALLEY FONTANA AREA060710024022 VALLEY FONTANA AREA060710026011 VALLEY FONTANA AREA060710034033 VALLEY FONTANA AREA060710118002 BARSTOW HINKLEY060710113002 MOUNTAINS HOLCOMB VALLEY060710104032 TWENTYNINE PALM JOSHUA TREE060710104033 TWENTYNINE PALM JOSHUA TREE060710104035 TWENTYNINE PALM JOSHUA TREE060710104036 TWENTYNINE PALM JOSHUA TREE060710097083 LUCERNE VALLEY LUCERNE VALLEY060710121006 LUCERNE VALLEY LUCERNE VALLEY060710121007 LUCERNE VALLEY LUCERNE VALLEY060710003031 VALLEY MONTCLAIR AREA060710103004 DESERT MOUNTAIN PASS060710106002 DESERT NEEDLES AREA060710107004 DESERT NEEDLES AREA060710091064 VICTOR VALLEY PHELAN060710092004 VICTOR VALLEY PHELAN060710078002 VALLEY REDLANDS AREA060710080013 VALLEY REDLANDS AREA060710104111 MORONGO/YUCCA V RIMROCK060710063014 VALLEY SAN BERNARDINO AREA060710063026 VALLEY SAN BERNARDINO AREA060710064012 VALLEY SAN BERNARDINO AREA060710065001 VALLEY SAN BERNARDINO AREA060710097072 VICTOR VALLEY TOWN OF APPLE VALLEY060710097121 VICTOR VALLEY TOWN OF APPLE VALLEY060710097142 VICTOR VALLEY TOWN OF APPLE VALLEY060710097155 VICTOR VALLEY TOWN OF APPLE VALLEY060710097171 VICTOR VALLEY TOWN OF APPLE VALLEY060710121004 VICTOR VALLEY TOWN OF APPLE VALLEY060710104112 MORONGO/YUCCA V TOWN OF YUCCA VALLEY060710104113 MORONGO/YUCCA V TOWN OF YUCCA VALLEY060710104121 MORONGO/YUCCA V TOWN OF YUCCA VALLEY060710104133 MORONGO/YUCCA V TOWN OF YUCCA VALLEY060710104143 DESERT TWENTYNINE PALMS AREA060710104164 DESERT TWENTYNINE PALMS AREA

San

Bern

ardi

no C

ount

y Po

int i

n Ti

me

2009

page

114

060710104092 DESERTTWENTYNINE PALMS MARINE BASE

060710104162 DESERTTWENTYNINE PALMS MARINE BASE

060710099014 VICTOR VALLEY VICTORVILLE AREA060710092003 MOUNTAINS WRIGHTWOOD

060710100221 VICTOR VALLEY APPLE VALLEY AREA060710089012 DESERT BAKER060710103006 DESERT BAKER060710115003 MOUNTAINS BARTON FLATS060710120005 BARSTOW CITY OF BARSTOW060710001072 060710001071 VALLEY CITY OF CHINO CITY OF CHINO060710001083 VALLEY CITY OF CHINO HILLS060710001121 VALLEY CITY OF CHINO HILLS

060710066002 060710066001 VALLEY CITY OF COLTONCITY OF COLTON

060710067004 060710067003 VALLEY CITY OF COLTONCITY OF COLTON

060710026022 060710026021 VALLEY CITY OF FONTANACITY OF FONTANA

060710026024 060710026023 VALLEY CITY OF FONTANACITY OF FONTANA

060710026037 060710026036 VALLEY CITY OF FONTANACITY OF FONTANA

060710028006 060710028005 VALLEY CITY OF FONTANACITY OF FONTANA

060710031006 060710031005 VALLEY CITY OF FONTANACITY OF FONTANA

060710034014 060710034013 VALLEY CITY OF FONTANACITY OF FONTANA

060710071072 060710071071 VALLEY CITY OF GRAND TERRACE

CITY OF GRAND TERRACE

060710100132 VICTOR VALLEY CITY OF HESPERIA060710100181 VICTOR VALLEY CITY OF HESPERIA060710100202 VICTOR VALLEY CITY OF HESPERIA060710100231 VICTOR VALLEY CITY OF HESPERIA

060710079023 060710079022 VALLEY CITY OF HIGHLANDCITY OF HIGHLAND

060710073016 060710073015 VALLEY CITY OF LOMA LINDACITY OF LOMA LINDA

060710018022 060710018021 VALLEY CITY OF ONTARIOCITY OF ONTARIO

060710013044 060710013043 VALLEYCITY OF RANCHO CUCAMONGA

CITY OF RANCHO CUCAMONGA

060710020073 VALLEYCITY OF RANCHO CUCAMONGA

060710020074 VALLEYCITY OF RANCHO CUCAMONGA

060710084039 VALLEY CITY OF REDLANDS060710085002 VALLEY CITY OF REDLANDS

060710047003 060710047001 VALLEY CITY OF SAN BERNARDINOCITY OF SAN BERNARDINO

060710051007 VALLEY CITY OF SAN BERNARDINO

060710054002 060710054001 VALLEY CITY OF SAN BERNARDINOCITY OF SAN BERNARDINO

060710072005 VALLEY CITY OF SAN BERNARDINO060710075001 VALLEY CITY OF SAN BERNARDINO

San

Bern

ardi

no C

ount

y Po

int i

n Ti

me

2009

page

115

060710087033 VALLEY CITY OF YUCAIPA060710020101 VALLEY FONTANA AREA060710103007 DESERT HARVARD060710117001 DESERT HELENDALE060710109006 MOUNTAINS LAKE ARROWHEAD060710086008 060710086007 VALLEY MENTONE MENTONE060710091063 VICTOR VALLEY PHELAN060710091067 VICTOR VALLEY PHELAN

060710020094 VALLEYRANCHO CUCAMONGA AREA

060710027017 060710027016 VALLEY RIALTO AREA RIALTO AREA

060710008042 060710008041 VALLEY SAN ANTONIO HEIGHTSSAN ANTONIO HEIGHTS

060710100042 060710100041 VICTOR VALLEY SPRING VALLEY LAKESPRING VALLEY LAKE

060710097091 VICTOR VALLEY TOWN OF APPLE VALLEY060710097113 VICTOR VALLEY TOWN OF APPLE VALLEY060710097132 VICTOR VALLEY TOWN OF APPLE VALLEY060710087031 VALLEY YUCAIPA AREA

San

Bern

ardi

no C

ount

y Po

int i

n Ti

me

2009

page

116

Appendix JAcknowledgmentsThe County of San Bernardino Office of Homeless Services wishes to thank all of the individuals and organizations that contributed to the implementation of the 2009 Point-In-Time Homeless Count and Survey process. A project of this magnitude could not be successful without the support and tireless efforts of the many volunteers, County and city employees, service providers, law enforcement and elected officials. All of your hard work was central to the success of this project.

• Volunteers – Over 500 community volunteers from all across San Bernardino County canvassed hundreds of census block groups and conducted many surveys on the night of the Count. Volunteers came from a variety of agencies, including County departments, city departments, faith-based organizations, social service providers, and local colleges and universities.

• County Departments – Many County departments contributed their staff time to this effort during the planning process. Several departments also offered the use of their offices as Deployment Center sites on the night of the Count:

o County of San Bernardino Board of Supervisorso County of San Bernardino Human Services o County of San Bernardino Department of Behavioral Healtho County of San Bernardino Transitional Assistance Departmento County of San Bernardino Information Services Department

San

Bern

ardi

no C

ount

y Po

int i

n Ti

me

2009

page

117

o County of San Bernardino Probation Departmento County of San Bernardino Department of Veterans Affairso County of San Bernardino Public Health Departmento County of San Bernardino District Attorneyo County of San Bernardino Public Defendero County of San Bernardino Department of Public Workso County of San Bernardino Land Use Services Departmento County of San Bernardino County Counselo County of San Bernardino Department of Aging and Adult Serviceso County of San Bernardino Department of Children and Family Serviceso County of San Bernardino Performance Education and Resource Centers

• Public Agencies – Several communities and other public agencies outside of the County network were involved in the planning process and offered the use of their staff and facilities resources for the Count:

o City of Adelantoo Fontana Unified School Districto City of Ontario – Code Enforcemento Ontario-Montclair School Districto City of San Bernardinoo San Bernardino City Unified School Districto City of San Bernardino Police Departmento County of San Bernardino Superintendent of Schoolso Upland Unified School Districto City of Victorville

• Community Organizations – Many homeless service providers and other community groups offered their support and counsel during the planning process, and some agencies also offered their facilities for use as training sites:

o 2-1-1 San Bernardino Countyo Arrowhead Unity Wayo The Agape Houseo California State University at San Bernardinoo Catholic Charitieso Central City Lutheran Missiono Community Action Partnershipo Desert Sanctuaryo Doves of Big Bear Valleyo Desert Manna Ministrieso Ecclesia Christian Fellowship o Family Services Association of Redlandso First 5 San Bernardinoo Foothill Family Shelter

San

Bern

ardi

no C

ount

y Po

int i

n Ti

me

2009

page

118

o Frazee Community Centero Goodwill Southern Californiao High Desert Domestic Violence Programo High Desert Homeless Serviceso House of Prayero House of Rutho Inland AIDS Projecto Inland Behavioral and Health Serviceso Inland Counties Legal Serviceso Inland Empire United Wayo Inland Temporary Homeso Loma Linda Universityo The Loving Home of Lady Magdaleneo Lucerne Valley Domestic Violenceo Mary’s Mercy Centero Mental Health Systems, Inc.o Mercy Houseo Mimi’s Houseo Morongo Basin ARCHo Morongo Basin Unity Homeo New Hope Villageo Oasis House o Option Houseo Pacific Lifelineo Restore to Hopeo The Salvation Armyo Seventh Day Adventist - Inland Spanish Churcho St. Bernardine Medical Centero Time For Change Foundationo Veteran’s Affairs Loma Lindao Veteran’s Alliance

• Research Consultants – The consultants and Planning Committee provided guidance and expertise to the methodology development, logistics support, and the development of the final report. Their assistance throughout the process was incalculable.

o Bob Cerince, M.Div., MPA, CLA & Associateso Jennifer Lee-Anderson, M.A., CLA & Associateso Gregory Robinson, Ph.D., Applied Research Solutionso Brent Rolf, Director, San Bernardino County Information Services

Department GIMS Teamo Jay Sumner, Ph.D., Professor Emeritus, University of California, Los Angeles

San

Bern

ardi

no C

ount

y Po

int i

n Ti

me

2009

page

119

• Planning Commitee – The consultants and Planning Committee provided guidance and expertise to the methodology development, logistics support, and the development of the final report. Their assistance throughout the process was incalculable.

o Lori Ciabattini, County of San Bernardino Human Serviceso Issac Jackson, County of San Bernardino Human Services Office of

Homeless Serviceso Phalos Haire, County of San Bernardino Human Services Office of

Homeless Serviceso Regina Linares, County of San Bernardino Human Services Office of

Management Serviceso Harriet White, County of San Bernardino Human Services Program

Integrity Divisiono Heidi Mahlum, County of San Bernardino Human Services Program

Integrity Divisiono Barbara Rigg, County of San Bernardino Human Serviceso Kelly Tobin, County of San Bernardino Human Services Administrative

Support Divisiono Brenda Bovee, County of San Bernardino Human Services Program

Development Division

San

Bern

ardi

no C

ount

y Po

int i

n Ti

me

2009

page

120