Measuring Transportation Accessibility for the Homeless Populations in Gainesville, Florida Using Network Analyst

Material Information

Measuring Transportation Accessibility for the Homeless Populations in Gainesville, Florida Using Network Analyst
Salvatore, Kaysie A
Place of Publication:
[Gainesville, Fla.]
University of Florida
Publication Date:
Physical Description:
1 online resource (82 p.)

Thesis/Dissertation Information

Master's ( M.U.R.P)
Degree Grantor:
University of Florida
Degree Disciplines:
Urban and Regional Planning
Committee Chair:
Committee Co-Chair:


Subjects / Keywords:
accessibility -- arcgis -- homeless -- transportation
Urban and Regional Planning -- Dissertations, Academic -- UF
bibliography ( marcgt )
theses ( marcgt )
government publication (state, provincial, terriorial, dependent) ( marcgt )
born-digital ( sobekcm )
Electronic Thesis or Dissertation
Urban and Regional Planning thesis, M.U.R.P


Despite the importance placed on mobility in the US, the ability to move does not benefit all populations equally. The homeless population has been found to be the most negatively impacted in regards to transportation accessibility given their low economic standing. Research shows that providing transportation services and programs to the homeless would not only increase their mobility, but also their access to jobs and services. This paper intends to supplement current research on transportation accessibility for the homeless by providing a spatial analysis of transportation disparities in populations with access to public transit. Using research interviews and questionnaires, this study answers the questions of where and how the homeless are traveling. Additionally, the Esri Network Analyst extension is used to map how the homeless travel and compares it to other modes of travel. Lastly, this study uses the questionnaire and spatial analysis to determine where any disparities might be occurring in the current transit system and provides recommendations of route updates to better service the homeless population in Gainesville, Florida. The questionnaire found that an overwhelming majority of the homeless population in Gainesville travel using bus. The network analysis found that travel by car was the fastest and most efficient, while traveling by bus took on average eight minutes longer when excluding transfer wait times and twenty minutes longer when including transfer wait times. Based on the data collected, two bus routes servicing GRACE Marketplace were updated to reflect realistic travel patterns of the homeless. ( en )
General Note:
In the series University of Florida Digital Collections.
General Note:
Includes vita.
Includes bibliographical references.
Source of Description:
Description based on online resource; title from PDF title page.
Source of Description:
This bibliographic record is available under the Creative Commons CC0 public domain dedication. The University of Florida Libraries, as creator of this bibliographic record, has waived all rights to it worldwide under copyright law, including all related and neighboring rights, to the extent allowed by law.
Thesis (M.U.R.P)--University of Florida, 2017.
Co-adviser: ZWICK,PAUL D.
Statement of Responsibility:
by Kaysie A Salvatore.

Record Information

Source Institution:
Rights Management:
Applicable rights reserved.
LD1780 2017 ( lcc )


This item has the following downloads:

Full Text




2017 Kaysie Salvatore


To my loving family boyfriend, and all the homeless individuals current ly struggling to survive. Just know, most of us see you and are fighting for you. Keep on keeping on.


4 ACKNOWLEDGMENTS I thank Jon De Carmine, the Operations Director of GRACE Marketplace, for making this study possible and for being the voice of the homeless in Gainesville. I would also like to thank Dr. Ruth Steiner for her contribution to my study as well as for supporting me and pushing me to reach my true potential.


5 TABLE OF CONTENTS page ACKNOWLEDGMENTS ................................ ................................ ................................ .. 4 LIST OF TABLES ................................ ................................ ................................ ............ 7 LIST OF FIGURES ................................ ................................ ................................ .......... 8 ABSTRACT ................................ ................................ ................................ ..................... 9 CHAPTER 1 INTRODUCTION ................................ ................................ ................................ .... 11 2 LITERATURE REVIEW ................................ ................................ .......................... 14 2.1 Dimensions of Homelessness ................................ ................................ ........... 14 2.1.1 Housing Dimension ................................ ................................ ................. 14 2.1.2 Temporal Dimension ................................ ................................ ............... 14 2.1.3 Working Dimension ................................ ................................ ................. 15 2.1.4 Veterans and Disability Dimension ................................ .......................... 17 2.2 Homeless and Transportation Disparities ................................ ......................... 18 2.3 Transportation as a Civil Right ................................ ................................ .......... 24 3 STUDY AREA AND DATA COLLECTION ................................ .............................. 27 3.1 Study Area ................................ ................................ ................................ ........ 27 3.1.1 Selection Process ................................ ................................ .................... 27 3.1.2 Demographic Data ................................ ................................ ................... 30 3.2 Data Collection and Preprocessing ................................ ................................ ... 33 4 METHODOLOGY ................................ ................................ ................................ ... 38 4.1 Travel Data Collection ................................ ................................ ....................... 38 4.1.1 Interviews ................................ ................................ ................................ 39 4 .1.2 Homeless Questionnaire ................................ ................................ ......... 39 4.2 Network Analysis ................................ ................................ .............................. 40 5 RESULTS AND DISCUSSION ................................ ................................ ............... 51 5.1 Results ................................ ................................ ................................ .............. 51 5.1.1 Questionnaire Statistics ................................ ................................ ........... 51 5.1.2 Network Analyst Analysis ................................ ................................ ........ 52 5.1.3 Creation of a New Route ................................ ................................ ......... 54 5.2 Discussion ................................ ................................ ................................ ........ 55 5.2.1 Questionnaire Responses ................................ ................................ ....... 55


6 5.2.2 Travel Statistics ................................ ................................ ....................... 57 5.2.3 Network Analyst Analysis ................................ ................................ ........ 58 5.2.4 Creation of a New Route ................................ ................................ ......... 5 9 6 LIMITATIONS AND CONCLUSION ................................ ................................ ........ 69 6.1 Limitations ................................ ................................ ................................ ......... 69 6.2 Conclusion ................................ ................................ ................................ ........ 72 APPENDIX A INTERVIEW QUESTIONS ................................ ................................ ..................... 74 B HOMELESS TRAVEL QUESTIONNAIRE ................................ .............................. 76 LIST OF REFERENCES ................................ ................................ ............................... 78 BIOGRAPHICAL SKETCH ................................ ................................ ............................ 82


7 LIST OF TABLES Table page 3 1 Organizations Contacted for Interviews ................................ .............................. 35 4 1 Travel Destinations as Determined by the Homeless Questionnaire .................. 46 5 1 Trav el Time Given in Minutes and Seconds ................................ ....................... 62 5 2 Destinations within 0.5 Miles of the Suggested Bus Stop ................................ ... 64


8 LIST OF FIGURES Figure page 3 1 Study Area ................................ ................................ ................................ .......... 36 3 2 GRACE Marketplace Boundary & Bus Routes ................................ ................... 37 4 1 Methodological Design ................................ ................................ ....................... 48 4 2 Alachua County Roads Network Dataset ................................ ........................... 49 4 3 City of Gainesville Bus Network Dataset ................................ ............................ 50 5 1 Homeless Questionnaire Statistics ................................ ................................ ..... 66 5 2 Alachua County Jail Shortest Path Example ................................ ...................... 67 5 3 S uggested Bus Route ................................ ................................ ......................... 68


9 Abstract of Thesis Presented to the Graduate School of the University of Florida in Partial Fulfillment of the Requiremen ts for the Degree of Master of Urban and Regional Planning ME ASURING TRANSPORTATION ACCESSIBILITY FOR THE HOMELESS POPULATIONS IN GAINESVILLE, FLORIDA USING NETWORK ANALYST By Kaysie Salvatore December 2017 Chair: Ruth L. Steiner Cochair: Paul Zwick Major: Urban and Regional Planning Despite the importance place d on mobility in the US the ability to move does not benefit all populations equally. The homeless population has been found to be the most negatively impacted in regards to t ransportation accessibility given their low economic s tanding. Research show s that p roviding transportation services and programs to the homeless would not only increase their mobility, but also their access to jobs and services. This paper intends to supplement current research on transportation accessibility for the homeless by pr oviding a spatial analysis of transportation disparities in populations with access to public transit Using research interviews and questionnaires, this study answers the questions of where and how the homeless are tra veling. Additionally, The Esri Netw ork Analyst exte nsion is used to map how the homeless travel and compares it to other modes of travel. Lastly, this study uses the questionnaire and spatial analysis to determine where any disparities might be occurring in the current transit system and p rovides recommendations of route updates to better service the homeless population in Gainesville, Florida.


10 The questionnaire found that an overwhelming majority of the homeless population in Gainesville travel using bus. The network analysis found that t ravel by car was the fastest and most efficient, while trave ling by bus took on average eight minutes longer when excluding transfer wait times and twenty minutes longer when including transfer wait times Based on the data collected, two bus routes servi cing GRACE Marketplace wer e updated to refle ct realistic travel patterns of the homeless.


11 CHAPTER 1 INTRODUCTION plays a critical role in the lives of all Americans, especially th ose who may be homeless The Bureau of Labor Statistics (BLS) Consumer Expenditure Survey, further quantifies the importance of transportation by reporting that household s spent an average of $9,073 on transportation in 2014; the second largest household expenditure behind housing ( Bureau of Transportation Statistics, pg 71, 2016 ). Despite the importance placed on mobility in the United States, the ability to move does no t benefit all populations equally. In 2016, BLS reported that low and moderate income household s spent approximately 30% of their annual income on transportation expenditures compar ed to 16% by middle income households ( Bureau of Transportation Statistic s, pg 75, 2016 ). Further, the Consumer Expenditure Survey found that households in the top income quintile bracket ($99,621+) owned an average of 2.8 vehicles per household, whereas households in the bottom income quintile bracket (<$18,362) owned an aver age of 0.9 vehicles per household ( Bureau of Transportation Statistics, pg 74, 2016 ). Although specific organizations dedicated to assisting the homeless, such as the National Coalition for the Homeless (NCH), have spent countless hours and money providin g access to transportation, there continues to be a lack of transportation options for the homeless. A study conducted by the US Department of Housing and Urban Development (HUD) concluded that despite increased awareness of homelessness and transit barri ers, not enough fund ing has been dedicated to serv ing the homeless (Burt et


12 al., pg 79, 2010). Further, the US DOT states that the homeless lack the resources necessary for mobility, thus increasing their need of transit services (US DOT, np, 2016). Beca use the homeless have significantly less to spend on transportation than other economic groups, they can be thought of as being the most needing population of transportation programs and services. Providing transportation services and programs to the home less would not only increase their mobility, but also their access to jobs and services. A survey conducted by the Employment Committee of the Sacramento Ending Chronic Homeless Initiati ve (ECHI) found that 30% of the homeless interviewed cited a lack of transportation as a barrier to finding work (Acuna, J. & Erlenbusch, B., np, 2009). Further, studies by Harvard and New York University have found that accessibility to transit increased job prospects and social mobility. The increased awareness of mobili ty impacts on all individuals has led to a movement focused on transportation access as a civil right. Support for this movement comes from Article 13 of The Universal Declaration of Human Rights which states that veryone has the right to freedom of mo vement and residence within the borders of United Nations General Assembly pg 4, 1948) and the Fourteenth Amendment to the U.S. Constitution recognizing the right to travel as a fundamental right. Furthermore, the Leadership Conference on Ci vil and Human Rights believes that succeed. This paper intends to supplement current research on transportation accessibility for the homeless by providing a spatial ana lysis of transportation disparities in


13 populations with access to public transit. This study has four main objectives and research questions: 1. Where are the homeless traveling in Gainesville, Florida? According to HUD, the homeless are traveling to ac cess health services, education facilities, social service administers, career centers, and shelters. This study will use interviews and questionnaires to determine which locations the homeless are traveling in Gainesville, Florida. 2. How are the homel ess traveling in Gainesville, Florida and what transit difficulties are they facing? A study conducted by California State University fo und that approximately 94% of the homeless in Long Beach, CA used public transit to travel. Other data shows that the homeless travel via bicycle, carpooling, walking, and driving personal vehicles. This paper will use data collected from interviews and questionnaires to determine which modes of transportation the homeless in Gainesville are using. 3. Are there transit disparities between where the homeless travel and the modes in which they use to travel? In particular, this study will use Network Analyst to model the most efficient routes for bus and car between origin and destination (O D). Additionally, Network An alyst will determine where disparities exist in bus transit compared to car travel. 4. If disparities exist between popularly visited locations by the homeless and bus networks, how can these disparities be remedied? This study will conclude with recomm endations for better transit location connectivity. The remainder of this thesis is organized into five chapters. Chapter 2 covers a literature review based on current research of homeless travel. Chapter 3 introduces the study area and data sources. Chapter 4 outlines the methodological de sign of the study Chapter 5 provides t he results and Chapter 6 provides the conclusion of the study along with implications and limitations.


14 CHAPTER 2 LITERATURE REVIEW This lit erature review will be divided into three sections. The first section explores the many dimensions of homelessness, the second section outlines the disparities associated with homeless mobility, and the concluding section discusses transportation as a hum an right. 2.1 Dimensions of Homelessness 2.1.1 Housing Dimension The definition of homelessness, as denoted by the US Department of Health and an individual whose primary residence during the night is a supervised public or private facility that provides temporary living accommodations, and an individual who is a n.d.). Although this de finition may appear straightforward, much of the concepts this definition, a person squatting in an abandoned apartment building could be classified as having housing; a classification most humans would assume as being incorrect. According to HHS, a person residing in transitional housing is considered housing is not considered homeless. As this example shows, there is no singular definition for homelessness. 2.1.2 Temporal Dimension The New Homelessness Revisited a peer reviewed article authored by Barrett Lee, Kimberly Tyler, and James Wright, discusses the complexity associated with


15 cl assifying homelessness. In addition to the housing dimension, homelessness also has a temporal dimension. Lee et al. identifies three major types of homeless patterns: transitional or temporary describing individuals who are in transition between s table housing situations and whose brief homeless spells often amount to once in a lifetime events, (2) episodic which entails cycling in and out of homelessness over short periods, and (3) chronic yler, K. A., & Wright, J. D., pg 3, 2010). Statistically, most homeless studies show much of the homeless population as being chronically homeless. However, this figure is drastically inflated because most individuals falling within the temporary or epi sodic clusters of homelessness are often cities), and may not appear as being homeless given their short stint without housing. According to the Department of Hous ing and Urban Development (HUD), long term or homeless at some point in 2009, but on any given day, only about 112,000 people fit the federal definition of chronic hom Additionally HUD found that most people entering the shelter system leave within thirty days and never re turn (Culhane, D., n.p., 2010). 2.1.3 Working Dimensio n In 1967, an individual making minimum wage could supp ort a family of three while living above the poverty line ( National Coalition for the Homeless n.p., 2009). Between 1981 1990, the cost of living increased b y 48% while the minimum wage stagnated at $3.35 an hour. In 1996, Congress increased the minimum wage to $5.15 an hour and then again in 2007 to $7.25 an hour. The national minimum wage has not


16 increased since 2007. Although the national minimum wage has increased between alue of the National Coalition for the Homeless n.p., 2009). The belief is that minimum wage is designed for those in high school, college, or just entering the work environment. However, the Economic Policy Institute fou nd that approximately 79% of minimum wage workers are 20 years of age or older ( National Coalition for the Homeless n.p., 2009). With more adults finding themselves working minimum wage jobs; the number of individuals able to afford housing is decreasing two fold. In order to afford a two bedroom apartment at 30% of their income, a minimum wage worker would be forced to work 87 hours per week ( National Coalition for the Homeless n.p., 2009). As housing prices increase and minimum wage sta gnates, it is no surprise that homelessness is a problem in the United States. about 45% of it is for the work ing homeless to escape homelessness, it is virtually impossible for those without a job to escape homelessness. Those with limited skills struggle to find as limited ( National Coalition for the Homeless n.p., 2009). To combat the educational and training barriers, many homeless shelters are now offering free or reduced rate assistance for career centers and colleges. In addition, many local businesses provide resources that increase the chances of landing employment (such as appropriate dress wear and transportation) and making a good first impression. An evaluation of the Job


17 Training for the H omeless Demonstration Program found that successful employment programs increase access to services such as housing and transportation ( National Coalition for the Homeless n.p., 2009). 2.1.4 Veterans and Disability Dimension Those who are veterans or thos e with disabilities are often characterized as being chronically homeless. A paper written by Laura Blankertz, Ram Cnaan, and Marlene Saunders (1992) population comprised of the long term mentally di sabled, because of their recognized often overwhelming needs, their public visibility, and the controversy surrounding policy changes in institutionalization, housing, and disability benefits, which have repeatedly been targeted a s causes of their homeless A 2011 study by the National Health Care for the Homeless Council found that people with di sabilities constitute over 40% of the homeless population in America ( National Health Care for the Homeless Council, pg 1, 2011). In addition, a study condu cted by the National Coalition for Homeless Veterans determined t hat approximately 11% of the homeless population are veterans (National Coalition for Homeless Veterans, n.p., n.d.). Of those 11% the majority suffer from mental illnesses or co occurring disorders. Fortunately, both veterans and those with disabilities are afforded resources that the rest of the population is unable to access. These resources come in the form of monetary assistance, increased housing and work assistance, and transportat ion assistance. However, accessing these resources can be quite difficult and time consuming. Often, the disabled and veterans struggle to obtain the necessary paperwork due to transportation, physical, and mental barriers. To decrease the number of vet erans and disabled living as homeless, programs have been developed to


18 simplify the assistance applications. For example, GRACE Marketplace, a homeless shelter located in Gainesville, Florida, brings in volunteers to help fill out assistance forms so that veterans and the disabled are able to access transportation, housing, work, and monetary benefits. 2.2 Homeless and Transportation Disparities it has been treated academically (Arnold, K. R., np, 2004). In an article written by Kathleen Arnold, she analyzes how the non homeless population views the homeless population. Arnold states that the views of the homeless originat e in myth and stereotype which often demonizes and incriminates the homeless population. It is through the confluence of media, policy makers, and gossip that these myths and stereotypes continue to spread and stigmatize the homeless population as being r esponsible for their situation. However, this radicalized perception could not be further from the truth. According to the US Department of Housing and Urban Development (HUD), an at risk of homelessness individual or family must fall within the following criteria: 1. Having an annual income below 30% of median family income; AND 2. Does not have sufficient resources or support networks available preventing them from moving to a shelter or another place defined within the homeless definition; AND 3. Meets ONE of th e following conditions: a) Has moved because of economic reasons two or more times in 60 days; OR b) Is living in the home of another because of economic hardship; OR c) Has been notified of that their right to occupy their current housing or living situation will be terminated within 21 days; OR d) Lives in a hotel or motel where the cost is not paid for by a charitable e) organization or governmental program; OR f) Lives in an efficiency apartment unit in which there reside more than 2 persons or lives in a larger housing unit in which there reside more than one and a half persons per room; OR g) Is exiting a publicly funded institution or system of care; OR


19 h) Otherwise lives in housing that has characteristics associated with instability and an increased risk of homelessness (U S Department of Housing and Urban Development, pg 1, 2012) In addition to the technical definition given above, further research outlines specific demographics as having an increased risk of homelessness. These demographics include African American indiv iduals, those between the ages of 50 64, the extremely low income, those with prior drug/alcohol abuses or imprisonment, the disabled, and the victimized (Cohen, C., n.p., 1999). When combining the technical definition with those demographics more at risk of homelessness, the actual number of people in the United States that are at risk of becoming homeless is extremely high. What is most important to note, however, is that the majority of those at risk and those that are homeless are not homeless by choi ce; a stereotype thought to be true by their non homeless counterparts. The statistics in section 2.1 show that the homeless population is mostly composed of veterans, disabled, and the working homeless. Section 2.2 further outlines that those at risk of homelessness fall within similar demographic groups. Much of the research on these populations point to a lack of federal and local assistance programs as the main reason these populations fall prey to homelessness. A report by Phil Stewart of Reuters fo u nd that approximately 45% of the 1.6 million veterans from the wars in Iraq and Afghanistan are ineligible for compensation and benefits and that over half of the veterans who have requested mental health evaluations have not received them within two week s of initial contact (Kelley, M. B., n.p., 2012). Further the unemployment rate for veterans was 12.1% in 2011; a staggering 3.2 % increase over the national unemployment rate (Kelley, M. B., n.p.,


20 2012). The National Coalition for Homeless Veterans stat es that in order to decrease homelessness amongst veterans, there much be a coordinated effort between the public and private spheres to provide housing, meals, health care, abuse care and aftercare, mental health counseling, job training, and transportati on (National Coalition for Homeless Veterans, n.p., n.d.). When it comes to those with physical and mental disabilities, The Center for earnings, barriers to education a nd skills development, significant additional expenses, S., & Ekman, L., pg 1, 2015). In addition, th e report found that 61.2% of disabled adul ts had incomes below 200% o f the federal poverty line as compared to 28.8% of their nondisabled counterparts (Vallas, R., et al., pg 3, 2015). Although progress has been made in removing the barriers faced by the disabled, much work remains. The Center for American Progress report s that an increase in public policies directed towards the disabled could decrease the barriers faced by the disabled and ultimately decrease the number of disabled homeless. The barriers that increase the chance of homelessness for the disabled include: employer reluctance, additional living costs, transportation difficulties, insufficient affordable and accessible housing, and lack of access to necessary support and services (Vallas, R., et al., pg 3 6, 2015). As mentioned previously, wages have not inc reased proportionately with the cost been equally distributed; instead, they have been concentrated at the top of the income


21 National Coali tion for the Homeless n.p., 2009). Both have led to an increase in the population known as the working poor. On the other end of the spectrum, are those who become homeless due to losing their jobs. The last ten years have been characterized by a high unemployment rate as a result of the recession. Although the government provides compensation for those who become unemployed, the amount of money disbursed does not provide the necessary funds for an individual or family to survive. In addition, unempl oyment is only disbursed for a brief period; not always allowing an individual the necessary time needed to apply, interview, and accept a new job. As a result, many unemployed individuals become homeless. Both the working homeless and the unemployed hom eless face similar barriers to escaping homelessness. Limited transportation, reduced access to education and training programs, and a competitive environment for employment are only a few of the barriers ( National Coalition for the Homeless n.p., 2009). While the government has invested significant amounts of money in not necessarily end homelessness ( National Coalition for the Homeless n.p., 2009). Policies focused on hou sing, transportation, health care, and wage increases are necessary in decreasing the number of homeless individuals. There are many barriers that homeless individuals face; many of which are mentioned above. Some of the most common barriers faced by the homeless include the excessively high cost of housing, inability to find stable employment, and transportation disparities. Although these are some of the most common barriers associated with homelessness, they have not all been addressed equally. In rec ent


22 years, many policies directed towards education/employment and housing for the homeless have been passed through Congress as to remedy the homeless crisis. However, little time has been spent on increasing mobility for the homeless. Dedicating policie s and programs to increasing mobility for the homeless has proven to decrease homelessness while increasing the quality of life. In 2012, California State University conducted a mobility study interviewing the homeless population of Long Beach, California. The study fou nd that approximately 13% of the homeless population owned motor vehicl es; whereas the other 87% relied on walking or public transit for mobility (Jocoy, C., Del Casino, V., Jr., pg 32, 2012). Further analysis foun d that approximately 94% o f the homeless population interviewed crucial to the ability of homeless people to move between stigmatized and non coy, C., et al., pg 2, 2012). In addition, a study conducted by R. David Parker and Shana Dykema found that although the homeless are stereotypically considered to be a mobile population, in reality they were less transient than the rest of the population (Parker, R. D., & Dykema, S., pg 1, 2013). The study revolving around Long Beach, California outlines how the homeless travel while the Study by Parker and Dykema shows how much the homeless travel. Due to the lack of public transit in America, it is no surprise that those areas with public transit have higher homeless populations and vice versa. However, public transit is designed for a different population; resulting in a disconnect between where public transit goes and where the homeless need to g o.


23 Transportation, about 6 million Americans with disabilities report trouble accessing the transportation they need, and more than 500,000 report never leaving home because of transportati 5, 2015). Another study by the National Coalition for the Homel ess found that nearly 30% of homeless interviewed stated that lack of transportation was a significant barrier to finding work ( National Coalition for th e Homeless n.p., 2009). Additionally, the study by Parker and Dykema found that the homeless often seek emergency services for conditions that could be addressed through outpatient care, simply because they do not have access to the transportation necess ary for reaching a medical facility (Parker, R. D., & Dykema, S., pg 1, 2013). Studies and personal interviews show that the homeless are traveling primarily to medical facilities, education and work facilities, multi purpose stores, and public facilities. However, accessing the transportation to reach these facilities is often difficult. For example, the study area of Gainesville, Florida has three bus routes that travel to the largest homeless shelter in the county, however the buses only travel at cert ain hours of the day and very seldom on the weekend; significantly decreasing the number of places a homeless individual can apply for work. In addition, the bus stop is located a half mile walk from the shelter; proving difficult for the physically disab led and elderly to access. Once getting on the bus, homeless often have to transfer multiple times to actually reach the facility they are trying to access. This real world example shows the transportation disparities felt locally but also outlines the ne ed for increased transportation and transportation related programs for the homeless in Gainesville, Florida. Although the shelter is working with many local


24 businesses and health facilities to bring necessary amenities to the homeless, this is only a tem porary remedy and does not address the problem of mobility. If anything, bringing services to the homeless decreases mobility and increases a homeless decrease the number of homeless individuals, it is imperative to provide adequate transportation services. 2.3 Transportation as a Civil Right According to Encyclopedia Britannica, civil rights guarantee equal social opportunities and protections under the law, regardless o f race, religion, or other personal characteristics (Hamlin, R., n.p., 2016). As mentioned previously, the Fourteenth Amendment and the Universal Declaration of Human Rights recognizes the right to travel as a civil right. Given, the importance of transp policymakers, advocates and users battle over transportation policy and its transportation has turned to arguments over funding and constru ction; forgetting why transportation is essential to begin with. Lexer Quamie, a Counsel for the Leadership Conference on Civil and Human Rights, states that the choices made with respect to transportation policy impact the economy, our health, and the cli mate enormously (Quamie, L., pg 59, 2011). However, these decisions rarely benefit all populations equally; especially those who are low income or those who rely most heavily on public transit. The struggle to end what is often referred to as transportat Ferguson (Quamie, L., pg 59, 2011). Approximately half a century after Plessy v.


25 Ferguson, bus boycotts (such as the Montgomery Bus Boycott, the Rosa Parks Boycott, and the Freedom Rides Boycott), re ignited the discussion of transportation as a civil right. Nearly 50 years later, transportation remains a civil rights issue. The majority of transportation funding has been dedicated to road development resulting in a car dependent infrastructure. This car dependent infrastructure has led to lower quality of life aspects for those dependent on public transit, the low income, the disabled, seniors, and the homeless. An example is a decrease in health care access that poses severe health hazards. Inadequate access to transportation often forces those with limited transportation options to miss doctor appointments, often worsening medical problems ( The Leadership Confere nce Education Fund pg 2, 2011). To avoid missing appointments, those with limited access to transportation may be forced to pay for expensive transportation services such as taxis or other car services such as Uber nsportation policy needs to shift a portion of investment away from new highway construction towards expanding public transportation and building bicycle and pedestrian friendly roads to promote greater parity in health care access, as well as to decrease health hazards, such as pollution Further, a report conducted by the Leadership Conference Education Fund, paucity of pedestrian and bicycle accessible thoroughfares have isolated urban and low The Leadership Conference Education Fund pg 2, 2011). In order to increase access to transportation, it is essential to have a job. However, if som


26 because of limited transportation options, it is unlikely they will be able to pull themselves out of this situation. Additionally, those with disabilities and senior citizens often lack t he option to many individuals with disabilities have increased health care needs -such as physical therapy, medication monitoring, and other medical services -isolati on from providers The Leadership Conference Education Fund pg 2, 2011). In conclusion, providing access to transportation is essential in connecting the poor, disabled, seniors, and hom eless to jobs, education, health care, and other resources necessary to survival. Many Americans take transportation for granted, but millions of people struggle to access viable transportation every day. The last fifty years have led to where we are tod ay -fighting for the right to equal transportation opportunities, not just for those with money, but for everyone. Wade Henderson, the playing field, but our transport ation choices have effectively barred millions of The Leadership Conference Education Fund n.p., 2011). It is beca use of these unprotected people that society must fight for the right to transportation for all.


27 CHAPTER 3 STUDY AREA AND DATA COLLECTION 3.1 Study Area 3.1.1 Selection Process The study area contains the city of Gainesville Florida (Figure 3 1). Gainesville by the National Coalition for the Homeless twice ( National Coalition for the Homeless n.p, 2004). Gainesville was selected for the list in 2004 because of its harsh criminalization of the homeless and again in 2009 for restricting the amount of meals soup kitchens could offer daily (Cunningham, R., n.p., 2010). An interview with Jon DeCarmine, the Operations Director for GRACE Marketplace, reveal ed that although the City of Gainesville was determined to change this ranking, the homeless continue to struggle with receiving basic necessities such as housing, medical assistance, and transportation (J. DeCarmine, personal communication May 31, 2017). As part of the City of Gainesville/Alachua County 10 Year Plan to End Homelessness, GRACE Marketplace opened its doors in 2014 to provide shelter and services for single homeless individuals. The opening of GRACE was monumental for the homeless populati on within Gainesville and the surrounding areas; taking in the homeless that were often turned away for drug addictions, unemployment, and criminal backgrounds. Although GRACE was deemed as a win for the homeless, little knew what consequences would arise from this project. In 1988, the Miami Chapter of the American Civil Liberties Union brought litigations against the City of Miami and police over the treatment of a homeless person (The Pottinger Agreement and Your Rights, n.p., n.d.). Nearly a decade la ter, the


28 Pottinger Agreement was established to protect the rights of homeless individuals (The Pottinger Agreement and Your Rights, n.p., n.d.). The Pottinger v. City of Miami case ruling is as follows: ls for performing inoffensive conduct in public when they have no place to go is cruel and unusual in violation of the eighth amendment, is overboard to the extent that it reaches innocent acts in violation of the due process clause of the fourteenth amend ment and infringes on the fundamental right to travel in violation of the equal protection clause of the fourteenth amendment. (Michael Pottinger, Peter Carter, Berry Young, et al. v. City of Miami, Pg 40, 1992). In accordance with this Supreme Court decision and the Pottinger Agreement, the development of GRACE Marketplace allowed the City to legally move the homeless from their residences throughout the city (w hether it be a tent city or park bench) to GRACE Marketplace. Within two months of GRACE Marketplace opening, the City set forth on closing tent cities within city limits, including Gainesville Tent City; the largest tent city at the time. Since then, th e City has continued in their pursuit to move the homeless to the outskirts of Gainesville where less services, transportation options, and housing exists. Further, interviews with both Jon DeCarmine and Tom Tonkavich, the Assistant Director of Community S upport Services for the Alachua County Health Department, revealed that although GRACE is essential in providing services and housing for the homeless, its location is less than ideal. GRACE is located on the eastern side of Gainesville and can be accesse d by the bus routes 24A and 25. Figure 3 2 shows an


29 aerial view of GRACE Marketplace. As Figure 3 2 shows, the bus stops located closest to GRACE Marketplace are located within at most a mile walk from the site. The bus stop denoted by a blue mark trave ls northwest into the City and can be accessed via GRACE Marketplace using two different pedestrian routes. The first route, as denoted in dark blue, is the safest pedestrian route which uses a crosswalk located at the intersection of 39 th Avenue and Wald o Road. This route takes significantly longer time given the light cycle and extended walking distance, but is the safest route. The second route, denoted in light blue, is the fastest and most efficient. This pedestrian routes travels differently acros s 39 th Avenue; avoiding proper crossing techniques. There is no crosswalk at this point so the individual must cross at their own risk. The purple bus stop travels southeast out of the city. This bus stop is located on the same side of the road and is t he quickest/easiest to get to. Despite the multiple pedestrian routes that can be taken to reach the bus stops, these routes pose a challenge for those homeless that are disabled and elderly; especially the bright blue route which crosses 39 th Avenue with out a crosswalk. As Figure 3 2 shows, GRACE Marketplace is located on a dead end street. Both Jon DeCarmine and Tom Tonkavich discussed the idea of placing a bus stop at the dead end (as denoted as the green mark on Figure 3 2, but talks with the Gainesv ille Regional Transit Service (RTS) have been futile in that RTS deems the dead Because of this, local businesses have partnered with GRACE Marketplace to come out on cer tain days of the week to provide services such as medical care. Despite the action taken by GRACE Marketplace, the Alachua County Coalition for the Homeless,


30 the City of Gainesville, and others around town, it is apparent that services, including transpor tation, for the homeless population continue to be an issue. 3.1.2 Demographic Data Gainesville is located in the North Central portion of the state; housed within Alachua County. Gainesville is the largest city in North Central Florida with an estimated population of 131,591 people as of 2016 (US Census Bureau, n.p., n.d.). Further, Gainesville is contained within the Gainesville, Florida Metropolitan Statistical Area; attracting work from all around the North Central Florida region. The median age of t he City is much less than the state average at 25.3 years, respectively (US Census Bureau, n.p., n.d.). Gainesville also has a median income lower than the county and state average totaling $31,818 (US Census Bureau, n.p., n.d.). The low median age and i ncome of Gainesville can be attributed to its designation as a college town; housing the University of Florida and Santa Fe College. Because student incomes are lower than their working aged peers, the poverty level within Gainesville proper is sign ifican tly inflated at 35% respectively. As to account for the college population, the county has conducted individual studies findi ng that approximately 15% of the population was living in poverty as of 2013. ( Alachua County n.p., 2013). Traditionally, socie ty associates poverty with homelessness. However, it is important to mention that a person can be living in poverty without being homeless and someone can be homeless without being in poverty. This study focuses not so much on poverty as it does homeless ness but mentions poverty as being an indicator of homeless percentages. For example, as poverty increases, traditionally, so will homelessness. To monitor homelessness, Alachua County conducts a yearly Point in Time Survey.


31 The intent of the Point in Ti me Survey (PIT) is to provide a snapshot of a Housing and Urban Development for communities across the country to receive e, n.p., n.d.). The most recent PIT Survey for Alachua County was conducted in April 2017. The PIT Survey revealed that there was a total of 702 homeless individuals living in Alachua County in 2017 (City of Gainesville, pg 15, 2017). Of those 702 indiv iduals, 409 were unsheltered living in places such as tents, cars, and storage units (City of Gainesville, pg 15, 2017). Another 293 were living in shelters such as GRACE Marketplace, Family Promise, Peaceful Paths, and VetSpace (City of Gainesville, pg 1 5, 2017). Further, 252 of those individuals were deemed as being chronically homeless (City of Gainesville, pg 15, 2017). Alachua County defines chronically homeless as being homeless for one or more years or having a disabling condition (physical or beh avioral). Like much of the 702 homeless individuals, 58% are aged 50 plus, 22% are aged 60 plus, and 3% are aged 70 plus (City of Gainesville, pg 15, 2017). Despite the substantial number of homeless individuals living in Alachua County, the number of homeless individua ls has decreased by 15.4% between 2016 and 2017 (City of Gainesville, pg 16, 2017). The decrease in homelessness is attributed to the county focusing thei r efforts in programs such as Rapid Re Housing (RRH) and Permanent Supportive Housing (PSH). Both programs are based on the theory that providing housing to the homeless is the most effective way to move out and stay out of homelessness. These programs h ave


32 resulted in a 19.6% decrease in unsheltered homeless between 2016 and 2017 (City of Gainesville, pg 16, 2017). In addition to the PIT Survey, both Dignity Village and GRACE Marketplace report on the homeless they serve. Dignity Village, a city run te nt city located on the east side of Gainesville, reported a total of 195 homeless individuals in April 2017 (City of Gainesville, pg 5, 2017). The number of individuals at Dignity Village has stayed constant over the last two years reporting a decrease of nine individuals between May of 2015 and April of 2017. GRACE Marketplace provided the county with a Monthly Services Report for the month of March 2017. In the report, a total of 21,398 services were provided to 940 individuals (City of Gainesville, pg 8, 2017). The most requested service was a meal; attributing to 54% of all services. Of those receiving services fr om GRACE Marketplace, 64% were males and 39% were between the ages of 31 50 followed by 30% between the ages of 51 61 (City of Gainesville pg 8, 2017). Of the 940 individuals receiving services, 521 fell in to the special populations cat egory. Approximately 19% were chronically homeless, 20% were veterans, and 61% were disabled (City of Gainesville, pg 8, 2017). Although the PIT Survey r eveals that the total number of homeless individuals in Alachua County has decreased, there are still 702 individuals in need of help and services. Providing free or reduced cost housing to the homeless is essential in ending homelessness; however, it is only one portion of the fight. While speaking with Jon DeCarmine, a prior GRACE Marketplace resident politely interrupted the conversation. Mr. DeCarmine introduced this individual stating that despite finding housing he was forced to spend two hours eac h day getting to work; having to transfer buses three


33 times or walk long distances as to not miss transfers due to buses running late. This interaction proved that providing housing is essential to eliminating homelessness, but unless the homeless have tr ansportation to and from the housing, the attempt to move the homeless out of homelessness will be fruitless. 3.2 Data Collection and Preprocessing All data was collected using interviews and questionnaires. Before starting this study, the appropriate pap erwork was filed to allow for interviews and questionnaires to be conducted. The intent of the interview was to talk with members of the community who had an extensive knowledge of homelessness in Gainesville. The people selected for interviewing include d those running shelters and health service facilities for the homeless. Some of the locations selected for interviews include GRACE Marketplace, Helping Hands Clinic, and the Veterans Medical Center. Further, other local service administrators that deal with homeless regulation or transportation were contacted. These organizations include the Gainesville Police Department, RTS, and the City of Gainesville Table 3 1 provides a list of organizations contacted for interviews. During the interviews, a gen eralized set of questions were asked having to do with the homeless and their travel habits. The information from these interviews was to be used as the research portion of this study. However, upon conducting interviews, the opportunity to collect trave l data from the homeless themselves arose. A simple questionnaire was drawn up asking the homeless the locations in which they traveled over a one week period and how they got to those locations. Appendices 3 1 and 3 2 show both the interview questions a nd the homeless questionnaire. Answering the questionnaires was strictly voluntary. To increase participation, gift bags filled with


34 hygienic items such as soap and socks, were given to the first 100 participant. No children were asked to participate in the questionnaire. Once the questionnaires were complete, the data was compiled in a spreadsheet. The interviews were recorded and then transcribed into a text document for later use. The questionnaire data was used for quantitative analysis purposes wh ile the interview data was used for qualitative purposes. After collecting the transportation data, a road network dataset was developed using Network Analyst in ArcGIS. A 2017 Alachua County roads shapefile was used to create the road network dataset. T he shapefile contained speed limit information; allowing for the development of a travel time algorithm embedded within the road network. Additionally, a bus route shapefile was obtained from Gainesville Regional Transit Bus System (RTS). The shapefile c ontained all bus routes traveling from GRACE Marketplace and the various locations in which the homeless traveled; as collected in the homeless travel questionnaire. Important attribute data from the roads shapefile was combined with the bus route shapefi le in order to calculate travel time for future analysis purposes.


35 Table 3 1. Organizations Contacted for Interviews GRACE Marketplace Family Promise of Gainesville, Inc St Francis House Gainesville Housing Authority City of Gainesville North Centra l Florida Alliance for the Homeless and Hungry Shimberg Center RTS CDS Family & Behavioral Health Services Alachua County Coalition for the Homeless and Hungry Career Source North Central Florida Gainesville Police Department Helping Hands Clinic V olunteers of America The Salvation Army Veterans Industries Gainesville Goodwill Industries Job Junction Equal Access Clinic Alachua County Health Department NF/SG Malcolm Randall VA Medical Center


36 Figure 3 1. Study Area


37 Figure 3 2. GRACE Ma rketplace Boundary & Bus Routes


38 CHAPTER 4 METHODOLOGY The workflow of this project was divided into four questions (Figure 4 1 ). The first and second research questions were addressed using interviews and questionnaires. The third and fourth questions Analyst extension of ArcGIS. Details of the methodological design are described below. 4.1 Travel Data Collection L ittle data has been collected on homeless travel. Not only is it difficult to find the homeless (given their transient lifestyle), but many homeless prefe r to live their lives in solitude History shows that the homeless have consistently been discriminated against by the government, other non homeless individuals, and police. It is because this, that man y of the homeless fear divulging information about themselves including their residence, travel patterns, and any personal information. Additionally, many of the homeless are disabled and may not have the social, mental, or physical skills to communicate properly with those conducting research. La stly, the Institutional Review B oard, a committee designated by the US Government to monitor, review, and approve increasing the diff iculty in studying the homeless. A study conducted by the Department of Geography at California State University provides the most comprehensive research on how the homeless travel. This study focused on the mobility of the homeless around Long Beach, Cal ifornia using public transit This study fou nd that approximately 87% of the homeless in Long Beach relied on walking or public transit for mobility ( Jocoy, C., et al., pg 2, 2012 ). In addition to tracking the type of transit, this study tracked where th e homeless traveled. Of the most


3 9 pop ular places traveled, 24% of trips were to hospital/med ical service centers, 22% of trips were to so cial service centers, and 14% of trips were for job interviews or looking for work ( Jocoy, C., et al., pg 33, 2012). G iven its prominence as the leading study in homeless mobility, the Long Beach study will be used as the basis for the methodology of this study. 4.1.1 Interviews To understand the relationship between the homeless, government, and businesses in Gainesvill e, interviews were conducted. Additionally, the interviews were used to collect information on travel patterns of the homeless, such as where the homeless were traveling and how they were traveling. Table 3 1 provides a list of organizations contacted fo r interviews and Appendix 3 1 shows the list of questions asked. Notes were taken during each interview and upon approval, each interview was recorded. The recorded interviews were transcribed and used as supplemental information for this study. 4.1.2 Ho meless Questionnaire Initially, the interviews were going to be used in place of communication with the homeless. However, an interview with the Operations Director of GRACE Marketplace allowed this study to further enhance the data collection by providi ng questionnaires on travel patterns to the homeless. On September 20 th 2017, Jon DeCarmine disbursed homeless travel questionnaires during breakfast to the homeless. To guarantee receiving responses, gift bags filled with hygienic items were offered to those who completed the questionnaire. The homeless were given one week to complete these questionnaires and turn them in to the administrative office of GRACE Marketplace. All


40 questionnaires collected after the one week period were added to the travel data Jon DeCarmine collected, but were not used in the study. At the end of the week DeCarmine releas ed the travel data Appendix 3 2 provides a copy of the homeless travel questionnaire. Once the questionnaires were released, each questionnaire was co mpiled to create a spreadsheet detailing each travel location and method. The homeless questionnaire data was used for the analysis section of this study. 4.2 Network Analysis to compare the travel time using the two most common modes of travel in Gainesville: driving and bus. edges (lines) and connecting junctions (points), that represent possible r outes from one ESRI, n p n d .). The ArcGIS Network Analyst extension makes it possible to model potential travel paths along diverse types of networks. Esri has categorized networks into two categories: geometric networks and netw ork datasets. Examples of geometric networks include river and utility networks. Geometric networks are usually influenced by external forces; rather, they are unable to choose which direction to travel. Transportation networks such as streets, sidewalk s, and railroads are examples of network datasets. Compared to geometric networks, the agent using a ESRI, n p n d .). This study will only use network datasets and will thus be explained in detail below.


41 Network datasets must be created using the Network Dataset wizard located within ArcGIS. The Network Dataset Wizard uses a shapefile or geodatabase dataset that is composed of an edge source and an opt ional turn source. This study uses a 2017 Alachua County roads shapefile taken from the Alachua County Property Appraiser (ACPA) along with a shapefile of RTS bus routes taken from Gainesville RTS. The bus routes shapefile was combined with the roads sha pefile to provide it with attribute data such as road speed and direction. When creating the network dataset for both the roads and bus shapefiles, a drive time (DT) attribute was added to the network. This attribute was created using Equation 4 1 locate d below: (4 1) Where road length is in miles (m) and speed limit is given in miles per hour (MPH). The travel time is then converted to minutes as minutes better represent travel time within a small city such as Gainesville. The streets network dataset can be seen in Figure 4 2 The bus network dataset can be seen in Figure 4 3 respectively. Once the network datasets are created, the Network Analyst extension can be used to calcula minimum cumulative impedance between nodes on a network. The path may connect ( Rajput, S., pg 8, 20 14 nodes so that after exactly n iterations, the shortest paths are foun d (Moser, T., pg 63, 1991). Equat ion 4


42 (4 2) (I) Initialization Q : = N tt(s) : = 0 (II) Selection Find I (III) Updating Tt(j) : min {tt(j) + d ij } for all j Transfer I from Q to P (IV) Iteration check If P = N stop Else go to 2. (Moser, T., pg 63, 1991) P of nodes with known traveltimes and a set Q of nodes with not yet known traveltimes along shortest path from s Initially, P is empty a nd Q = N The minimum traveltime node of Q is s It has a known traveltime [tt(s) = 0] so it can be transferred to P The traveltime of all nodes connected with s all j FS(s) are then updated in agreement with the equation The node in Q with the smallest tentative traveltime will not be updated any more. It can therefore be transferred to P and the nodes from Q co nnected with it are again updated. This process of finding the minimum tentative traveltime node, transferring it to P and updating its forward star is repeated exactly n (Moser, T., pg 63, 199 1). Although this algorithm appears to be difficult, it is quite easy to calculate by is most often computed using scripting language. For this research, the algorithm is built into the software; manually computing the shortest path with the click of a button. As mentioned previously, this study equates travel time with the shortest path. However, calculating shortest path is different for each mode of transportation. For example, finding the shortest path for the roads network dataset is different than calculating the shortest path for the bus network dataset because the bus network


43 dataset contains a set of pre designed routes as well as additional travel time aspec ts. The additional travel time aspects include both walking and transferring. The walking aspect considers the time it takes to walk from the origin location (GRACE Marketplace) to the bus stop, the final bus stop to the destination, and the distance bet ween transfer bus stops if applicable. The transfer aspect focuses on the time the individual must wait before getting on a second or third bus. This study denotes the travel time from the location of origin to the bus stop as PT O, the travel time from t he bus stop to the destination as PT D and the travel time from bus stop to bus stop as PT B The equation for both PT O PT D and PT B is explained as Equation 4 3 below. (4 3) Where 60 is the number of minutes in an hour, distance is in miles (m) and 3.1 is the average walking pace in miles per hour. PT O, PT D and PT B all follow the same equation, the only difference is that the distance will change between the two. The transfer aspect was calculated using the Departure Board function on Google Maps. The Departure Board function shows the arrival time of each bus at a stop for that day. For this study, it was assumed that the arrival time was the same as the d eparture time. In order to follow consistency, all the arrival times were collected on the same time and day. The arrival times at bus stop 1 was subtracted from the departure times of bus stop 2 and then divided by each pair as to calculate an estimated transfer time. For example, many individuals took bus route 26 to the Rosa Parks Transfer Station and then transferred to another bus at that station. Bus 26 arrived at Rosa Parks at 11:00 a m 12:00 p m 1:00 p m 2:00 p m and 3:00 p m . Suppose the individual was transferring to a bus that arrived at Rosa Parks Transfer Station at


44 11:30 a m 12:30 p m 1:30 p m 2:30 p m and 3:30 p m The arrival times of bus 26 were subtracted from the arrival times of the transferring bus (11:30 11:00 = 3 0, 12:30 12:00 = 30, etc.) Those times were then added up and divided by the total number of paired arrival and departure times ((30+30+30+30+30)/5 = 30 minutes average transfer time). If the individual getting off at the stop had to travel from one bus stop to another, the time it took to travel between the bus stops was subtracted from the average transfer time. This part of the equation was added in to remove overcounting of time. The formula for transfer travel time i s denoted as BT and is shown bel ow as Equation 4 4. (4 4) Where AT is equal to arrival time at the first bus stop, DT is the departure time of the transfer stop, P is the number of paired transfer times, and PT is the pedestrian walking time calculated above. A s the formula for calculating travel time (TT) differs between the streets network and the bus network, for clarification purposes, both the street and bus travel time formulas are g iven below as Equations 4 5, 4 6, and 4 7 respectively. (4 5) (4 6) (4 7) Where travel time for a personal car only includes the time in which it takes to drive from the location of origin to the destination and travel time for bus includes the drive time (along a specific route), the pedestrian walking time, and the transfer time.


45 incidents. In this instance, facility is equivalent to an origin location and incident is equivalent to the destinat ion location. An origin is the location in which a trip begins. A destination is the location in which a trip ends. For this research, the origin location is always GRACE Marketplace. The destination locations are all the locations the homeless travele d to from the origin locations. A list of destination locations can be found in Table 4 1 Both the origin and destination locations were collected using the homeless travel survey mentioned previously. After inputting facility and incidents in Network Analyst, specific criteria can be set. For this analysis, the settings have been changed to travel from the facility (origin) to the incident (destination), U turns have been allowed for the streets network dataset but not for the bus network dataset, th e impedance has been set to travel time, the distance units have been set to miles, and the finding network locations search tolerance has been set to 3.11 miles. Once the network dataset has been created, Network Analyst shortest path can be run. An out put shapefile is created modeling each of the shortest path routes and the time in which it takes to get from the origin location to the destination location (drive time). The travel time was then calculated for both modes of travel using the formulas abo ve. The travel times were compared and the difference was calculated. After completing the shortest path analysis, the data was evaluated for potential improvements and route changes. Any potential changes and reasons fo r changes will be discussed in C h apter 5.


46 Table 4 1. Travel Destinations as Determined by the Homeless Questionnaire Number of times recorded as a destination Destination 1 Harry's Restaurant 1 16th Avenue 1 First United Methodist Church 1 2nd and Charles 1 Alachua County Airport 1 Alachua County Courthouse 1 Alachua County Jail 1 Alachua County Tax Collector off University 1 Aldi at Butler Plaza 1 Apple Market 1 Arredondo Estates 1 Bank of America on 13th Street 1 Big Lots 1 Bly's School of Cosmetology 1 BP Gas Station on University Avenue 1 Checkers on University Avenue 1 Children's Home Society 1 Church 1 Circle K 1 Community Overdrive 1 Davis Chevrolet 1 Dollar Tree 1 Domino's 1 Epilepsy Foundation of Florida 1 Family Dollar 1 Florida Recovery Center 1 Gas Station on Waldo Road 1 Goodwill 1 GRU 1 Kangaroo 1 Labor Pool 1 Library Partnership 1 Little Caesars 1 Pawn Shop on 6th Street 1 People Ready 1 Salvation Army on University Avenue 1 Sonny's Waldo Road 1 SS UF 1 Sunoco Gas Station on 39th Aven ue 1 Sunoco Gas Station on Waldo Road 1 SW 13th Street


47 Table 4 1 Continued Number of times recorded as a destination Destination 1 Taco Bell on 13th Street 1 VA Hospital 1 Walgreens 1 Wells Fargo on University Avenue 1 Work 2 Citgo Gas Station 15 th Street & 39th Avenue 2 North Florida Regional Hospital 2 Plasma Donation on 6th Street 2 Rural King on 13th Street 2 Save A Lot 2 Shands Medical Plaza 2 St Francis House 2 Store 2 Triangle Club 3 Dollar General on 39th Avenue 3 Food Stamp Offi ce 3 Health Department 3 Humane Society 3 Job Interview 3 Publix on Main Street 3 Salvation Army on 23rd Avenue 3 Thrift Store 3 UF 4 Helping Hands 4 Shands Hospital 4 SS Office 4 Taco Bell on University Avenue 4 Walmart at Butler Plaza 5 Butl er Plaza 5 Career Source 5 Meridian on 13th Street 5 UF Football stadium 6 Unknown 7 Mall on Newberry Road 8 Downtown Gainesville 8 Library on University Avenue 10 Rosa Parks 25 Walmart on Waldo Road 196 TOTAL


48 Figure 4 1. Methodological Desi gn


49 Figure 4 2. Alachua County Roads Network Dataset


50 Figure 4 3. City of Gainesville Bus Network Dataset


51 CHAPTER 5 RESULTS AND DISCUSSION 5.1 Results 5 1.1 Questionnaire Statistics The results from the homeless questionnaire provided an abundant amou nt of demographic statistics on the homeless population surveyed. In total, 44 individuals completed the questionnaire. Of the 44 respondents, 29 were male, 13 female, and 2 chose not to identify. Compared to the demographic composition of the homeless population around the country, the percentage of males who responded to this survey was slig htly below average at 66% respectively. In general, the composition of homeless males to females in America is approximated between 67% and 75% Of the 44 respon dents, approximately 16% fell within th e age group of 18 30, 18% wer e 31 40 years of age, 34% belonged to the age cohort of 41 50, 14% within the 51 60 cohort, and 18% fell within the 61 and older cohort. When looking at these numbers graphically, it is a pparent that the population of respondents was varied and fell along a bell curve. The age statistics of the respondents can be seen in Figure 5 1 Much like the rest of the nation, Gainesville is experiencing an increase in the homeless elderly populati on. However, the statistics collected from this analysis reflect a population that is somewhat homogenous. Some of the most interesting information collected from this study show that an overwhelming majority of the homeless population surveyed use the b us as their primary mode of t ravel. Approximately 87% of those surveyed used the bus to get around ; followed by walking (6%), biking (4%), and car (2% ). In total 196 trips were recorded with each homeless individual averaging four trips a week. These


52 pe rcentages are similar to those collected in Long Be ach, California where 87% of the trips made by the homeless were by bus or bike. A visual representation of these statistics can be found in Figure 5 1. Interviews with those knowledgeable about the ho meless in Gainesville resulted homeless questionnaire was conducted that actual destinations of the homeless were discovered. Of the 196 trips recorded, 25 were made to the Super Walmart off of Waldo Road. Additionally, 10 trips were taken to the Rosa Parks Transfer Station, 8 to the Alachua County Library on University Avenue, 5 to Meridian Behavioral and Psychiatric Care Center and Car eer Source, 4 to the University of Florida Shands Hospital and Helping Hands Clinic, and various other trips to thrift stores, shelters, government offices, gas stations, and fast food restaurants. A detailed list of locations visited by the homeless can be found in Table 4 1 of the pr evious chapter. 5.1.2 Network Analyst Analysis The data from the Network Analyst results were divided into three scenarios: travel time via car, travel time via bus with transfer time not included, and travel time via bus with transfer time included. B ecause transfer time can significantly vary due to factors such as driving speed, road congestion, and accidents, two travel times for bus have been provided. The data was compiled in table format showing the travel times for each of the three scenarios. This table can be found below in Table 5 1, respectively. In the table, some destinations have been listed twice. This is because questionnaire respondents cited multiple routes for one location. The differentiation in routes can be found at the end of each destination located in the destination column of Table 5 1.


53 Comparing the three scenarios tells a compelling story. The first scenario, or travel time by car, had the lowest travel time of all three scenarios. The shortest travel time is a mere 55 seconds while the longest travel time is approximately 17.5 minutes. On average, the travel time by car was 6 minutes and 24 seconds. Although six and a half minutes on average is a relatively quick travel time, the majority of questionnaire responden ts revealed that travel by car was uncommon. Travel time by bus without transfer time shows an increase in travel time of approximately 8 minutes. The shortest travel time by bus without transfer was 3 minutes and 43 seconds while the longest travel tim e was approximately 28 minutes. The final scenario, or travel time by bus with transfers, resulted in the longest average travel time. Travel time by bus with transfer increased by an average of 19 minutes when comparing it with car and by approximately 11 minutes when comparing it to the travel time by bus without transferring. The shortest travel time by bus with transferring was 3 minutes and 43 seconds, while the longest travel time by bus with transfers approximated 68 minutes with two transfers. T his data can be seen in Table 5 1 below. For destinations located along the bus routes that service GRACE Marketplace, travel times did not differ drastically. An example of this is the trip made from GRACE Marketplace to the Alachua County Jail. The tot al travel time by car (without lights and traffic) was just shy of a minute. When taking the bus, the travel time was approximately 3.5 minutes. The increase in travel time between car and bus is a result of having to walk to the bus stop. When comparin g the drive time of the car to the drive time of the bus, the drive time for bus is less than car; 55 seconds by car and 34 seconds by bus, respectively. Because there is no transfer time associated with this


54 example, the total travel time by bus includin g transfers is the same as the travel time by bus without transfers. This example can be found in Figure 5 2 Trips where a bus transfer was required, significantly increased the travel time, as mentioned above. The travel questionnaire showed that when traveling to Career Source, located at the intersection of 6 th Street and University Avenue, the homeless took one of two routes. One route involved transferring at the Rosa Parks Transfer Station while the other route involved no transfers but instead a quarter mile walk from the bus stop to the destination. According to the analysis, the drive from GRACE Marketplace to Career Source is approximately six minutes by car. When taking bus route 25 with no transfer, the travel time increases to approximate ly 12 minutes. Comparatively, the route with a transfer increased the trip by 2 minutes (not accounting for transfer time) and by approximately 11 minutes when including transfer time. To conclude, when traveling to Career Source it takes approximately 6 minutes by car, 12 minutes by route 25 with no transfer, 14 minutes by route 26 with no transfer time included, and 23 minutes when including the approximated transfer time for route 26. 5.1 .3 Creation of a New Route The new route created in ArcGIS for t he homeless is within a half mile of 74% of the locations visited by the homeless. A list of the destinations within a half mile of the route can be found in Table 5 2 The new route, along with the stops located within a half mile of the route can be se en in Figure 5 3 The route is approximately 12 miles long and travels south on Waldo Road before turning onto 16 th Avenue where both the Library Partnership and Alachua County Food Stamp Office are located. The route then travels north on 6 th Street bef ore turning west onto 23 rd Avenue. Once the bus reaches 13 th Street, it then turns south. The intersection of 23 rd Avenue and 13 th Street is


55 important in which it encompasses many of the shops the homeless like to travel as well as banks, pharmacies, and the Epilepsy Foundation. The bus would then turn east onto University Avenue until reaching Main Street. Traveling on University Avenue and south Main Street allow the homeless to easily reach Helping Hands Clinic, the Library, the Alachua County Courth ouse, Career Source, People Ready/Labor Pool, and St. Francis House Homeless Shelter. The bus would continue its trek down Main Street until reaching Williston Road where it would then turn west and travel to the intersection of Williston Road and 13 th St reet. This intersection is crucial in that it allows the homeless easy access to Meridian Behavioral and Psychiatric care. The bus would then turn north onto 13 th Street until it reaches SW 16 th Avenue. The bus would then follow 16 th Avenue to Center Dr ive where it would then turn north until reaching SW Archer Road where it would conclude on Archer Road at the University of Florida Shands Hospital. 16 th Avenue and Archer Road are both very important locations where the homeless can then visit the Veter ans Hospital as well as all the practices located at and around Shands Hospital. 5.2 Discussion 5.2.1 Questionnaire Responses Going through the travel questionnaire was the most interesting part of the analysis. Everybody looks at travel differently and i t is difficult to provide one solid definitive definition for it. One respondent wrote every single city that they had visited over their lifetime. Although this was not part of the instructions, it was interesting that the respondent included it in her questionnaire. Another respondent rode his bike all the way from Sarasota County to Alachua County to avoid the hurricane that made landfall in Florida a week prior. Lastly, a respondent wrote that they had moved their


56 tent around GRACE Marketplace three times within the survey period. The uniqueness of the data showed just how difficult it is to track travel and mobility. Further, the responses really opened my eyes to how different classes of people may view travel. For me, moving a tent from one loc travel but for this individual it did. When thinking about it further, this person had to pick up their entire life and move it; something that could be equated with moving from one city to another. Al though the scale in which this person moved is less than the average relocation distance, for them it was still a movement and it still represented change. Additionally, the questionnaires were only passed out to those getting breakfast at GRACE Marketplac e. Although GRACE is the most popular location for homeless individuals in the Gainesville MSA, it is not the only spot in which the homeless go. If I were to continue with this study, I would like to expand to multiple locations throughout Gainesville t o collect homeless travel data. Gainesville has many locations in which the ho meless congregate including t he city run Tent City, the St. Francis House and Downtown/Bo Diddley Plaza. Collecting data at multiple locations would expand upon my research of the homeless as well as provide an abundant sample of data from all over Gainesville. Lastly, if conducting this study again, I would change the homeless questionnaire. When composing the questionnaire, John DeCarmine and I were tasked with creating a sho rt and easily comprehensible document. If given free reign of the questionnaire, I would have asked more demographic questions, added children to the survey population, and increased the number of questions on mobility. Increasing the amount of demograph ic questions provides the City of Gainesville with information on a


57 population that is seldom known about. Although Alachua County collects demographic data on the homeless for the Point In Time Survey, it is very common for the homeless to avoid respondi ng. Writing the questionnaire with the idea of helping the homeless instead of solely collecting data on them could potentially increase the number of respondents. Adding children to t he survey population would increase the number of respondents as well a s provide a type of data that has never been collected before. When starting my study, I intended to look at homeless child mobility, but due to protection restraints, I was unable to study children. However, I think it is important to acknowledge how ch ildren get to school, doctors, and participate in social activities. Simplifying the travel questions as well as increasing them would increase the amount of data collected on homeless mobility. Additionally, it could potentially provide data on exact tri p routes, times, and mode. I also think that simplifying the questions (I.E. providing columns for the respondents to give origin location, destination location, bus route, and time) would increase the data collected. It lengthens the survey overall but increases the amount of data collected. 5.2.2 Travel Statistics Comparing the interviews with the questionna ire data showed that there is a disconnect between those serving the homeless and the homeless themselves. As shown in Appendix 3 1, interviewees were asked where they believed the homeless to be traveling. Many responded with no knowledge while others pointed out typical locations such as the Alachua County Library, the University of Florida, and GRACE Marketplace. Although the homeless did trave l to these locations, these were only a small portion of the complete list of destinations. In addition to these locations, the


58 homeless travel to Career Source, government assistance offices, the Alachua County Humane Society, different job locations, ph armacies, clinics, fast food restaurants, and dollar stores. If the interviewees intend to help the homeless I think it would be help meet the travel needs of the home Perhaps the data from this can be used to start research on homeless travel in Gainesville. When looking at the mode statistics from the questionnaire, I was shocked at the percentage of homeless using the bus to travel. Consistently throughout the interview process interviewees would mention the need for subsidized bus passes for the homeless. Additionally, based on my own observations of the homeless in Gainesville over the last seven years, I was under the impression that the homeless were mostly traveling by foot or bike. However, an o verwhelming majority (87% ) used the bus for travel. This statistic left me wanting more information and data on just how many homeless individuals have bus passes and if su bsidizing more bus passes would increase the mobility of the homeless in Gainesville. If I were to conduct this study again, I would like to focus more attention on acquiring a bus pass as a homeless individual. 5.2.3 Network Analyst Analysis Looking at the comparison between car, bus without transfers and bus with transfers, I was not shocked by the results; these were the results I expected. Someone driving their own vehicle is traveling solely to their destination while a bus has multiple destination s. When taking that into account, it was somewhat interesting to see that the average increase in travel time was only 8 minutes for bus without transfers.


59 Given the modes in which an individual must use to get to the destination (bus and walking), it su rprised me that the difference was a mere 8 minutes. I assumed that walking would increase the overall travel time significantly; however, this was not the case. I was also surprised that the travel time for bus with transfers only increased by 11 second s. Given my experience with the bus system I was expecting the average transfer time to be higher than the calculated 16 minutes. I believe the low transfer time between buses for this study can be attributed to the number of people transferring at trans fer stations increasing the amount of options an individual has to travel and decreasing the amount of time waiting for a bus. 5.2.4 Creation of a New Route I believe there to be many issues with the routes servicing GRACE Marketplace. First, two of th e routes (route 25 and 26) make similar trips. On the weekend, route 25 extends its trip to the Rosa Parks Transfer Station; almost completely mimicking route 26. The reason route 25 changes on the weekend is because route 26 is not offered. I assume th e change in route is to provide access to a transfer station on the weekend. Second, there is only one bus that stops at GRACE Marketplace on the weekend; route 25. Both routes 26 and 39 halt services on the weekend; severely decreasing the locations the homeless can travel. Third, the routes are limited in the times and frequency in which they travel ( route 25 is in service from 7 a.m. to 5 p.m. and runs ever y 65 minutes, r oute 26 is in service from 6 a.m. to 8 p.m. and runs every 60 minutes, and route 39 is in service from 8 a.m. to 5 p.m. and runs every 60 minutes). On the weekend, service to and from GRACE is even more limited in time and frequency. The limitations in time and frequency often increase the travel time for the homeless resulting in an increase in the amount of time an individual must dedicate to


60 traveling via bus. Additionally, the lack of travel options limit the jobs which the homeless can take adding another limitation to the already arduous process of finding work. When creatin g a new route, I considered the three problems most linked to the bus routes servicing GRACE Marketplace. Additionally, I looked at the travel destinations from the questionnaire and created a route that serviced as many of those locations as possible. I wanted my route to make sense as well as take advantage of already existing routes and bus stops. Lastly, I wanted my bus stop to replace one of severely overlapping routes to increase efficiency and the number of locations visited. My first suggestion is to adopt the route 25 weekend schedule for weekday service. The weekend schedule overlaps greatly with the route 26 schedule and stops at both the Rosa Parks Transfer Station as well as the University of Florida. I would extend the hours of t his bus t o start service at 6 a.m. and end service at midnight. The bus would have a frequency of 30 minutes. The reason for extending the service and changing the frequency is to increase the hours in which an individual living at GRACE can work. Because route 39 is the least traveled route, I would leave it how it is. Additionally, route 39 travels from the east side of Gainesville to the west side; completely different from both routes 25 and 26. This route increases the amount of locations the homeless can travel as well as provides access to many of the locations in which the homeless are currently traveling. Lastly, I would change route 26 to reflect the route shown in Figure 5 3 As mentioned previously, this route is within a half mile of 74% of the de stinations recorded by the homeless questionnaire. This route was selected mainly for its proximity to the homeless questionnaire destinations but also


61 because it overlaps with existing routes and bus stops; decreasing the overall costs associated with im plementing a new bus route. I believe this route would best service the homeless population of GRACE Marketplace by running every 30 minutes on weekdays from 7 a.m. to midnight and e very hour on weekends from 10 a.m. to 5 p.m. The reason for these servi ce times are to provide access to important clinics and governmental services such as Helping Hands and the Disabilities Office as well as provide access to the hospital primarily on the weekend when most clinics are closed.


62 Table 5 1. Travel Time Given in Minutes and Seconds Destination Travel Time by Car Travel Time by Bus without Transfer Travel Time by Bus with Transfer Dollar General on 39th 3:01 5:17 5:17 Salvation Army on 23rd 4:00 9:41 9:41 Humane Society Bus 39 4:03 12:40 12:40 Humane S ociety Bus 26 4:03 15:39 30:39 Downtown Gainesville Bus 26 5:00 8:47 8:47 Downtown Gainesville Bus 25 5:00 16:32 16:32 Health Department 5:01 16:55 46:55 GRU 5:03 13:57 13:57 Alachua County Tax Collector off University 5:06 8:35 8:35 Publix on M ain St 5:06 10:25 25:25 SS Office Bus 39 5:08 10:04 65:04 SS Office Bus 26 5:08 20:10 29:10 Career Source 6:01 11:42 11:42 Career Source 6:01 13:57 22:57 Walgreens 6:02 17:42 37:42 Plasma Donation on 6th St Bus 25 6:08 12:52 12:52 Plasma Donat ion on 6th St Bus 26 6:08 13:46 22:46 Shands Hospital Bus 1 8:06 26:23 34:23 Shands Hospital Bus 17 8:06 15:57 34:57 Mall on Newberry Bus 25 14:09 21:32 32:32 Mall on Newberry Bus 26 14:09 21:52 39:52 Alachua County Jail 0:55 4:03 4:03 Sonn y's Waldo Road 1:16 3:43 3:43 Citgo Gas Station 15th & 39th Ave 1:38 4:24 4:24 Aldi at Butler Plaza 12:12 23:09 31:09 Butler Plaza Bus 25 12:50 17:03 24:03 Butler Plaza Bus 26 12:50 21:52 29:52 North Florida Regional Hospital 13:50 23:04 65:04 Ar redondo Estates 15:29 24:42 68:42 Tower Road Library 17:21 28:36 58:36 Walmart at Butler Plaza 2:42 22:27 30:27 Library Partnership 3:15 7:02 7:02 Food Stamp Office Bus 3 3:22 8:28 55:28 Food Stamp Office Bus 24 3:22 9:06 39:06 Davis Chevrolet 3: 58 20:35 28:35 Triangle Club 4:12 9:49 9:49 Salvation Army on University 4:25 9:01 9:01 Little Caesars 4:32 8:30 8:30 Library on University 4:45 10:20 10:20


63 Table 5 1 Continued Destination Travel Time by Car Travel Time by Bus without Transfer Trave l Time by Bus with Transfer Firs t United Methodist Church 4:49 15:53 30:53 Helping Hands 4:50 15:32 30:32 Children's Home Society 5:13 14:09 38:09 Bank of America on 13th 5:15 20:05 28:05 Walmart on Waldo Bus 25 5:15 8:04 8:04 Walmart on Waldo Bu s 26 5:15 8:04 8:04 Goodwill 5:16 14:10 22:10 Alachua County Courthouse 5:17 18:36 18:36 Big Lots 5:17 11:01 26:01 Family Dollar 5:18 12:38 20:38 Labor Pool 5:20 19:57 19:57 People Ready 5:20 8:58 8:58 Save A Lot 5:22 15:27 30:27 Bly's School of Co smetology 5:23 16:03 25:03 Rosa Parks 5:26 11:03 11:03 Taco Bell on 13th 5:26 11:59 37:59 St Francis House 5:29 13:28 13:28 2nd and Charles 5:31 20:11 29:11 Dollar Tree 5:41 11:59 37:59 Pawn Shop on 6th St 5:46 13:38 13:38 Epilepsy Foundation of Flo rida 5:54 12:06 38:6 Rural King on 13th Bus 15 5:59 14:14 29:14 Rural King on 13th Bus 6 5:59 21:21 30:21 Checkers on University 6:16 15:27 33:27 Taco Bell on University 6:16 15:50 33:50 BP on University 6:21 14:28 32:28 UF 7:39 10:56 10:56 UF F ootball stadium 8:25 15:58 15:58 Circle K 8:29 18:10 25:10 Florida Recovery Center 8:35 19:08 26:08 Meridian on 13th 8:36 17:36 24:36 VA Hospital 8:45 15:35 34:35 Shands Medical Plaza 8:52 16:44 23:44 Wells Fargo on University 9:11 15:06 30:06 Millh opper Library 9:18 18:37 28:37


64 Table 5 2. Destinations within 0.5 Miles of the Suggested Bus Stop Harry's Restaurant First United Methodist Church 2nd and Charles Alachua County Courthouse Alachua County Tax Collector off University Bank of Ameri ca on 13th Street Big Lots Bly's School of Cosmetology BP Gas Station on University Avenue Career Source Checkers on University Avenue Children's Home Society Circle K Community Overdrive Dollar Tree Downtown Gainesville Epilepsy Foundation of F lorida Family Dollar Florida Recovery Center Food Stamp Office Goodwill GRU Helping Hands Labor Pool Library on University Avenue Library Partnership Little Caesars Meridian on 13th Street Pawn Shop on 6th Street People Ready Plasma Donation on 6th Street Publix on Main Street Salvation Army on University Avenue Rosa Parks Rural King on 13th Street Salvation Army on 23rd Avenue Salvation Army on University Avenue Save A Lot


65 Table 5 2. Continued Shands Hospital Shands Medical Plaza So nny's Waldo Road St Francis House Taco Bell on 13th Street Taco Bell on University Avenue UF VA Hospital Walgreens Walmart on Waldo Road Wells Fargo on University Avenue


66 Figure 5 1. Homeless Questionnaire Statistics


67 Figure 5 2. Alachua Coun ty Jail Shortest Path Example


68 Figure 5 3. Suggested Bus Route


69 CHAPTER 6 LIMITATIONS AND CONCLUSION 6.1 Limitations There were limitations with this study. The first limitation was access to communication with the homeless. When conducting a study wit h human subjects, the United States mandates submitting research design to an institutional review board (IRB) that approves and monitors the research; making sure it follows an ethical standard that protects human subjects from physical or psychological h arm. Initially, the IRB was submitted with a protocol interested in interviewing individuals who worked with the homeless. However, this did not provide enough data on homeless transport so the IRB protocol was amended to include surveying the homeless. IRB considers the homeless as an endangered population given their compos itional makeup as discussed in C hapter 2. Further, the homeless can contain other endangered populations such as pregnant women, children, and the mentally/physically disabled. Beca use of this, the protocol of this study had to be modified allowing a secondary figure to collect the data. In this case, the secondary figure was Jon DeCarmine, the Operations Director of GRACE Marketplace. Jon DeCarmine communicates and works with the homeless regularly and has collected his own personal data on the homeless he serves. In this case, Jon was deemed as appropriate for conducting the questionnaire as he has made his profession working with these homeless individuals and has therefore deve loped a relationship with these people in a way that does not physically or psychologically harm them. Getting IRB to approve a second researcher was complicated, but the protocol was formulated in such a way that allowed Jon DeCarmine to collect all the d ata and


70 donate it to the study. Once the protocol was finalized, it was submitted for a second time to IRB. After approximately eight months, this study was approved. For this study, it was imperative to speak with the homeless. However, IRB limited thi s. Instead of going directly to the source, this study was stalled for eight months and was constantly being redesigned to find a way to communicate with the homeless. IRB protects the rights and ethical treatment of individuals, but it also creates a ba rrier to collecting research on the populations that need studying the most. Writing the literature review for this research was difficult because very little data has been collected and published on homeless travel. I believe one of the factors limiting this research is the guidelines set by IRB on communicating with the homeless. Being able to freely communicate with and survey the homeless would have resulted in a quicker turn a round time on data collection as well as an increased personal knowledge of how and where the homeless are traveling. Again, it is great that the US ha s an ethical review board in place to protect individuals, but it does hinder data collection for those who are estimated to be the most affected by the lack of services such as transportation and accessibility to housing. The second limitation of this study was the lack of organizational respondents and a survey population that was not truly random. As mentioned in the methodology chapter, different organizations were contacted in order to collect travel data on the homeless. It was up to that organization to respond back agreeing on a date and time in which to meet and discuss the homeless. Very few organizations chose to respond and of the few that did, some felt as though th ey were not informed enough to discuss


71 homeless travel. In this instance, this portion of the study revolved around personal responses. Additionally, the second portion of the study where data was collected using questionnaires given to the homeless was n ot statistically random. The questionnaires were handed out to all homeless individuals attending the breakfast service offered at GRACE Marketplace on September 20th, 2017. The individuals were asked to respond to the questionnaires, but responding was not required. An incentive was offered to increase the number of respondents. It was completely up to the individual to respond or not. The ability to choose whether or not to respond is one of the reasons why the study was not truly random. The questi onnaire also relied on the individual having the physical and mental ability to write and read. Third, the questionnaire offered an incentive; something that which does not satisfy everyone equally. For example, someone who has absolutely nothing is more likely to answer the questionnaire and collect the incentive than someone who already has socks or a toothbrush. Lastly, the questionnaire was only administer ed at GRACE Marketplace. GRACE Marketplace does not encompass the entire homeless population of Gainesville and certainly does not account for those living off the grid and those homeless that are traveling most by foot or bike. Although this study had geographical and individual bias, it is still a study worthy of attention. The majority of the ho meless in Gainesville do reside or attend GRACE for services and only a small percentage avoid GRACE completely. Additionally, as it has been mentioned previously, very little travel data has been collected on the homeless so any type of data collection i s momentous.


72 The final limitation involved a limited knowledge of ArcGIS Network Analyst combined with limited RTS data. The shapefile received for the bus data contained many attribute fields but had no metadata providing explanations of the data. Also, the bus network did not accurately line up with the street network provided by Alachua County Property Appraiser. After multiple attempts to reproject the data, it was clear that the bus routes were created separate from the streets data. Because of thi s, each individual bus route had to be recreated using the features of the streets shapefile. By using the streets shapefile to recreate the bus network, I was able to redesign my study and simplify the analysis process. Although simplifying the process removed the realistic factors associated with driving and bus transit (IE time of day, day of the week, frequency, etc.) it created a more uniform analysis that decreased the margin of error and number of factors that would have needed to be considered whe n comparing the two different networks. Instead, this analysis focused on an individual factor -travel time. 6.2 Conclusion The purpose of this study was to evaluate the travel patterns of the homeless in Gainesville, Florida using questionnaires. The travel patterns revealed that an o verwhelming majority (87% ) of the homeless population in Gainesville use bus to travel. Looking at bus travel compared to car travel showed that on average, bus travel without including transfer time takes approximately e ight minutes longer than car travel while bus travel that includes transfer times takes, on average, twenty extra minutes. Additionally, the routes for GRACE Marketplace were individually studied and compared to each other as well as with the locations in which the homeless are traveling. After completing a full travel time analysis, an updated to bus route 36 was created to


73 increase the overall efficiency in how and where the homeless travel. The new route was designed to utilize existing bus stops and bus routes. Not only does this keep the costs associated with route development down, but also takes advantage of updating an existing route instead of completely developing a new one. Additionally, this updated route would increase frequency and extend service hours. The second suggestion for maximum utility and efficiency is to retire the weekday route for bus 25 and instead, adopt the weekend route. The rationale behind this change is that the bus 25 weekend route travels to both UF and the Rosa Parks Transfer Station; encompassing the routes for both bus 25 and 26. Changing the route from the weekday route to the weekend route allows for the updating of route 26 as mentioned above. Further, it is recommended that the frequency be increased and servi ce hours be extended. As mentioned previously, providing transportation to the homeless can significantly increase their chances of finding work, housing, and moving out of homelessness. Given the information collected during interviews, the questionnaire and discovered during the analysis it is predicted that increasing bus services and updating routes would not only increase efficiency, but also increase the standard of living for the homeless by providing greater access to government services, health s ervices, jobs, and recreational locations.


74 APPENDIX A I NTERVIEW QUESTIONS Organization/demographic information: What is your job title? What sort of demographic does this organization serve? How many homeless does this organization typically se rve? What is the average age of a homeless person at your organization? Approximately how many of the homeless population at this organization are male? Approximately how many of the homeless population at this organization are female? Which gender would you say travels more? Male or Female? Approximately how many children come to this organization? What do you do with the children that come to this organization? Homeless work and living information: Do you know the approximate location in which these peo ple live? If so, is it possible to have this information or a general description of the major housing locations of the homeless? About how many live in their vehicle? What percentage reside in shelters? What percentage live outside? Approximately what p ercentage of the homeless relocate to another city or state? How do the homeless get to these new residencies? Approximately what percentage of the homeless relocate from one housing location to another within Gainesville? What type of jobs do the homeless in this organization work? Where do the homeless go to find work? Do they go to work camps? Are there companies around the city that traditionally hire the homeless? Homeless transit information: Approximately how many of the children at this organizatio n go to school? How do they get to school? How many take the bus? How many are driven? How many walk? Does this organization provide bus or transportation vouchers for the homeless? If yes, approximately how many vouchers does this organization give out da ily/monthly? Approximately how many of the homeless population at this organization are classified as working homeless? How do they get to work? o Approximately how many take the bus? o How many drive? o How many walk? Besides school and work, where specifically do the homeless travel? Could you provide me with a list of locations that the homeless travel? How do they get to these locations? Approximately how many take the bus?


75 Approximately how many drive? Approximately how many walk? What is the most traveled d ay of the week by the homeless at this organization? Do you have any data showing that better transit and/or the owning of a vehicle increases the welfare of the homeless? Do you believe that better transit increases the welfare of the homeless? What about owning a car? Do you think that owning a car increases the welfare of the homeless? Data collection and mitigation strategy information: How does this organization collect data on homeless transportation? Does this organization have tangible data on home less transportation? If yes, could I have this data? If no, how do you recommend I collect this data? Is it possible to interview or survey the homeless at this organization? Have you worked with the City of Gainesville or (Regional Transit System) to make transportation more accessible to the homeless? What procedures, if any, are in place to help the homeless travel? What improvements would you make to the Gainesville transportation system to better serve the needs of the homeless? Do you know of any othe rs with knowledge on this topic to talk with? Additional Questions for RTS Approximately how many homeless use the bus service? What bus line do the homeless typically ride? At which bus stops do the homeless get on at? At which bus stops do the homeles s get off at? Do the homeless use the covered bus shelters as places to sleep? Do the homeless use the bus shelters as spots to congregate? Do you offer the homeless lower fares compared to other riders? If so, how do you determine the person is homeless? Do you collect data on where the homeless travel? If so, could I have that data? Do you locate bus stops in areas where the homeless can easily access it?


76 APPENDIX B HOMELESS TRAVEL QUESTIONNAIRE Please read this consent document carefully before you de cide to participate in this questionnaire. What you will be asked to do in the study You will be asked to participate in a questionnaire that tracks your travel patterns throughout Gainesville over the past week. The data collected from this questionnaire will be used in a study intending to improve and increase bus services to GRACE Marketplace. Compensation Gift bags will be given to those who complete this questionnaire. Confidentiality Your identity will be kept confidential to the extent provided by l aw. No personal information including your name, age, and gender will be released. Voluntary participation Your participation in this questionnaire is completely voluntary. There is no penalty for not participating. Who to contact if you have questions ab out the study Jon DeCarmine, Operations Director of GRACE Marketplace, phone 352 792 0800 Who to contact about your rights as a research participant in the study IRB02 Office, University of Florida Gainesville, FL 32611 2250 phone 352 392 0433 Agreement I have read the procedure described above. I voluntarily agree to participate in the procedure and I have received a copy of this description. I give my consent to publish my travel data. I DO NOT give my consent to publish my travel data.


77 Participant: ______________________________________ Date: _________________ TRAVEL QUESTIONNAIRE Age: _______ Gender: Male / Female / Other / Prefer not to answer Please list all the locations you have traveled to in the past week and how you got that location. Ple ase be as specific as possible by including bus routes (if you can remember them). Example: Traveled to Walmart on Waldo Road using the bus route 25 1. ________ __________________________________________________ 2. ___________ _______________________________ ________________ 3. ________________ __________________________________________ 4. ____________ ______________________________________________ 5. _________________ _________________________________________ 6. _____________________ _____________________________ ________ 7. ________ ______ ____________________________________________ 8. _________ _________________________________________________ 9. _____________ _____________________________________________ 10. _____ ____________________________________________________ 11. ____ _____________________________________________________ 12. _______ __________________________________________________ 13. _______________ __________________________________________ 14. ____________ _____________________________________________ 15. ___ _____ _________________________________________________ I f you have traveled more than 15 locations, please continue the list on the back of this paper. Please return this questionnaire t o Jon DeCarmine in Building 7.


78 REFERENCES Acua, J., & Erlenbusc h, B. (2009, August). HOMELESS EMPLOYMENT REPORT: Findings and Recommendations (Rep.). Retrieved August 6, 2017, from National Coalition for the Homeless website: http:// Alachua County. (2013, December 3). Alachua County's Economic, Demographic & Fiscal Trends [PPT] Gainesville: Alachua County. Arnold, K. R. (2004). Homelessness, Citizenship, and Identity: The Uncanniness of Late Modernity. Albany: State University of New York Press. Retrieved August 6, 2017, from Project MUSE database. Blankertz, L. E., Cn aan, R. A., Saunders, M. (1992). Assessing the impact of serving the long term mentally disabled homeless. Journal of Sociology and Social Welfare 19(4), 199 220. Bureau of Transportation Statistics. (2016). Transportation Economi c Trends (Publication). Retrieved August 6, 2017, from economic trends Burt, M., Carpenter, J., Hall, S., Henderson, K., Rog, D., Hornik, J., . Moran, G. (2010, March). Strategies for Improving Homeless People's Access to Mainstream Benefits and Services [PDF]. US Department of Housing and Urban Development. City of Gainesville. (2017, April 26). Empowerment Center Oversight Advisory Board Alachua County Administration Building, Gainesville, FL. Cohen, C. (1999). Http:// In Medicin e Encyclopedia J Rank. Retrieved August 6, 2017, from Corporation for National & Community Service. (n.d.). National Point in Time Count of People Experiencing Homelessness. Retrieved August 6, 2017, from initiatives/days service/martin luther king jr day service/toolkits/national point time Council on Homelessness. (2016). Council on Homelessness 2016 Annual Report (Rep.). Retrieved August 6, 2017, from Culhane, D. (2010, July 11). Five myths about America's homeless. Washington Post Retrieved August 6, 2017, from dyn/content/article/2010/07/09/AR2010070902357.html


79 Cunningham, R. (2010, July 7). Parks as soup kitchens. The Gainesville Sun Retrieved August 6, 2017, from 10272/parks as soup kitchens/ ESRI. (n.d. ). What is the ArcGIS Network Analyst extension? Retrieved August 21, 2017, fro m analyst/what is network analyst .htm Hamlin, R. (2016, July 28). Civil Rights. In Encyclopedia Britannica Retrieved August 6, 2017, from il rights#toc119317main Jocoy, C., & Del Casino, V., Jr. (2012). The Mobility of Homeless People and Their Use of Public Transit in Long Beach, California (Rep. No. 06 13). Long beach, CA: California State University. Kelley, M. B. (2012, July 26). The US Government Is Failing Miserably at Helping Veterans. Retrieved August 6, 2017, from the number s the us government is failing miserably at helping veterans 2012 7 The Leadership Conference Education Fund. (2011, March). Where We Need to Go: A Civil Rights Roadmap for Transpor tation Equity (Publication). Retrieved August 6, 2017, from s/transportation/52846576 Where We Need to Go A Civil Rights Roadmap for Transportation Equity.pdf Lee, B. A., Tyler, K. A., & Wright, J. D. (2010). The New Homelessness Revisited Manuscript submitted for publication, National Inst itute of Health. Retrieved 2016, from 583582.pdf Michael Pottinger, Peter Carter, Berry Young, et al. v. City of Miami, 810 F. Supp. 1551 (S.D.F.L. 1992). Moser, T. (1991). Shortest path calculation of seismic rays. Geophysics, 56 (1), 59 67. Re trieved August 21, 2017, from http://library.seg .org/doi/abs/10.1190/1.1442958 National Coalition for Homeless Veterans. (n.d.). Background & Statistics. Retrieved August 6, 2017, from National Coalition for the Homeless. (2009, July). Employment and Homelessness (Rep.). Retrieved August 6, 2017, from National Coalition for the Homeless. (2004). Illegal to be Homeless (Rep.). Retrieved August 6, 2017, from


80 National Health Care for the Homeless Council. (2011). Disability, Employment & Homelessness (Publication ). Retrieved August 6, 2017, from content/uploads/2011/09/disability2011_ final.pdf National Health Care for the Homeless Council. (n.d.). Wha t is the official definition of homelessness? Retrieved August 6, 2017, from definition homelessness/ Parker, R. D., & Dykema, S. (2013). The Rea lity of Homeless Mobility and Implications for Improving Care. Journal of Community Health, 38 (4), 685 689. Retrieved August 6, 2017, from 7%2Fs10900 013 9664 2 The Pottinger Agreement and Your Rights. (n.d.). Retrieved August 6, 2017, from Quamie, L. (2011). Transportation Equity a Key to Winning Full Civil Rights (Rep.). Retrieved August 6, 2017, from The Leadership Conference on Civil and Human Rights website: 2.quamie.pdf Rajput, S. (2014, March 12). Network Analysis [PPT]. Retrieved August 21, 2017, from analysis in gis Sanchez, T. W., & Brenman, M. (2007). Transportation and Civil Rights Retrieved August 6, 2017, fro m 50th United Nations General Assembly. (1948, December 10). Universal Declaration of Human Rights [PDF]. Paris: United Nations General Assembly. US Census Bureau. (n.d.). Gainesville city, Florida. Retrieved August 6, 2017, from mk US Department of Housing and Urban Development. (2012). Criteria for Definition of At Risk of Homelessness. Retrieved August 6, 2017, from https:/ / for definition of at risk of homelessness/ US Department of Transportation. (2016). Transit's Role in Preventing and Ending Homelessness [Editorial]. How should FTA prioritize and invest in research activities over the next several years? Retrieved August 6, 2017, from national online dialogue on fta s 5 year strategic/suggestions/15883482 transit s role in preventing and ending homelessne Vallas, R., Fremstad, S., & Ekman, L. (2015, January 28). A Fair Shot for Workers with Disabilities (Publi cation). Retrieved August 6, 2017, from Center for American Progress website:


81 y/reports/2015/01/28/105520/a fair shot for workers with disabilities/


82 BIOGRAPHICAL SKETCH Kaysie Salvatore is a transportation planner for a private consulting firm located in the state of Florida. In 2015, s he earned her Bachelor of Arts d egree from the University of Florida in geography and political science. In 2017, Kaysie graduated again from the University of Florida with a Master of Urban and Regional Planning d egree. Although Kaysie is currently a transportation planner, she has many academi c passions such transportation engineering, geographic information system s and demography. Throughout her time at the University of Florida, Kaysie has worked with both UNICEF and the American Cancer Society helping to raise money for those less fortunat e. Kaysie is a dedicated traveler and lover of dogs. She can list all 67 counties in the state of Florida; a skill she picked up from working for the University of Florida GeoPlan Center. She has a passion for do it yourself projects, interior design, a nd baking.