<%BANNER%>

The Homeless as a Social Movement Group

Permanent Link: http://ufdc.ufl.edu/UFE0024134/00001

Material Information

Title: The Homeless as a Social Movement Group Group Identity, Group Cohesion, and Goal Orientation among the Homeless in Gainesville, Florida
Physical Description: 1 online resource (101 p.)
Language: english
Creator: Cadavieco, Norma
Publisher: University of Florida
Place of Publication: Gainesville, Fla.
Publication Date: 2009

Subjects

Subjects / Keywords: Family, Youth and Community Sciences -- Dissertations, Academic -- UF
Genre: Family, Youth and Community Sciences thesis, M.S.
bibliography   ( marcgt )
theses   ( marcgt )
government publication (state, provincial, terriorial, dependent)   ( marcgt )
born-digital   ( sobekcm )
Electronic Thesis or Dissertation

Notes

Abstract: This study was an attempt to explore the precursors to social movement action by examining the presence of those precursors within the homeless population in Gainesville, Florida. New social movements focus on the influence of identity, culture, and emotion in movement dynamics. Group identity, group cohesion, and goal orientation are all factors that contribute to collective action in the form of social movements. A model for new social movements based on the literature is proposed and tested. A two-phase, cross-sectional and case study design was used to determine the presence of the above precursors to social movement within the Gainesville homeless population. Results indicate that group identity, group cohesion, and goal orientation are present within the study sample. Group cohesion and group identity are closely related but distinctly different concepts leading to the formation of goals and eventually collective action within a social group. Multiple regression analysis indicates that group cohesion is predictive of goal orientation, but group identity is not, suggesting that a model for new social movements would move from group identity, to group cohesion, and finally to goal orientation, then collective action. In-depth analysis of qualitative interviews indicates strong mutual respect within the population as well as self-leadership and a variety of circumstances leading to first homeless episode. Therefore, the models used for delivery of homeless services should be adjusted to account for the role of homeless persons in their own recovery.
General Note: In the series University of Florida Digital Collections.
General Note: Includes vita.
Bibliography: 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.
Statement of Responsibility: by Norma Cadavieco.
Thesis: Thesis (M.S.)--University of Florida, 2009.
Local: Adviser: Swisher, Marilyn E.

Record Information

Source Institution: UFRGP
Rights Management: Applicable rights reserved.
Classification: lcc - LD1780 2009
System ID: UFE0024134:00001

Permanent Link: http://ufdc.ufl.edu/UFE0024134/00001

Material Information

Title: The Homeless as a Social Movement Group Group Identity, Group Cohesion, and Goal Orientation among the Homeless in Gainesville, Florida
Physical Description: 1 online resource (101 p.)
Language: english
Creator: Cadavieco, Norma
Publisher: University of Florida
Place of Publication: Gainesville, Fla.
Publication Date: 2009

Subjects

Subjects / Keywords: Family, Youth and Community Sciences -- Dissertations, Academic -- UF
Genre: Family, Youth and Community Sciences thesis, M.S.
bibliography   ( marcgt )
theses   ( marcgt )
government publication (state, provincial, terriorial, dependent)   ( marcgt )
born-digital   ( sobekcm )
Electronic Thesis or Dissertation

Notes

Abstract: This study was an attempt to explore the precursors to social movement action by examining the presence of those precursors within the homeless population in Gainesville, Florida. New social movements focus on the influence of identity, culture, and emotion in movement dynamics. Group identity, group cohesion, and goal orientation are all factors that contribute to collective action in the form of social movements. A model for new social movements based on the literature is proposed and tested. A two-phase, cross-sectional and case study design was used to determine the presence of the above precursors to social movement within the Gainesville homeless population. Results indicate that group identity, group cohesion, and goal orientation are present within the study sample. Group cohesion and group identity are closely related but distinctly different concepts leading to the formation of goals and eventually collective action within a social group. Multiple regression analysis indicates that group cohesion is predictive of goal orientation, but group identity is not, suggesting that a model for new social movements would move from group identity, to group cohesion, and finally to goal orientation, then collective action. In-depth analysis of qualitative interviews indicates strong mutual respect within the population as well as self-leadership and a variety of circumstances leading to first homeless episode. Therefore, the models used for delivery of homeless services should be adjusted to account for the role of homeless persons in their own recovery.
General Note: In the series University of Florida Digital Collections.
General Note: Includes vita.
Bibliography: 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.
Statement of Responsibility: by Norma Cadavieco.
Thesis: Thesis (M.S.)--University of Florida, 2009.
Local: Adviser: Swisher, Marilyn E.

Record Information

Source Institution: UFRGP
Rights Management: Applicable rights reserved.
Classification: lcc - LD1780 2009
System ID: UFE0024134:00001


This item has the following downloads:


Full Text

PAGE 1

THE HOMELESS AS A SOCIAL MOVEMENT GROUP: GROUP IDENTITY, GROUP COHESION, AND GOAL ORIENTATION AMONG THE HOMELESS IN GAINESVILLE, FLORIDA By NORMA CADAVIECO A THESIS PRESENTED TO THE GRADUATE SCHOOL OF THE UNIVERSITY OF FLOR IDA IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF MASTER OF SCIENCE UNIVERSITY OF FLORIDA 2009 1

PAGE 2

2009 Norma Cadavieco 2

PAGE 3

To my sisters. 3

PAGE 4

ACKNOWLEDGMENTS I would like to thank my family, my friends and my graduate committee members for all of their guidance and support thr oughout this strenuous process. 4

PAGE 5

TABLE OF CONTENTS page ACKNOWLEDGMENTS ...............................................................................................................4 LIST OF TABLES ...........................................................................................................................8 LIST OF FIGURES .........................................................................................................................9 ABSTRACT ...................................................................................................................................10 CHAPTER 1 INTRODUCTION AND PROBLEM STATEMENT............................................................12 Homelessness in the United States .........................................................................................12 Homelessness in Gainesville ..................................................................................................14 Homeless Programs and Policies ............................................................................................15 Housing First...................................................................................................................15 Categorical Programs ......................................................................................................16 Comprehensive Care Centers ..........................................................................................17 Policy and Programming in Gainesville ..........................................................................18 Contribution of Research ........................................................................................................18 Research Question and Objectives .........................................................................................19 2 LITERATURE REVIEW.......................................................................................................21 Social Movements and Social Change ...................................................................................21 Classical Social Movement Perspectives ................................................................................22 Resource Mobilization Model ................................................................................................23 Political Process Model ..........................................................................................................25 Theoretical Perspectives on New Social Movements .............................................................27 Manuel Castells ...............................................................................................................27 Alaine Touraine ...............................................................................................................28 Jurgen Habermas .............................................................................................................29 Alberto Melucci ...............................................................................................................30 Culture, Identity, and Emotion in New Social Movements ....................................................31 Critical Components of New Social Movement Theories ......................................................34 Group Identity .................................................................................................................34 Group Cohesion ...............................................................................................................36 Goal Setting & Achievement ...........................................................................................37 Purpose of Research ...............................................................................................................39 A Model for New Social Movements .....................................................................................40 5

PAGE 6

3 METHODS...................................................................................................................... .......42 Research Design .....................................................................................................................42 Sample Selection ....................................................................................................................42 Instrumentation .......................................................................................................................44 Group Identity .................................................................................................................44 Group Cohesion ...............................................................................................................45 Development of group cohesion index .....................................................................46 Pretesting of group cohesion index ..........................................................................49 Goal Orientation ..............................................................................................................50 Qualitative Interview .......................................................................................................51 Data Collection .......................................................................................................................53 Data Analysis ..........................................................................................................................53 Quantitative Data .............................................................................................................53 Qualitative Data ...............................................................................................................54 Limitations ..............................................................................................................................54 4 RESULTS...................................................................................................................... .........58 Sample ....................................................................................................................................58 Demographic Characteristics ..................................................................................................58 Tests of Normal Distribution ..................................................................................................59 Hypothesis 1 ...........................................................................................................................60 Hypothesis 2 ...........................................................................................................................61 Hypothesis 3 ...........................................................................................................................61 Qualitative Analysis ................................................................................................................62 Sample .............................................................................................................................62 Coding and Grouping ......................................................................................................62 5 DISCUSSION AND CONCLUSIONS..................................................................................67 Research Question and Hypotheses ........................................................................................67 Hypothesis 1 ...........................................................................................................................67 Hypothesis 2 ...........................................................................................................................69 Hypothesis 3 ...........................................................................................................................70 Interview Data ........................................................................................................................71 Group Identity .................................................................................................................71 Individuality .............................................................................................................72 Group concept ..........................................................................................................72 Acceptance of homelessness ....................................................................................73 Group Cohesion ...............................................................................................................74 Interactions with homeless persons ..........................................................................74 Mistrust .....................................................................................................................75 Goal Orientation ..............................................................................................................76 Outlook .....................................................................................................................76 Goals .........................................................................................................................77 Self-efficacy amd motivation ...................................................................................77 6

PAGE 7

Leadership .......................................................................................................................78 Identification of leaders ............................................................................................78 Definitions of leadership ..........................................................................................79 Personal Experience ........................................................................................................79 Homeless experience ................................................................................................80 Coping......................................................................................................................80 Support systems ........................................................................................................81 A New Model .........................................................................................................................81 Limitations ..............................................................................................................................82 Implications for Research .......................................................................................................84 APPENDIX A ITEMS IN GROUP IDENTITY SCALE...............................................................................87 B ITEMS IN GROUP COHESION INDEX..............................................................................88 C ITEMS IN GOAL ORIENTATION SCALE.........................................................................89 D INTERVIEW SCHEDULE....................................................................................................90 E INTERVIEW PROCESSING FORM....................................................................................91 LIST OF REFERENCES ...............................................................................................................92 BIOGRAPHICAL SKETCH .......................................................................................................101 7

PAGE 8

LIST OF TABLES Table page 3-1 Reliability test on group identity scale...............................................................................56 3-2 Major dimensions of cohesion across three different cohesion measures. ........................56 3-3 Preliminary reliability test on group cohesion index. ........................................................57 3-4 Final reliability test on group cohesion index. ...................................................................57 3-5 Reliability test on goal orientation index. ..........................................................................57 4-1 Gender and age of homeless persons, Gainesville, Fl, in 2008. ........................................64 4-2 Wilkes-Shapiro test of normal dist ribution for the outcome variable of goal orientation among homeless persons in Gainesville, Fl, in 2008. .....................................64 4-3 Wilkes-Shapiro test of normal distribu tion for the predictor variables group cohesion and group identity for homeless persons in Gainesville, Fl, in 2008. ................................64 4-4 Summary statistics for goal orientation among homeless persons in Gainesville, Fl, in 2008. ..............................................................................................................................64 4-5 Mann Whitney U 2-sample test to comp are goal orientation scores between short term and chronically homeless persons in Gainesville, Fl, in 2008. ..................................64 4-6 T-tests for predictor variables group identity and group cohesion for homeless persons in Gainesville, Fl, in 2008. ....................................................................................65 4-7 Spearman rank order correlations fo r group identity, group cohesion, and goal orientation among homeless persons in Gainesville, Fl, in 2008. .....................................65 4-8 Multiple regression results. The rela tionship between group identity and group cohesion on goal orientation for homeless persons in Gainesv ille, Fl, in 2008. ................65 4-9 Regression summary for the relationship between group identity and group cohesion on goal orientation among homeless persons in Gainesville, Fl, in 2008. .........................65 4-10 Categories organized by themes from interviews with homeless persons in Gainesville, Fl, in 2008. .....................................................................................................66 8

PAGE 9

LIST OF FIGURES 2-1 Proposed new social movement model. .............................................................................41 5-1 Revised model for new social movements. ........................................................................86 9

PAGE 10

Abstract of Thesis Presen ted to the Graduate School of the University of Florida in Partial Fulfillment of the Requirements for the Degree of Master of Science THE HOMELESS AS A SOCIAL MOVEMENT GROUP: GROUP IDENTITY, GROUP COHESION, AND GOAL ORIENTATION AMONG THE HOMELESS IN GAINESVILLE, FLORIDA By Norma Cadavieco May 2009 Chair: Marilyn E. Swisher Major: Family, Youth, and Community Sciences This study was an attempt to explore th e precursors to social movement action by examining the presence of those precursors wi thin the homeless population in Gainesville, Florida. New social movements focus on the influence of identity, culture, and emotion in movement dynamics. Group identity, group cohesion, and goal orientation ar e all factors that contribute to collective action in the form of social movements. A model for new social movements based on the literature is proposed and tested. A two-pha se, cross-sectional and case study design was used to determine the presence of the above precursors to social movement within the Gainesville homeless population. Resu lts indicate that group identity, group cohesion, and goal orientation are present within the study sample. Group cohesion and group identity are closely related but distinctly different concepts leading to the fo rmation of goals and eventually collective action within a social group. Multiple regression analysis indicates that group cohesion is predictive of goal orientati on, but group identity is not, sugges ting that a model for new social movements would move from group identity, to group cohesion, and finally to goal orientation, then collective action. In-depth analysis of qualitative interviews indicates strong mutual respect within the population as well as self-leadership and a variety of circumstances leading to first 10

PAGE 11

homeless episode. Therefore, the models used for delivery of homeless services should be adjusted to account for the role of homeless persons in their own recovery. 11

PAGE 12

CHAPTER 1 INTRODUCTION AND PROBLEM STATEMENT Homelessness in the United States Homelessness is a condition in which a person is without permanent a nd/or stable shelter for any period. According to the U.S. Depart ment of Housing and Urban Development, a homeless person is defined as an individual w ho lacks a fixed, regular, and adequate nighttime residence, and a person whose primary nighttime residence is a public or private shelter, an institution, a public or private place not desi gned for a regular sleeping accommodation for human beings (U.S. Department of Housing and Urban Development[HUD], 2007). However, current social research about homelessness suggests that this definition is problematic contingent with the actual nature of homelessness in the U.S. According to the research, a large proportion of homeless people are part of a group known as th e hidden homeless, or those who live in the homes of friends or family members and often do not seek public assi stance for their condition (Rollinson, 2007). These people often include ma ny of the rural homeless, which constitute nearly half of the homeless population, and ma ny times women with children (HUD, 2007). Current definitions vary in specificity and orie ntation; some include those who choose to be homeless, some include children, and some include social isolates, while others do not. It is clear that the classic conception of the homeless person as an older white male white veteran with mental illness actually constitutes a small proportion of the homeless population nationwide. Two major historical events, the Great Depression of the 1920s and 30s (Miller, 1991) and the deinstitutionalization of mental health care in the 1970s (Marvasti, 1999) are key points in history. Both significantly changed the face of homelessness in the U.S. The Great Depression put many previously employed, middle and lower-cla ss persons on the street, contributing to the 12

PAGE 13

belief that anyone could end up homeless. The de institutionalization of me ntal health care put many persons with severe mental health disord ers, such as schizophrenia, schizoaffective disorders, bipolar disorder, and major depression, on the streets. This created a problem that still exists today. A significant numbe r of homeless persons deal with untreated, debilitating mental illness. The most current survey of homelessness in the U.S. puts the number of homeless around 754,000 nationwide (National Alliance to End Ho melessness[NAEH], 2005, The Associated Press, 2007). The number of homeless people with severe mental health disorders has been consistent over the years, but current research points to a greater prevalence of these conditions in the chronically homeless (Cat on et al. 2005; Weinreb, 2006). However, it is clear that poverty and market forces are significant contributors to homelessness today just as they were in the post-depression era (Dale, 2004; Rollinson, 2007). Work opportunities are declining consistently despite increases in real wages. Many jobs offer fewer benefits and pay less co mpared to the cost of living than they did in the past. The real value of the U.S. minimum wage was 26% lo wer in 2004 than it was in 1979 (The Economic Policy Institute, 2005). Falling real wages reduce an individuals ability to secure housing. In addition to the ballooning problem of low wages, public assistance has declined since the mid 1990s (National Coalition for the Homeless[NCH], 2007). The current level of assistance from several major sources, including Temporary Assistance for Needy Families (TANF), is below the poverty level in every state by an average of 29% (Nickelson, 2004). In addition, fewer people are enrolled in and/or qualif y for welfare benefits as a result of current welfare reform laws. These families struggle to obta in medical care and even to meet basic needs like housing. 13

PAGE 14

Current economic and real estate trends are causing a drastic decrea se in affordable and low-cost housing units throughout the U.S. Betw een 1973 and 1993, the availability of low-rent housing units decreased by over two million units, desp ite the fact that the number of people in need of these units has soared (Daskal, 1998). In addition, the amount of federal support for lowincome housing decreased by 49% between 1980 and 2003 (National Low Income Housing Coalition, 2005). These trends ha ve created waiting lists for a ffordable housing units supported by government subsidy or private agen cies like Habitat for Humanity. Other factors that contribute to the prevalence of homelessn ess are inability to secure healthcare or healthcare benef its, domestic violence, substance abuse, and mental illness. Homelessness is a complex issue that forces pe ople to choose between basic needs (NCH, 2006). Homelessness in Gainesville The Gainesville Office on Homelessness conduc ts a yearly point-in-time survey in conjunction with the Alachua County Coalition for the Homeless and Hungry. According to the most recent release on April 12, 2007, there are 952 identified ho meless men, women and children in Alachua County, 317 of them under the ag e of 18. This represents a 20% decrease in the number of homeless persons from the 2006 su rvey (Gainesville/Alachua County Office on Homelessness & Alachua County Coalition for the Homeless and Hungry, 2007). Some key findings of the survey are listed below. 70.5% are male and 29.5% percent are female 87.6% are individuals and 12.4% families with children 34.7% are U.S. military veterans The mean age of homeless adults is 44.1 years About 2/3 are between the ages of 18-59 Nearly 60% have been homeless for less than a year. 14

PAGE 15

28% are considered long-term or chronically homeless 43.7% are unsheltered on any given night More than half lived and worked in Alachua County prior to becoming homeless. The largest cause of homelessness among wome n in Alachua County is male violence against women. 60% self-reported having some physical medical, or mental health problem Almost 2/3 of homeless adults have at least some high sc hool education, 44.6% completed high school, 28.3% have some college edu cation, and 10% hold a college degree. 19% report having been arrested for not havi ng a place to stay within the last year. These findings help to identify severa l homeless subgroups which exist based on common needs, experiences, and length of homelessn ess. The two most dis tinct groups are the short-term homeless, those who have been homel ess less than three years and the chronically homeless, those who have been homeless three or more years. The longer a person is homeless, the more difficult it becomes to become housed. Given this difficulty, it can be implied that services, needs, and characteristics of short-term versus chronically homeless persons differ. Homeless Programs and Policies In 1987, the McKinney-Vento Homeless Assistance act was passed as the first federal policy aimed at reducing homelessness (HUD, 2007). Programming supported through McKinney-Vento grants is heavily focused on creating housing opportunities for homeless persons through housing vouchers and program gran ts to state and local agencies (HUD, 2007). Currently, over $1.5 billion are allocated to McKinney-Vento programming with increased funding over the past seve ral years (NAEH, 2006). Housing First Housing first is quickly becoming one of the most popular models for homeless assistance service delivery and has been empirically supported by research (Fitzpat rick, 2004; Jensen, 2005; 15

PAGE 16

Padgett, Gulcure, & Tsemberis, 2006). The sequen ce of services in a housing first program is usually as follows: housing, case management, mental health care, substance abuse treatment, employment training, then any other services de emed necessary to help sustain stability and self-sufficiency (Legander, 2006). New York and Seattle have both implemented successful programs using the housing first model. Homelessness in New York has been on a steady decline since 2002 because of the widespread implemen tation of the model in public and private agencies (New York City Department of Home less Services, 2007). Pathways to Housing, Inc., was the first agency in New York to try the housing first model in 2002; since 2004, 85% of the programs participants have remained in thei r permanent housing placem ents (Fitzpatrick, 2004; Tsemberis, Gulcur, & Nakae 2004). In 2005, Kings County, Washington, which includes Seattle, moved towards using a housing first approach by developing a 10-year plan to end homelessness focused on creating housing opportunities (Committee to End Homelessness in Kings County, 2005). As of June 2006, over 900 homeless people were moved into permanent housing and 1,300 new housing units were create d in Kings County (Chan, 2006). Categorical Programs Categorical programs attempt to tackle a singl e aspect of homelessness that is usually identified as the major contribu tor to an individuals situatio n. Most categorical programs are focused on mental health and substance abuse re covery (Caton et al., 2005; Fisk, Rakfeldt, & McCormack, 2006). Programs associated with other risk factors such as lack of affordable housing and domestic violence usually attempt to tackle these specific problems in conjunction with other intervening contributing factors. Th ese problem-specific programs are the most researched and validated, because they follow a me dical approach to recovery where quantitative data are used almost exclusively to support the effectiveness of programs. While these programs are validated by extensive evaluation, taking a onesided approach to eliminating drug abuse and 16

PAGE 17

therefore, homelessness, tends not to work (C aton et al., 2005; Fisk et al., 2006). The reason these programs may not be as effective as indicated by statistics is because they ignore the larger social forces that contribute to homelessness; wh ile these programs are su ccessful, the number of homelessness in the U.S. has not significantl y decreased since the 1980s (Mowbray, Cohen, & Bybee, 1993). Comprehensive Care Centers Comprehensive care centers are one-stop as sistance centers where all homeless services are consolidated in one area, us ually one building. This is also a popular method of dealing with homelessness because, ideally, it moves homeless people to one place and keeps them there until they can return to the community in perm anent housing. However, many communities that implement comprehensive care centers do not have the resources, infrastru cture, or political influence to make the centers tr uly comprehensive. These kinds of centers have become popular in Florida, after the success of several centers in Broward Count y (Broward Outreach Centers, 2007). Gainesville, Florida is currently in the process of approving and constructing its own onestop center, the GRACE Marketplace. However, because these programs are so large in scope, setting up evaluation often become s a daunting task and is not planned for from the beginning. Little to no research exists on evaluation practices of comprehe nsive programs such as these (Crook, Mullis, Cornell, & Mull is, 2005); an overview of the comprehensive plans for the programs in Broward County and Gainesville indicat e evaluation as part of the plan, but, it is unclear what these evaluations consist of. In Ga inesville, a comprehensive evaluation plan has not been developed as the proposal for the center has been delayed and ed ited in several battles at city commission (Gainesville Offi ce on Homelessness, 2005; Tinker, 2007). 17

PAGE 18

Policy and Programming in Gainesville The city of Gainesville and Alachua Count y have undergone rapid changes in homeless policies over the past five years. The city and county commissions in 2005 approved the 10-Year plan to end homelessness (Gaine sville Office on Homelessness, 2005) The passing of this bill allowed for the creation of the Office on Homelessn ess, a department jointly funded by the city and county governments. This office works closely with the Alachua County Coalition for the Homeless and Hungry. On October 30, 2006, the construction and funding of the GRACE Marketplace, a comprehensive one-stop care center for the homeless, was approved in a joint session of the city and county commissions (Gainesville City Commission, 2006). The GRACE Marketplace is intended to bring all homeless services curren tly provided throughout the county to one place. All of the major homeless agencies, providers of meals, case management, and behavioral healthcare, in the county have agre ed to relocate all or some of their services to the center once it is opened (DeCarmine, 2007). After the center was approved, numerous additional policies, including a ban on panhandling, evacuation of tent ci ty, and restrictions on use of th e downtown plaza, were passed through the commissions. Many of the policies were considered negative action against homeless persons, and created significant controversy be tween local activists and policymakers (Adelson, 2007a, 2007b; Tinker, 2007). As of May 2008, c onstruction has not begun on the GRACE Marketplace due to proble ms with site selection. Contribution of Research Although programs addressing the needs of ho meless Americans are widespread and well supported locally and federally, it seems there are still major gaps in service provision that are not successfully being addressed by current prac tices. As previously stated, the categorical 18

PAGE 19

approach, tackling one issue as th e significant contributi ng factor, is not eff ective in eliminating homelessness and associated issues in the long-t erm. Comprehensive approaches are newer, but tend to have more positive long-lasting results and many have fewer problems with follow-up as clients develop deeper connectio ns with service providers. In recent years, groups of homeless persons throughout the U.S. have acted as empowered groups able to work towards their own pr ogress (Cononie, 2003; Swithinbank, 1997; Williams, 2005). Studies on these groups have found that although homeless groups lack some of the characteristics of social movements, particularly organized protes t, other factors such as clear organization, goals, and access to resources cl assify these groups as social movement organizations. This approach to homeless services also serves to minimize the top-down perception of outsiders providing services that can hinder effec tiveness of service (Thompson, McManus, Lantry, Windsor, & Flynne, 2006). Perhaps the most necessary change in homeless services is a shift from needs-assessment based programming to asset-based programmi ng, which would work with the homeless to determine what they can do for themselves befo re creating services wh ich may or may not be effective. If the homeless can be viewed as a special interest community, then surely the principles of asset-based community development can be applied to work towards improvement (Bergdall, 2003; Future generations n.d.). In this research, I aim to find out if homeless persons have the ability to behave as a group, and th erefore organize, by expl oring homeless persons perceptions of the homeless as a community and of themselves as homeless persons. Research Question and Objectives This study will attempt to propose and explor e a model for new social movements based on the presence of absence of three key precu rsors to social action within the homeless population in Gainesville. These precursors are group identity, group cohesion, and goal 19

PAGE 20

orientation. The role of these concepts will be expl ained in further detail in the literature review. The three research questions that will guide this study are listed below. Are group identity, group cohesion, and goal orientation appropria te precursors to collective action and thus social movements? Are the precursors described above present within the homeless population in Gainesville? Does a difference in the presence of group ch aracteristics exist be tween short-term and chronically homeless persons? At the end of the study, I hope to be able to determine the presence of group identity, group cohesion, and goal orientation within the homeless population in Gainesville, Florida. If the precursors are present, I will use the results of the quantitative and qualitative data to determine if the model proposed at the end of th e literature review for ne w social movements is appropriate. The directionality and relationship between each concept will be determined during the data analysis. A difference between short-term and chronically homeless persons will indicate that each of the charac teristics is influenced by length of group membership despite membership in this case being involuntary. 20

PAGE 21

CHAPTER 2 LITERATURE REVIEW Social Movements and Social Change Social movements are a significant part of hum an social life because they influence norms and policies from the local level to the global leve l. In recent years, studies of social movements have focused on events such as the Civil Rights and Womens movement in the U.S., labor union strikes, migrant workers rights, the socialist revo lution in Eastern Europe and, more recently, in Latin America, and more societal changes su ch as globalization and environmental advocacy (Russell, 2005). With a wide range of social change under the social movement umbrella, generalizing the dynamics of social movements ha s challenged scholars. Social movements have been described as collective ev ents in which institutions a nd groups are loosely bound by the same ideology (or more accurately, frames) and symbols and connected by overlapping, primarily social, networks (Clark, 2004, p. 942), a nd one of the principal social forms through which collectivities give voice to their grievances and concerns about the rights, welfare, and well-being of themselves and others by engaging in various types of collective action (Snow, Soule, & Kriesi, 2006, p. 3). Scholars have differed on the defining traits of social movements, but most of the differences can be attributed to semantics. P openoe, Merton, Jasper, and Heywoods perspectives all differ in the number and specificity of character istics. Popenoes four ch aracteristics of social movements are (1) a new perspectiv e to see things differently; (2 ) an ideology maintaining group loyalty; (3) a commitment to action; and (4) a dispersed or decentralized leadership (Simsek, 2004, p. 113; Touraine, 1988, p. 532). The definitions provided by other scholars include the above characteristics, in some cases, adding or modifying traits of social movements. Merton, for ex ample, emphasizes goal seeking, 21

PAGE 22

cohesive organizational struct ures, and unifying ideology shared among movement members (Merton, 1983); Jasper suggests that social movements ar e also long-lasting and noninstitutionalized (Jasper, 1997); Heywood suggests that social m ovements are connected to the New Left, based in new middle classes, and innovative (Heywood, 1997; Simsek, 2004). Despite all of the different terms used, it can be gathered from these various characteristics that social movements are movements of groups of people committed to an ideology and goals, whether explicitly stated or not, who are committed to action to correct perceived social injustices. Even in the earliest incarnations, explora tions of social movements bore disagreement among scholars as evidenced by the previously me ntioned definitions. There are currently two dominant perspectives on social movements, th e resource mobilization pe rspective, including the political process model and political opportunity structure as a theoretical offshoot, and the new social movement perspective. The resource mob ilization and political process perspectives are most easily described as variati ons on a structural, goal-oriented model of social movements. The new social movement perspective addresses movements from a less structural perspective, focusing on the role of emotion, identity, and culture in movement success. Classical Social Movement Perspectives The earliest research on social movements is rooted in a Marxian view of class-based conflict leading to change in the distribution of power by a radical and forceful transfer in ownership of the means of industrial production. This perspective grew from the belief that social movements are a particular type of soci al conflict and, ultimately, a method of resolving social conflict and the driving force behind protest and collective action (Touraine, 1985; Simsek, 2004). These movements were seen as largely focu sed on bringing about ch ange in policies and the social order. Issues present in movements that fall into this category include economic 22

PAGE 23

growth, income distribution, military and social se curity, and social control (Simsek, 2004). The perspective grew out of research on the worki ng class protest movements of the late 1960s (Simsek, 2004; Wievorka, 2005). Two or more clearly defined, opposing actors are identified and battling over competing interests. Conflict is resolved by collective action, primarily in the form of protest. Additionally, social movements differ from institutionalized behavior in that they are relatively spontaneous a nd unstructured reactions to soci etal strain (Morris, 1981). Resource Mobilization Model The resource mobilization model was the most prominent social movement theory in the United States during the 1970s, wh en scholars began to move away from the socioeconomically based classical perspectives on social movements that dominated research in the 1960s (Simsek, 2004). This perspective posits that access to reso urces and well-defined, well-organized protest groups are the necessary components to a successful movement. The resource mobilization model is heavily structural in its basis, underlining the im portance of organizational and institutional structures as well as networks in development of a successful movement (McAdam & Snow, 1997; Morris, 1981; Walsh & Warland, 1983). The central focus of the resource mobilization perspective is the social movement organization, or the protest organization, which functions in a highly institutionali zed manner; this is contrary to the classical perspective in which movements start out disorganized and irra tional (Morris, 1981). Without the presence of a well-organized group of activists, the mobilization and unrest necessary to create change cannot be developed (Useem, 1980). The idea that access to resources and orga nizational capacity determines movement success has been evidenced in studies focusing on the urban riots of the 1960s (Useem, 1980). Rioters, in comparison to non-ri oting counterparts, were more racially conscious, had knowledge and access to the political structure, and more likely to have been victims of racial 23

PAGE 24

discrimination. On factors such as educati on, income, and occupati on, the rioters were comparable to non-rioters. This evidence, in additi on to the fact that most riots occurred in cities where political opportunities had been blocked for African Americans, established resource mobilization as the most prominent social move ment model of the 1970s, displacing classical perspectives and the theoretically w eak breakdown model (Useem, 1980). Another study on the sit-in protest phe nomenon of the 1960s found that pre-existing organizational structures contri buted to the success of the mo vement (Morris, 1981). This is contrary to the classical social movement perspective, which posits that protest actions such as sit-ins would have emerged later in the movement as a product of the action. In fact, the sit-ins were found to be the catalysts to later, more highly organized fo rms of protest during the civil rights movement. The resource mobilization model is effective as a modern interpretation of the classical social movement perspective. The incorporati on and necessity of organizational structure and clearly defined goals lends itself to logical explanations of m ovement success. It seems obvious that organizations with effective structures a nd clearly defined, specific goals would succeed in achieving those goals. Thus, social movements th at incorporate these fa cets of organizational success will effectively enact chan ge at the social level. The major caveat with this perspective lies in the absence of identity, culture, and emotion in movement dynamics. As governmental structur es become less relevant and movements focus less on enacting policy change, the rigidity of structural approaches fails to accommodate these more sublime elements. A second fault of this perspective is the assump tion that persons with little or no access to resources and lack of clear organization will not succeed. However, in the instances where 24

PAGE 25

disenfranchised persons with li ttle access to resources do en gage in protest, resource mobilization falls short. This is especially evident in the case of the 1980 prison riots in New Mexico, where prison inmates managed to take ove r a prison resulting in the most brutal and costly U.S. prison riot (Useem, 1980). However, the no social stru cture or leadership that would hint at organization was found by Useem in a 198 0 case study on the riots. This challenges the dominance of resource mobilization, and suggests that factors othe r than structure and influence can drive protest. Political Process Model The political process model, including the political opportunity stru cture, emerged as a social movement theory shortly after the re source mobilization mode l. In many schools of thought, it is considered to be a more politically centered variation on th e resource mobilization model. While both share a struct ural approach to social move ment processes, the political process model focuses more on the availability and exploitation of politic al opportunities. Within this perspective institutions that usually take the form of governments and political bodies hold the majority of power. The four main variables of political opportunities, as defined by Tarrow (1988) are outlined below (McAdam& Snow, 1997, p. 253; Noonan, 1995, p. 83): Degree of openness in the polity Stability and instability of political alignments Presence or absence of a llies and support groups Divisions within the elite or its tolerance for protest Policymaking capacity of the government Many movements originally considered to be of the organizational-type promoted by the resource mobilization model are now seen as mo re political in nature. On further examination, movement goals and actions are influenced and shaped by political structures, movements seek political support and political influence, and changes in policy are considered successes. One 25

PAGE 26

example of this is the case of protests by labor unions; originally seen as an expression of sociopolitical inequity, these movements are be tter explained by viewing unions as political actors attempting to exert influence over the political structure (Touraine, 1985). Recent scholarship has taken note of several th eoretical weaknesses of the political process model when used to examine modern movements. A major weakness is the fact that this theory was built almost entirely on an understanding of the processes inherent in western democracies. Scholars have shown that large-scale social unrest is closely linked to shifts in political power in Western Europe (Kriesi, Koop mans, Duyvendak, & Guigni, 1992) This western orientation makes it difficult to apply the model to areas of th e world with different political structures such as autocracies, monarchies, etc. It has been empirically proven that civil unrest in western industrialized nations is strongly lin ked to changes in the political st ructure. It would be useful to apply this model to the political and social turm oil in newly democratized nations in Asia and Africa, as well those countries striving for po litical change. Unfortuna tely, in the current incarnation, it is not adaptable. This perspective also widely excludes persons with little or no influence over the polity, namely disenfranchised persons such as the hom eless, prisoners, illegal immigrants, or the mentally ill. Despite the seeming incompatib ility with influencing political change, these individuals are often the most likely to feel the effects of changing political structures. Disenfranchised persons are often the most closely ti ed to the political structure in terms of being controlled by it. Homeless persons and the menta lly ill, as a group rely on the government to provide services that a llow them to have their basic human needs met. When political structures change, allocation of resources to aid these groups of people cha nges, and thus their lives are 26

PAGE 27

affected. It seems illogical that disenfranchised persons would not be able to participate in protest just because they have the least influence over the polity. Finally, the lack of accommodation for processes of emotion, culture, and identity in social movements limits the scope of the political process model, particularly with the more recent, apolitical movements to be discussed below. Theoretical Perspectives on New Social Movements The new social movement perspective emerged in the 1980s as a contrasting approach to the previously mentioned approaches to collec tive action and protest. Where social conflict attempted to explain mobilization as based on between group economic discrepancies, new social movement theory used cultural and moral perspectives as well as group identity (Williams, 2006). Manuel Castells Castells approach to new social movements is the most compatible with the early neoMarxist explanations of social movements. This perspective focuses on capitalist structures and dynamics in urban movements, but acknowledges, unlike classica l social movement perspectives, that the ro les of movement actors are based in socioeconomic factors in addition to factors such as culture and identity. Three majo r themes of protest movements were identified by Castells (Buechler, 1995): Modes of collective consumpti on supported by the state. Community culture, defined as maintenance of cultural identity and territoriality, preserving genuine forms of community. Political self-management, defined as aut onomy and decentralization of decision-making processes. For Castells, the locality is th e action point for protest movement s. Much of this theoretical framework is based on an attempt to bring power back to the community level and preserve a 27

PAGE 28

local culture in urban settings. Cities are the centers of mass consumption and the form of community most closely tied to pol itical structures. The state attemp ts to exert influence over the community by diluting and nationalizing local culture, identity, political processes, and economic processes. Urban social movements, acco rding to Castells, are not exclusively fueled by emotion, culture, and identity as many new so cial movements theorists propose. Rather, the socioeconomic struggle and the social-political st ruggle are intertwined but distinct arenas of social unrest. Alaine Touraine Touraines perspective on new social move ments is heavily focused on culture and the battle over control of society in a post-industrial world. In this perspective, protest has shifted from the economic to the cultural realm. Because the cultural realm is not concrete like the economic realm, and contention in the cultural r ealm is more nuanced, collective action is often more difficult to organize and t hus takes newer forms such as struggles for identity (Touraine, 1985). In addition, Touraine suggests that society as a concrete sociological concept should no longer be a part of the examination of human social life, rather, social movements should be used to understand human social behavi or (Simsek, 2004; Touraine, 1988). Touraines body of research focuses primarily on the workers movement in France and later, the Polish solidarity movement. His work with these social movement groups led him to propose that the role of new social movements in postmodern society is to create spaces for two new classes, consumers and clients vs. managers and technocrats, to construct both a system of knowledge and the technical tool s that allow them to intervene in their own functioning (Buechler, 1995, p. 444). Touraine us es the term historicity to refe r to this process of knowledge and tool development. Conflicts in modern so ciety are based in the opposing interests between these groups. A major theme in Touraines work is what he terms the privatization of social 28

PAGE 29

problems: making social issues re levant at the individual level, therefore creating an emotional relationship to the issue and perhaps developing culture and identity based on this emotional connection (Buechler, 1995). The struggle for individuality within Touraines perspective is typical of the new social movement theories. Rather than viewing protest groups as inherently cohesive, groups are seen as composed of individuals struggling to mainta in and develop identities in an increasingly monotonous social field. This perspective is echoe d in the work of other new social movement theorists, most notably the work of Alberto Melu cci, to be discussed in further detail below. Jurgen Habermas The influence, or colonization, of the indi vidual and personal environment by the media, money, and social power is the focus in Haberm as perspective on new social movements. He posits that these actors are regulating the lifewo rld to such an extent that it goes beyond the economy and government, infringing upon culture, identity formation, and other forms of symbolic reproduction. This perspective focuse s the most on post-industrial elements of consumers versus producers. With the expanding nu mber of roles an individual can take on, the producers and governments take more control over individuals because they create and alter these new identities (Buechler, 1995). Habermas identifies two characteristics of ne w social movements. First, he asserts that social movements are defensive in nature, and do not generally bring about broad social change. This goes against most early th eoretical perspectives, as clas sical perspectives, resource mobilization, and political process models all propose the purpose of social movements is some broad-based social change, soci oeconomic or otherwise. Seco nd, the conflicts on new social movements are rooted in changes in the quality of life and self-re alization, with goals primarily focused on movement participation and identity formation. The foci of new social movements 29

PAGE 30

according to Habermas framework are based on cultural reproduction, social integration, and socialization (Buechler, 1995, p. 446). That is, individuals struggle to find meaning by forming identities, finding means of participati on, and improving the in dividual lifeworld. Alberto Melucci Alberto Melucci is the most widely recognized of new social move ment scholars. Most often, references to new social movement theories in research are heavily or exclusively based in Meluccis framework. Of all four perspectives ex plored here, Meluccis is the most explicitly related to emotion, culture, and identity format ion. Where other theorists focus heavily on the struggle between the individual and the system in a postmodern society, Melucci takes the examination a step further to explain that the ro le of new social movements in society is to convey messages about the rigid nature of inst itutional power practices to propose alternative methods of social functioning through identity and culture (Buechler, 1995). The conflicts of daily life and the importance of submerged networks are the main foci of Meluccis perspective. He posits that the major fau lt with earlier theories of social movements is the false belief, stemming from th e Marxian roots of social moveme nts, that movement actors are a unified force with a single, solid collective id entity. Instead, he explai ns, movement actors are drawn together by personal, indi vidual-level interactions base d on the everyday misfortunes and injustices of modern life (Mel ucci, 1988, 1989; Mueller, 1994). Subm erged networks play a role in the creation of collective identities, asse ssing social grievances and determining the effectiveness and usefulness of collective action (Mueller, 1994) According to Melucci, new social movements rely heavily on symbolic me ans and actions, with action itself a value and possibly a goal of social movements (Simsek, 2004). Mueller (1994, p. 342) succinctly outlines the three major characteristics of Meluccis framework: 30

PAGE 31

The content or outcome of the process of social construction, the collective identity of the movement that comes to exist as a part of the movement culture The social processes by which the collective id entity is created in submerged networks of small groups concerned with the ongoing routines of everyday life The emotional investments that enable individuals to recognize themselves as the we in a collective identity. Meluccis framework most neatly explains th e ultimate role of new social movements in postmodern society. By placing value on movement participation processes such as identity formation and symbolic action, new social movements create a space where individuals can both assert their rights to individuality and in some cases, political and social equity, as well create social spaces and alternative means of so cial action within those social spaces. Culture, Identity, and Emotio n in New Social Movements New social movements have three key components that were ignored by previous formulations of social movement theory. Culture, emotion, and identity are closely tied components of and often influence each others presence in within new social movements. The study of culture and identity in social movements is based heavily on the perspectives on new social movements of Alaine Touraine and Alberto Melucci. As explained above, Touraine and Melucci both emphasized the sh ift in social movements since the 1970s to movements based in the cultural sphere, where individuals attempt to de velop identities through common grievances against the perception of impo sed autonomy from those in power. The role of identity development in new social movements has been the most researched aspect of new social movments over the past ten years. New c oncepts such as identity movements and identity politics have emerged in an attempt to explai n how movement actors de velop group identities, and how these are used to streng then movements. However, the role of identity in social movements has been recognized since the inception of social movement research. Theorists now 31

PAGE 32

agree that participating in social movement s through activism, mobilization, and protest can significantly and permanently alter a movement particip ants sense of identity. Mueller (1994) explores the role of collective identities in the U.S womens movement using Meluccis perspective that part of the purpose of social m ovements is in creating collective identities for movement actors. She explores th e creation of new feminist identities following the split in the American womens movement after the Womens Equality Day Strike in 1970. The study found that creative tensions within the gr oup led women to forge new identities based on grievances both within and outside the m ovement. Identity was central in the womens movement as a whole because women sensed a discrepancy between their high levels of education and the low social status granted to them by society. Simsek (2004) explores the womens movement in Turkey from Meluccis perspective. The womens movement in Turkey did not pick up until the 1980s, a decade or more after womens movements in western na tions. Simsek asserts that feminism in Turkey, despite the varied types of feminist movements, maintains a post-industrial, identi ty-based focus, thus qualifying as a new social movement; all of th eir demands are basically of a post-material nature, and call for identity politics (Simsek, 2004, p. 127). More recently, a study by Cherry (2006) ch allenges the assumpti on that identification with a social movement group is essential to ac tivism and engagement in movement goals. This study on vegans found that those vegans who were not members of vegan groups were stricter in their lifestyle than thos e participants who were members. The findings of this study support the idea that group identity itself can be the str ongest motivator towards personal change as a member of a cultural group. This challenge s the resource mobilization proposition that movements must be highly organized to succeed Additionally, Cherry f ound that the pervasive 32

PAGE 33

identity need not be the dominant identity in the indi viduals perspective; most of the strict vegans in his study identified primarily as punks, while the less strict vegans who were members of vegan organizations, identif ied primarily as vegans. In a study of Jewish resistance fighters in Nazi occupied Warsaw, Einwohner (2006) explores the role of high-risk id entities in social movements. Hi gh-risk identities are forged in environments where identifying as a group member can create personal danger for the movement participant. Despite the danger inherent in id entifying as a group member, high-risk identities can often forge stronger connections, particularly emotional connections between group members due to the perception of a struggle to eliminate the danger and make the identity acceptable. The role of emotion in social movements is inhe rent in the new social movement theories, but never was considered a catalyst for social m ovement action until recently. The new emotional movements emerged as scholars noticed the se emingly spontaneous organization of people following tragic events. Trigger events for emoti onal movements involve the victimization of an innocent person as a result of senseless viol ence by another individua l or group of persons. Walgrave & Verhulst (2006) conducted a study to compare four movements involving mobilization of common people following tragic events. The movements compared were the White Movement in Belgium, the Million Mom March in the U.S., the Snowdrop Campaign in the U.K., and the Movement Against Senseless Violence in the Netherlands. It can be concluded that while emotional movements may not be a new movement type, they can certainly be considered new social movements based on the a bove characteristics. The researchers found that four characteristics were presen t among the four movements: A basis in victims activism and identification mechanisms Low levels of organization capacity 33

PAGE 34

Support from the mass media Endorsement by elite groups, including politicians. Critical Components of New Social Movement Theories Group Identity The inclusion of group identity as a necessary component of social movements is one of the major differences between the new social m ovement perspective and older social movement theories. Because culture and emotion play so heav ily into the advent of new social movements, group level change often times seems more passi onate and value-driven. Despite the importance of identity development in new social movements, research remains highly descriptive and insulated within the social movement scholarship. Definitions of group identity vary between disc iplines, but they share a common thread of a personal cognitive connection between the individual and the work group (Riordan & Weatherly, 1999). Social identity theory defines soci al identity as that pa rt of an individuals self-concept which derives from his knowledge of his membership of a social group together with the value and emotional significance attached to that membership (Tajfel, 1978). While these definitions are similar, the second includes the emotional aspect of group membership that is central to new social movements. The literature on group identity has been dom inated by studies of small group action and beliefs. However, the idea of collective identity has been used by sociologists since the 1980s to fill the gaps in dominant resource mobilization an d political process models, (Polletta & Jasper, 2001) both of which were dominant social movement models at the time. Those models explain in great detail the how of soci al movements, but do not explai n the why, (Polletta & Jasper, 2001) which is the reason values and culture under the umbrella of identity surfaced as a way to bridge this gap. 34

PAGE 35

It is clear that collective identity is central to soci al movements (Bernstein, 2005; Einwohner, 2006), but New Social Movements formally incorporate identity into its theoretical base rather than using it to paste together the structures a nd processes described by previous models. The term identity politics has recently en tered the lexicon of social movement studies; while debate exists over exact definition, it is an expansion of the concept of class politics. Instead of ones central identity being based in social class a nd the Marxist class structure, identity politics focuses on the salie nt identities that were present in the social movements of the 1960s and 1970s, which seemed to be more concerned with culture and identity than with challenging the class structure (B ernstein, 2005, p. 49). In this se nse, identity politics has been defined as the belief that identity itselfits el aboration, expression, or affirmationis and should be a fundamental focus of political work (Bernstein, 2005, p. 49; Kauffman, 1990, p. 67). Many models and instruments have been deve loped to measure group identity, but few of them have been proven empirically sound by fact or analysis technique s (Cameron, 2001). It is widely acknowledged that group identity is multi -dimensional, but the dimensions used have varied between models (Cameron, 2001; Deaux, 1996; Ellemers, 1999). The dimensions that were derived from Tajfels So cial Identity Theory were: aw areness of group membership, group evaluation, and emotional aspects of belonging (Brown et al., 1986; Hinkle et al., 1989). However, these dimensions proved difficult to define, and even more difficult to measure because they represent broad and overlapping con cepts of group identity. In response to the lack of empirical data concerning th e specificity the dimensions of group identity, Cameron and his colleagues developed a 3-factor model of social identity that includes the dimensions: cognitive centrality, ingroup affect, and i ngroup ties (Cameron & Lalonde, 2001). This model has been 35

PAGE 36

tested among various social gr oups, and thus has been determin ed empirically sound (Boatswain & Lalonde, 2000; Cameron et al., 1997; Obst, Zinkiewicz, & Smith, 2002). Group Cohesion While sociology has embraced the importance of collective/group identities, the concept of group cohesion has not been addressed for large-scale social change processes. Within the community studies realm, studi es of community and neighborhood cohesion have surfaced to examine perceptions of cohesion (Brown & Brooks, 2006; Riordan & Weatherly, 1999). However, the concept of group cohesion has widely been left to social psychology and has been used most often to evaluate the effectiven ess of workgroups (Kipnes & Joyce, 1999). Group cohesion can be defined as the degree to which an individual believes that the members of his or her work gr oup are attracted to each other, willing to work together, and committed to the completion of the tasks and goa ls of the work group (Bass, 1960; Riordan & Weatherly, 1999, p. 312; Stogdill, 1972). Social cohesion, including group cohesion, has recently been identified as an integral part of the empowerment process (Speer, Jackson, & Peterson, 2001). New Soci al Movements foster empowerment through the creation of strong cultural identities, but a sense of perceived cohesion is necessary among group members for the group to feel they have the degree of efficacy necessary to enact change. The most recent studies on cohesion as a constr uct have highlighted the need to redefine cohesion as a more complex and multidimensional variable than it has been in the past. As a result, researchers have concluded that measures of cohesion should be developed congruently with the theoretical construct. In other words, the context of cohesion mu st be considered for appropriate measures to be used (Kipnes & Joyce, 1999). In a study by Kipnes (1998), it was found that levels of cohesion varied dependent upon the usage of indivi dual or group-level 36

PAGE 37

instruments. This is likely due to the lack of conceptualization concerni ng the affect of other group dynamics including identity, leadershi p, and power in the context of cohesion. While many of the current models are similar, disagreement on the dimensionality of cohesion has rendered researchers unable to de velop universally accepted models (Kipnes & Joyce, 1999). It is likely due to this fact that the plethora of frameworks and instruments available have not been replicated or empirically tested enough to create a dominant paradigm in cohesion research. Goal Setting and Achievement The effect of goals in various situations including depression, achievement (in both educational and work settings), and personality, has been examined in current research. There are various frameworks that have been adapted to suit particular fields of study, but in each area the basic idea is the same: setting goals motivates people. Achievement goal theory examines the way people function when places in an achie vement setting, (Baranik, Barron, & Finney, 2007) whether this is a work-group, academic envi ronment, or social movement organization. It suggests that goals, and therefore the outcomes of those goals, vary among individuals placed in achievement settings. There are currently four type s of goals identified by the Achievement goal framework: performance avoidance (PAV), performance-appr oach (PAP), mastery-avoidance (MAV), and mastery-approach (MAP). These four domains, fro m Baranik et al. (2007), are outlined below: Performance approach: striving to demonstrate competence relative to others Performance avoidance: striving to avoid incompetence relative to others (Elliot & Harackiewicz, 1996) Mastery approach: focus on developing his or her competence or mastering a task (Ames, 1992; Elliot, 2005) 37

PAGE 38

Mastery avoidance: focus on avoiding self-referential or task-referential incompetence (Elliot, 2005) This model has evolved from an early 2-fact or model that recognized general mastery and performance goals. Individuals formulated mastery goals to develop competence and skills, whereas performance goals were formulated to de monstrate competence in comparison to others (Barron & Harackiewicz, 2001). Research using the original two-factor framew ork indicated that individuals tend to react differently to achievement situations relative to th e type of goals they se t in the situation (Ames & Archer, 1988; Barron & Harack iewicz, 2001; Dweck & Legget, 1988; Nicholls, 1984). It was originally believed that mastery type goals resulted in higher achievement; research had shown that individuals pursuing mastery goals were mo re productive when faced with difficult tasks and were more positive about the achievement si tuation than those pursuing performance goals (Ames, 1992; Ames & Archer, 1988; Barron & Harackiewicz, 2001; Elliot & Dweck, 1988; Nolen, 1988). However, a second perspective, co mbining performance and mastery goals were used, was advocated by researchers and known as th e multiple goal perspective. Currently, it is unclear which perspective is the most beneficial as several outside factors seem to affect achievement outcomes in addition to goal orientation (Barron & Harackiewicz, 2001). Elliot and colleagues (Elliot & Church, 1997; Elliot & Harackiewicz, 1996) introduced the distinction between avoidance and performance a pproaches as a way to differentiate between performance-avoidance goals a nd performance-approach goals. The model was empirically supported until the introduction of the 2 x 2 Goal achievement framework, which split the mastery category into two categories (Ellio t & McGregor, 2001). Research supporting the validity of this model has been strong (Baranik et al., 2007), and provide s a balanced framework for understanding goals and their effects on achievement. 38

PAGE 39

Recently there has been a shif t from examination of individual goals to examination of group goals. This is likely a response to the accu mulation of research in work-settings, and the fact that most organizational goals are fo rmed by groups, not individuals (Austin & Bobko, 1985; OLeary-Kelly, Martocch io, & Fink, 1994). Goals influence group processes by encouraging division of labor, st imulating coordination of effort aiding in appraisal of group processes, and helping groups decide what task s must be accomplished (Ansoff, 1965; Klein & Mulvey, 1995). However, it has been noted that the models used to examine individual goal setting may not be accurate for groups, because elements of group dynamics such as cohesion, acceptance of norms, and social loafing have significant influence on both goal-setting and achievement (Austin & Bobko, 1985; Ilgen, Shapir o, Salas, & Weiss, 1987; Klein & Mulvey, 1995; Weldon & Weingart, 1993). The research a bout group goals has demonstrated a complex relationship between cohesion, group goal diffi culty, group goal acceptance/commitment, and group performance (Ronan, Latham, & Kinne, 1973; Zander, Forward, & Albert, 1969). However, it is evident that highly cohesive groups tend to be more committed to goal achievement and tend to devise relatively difficult goals, which in turn lends itself to high levels of goal achievement (Klein & Mulvey, 1995). Purpose of Research Despite the wide range of cu rrent social movement research, there seems to be a discrepancy in research concer ning the social viab ility of fringe gr oups, specifically the homeless. While considerable research is av ailable about more visible movements, the vagabond populations have not been c onsidered a viable social force. In recent years, several studies have surfaced which examine groups of homeless throughout the U.S. (Swithinbank, 1997; Williams 2005), from both a broad social movement perspective and as a new social movement. These studies have found that although homeless 39

PAGE 40

groups lack some of the characteristics of social movements, particularly organized protest, other factors such as clear organization, goals, and access to resources classify these groups as social movement organizations from a resource mobilization perspective. Because homeless persons are a group who traditionally has little access to resources and ab ility to mobilize, this finding contradicts the traditional view s of social movement groups. The purpose of this research is twofold. Firs t, I would like to determine if the homeless population in Gainesville, Florida can be consid ered a social movement group, drawing heavily from the culture and identity based new social movement frameworks proposed by Touraine and Melucci as well as many of the structural elements of the resource mobilization model. I will determine this by measuring the presence of the following precursory char acteristics of social movements: group identity, group cohesion, and goal orientation. The second purpose of this research is to explore how well the proposed model explain the observed relationships among these constructs Model for New Social Movements Below is a proposed model for new social move ments that is outlined in Figure 2-1. I have chosen to name this model a new social move ment model because it focuses heavily on the presence of individual characterist ics such as group identity and goal orientation rather than the group level, organizational charact eristics inherent in the reso urce mobilization and political process frameworks of social movements. Th e element of group cohesion has been included because a group level measure of connectedness s hould be included for a movement to be one entity engaging individuals for action,. The outcome of the model is action capacity in terms of the creation of a new social movement group. Action in this case is de fined not as collective action or protest, but as the crea tion of culture and iden tity spheres where m ovement participants 40

PAGE 41

can improve their quality of life through interaction, identity fo rmation, preservation of culture, legitimization of emotion, and eventu ally broad-based social change. Figure 2-1. Proposed new social movement model. 41

PAGE 42

CHAPTER 3 METHODS Research Design A cross sectional design was used for the firs t stage of this study. Cross-sectional designs are appropriate when the data to be collected requires no time dimension and the data rely on existing differences between groups (DeVaus, 2001). Cross sectional designs are used to determine if variation between tw o or more variables exists for two or more comparison groups (Bryman, 2004). In this study, the presence of group cohesion, group identity, and goal orientation among short-term and chronically homele ss individuals in Gaines ville at one point in time were compared. In the second stage of data collection, a qualitative interview was conducted to further investigate the presence of th e research variables among the homeless population. A case study design was used as no comparisons between groups were made a nd the overall purpose of the interview was to gain more descriptive and complex explanations on the presence of group identity, group cohesion, and goal orientat ion among the homeless population as a whole (DeVaus, 2001). Sample Selection The theoretical population for this study include d all homeless persons residing in Alachua County, Florida. The sampling frame was the 952 identified homeless persons in Alachua County, Florida as of January 2007. The antic ipated sample size for this study was approximately 80 homeless persons total, with 40 per comparison group. The two comparison groups were chronically homeless persons, those who have been homeless three or more years, and short-term homeless persons, those who have been homeless less than three years (HUD, 2007). Placement in either group was determined by a demographic questionnaire distributed 42

PAGE 43

prior to the other research instruments. The size of 40 per comparison group was chosen to ensure that regression analysis could be c onducted on the variables. Because the homeless population in Alachua County fluctu ates significantly from year to year and seasonally, it is difficult to determine a sample size based on the estimated sampling frame. A referral sampling method was used to recruit participants. After completing the quantitative portion of the study, part icipants were asked to list five other homeless persons who may be interested in participating and indica ting where these people could be found. The first contacts were made at the downtown plaza. The target number of 80 total participants was not achieved, due mostly to the fact that many partic ipants refused to refer others. There were 63 total respondents. Forty-five responses of one index and two scales were complete enough for quantitative analysis. The other 18 responses were not adequate because they were incomplete or improperly completed. Participants were sampled at St. Francis House soup kitchen and shelter during lunch on different days of the week and at the downtown plaza in Gainesville during different days and times. These sites were chosen for safety purpos es, as they are well-supervised and public locations. Other locations where homeless people spend time, such as the tent-city and womens shelters, as well as other soup kitchens, were excluded because of safety and privacy. Despite excluding these sites, St. Francis House and th e downtown plaza provide an excellent sampling frame because homeless persons as well as homeless experts assert that most homeless people in Gainesville spend time at St. Francis Hous e for lunch or at the downtown plaza. Data were collected at St. Francis house duri ng the general public lu nch, which is attended by 150-200 persons each day. Attendees change dail y. In the evening, St. Francis House provides shelter for 30 homeless men, women, and childre n. This group was sampled during one after43

PAGE 44

hours visit. At the downtown plaza, data were collected during mid-morning on the weekend, during the weekly farmers market Wednesday afternoons between 3pm and 7pm, and during a weekly dinner provided by a local organizat ion called The HomeVan. Approximately 200-300 persons attend the HomeVan dinner each week, and, as with St. Francis House, the group attending often changes. Many homeless person s who live in tent cities come to HomeVan handouts because clothing and othe r supplies are often provided. Instrumentation Group Identity Cameron's three factor model of social iden tity was used to measure group identity among the homeless community. The three factors used by Cameron to develop this instrument are ingroup ties, centrality, and ingroup affect. These are defined below (Cameron, 2004, p. 241): Ingroup ties: perceptions of similarity, bonds, and belongingness with other group members Cognitive centrality: the amount of time spent thi nking about being a group member (Gurin & Markus, 1989) Ingroup affect: the positivity of feelings associated with being a group member Three different indices were us ed to test the validity of the model, including the goodnessof-fit index, the incremental fi t index, and the comparaive fit i ndex. Five trials were performed, and each trial produced values of .90 or higher, wh ich is the standard acceptable cut-off for a good model (Cameron, 2004). A reliability test was performed on the group iden tity scale at the end of data analysis. The data from this test are shown in Figure 3-1. Th e first test on the full 12 item scale produced a Cronbach's alpha of 0.562. Items six and eight were removed for a second analysis, which slightly improved the alpha to 0.625. The scale was shortened twice more, but neither test 44

PAGE 45

produced significant changes in the alpha score. The original 12-item scale was used for data analysis. Group Cohesion Group cohesion is a latent characteristic of gr oups that is widely i gnored in research on social movements and even group identity. Desp ite this, it is a major component of group dynamics and ability to mobilize, as established in the literature review. For this reason, I chose to use an index to measure cohesi on in conjunction with a set of interview questions to probe the more complex dimensions of cohesion. An index is an appropriate measure for gr oup cohesion because this construct includes several dimensions that must be measured to ge t a complete picture the level of cohesion present (Bryman, 2004). According to Kipnes & Joyce (1999) the debate over whether cohesion is to be treated as unidimensional or multidimensional is prevalent in cohesion research because of the problem that cohesion can be measured at both th e individual and group leve l. Indices also allow a researcher to get true measures of characteristics which peop le cannot readily or accurately communicate in a questionnaire or interview. Further, much of the research on group cohe sion thus far has been done with scales and indices, (Kipnes & Joyce, 1999). This lends itself to high congruent validity because I can borrow dimensions from previous research to enhance my own instrument. This is important because the debate over whether cohesion is unidimensional or multidimensional raises questions about instrument va lidity (Kipnes & Joyce, 1999). There are limitations to the scope of data collected with indices. For this reason, I developed qualitative interview questions that expounded upon certain aspect s of group identity that are difficult to identify in an index. These aspects include: 45

PAGE 46

Differentiating between the two (or three) comparison groups and which groups the participants are answering the questions for Judgments about homeless people in general or about specific types of homeless people Finding out how homeless people define a homeless person, a nd whether they are using this definition to base their responses on the inde x or if they are using a different definition Qualitative interviewing allows data to emerge from interaction with the interviewee, with each interview giving new depth and meaning to the data (Rubin & Rubin, 1995). The interviews allowed me to derive more distinct mean ings from respondents concerning group cohesion. Qualitative interviews emphasize the social c ontext of constructs (Warren, 2001), and as cohesion is both an individual and a social characteristic, an in terview is an ideal method for gaining that contextual informati on which is missing from an index. Development of group cohesion index A number of scales and indices exist to measure group cohesion under different circumstances, but the research is focused mostly on the areas of work-group cohesion and neighborhood/community cohesion (Kipnes & Joy ce, 1999). These scales and indices are not easily adaptable for work with marginalized groups Most of these measures fail to discriminate between group cohesion and group identity. Upon close examination, many of these instruments prove to measure a combination of both constructs. Because I chose to measure both independently, it was important for both of the in struments to carefully discriminate between these two variables to get an accurate measure of the presence of either or both variables among homeless persons in Gainesville. In developing this index, I used dimensi ons of group cohesion as indicators for the presence of group cohesion and developed statemen ts based on these dimensions. Teh statements were rated by participants according to agreement and disagreement with the statement on a scale from strongly disagree to agree, with corresponding levels of one through five. After 46

PAGE 47

completion, scores were given to each participan t based on their ratings. This is known as the group cohesion score. When developing an index it is important to consider construct vali dity, content validity, and concurrent validity (DeVellis, 2003). Constr uct validity ensures that a measure is valid across several methodologies, (Carmines & Zeller, 1979; Weber, 1985) which in turn eliminates the limitation of having only one method to use fo r data collection. While scales and indices are primary indicators of group cohesion in the literature, these scores are often used in conjunction with qualitative data to create a more comple te understanding of the information gathered. Content validity ensures that the el ements of an instrument are re levant and representative of the construct being tested (Carmines & Zelle r, 1979; Haynes, Richard, & Cubany, 1995). For indices, content validity must be considered from the very beginning. A complete understanding of the construct being tested is necessary, or the instrument will surely be low on content validity (Carmines & Zeller, 1979; DeVellis, 2003; Sommer & Sommer, 2002). For this reason, I consulted the literature on group cohesion to ensu re that I had a complete understanding of how this concept has been used. As is to be expecte d, the dimensions indentified as relevant for my exploration of group cohesion are different from those identified by previous researchers. This difference is based largely on the differences between the groups researched in each study and whether the authors took the unidimensional or multidimensional approach. However, the reason that so many scales and indices exist for group c ohesion is because they are used in different settings and address cohesion at different dimensional levels. For my instrument, I found it important to include both the gr oup and individual pers pective for cohesion to ensure construct validity. New social movements, by theoretical definition, are based on in dividual feelings of 47

PAGE 48

connection to the group as well as a group dynamic that are an aggregate of the group members' commitment to the group. I achieved concurrent validity by adapting and re structuring existing sc ales and indices for my particular group, the homele ss population in Alachua County. The three measures I borrowed from the most were Buckner's neighborhood co hesion index (Brown & Brooks, 2006; Buckner, 1988), Seashore's group cohesiveness index (Mil ler & Salkind, 2002; Seashore, 1954), and the perceived cohesion scale (Bollen & Hoyle, 1990; Chin, Salisbury, Pearson, & Stollak, 1999). Table 3-2 lists the major dimensions of cohesi on explored by these three instruments. Several dimensions identified by these indices and scales overlap or are too cl osely related to group identity. To ensure the greatest degree of c ontent validity without si gnificantly compromising concurrent validiity, I had to combine some di mensions and eliminate others. Based on the current literature on group cohesion research, I have identified several key dimensions to group cohesion, which are listed below. Attraction to other group members Willingness to work together Commitment to completiom of or development of tasks and goals Shared values and norms Presence of social support networks Sense of belonging and attachment Feelings of trust and so lidarity among members Repsect and tolereance among group members Collective action Sense of morale Several measures must be taken to pre-test an index before it is used to collect research data. Outlined below are the testing steps which were taken to ensure validity, reliability, and precision of the index (Weems, 2001). 48

PAGE 49

Informal test using an expert panel. For th is, I indentified about ten people who have worked with the homeless in Gainesville, asked them to look over the index, and make suggestions as to the relevance of the statements or any statements which they feel should be included. A pretest with a group of people to make sure that the instructions a nd statements are clear and straighforward. A pretest with a representative group to make sure it works as intended as well as to determine any midpoint bias and steps to be taken to ensure maximum validity of findings. During Step 1, I eliminated the dimension commitment to completion of tasks and goals because it seemed to overlap with some of the other dimensions. I also added several statments to create a more inclusive instrument. Pretesting of group cohesion index A reliability test was performed on this index after 12 surveys were completed. The results of the test are shown in Table 3-3. The test pr oduced a Cronbach's alpha score of .918. Items 3, 8, 15, 17, and 22 had the lowest item total correl ations; all were below .43 and item 17 actually had a negative score. A second te st was performed after removi ng the five low scoring items. This raised the Cronbach's alpha score to .934 and item total correlations for all items were above .46. After this test it was determined that the full 22-item index should be used for the duration of the study because the alpha change wa s not significant enough to warrant removal of any items. A final reliability test was performed once the data collection was completed. The results of this test are shown in Table 3-4. The full index produced a Cronbach's alpha of 0.900. A second reliability test was run after removing items 1, 15, and 22, which had the three lowest reliability scores. The Cronbach's alpha produced by this test was 0.907. Because the difference was not significant, the full 22-item i ndex was used during data analysis. 49

PAGE 50

Goal Orientation An adaptation of the 2 x 2 framework of achieve ment goals for a work domain was used as the measure for goal orientation (Elliot & McGr egor, 2001). The four factors accounted for within this framework are mastery-approach (M AP), performance approach (PAP), performanceavoidance (PAV), and mastery-avoidance (MAV). Definitions of each of these factors from Baranik et al. (2007, p.298-99) are below: Performance approach: striving to demonstrate competence relative to others; Performance avoidance: striving to avoid incompetence relative to others (Elliot & Harackiewicz, 1996); Mastery approach: focus on developing his or her competence or mastering a task (Ames, 1992; Elliot, 2005); Mastery avoidance: focus on avoiding self-referential or task-referential incompetence (Elliot, 2005) Twelve items in this model were taken from VandeWalle's 2 x 2 framework that covered the first three factors (MAP, PAP, & PAV). Six additional items (adjusted from 11 originally) were generated by Elliot & McGregor to account for the master-avoidance approach, making the measure an 18-item scale. Reliability scores usi ng Cronbach's alpha have all been reported at or above .78 (Baranik et al., 2007). Correlations for each of the four factors found that they were related, yet distinct (Baranik et al., 2007). Several confirmatory factor analyses have supported the validity of this framework (Baranik et al 2007; Elliot & McGre gor, 2001; Finney, Pieper, & Barron, 2004). A reliability test was performed on the goal orientation scale at the e nd of data analysis. The results of this test are shown in Table 3-5. The full 17-item scale produced a Cronbach's alpha score of 0.907. A second test was performed after removi ng low scoring items one and 50

PAGE 51

fifteen. This test produced a Cronbach's alpha score of 0.923. Because the alpha score was not greatly improved, the full 17-item scale was used for data analysis. Qualitative Interview The fourth piece of instrumenta tion consists of an interview that lasted no longer than 30 minutes. There are limitations to the scope of data collected with indices. For this reason, I developed qualitative interview questions that expounded upon certain aspect s of group identity, group cohesion, and goal orientation that are di fficult to identify in an index. Qualitative interviewing allows data to emerge from intera ction with the interview ee, with each interview giving new depth and meaning to the data (Rubin & Rubin, 1995). Through the interviews, I was able to derive more distinct meanings from respondents concerning group cohesion. Qualitative interviews emphasize the social context of c onstructs (Warren, 2001), and as the research variables are both individual a nd social characteristics, an in terview is an ideal method for gaining the contextual information that is missing from an index. The questions included in the interview addressed group cohesion, group identity, leadership, and goal orientation. The interview questions built on information garnered from the group cohesion index, and were therefore open-response format and in an attempt to get participants to be more explicit about their fe elings concerning each of the dimensions of group cohesion. Several considerations must be ma de when developing appropriate interview questions, particularly when working w ith marginalized groups, including: Sensitivity of topics, particularly when it comes to reasons for homelessness and history as a homeless person (Lee, 1993; Tortu, Goldsamdt, & Hamid, 2002). Only one of the questions concerning group cohe sion addressed this aspect, bu t it is still an important consideration, especially if I want to achieve precision and validity in responses. 51

PAGE 52

Silence and the need for probing questions to improve validity by making sure participants have enough time to frame their responses, a nd questions properly ai med at getting the right kinds of answers (Poland & Peders on, 1998). Knowing how to treat silence can greatly improve the validity of interview data by assuring that the respondent was given adequate opportunity to respond and formulate their response. Insider-outsider problems, es pecially since I clearly am not a homeless person (Ryen, 2001; Tortu et al., 2002 ). Informal socializing and dress are especially important for this group. Keeping a consistent rapport with pa rticipants increases both the validity and reliability of data. Social meaning and influence of interview location (Herzog, 2004). Interview location can greatly affect precision, reliabi lity, and validity. If the partic ipant is comfortable at the location, then they will be more likely to gi ve detailed and accurate answers, increasing reliability and validity; precision is increased if each participant is comfrotable in the chosen setting. To build all of these considerations into a valid, reliable, and precise interview instrument, I used a semi-structured format, where most main questions and many probing questions are determined before the interview, but improvisation on the part of the interviewer to elicit more information from the respondent is allowed. Theref ore, my interview schedul e is structured with specific questions organized by broad topic, with spaces allowing for additional questions to be noted. Pretesting of interview instrume ntation included the following steps: Cognitive testing Individual (myself) With colleagues Pretest with group repres entative of sample During Step 1, I was advised to develop addi tional probing questions and to clarify some of the wording of statements. This was the only feedback received from participants in the interview pre-test. During all testing phases it is important to get feedback from participants as well as keep individual notes as to the types of answers and adequacy of answers for each stage. Each stage will result in the elimination and/or reconstruction of interview questions based on feedback and individual observations (Willis, 2005). 52

PAGE 53

Data Collection Data was collected over six weeks from Fe bruary to March of 2008. The location of research conducted on fringe groups and sensitive topics plays a significant role in the quality of data and willingness to participat e of prospective partic ipants (Herzog, 2005). The sites for data collection were chosen based on the safety of the researcher and the ease of access to homeless persons. Data were collected primarily at the St. Francis house and at the downtown community plaza. The time of day of data collection varied to reach a more diverse group. Sites such as tent city and the other homeless camps were excluded from data collection because of the danger present in visiting those sites and the belief that most homeless persons in Gainesville eat at St. Francis house or spend time downtown even if they live in a camp. Data Analysis Two comparison groups were determined using demographic data from a brief questionnaire that included que stions addressing length and toni city of homelessness. The two groups which were expected are chronic homeless, those who have been homeless three or more years, and short-term homeless, those homeless less than three years. The emergence of two groups relied heavily on the make-up of the cu rrent homeless population in Gainesville. Recent surveys conducted in the city only differentiate between chronically ho meless persons and those who are not considered chronically homeless (A lachua County Coalition for the Homeless and Hungry & Gainesville Office on Homelessness, 2007). Quantitative Data T-tests for independent samples were run on the independent variab les of group identity and group cohesion to determine the presence of statistically significant differences between the two comparison groups. I have chosen to perform a t-test on these data because the index will produce interval data that is comparable after scores are totaled. Because we will only have two 53

PAGE 54

comparison groups, a t-test is ideal as it can only be used for two groups (Miller & Salkind, 2002). A Mann-Whitney u-test, a non-parametric al ternative to the t-test was performed on the outcome variable of goal orientation because the data were not normally distributed. In addition to the means tests, a Spearman rank-order correlation and a multiple regression analysis were performed. The correlation was used to test the relationship between group identity and group cohesion. The regression an alysis was used to determin e the predictive relationship between group cohesion and group identity with the outcome variable of goal orientation. Qualitative Data An interview processing form was used outli ning the key points I wish to obtain from each interview (Miles & Huberman, 1994). The pr ocessing form was not limited to keywords and phrases because silence is often an important part of the story in qualitative interviews (Poland & Pederson, 1998). Includ ing nuances such as silence and other non-verbal commands increases the precision of the inst rument by making sure that the en tire message is received from the participant. Limitations The scales used to measure group identity a nd goal orientation were created and tested for use with more structured and smaller group s that the homeless population in Gainesville. Although both instruments were adapted and tested, it is possible that these measures do not fully reflect the scope and presence of group identity and goal or ientation among homeless persons. Participants considered several items on the group iden tity scale irrelevant and bad items. One notable example is the item I enjoy being a homeless person. The scale does not account for the negative impact of group member ship on the lives of individuals despite the fact that they may indeed display high levels of group ident ity otherwise. Being homeless is generally a negative experience, even for those persons who are homeless by choice. Therefore, despite the 54

PAGE 55

fact that a homeless person may strongly identity as homeless and feel strong connections with other homeless persons, items related to enjoymen t of group membership provide contrary data. Intoxication, mental health, a nd education status may affect the quality of data collected. While I attempted to filter for extreme mental illness and intoxication, it is often difficult to determine a persons cognizance. Many participan ts reported mental health diagnoses and substance abuse histories, but none appeared too intoxicated or me ntally unstable to accurately complete the instruments. Some participants were unable to understand the statements on the instruments without clarification. Only two part icipants were unable to read and required assistance from the researcher. The smaller than expected sample size limits the generalizability of the data. The referral sampling method proved difficult to accomplish with this population. Many participants refused to provide referrals or were unable to tell me where I could find the people they were referring. Safety issues and transience among the population also limited the access to homeless persons because of the exclusion of certain sites and pe rsons who live in Gainesville only part of the year. Because of these limitations, it is impossi ble to determine how deeply into the homeless population I was able to reach and how representa tive the study sample is, despite demographic similarities to the overall population. The sample size for the qualitative portion was much smaller than anticipated. I would attribute respondent fatigue to many of the refu sals, as many participants remarked that the quantitative portion was longer than expected. However, the 10 interviews that were conducted were thorough and candid. Participants were open to probing and were open about their backgrounds and opinions. 55

PAGE 56

Table 3-1. Reliability test on group identity scale. Index Number of items Deleted items Valid cases Mean Cronbach's alpha Standardized alpha 1 12 -56 33.368 0.562 0.568 2 10 6, 8 56 25.341 0.625 0.627 3 9 6, 8, 12 57 27.151 0.598 0.599 4 8 5, 6, 8, 12 57 21.617 0.598 0.597 Table 3-2. Major dimensions of cohesion ac ross three different cohesion measures. Index/scale Definition of cohesion Dimensions identified Buckner's neighborhoood cohesion index (Buckner, 1988) None specified If they feel they belong If they have neighbors as friends If they feel a we-ness with neighbors If they agree with neighbors on what is important in life If they feel loyal to neighbors If they are willing to work collectively If they are similar to neigbors If they feel fellowship with neighbors If they feel a sense of community If they find the neighborhood attractive If they want to move out of the neighborhood If they plan to stay If they visit in neighbors homes If they seek advice from neigbors If they borrow or exchange favors with neighbors If they have neighbors to their home If they regularly stop an d talk to their neighbors Seashore's group cohesiveness index (Seashore, 1954) Attraction to the group or resistance to leaving Feeling part of the work group Feelings about leaving the work group How group members get along How group memebrs stick together How group members help each other Perceived cohesion scale (Bollen & Hoyle, 1990) An individual's sense of belonging to a particular group and his or her feelings of morale associated with group membership Sense of belonging Sense of membership Feeling part of the community Enthusiasm about group Feeling of happiness about group Feeling that the group is one of the best 56

PAGE 57

Table 3-3. Preliminary reliability test on group cohesion index. Index Number of items Deleted items Valid cases Mean Cronbach's alpha Standardized alpha 1 22 -10 61.960 0.919 0.914 2 17 3, 8, 15, 18, 22 10 48.660 0.934 0.935 Table 3-4. Final reliability test on group cohesion index. Index Number of items Deleted items Valid cases Mean Cronbach's alpha Standardized Alpha 1 22 -54 64.057 0.900 0.900 2 19 1, 15, 22 54 54.918 0.907 0.907 Table 3-5. Reliability test on goal orientation index. Index Number of items Deleted items Valid cases Mean Cronbach's alpha Standardized alpha 1 17 -45 59.578 0.907 0.914 2 15 1, 15 45 54.444 0.923 0.927 57

PAGE 58

CHAPTER 4 RESULTS This chapter first provides demographic information and descri ptive statistics on the sample. Next, the results of te sts of central tendency are presented, followed by the results of multiple regression analysis. The qualitative analys is of interview data makes up the last portion of this chapter. Sample There were 63 respondents. Forty-five res ponses of one index and two scales were complete enough for quantitative analysis. The other 18 responses were not adequate because they were incomplete or improperly answered. Many of the respondents chose not to complete one or more of the three instruments because they did not see them as relevant or good. The same response pattern was present in individual items from each scale; many respondents thought items were stupid or did not understand them, making it impossible to calculate scores for the instrument as a whole for these res pondents. The respondents were split into two comparison groups: short term homeless persons and chronically homeless persons. There were 18 chronically homeless persons and 36 short-term homeless persons. Demographic Characteristics Table 4-1 provides the descrip tive statistics for 54 of the 63 respondents, including those who chose not to answer certain questions. Most respondents (68.5%) were between the ages of 40-59 years, while few (7.4%) were under 30 year s or over 60. Males were overrepresented (66.7%) in the short-term homeless category, wh ile the chronically homeless group was divided equally between males (50%) and females (50%). The age distribution for the two comparison groups was similar. 58

PAGE 59

While the sample was not robust enough to be considered statistically generalizable, demographic characteristics closel y mirror those of the entire ho meless population in Gainesville as of 2007 point-in-time survey (Alachua C ounty Coalition for the Homeless and Hungry & Gainesville Office on Homelessness, 2007). A bout 33% of respondents were chronically homeless, while 28% of the population is chronically homeless. In my sample, 66.7% of respondents are male, while 33.3% are female; in the overall population, 70.5% are male and 29.5% are female. Tests of Normal Distribution Tests of normal distribution were conducted to determine if parametric of non-parametric measures should be used. A Wilkes-Shapiro test of normality was run on the outcome variable of goal orientation and found that the sample was not normally distributed. Table 4-2 shows the results of this test. A Wilkes-Shapiro test wa s also run on the predicto r variables. The test determined that the distribution for these variable s was approximately normal. The results of this test are shown in Table 4-3. Because the goal orientation data for th e sample are not normally distributed, a transformation had to be performed. To determine if a log or square root transformation should be performed, the summary statistics for the outcome variable were obtained. A log transformation is performed when the ratio betw een the mean and standard deviation for all treatment groups are the same. A square root tran sformation is performed when the ratio between the mean and the variance are the same for all treatment groups (Sheshkin, 2003). It was determined that a log transformation should be performed on the goal orie ntation scores because the ratios for log transformation were closest for both groups. These results are shown in Table 4-4. 59

PAGE 60

Hypothesis 1 H1: There is a difference in goal orientation, group identity, and group cohesion levels between short-term and chronically homeless persons. T-tests are used in cross-sectional studies where comparison groups are formed post-hoc to determine if differences in mean scores exist between the two groups. The outcome variable for this study was goal orientation. To use a t-test, several conditions must be met (Sheshkin, 2003). The first is an assumption of representativene ss of the sample to th e general population. A comparison of demographic data for respondents co mpared to homeless census data shows that this sample closely reflects the makeup of the general population of homeless persons. The samples are independent because none of th e respondents are members of both comparison groups. Finally, the population should be normally di stributed. The Wilkes-Sha piro test shown in Table 4-5 illustrates that th e population was not normally distributed. Therefore, a nonparametric alternative to the t-test was run for goal orientation. A Mann-Whitney U-test was chosen. The p value was much greater than .05; therefore, the results were not significant. Because the design of this research is cross-s ectional, it is customary to run means tests on predictor variables to determine if the lack of significant difference on outcome variables between the two groups is based in a lack of diffe rence in predictor variables. To examine this possibility, t-tests were performed on the predic tor variables, group iden tity and group cohesion, to determine if differences existed. The conditions of representativeness of sample and independence of comparison groups were established before performing the t-test on the outcome variable, and are sufficient for thee tests as well. No significant difference was seen for group cohesion, but group identity was closer to significance with a p value of .075. Results from the t-test are shown in Table 4-6. 60

PAGE 61

Hypothesis 2 H2: There is a positive relationship betw een group identity and group cohesion. A Spearman rank order correlation was performe d to test the relationships between group identity, group cohesion, and goal orientation in groups of two. The Spearman rank order test was chosen as a nonparametric alternative to a regression analysis due to the sampling method and the non-normal distribution of the goal orientation data. The test showed that the only significant relationship between va riables exists between group identity and group cohesion. The outcomes of the correlation test are shown in Table 4-7. Hypothesis 3 H3: There is a positive relationship between gro up cohesion, group identity and goal formation. A multiple regression analysis was used to determine if the predictor variables of group identity and group cohesion predic t the variance for the outcome va riable of goal orientation. To run a multiple regression analysis, normal distri bution of data and representativeness of sample are assumed (Sheshkin, 2003). Both of these co nditions have been previously established. Because of the finding that goal orientation is not normally distributed, the regression was run against the log transfor mation of the data. The results of the multiple regression analysis are shown in Table 4-8. The regression summary is shown in Table 4-9. The regression analysis indicated that group identity and group cohesion significantly contributed to the multiple correlation coefficient of R 2 = .165 at a significant level of a=.05 (p= .021). The adjusted R 2 =. 126 indicates that most of the variance among goal orientation scores was predicted by gr oup cohesion and group identity. The analysis indicated that group cohesion was the significant predictor (p=.006), expl aining nearly 45% of the variability with beta= .44. 61

PAGE 62

Qualitative Analysis Sample In addition to quantitative measures gathered from the index and scales, ten 30-45 minute interviews were completed. After completing the survey, participants were asked if they would be willing to participate in an interview. Ma ny participants declined interviews for various reasons, including being busy and having somewhere to go, and general unwillingness to divulge more information. Ten total interviews were completed, with seven male participants and three female participants. The participants were from two sites, with six taking place at the downtown plaza and four at the St. Francis House sh elter. Of those who chose to participate in the interview, most were fort hcoming with information and no questions were refused, though some were insufficiently answered. Coding and Grouping Interviews were performed immediately after the participants completed the scales and index. Interviews were recorded and notes were taken using a thematic interview processing form. Qualitative data were analyzed using a thematic analysis. First, the interviews were coded with words or phrases pertaining to participants responses and intangibles such as demeanor, attitude, self-presentation, and eye contact. The codes were grouped into the following categories: Individuality Coping Survival Attitude toward homeless persons Fear and uncertainty Strife Hard work, trying Mutual respect (between homeless persons) Strong sense of self Outsider mistrust 62

PAGE 63

Insider mistrust Community leadership Institutional perceptions Traumatic personal experiences Substance abuse (alcohol and/or drugs) Guilt/shame Pride Self-motivation Empathy Acceptance of homelessness Homeless friends Family connection Outlook and perception of the future Attitude Categories were formed by grouping the word s, phrases, and expressions used by the participants during interviews These categories were grouped into subthemes representing aspects of the research variab les. Three of the themes, group identity, group cohesion, and goal orientation, are directly related to variables in the proposed social movement model. The theme of leadership was explored to determine if l eadership plays a role in homeless group dynamics, despite it not being part of the model. Action capacity, as pres ented in the theoretical model, was not examined at this study is intended to e xplore the presence of the precursors to action capacity. Respondents were eager to divul ge information related to their experience as a homeless person and the circumstances of their homelessn ess. Many of the topics addressed in these discourses did not relate well to the theoretical themes, which led to the development of the emergent theme of personal situation. The them es are shown in Table 4-10 with corresponding subthemes. 63

PAGE 64

Table 4-1. Gender and age of homeless persons, Gainesville, Fl, in 2008. Gender Age Male Female 18-29 30-39 40-49 50-59 60+ Short term homeless 18 9 2 5 13 14 2 Chronically homeless 9 9 0 7 5 5 1 Total 36 18 2 12 18 19 3 Table 4-2. Wilkes-Shapiro test of normal distribu tion for the outcome variab le of goal orientation among homeless persons in Gain esville, Florida, in 2008. Variable N W P Goal mean 51 0.875 0.000 Table 4-3. Wilkes-Shapiro test of normal distri bution for the predictor variables group cohesion and group identity for homeless persons in Gainesville, Fl, in 2008. Variable N W P Group identity 58 0.960 0.056 Group cohesion 55 0.961 0.072 Table 4-4. Summary statistics for goal orientatio n among homeless persons in Gainesville, Fl, in 2008. Group N Mean Variance Std. dev. Trans Log trans Short term homeless 32 3.53 0.977 0.988 3.59 3.55 Chronic Homeless 16 3.40 0.875 0.936 3.90 3.60 Transformation (Mean/variance) = x. Log transformation (Mean/S tandard Deviation)= log (x). x = goal orientation score. Table 4-5. Mann Whitney U 2-sample test to comp are goal orientation scor es between short term and chronically homeless persons in Gainesville, Fl, in 2008. Goal orientation score U 206.0 Z 0.943 p-level 0.346 Z adjusted 0.944 p-level 0.344 2*1 sided p 0.355 64

PAGE 65

Table 4-6. T-tests for predictor variables group id entity and group cohesion for homeless persons in Gainesville, Fl, in 2008. Variable Length of homelessness Mean N SD F-ratio t value P Group cohesion Short term 2.933 35 .779 1.005 1.008 .318 Chronic 2.701 17 .781 Group identity Short term 2.553 36 .721 1.310 -1.814 .075 Chronic 2.916 18 .630 Table 4-7. Spearman rank order correlations for group identity, group cohesion, and goal orientation among homeless persons in Gainesville, Fl, in 2008. Variable Identity mean Cohesion mean Goal mean Identity mean 1.00 .338* -.154 Cohesion mean .338* 1.00 .226 Log goal mean -.154 .226 1.00 Pairwise deleted. *Marked corre lations significant at p < .05. Table 4-8. Multiple regression results. The relationship between group identity and group cohesion on goal orientation for homeless persons in Gainesv ille, Fl, in 2008 R R squared Adjusted R Standard error of estimate P-value .406 .165 .126 .117 .021 Predictor variables: group identity and group cohesion. Table 4-9. Regression summary for the relations hip between group identity and group cohesion on goal orientation among homeless persons in Gainesville, Fl, in 2008. Variable Beta Standard error of beta B adjusted Standard error of B adj. T p-value Constant 0.399 0.079 5.072 0.000 Group cohesion 0.442 0.153 0.068 0.024 2.882 0.006 Group identity -0.124 0.153 -0.022 0.028 -0.808 0.423 Dependent variable: goal orientation. 65

PAGE 66

Table 4-10. Categories organize d by themes from interviews with homeless persons in Gainesville, Fl, in 2008. Theme Subtheme Group identity Individuality Group concept Acceptance of homelessness Group cohesion Interactions with homeless persons Personal beliefs Outsider and insider mistrust Goal orientation Outlook Goals Self-efficacy Leadership Identif ication of leaders Definitions of leadership Personal situation Homeless experience Coping Support systems 66

PAGE 67

CHAPTER 5 DISCUSSION AND CONCLUSIONS Research Question and Hypotheses The purpose of this study was to determine if the precursors to social movement action were present among homeless person s in Gainesville, Florida as illustrated on a proposed model for collective action based on new social movement theory. The data suggest the following: goal orientation, group identity, a nd group cohesion levels among homeless persons do not differ between chronically homeless and short term homeless persons; group identity and group cohesion are positive related to each other; and gr oup cohesion is predictive of goal orientation, while group identity is not. Hypothesis 1 H1: There is a difference in goal orientation, group identity, and group cohesion levels between short-term and chronically homeless persons. No significant differences between compar ison groups were found, suggesting that the distribution of scores on gr oup identity, group cohesion, and goal orientation among homeless persons in Gainesville are about the same rega rdless of how long a person has been homeless. This is contrary to my proposed hypothesis. It was believed that the longer a person has been homeless, the more time they have had to get to know other homeless persons and the system through shelters, soup kitchens, and camps. The da ta suggest that despite the possibility of increased exposure to other homeless persons and homeless institutions, the duration of homelessness does not negatively or positively impact the presence of group identity, group cohesion, or goal orientation. Group identity was the closest variable to sign ificance after st atistical analysis with a pvalue of .075 on the t-test. The mean was higher for chronically homeless individuals at 2.916 in 67

PAGE 68

comparison to the mean for short-term homeless at 2.553. While not statistically significant, the direction of difference on group cohesion scores was the opposite of the expected pattern with short-term homeless reporting higher scores ( 2.993) than the chronically homeless (2.701). While contrary to the original hypothesis, this su pports the theoretical belief that group cohesion and group identity are co ncepts that measure different attitudes. This lack of a difference in group dynamic s based on length of homelessness indicates that homeless persons do not become more imme rsed with other homeless persons despite their increased exposure to the group. This is proba bly related to the hi gh sense of individual responsibility and self-dependen ce indicated in the interviews. Although those participants who had been homeless longer knew more homeless persons than those who had not been homeless longer than three years, every in terviewee indicated that the people they spen t the most time with were themselves. This strong individuality wa s echoed when participants were asked about leadership. Most responded that they were their own leaders. Self-leadership lends itself to individuality, with some particip ants even seeming isolative in their self-dependence. None of the participants were able to identify any pers ons within the homeless community as leaders to others. Isolation and self-dependence limit the de velopment of group connections irrelevant of the length of time exposed to the group. Despite this strong sense of individuality, pa rticipants displayed high levels of mutual respect and exchange networks within the community, indicating th at group cohesion was present among participants. Those interviewees who had been homeless less than three years seemed more optimistic about the future, but were equally as respecting and understanding of homeless persons as those who had been homeless over three years. The goals expressed by the interviewees were similar despite length of homelessness. The majority of stated goals were 68

PAGE 69

related to becoming housed and seeking or ma intaining employment. This similarity in perceptions of homeless persons and goals supports the quantitative findings. Hypothesis 2 H2: There is a positive relationship betw een group identity and group cohesion. The Spearman rank order correlation performed during data analysis supports this hypothesis in suggesting that group identity and group cohesion are related to each other. No causative relationship can be inferred from th is. It is possible that group cohesion and group identity are mutually fostered by an unknown pr ecursor. However, this finding supports the theoretical proposition that these two concepts are related and s hould be examined together to acquire a complete picture of group dynamics. The relationship between group identity and group cohesion may be related to similarity between these two concepts. There is no research to date on social movements or collective action where group identity and group cohesion ha ve been measured in the same study. As explained in the literature review, group cohesion has been conspicuously absent from social movement literature. Because of this, it is impo ssible to explain the re ason for the relationship between the two variables. Because the directionality of the relationship between group cohesion and group identity cannot be determined, it is possible that strong group identity in either direction, positive or negative, can foster high group cohesion. This is likely the case with homeless persons and other groups where negative identities are forged, often from forced identities. Examples of groups with negative identity who have formed viable social movements include African-Americans and the poor (Cattell, 2001). Often th e negative identity is embraced and viewed as positive among group members. Other times, the negative iden tity is adamantly rejected, and the common struggle for the group becomes no longer being identified as a group member (Reicher, 2004). 69

PAGE 70

The latter may be more likely for homeless pe rsons, but in either case, the negative group identity can lead to high group cohesion. The interview questions related to group id entity and group cohesi on elicited different responses from participants. On questions relate d to group identity, participants conveyed that while they were homeless, they themselves we re not like other homeless persons. When probed, participants indicated strong resp ect for other homeless persons and were eager to point out that very few of them are bad people; they usually just had some bad luck. Hypothesis 3 H3: There is a positive relationship between gro up cohesion, group identity and goal formation. The regression analysis show ed that group cohesion was strongly predictive of goal orientation, but group identity was not predictive. Contrary to the original hypothesis, group identity and group cohesion are not equally predic tive. This indicates th at the proposed model may not be accurate in depicting the influen ce of group cohesion and group identity on goal orientation. It is possible that goal orientation and goal form ation are closely tied to elements of group cohesion explored in the qualita tive interviews, such as mutu al respect and mistrust. Group cohesion is a sense of mutual group member ship, while group identity is a personal characterization of group membershi p. Whether or not a person identifies as part of the homeless group is irrelevant to goal form ation; all interviewees articula ted similar goals despite their perception of group membership. But, participants seemed to find hope in the fact that homeless persons were all suffering and all needed to have certain needs met. Therefore, it is possible that group cohesion is more closely related to goal orientation because hom eless persons develop goals based on mutual needs. 70

PAGE 71

Additionally, the goals expressed were base d largely on mutually accepted reasons for homelessness. The most apparent of these is housing. Interviewees repeatedly expressed opinions that the best way to combat homelessness is to increase the availability of affordable housing; participants also repe atedly indicated obtaining housi ng as their primary goal. While obtaining housing can be perceived as having several steps, such as saving money and obtaining and maintaining employment, participants expres sed entitlement to bei ng housed. This fixation on housing fosters camaraderie am ong homeless persons by creating similar goals. This indicates that the relationship between group cohesion an d goal orientation is complex, but based on mutual explanations and solutions for homelessness. The above finding may be influenced by th e effectiveness of the instrumentation measuring group identity and group cohesion. The scale used to measure group identity was adapted from a well-tested scal e on group identity that had not been used on fringe populations; the index used to measure group cohesion was de veloped specifically for this study by the researcher. The group cohesion scale may have b een more sensitive to the presence of group cohesion than the group identity scale was. Howe ver, as indicated in the methods section, both instruments were pre-tested and appeared to perform well with the study sample. Interview Data Group Identity It was expected that homeless persons would exhibit a strong sense of group identity and personal connection to being homeless. The interview findings suggest a paradoxical and dichotomous perception of homeless persons am ong homeless persons. It seems that homeless people can easily characterize other homeless persons and descri be them as a group, but have a hard time incorporating themselves into the gr oup, which would lead to low group identity. Most participants were quick to assert that while they were homeless, they did not identify with other 71

PAGE 72

homeless persons in terms of values and life pers pectives. The three major subthemes that arose in discussions on group identity were a strong se nse of individuality, an overall positive group concept and sense of mutual respect among ho meless persons, and an unwillingness to accept ones own identity as homeless. Individuality There was a strong sense of i ndividuality and self-reliance am ong respondents. Participants believed that they could help themselves, but not necessarily that homeless people could do so as a group. There was a strong sense of detachment from other homeless persons despite the fact that all participants reported having homeless friends. All particip ants reported spending most of their time alone, with myself, as they must keep to themselves to protect themselves and any personal belongings. Group concept A positive group image emerged from initial negativity towards other homeless persons. When describing homeless persons, participants used words like resilient, tough, and down on their luck. Many repeatedly assured me that homeless people are not bad people, they just got into a bad situation. Peter: All of them try. I know th at. A couple of them have jobs, they just dont make enough money for ends to meet, you know. Jake: A lot of us are good, were just homele ss. Were like everybody else. Some of us dont have work. Were trying to get work. Joe: [Homeless people are] Str ong. They have to be strong or else they wont survive long out here. They were quick to point out the downfalls of the system, which tend to keep homeless people where they are because of lack of resources and political prior ities being with condos and other high-profit projects. 72

PAGE 73

Feelings towards other homeless persons we re mixed, with many respondents reporting both a negative perception and positive percepti on, dependent upon the situation. It seems that respondents tend to view most homeless people as outsiders, but their own close group of homeless friends as the good one s. Respondents maintained that many of the homeless persons they have met are good people. Ralph: They like, help you out a lot. So, I dont have nothing bad to say. This was contradicted by the remarks of two individuals, who reported that their group was large, consisting of about 100 homeless persons w ho mainly spend their days in the downtown plaza. These two individuals, Ada and Dusty, expressed strong feelings of empathy and understanding towards other persons, as well as being adamant about the fact th at others will help you out. Ada: Everybody knows Ada. In our group Id say there are about 100. When probed about what members of this gr oup were like, Ada and Dusty mentioned they were referring mainly to the homeless persons who spend most of the time at the downtown plaza. Acceptance of homelessness About half of respondents seemed hesitant to accept their homelessness, with only one respondent completely in denial despite being a resident at the St. Francis House. This denial took various forms for other participants, some st ated they are not like most homeless people, or that they dont really hang out with other homeless people. The language used when talking about homele ss persons was a significant indicator of identity. Most participants did not use we when talking about homeless persons, rather them when referring to homeless people (in response to questions like How would you describe a 73

PAGE 74

homeless person?) and I when describing their own situation. A lack identification as a group member indicates low levels of group identity within an identified group. Group Cohesion It was expected that homeless persons would exhibit low group cohesion based on their transient nature. While the interview findings suggest that there is consciousness among homeless persons of this transience, they still tend to believe that homeless people will help each other. Similar dichotomies were present in the group cohesion data as we re present in the group identity findings discussed previously. The majo r subthemes that emerged in this part of the interview were overall positive interactions with other homeless persons, including mutual respect and friendships, and, the contradictor y presence of mistrust among homeless persons towards other homeless persons and towards outsiders. Interactions with homeless persons A sense of mutual respect was observed among homeless persons in the shelter and on the downtown plaza. Homeless persons seemed to know each other by name or alias and know each others vices and backgrounds. Several partic ipants gave me more extensive background on other participants while describing what homel ess people are like. The stories ranged from people who tried and succeeded to people who were given chances repeatedly but failed. Ada: Like Jan over there. She got a house but lost it because all she wanted to do was party. They also know who to ask for things. With a so rt of barter system in place for things they cannot get from shelters such as drugs, alcohol, and cigarettes Food trade was also present but not as prevalent because of the many shelte rs in Gainesville serving meals daily. Ada: You know some people will always have cigarettes, or some will always have some alcohol, or that some people get food stamps so they can help you get food. 74

PAGE 75

But, keeping with their sense of pride, most homeless people will not accept freebies from each other, with reciprocity a strong norm presen t. Homeless people will use any resources of their own to barter with others, or simply to help others. Ada: If I have something, everyone else has some thing. We try to help each other out. All participants indicated that they had formed friendships with other homeless persons. For most, it was a small number two or three, or a significant other that they considered to be true friends. Ralph: Yeah I come out to the park w ith them. Yeah (I made friends). Jenny: The only person I really spend any tim e with other than my self is Bobby right there, weve been together a long time. Max: Its just me and her (referring to his girl friend). Were all weve got but it definitely helps. It gets hard out here. Joe: Im sure homeless people have friendsI think they bond, they have to bond, sure. Ive seen some really great things, people together just enjoying themselves, but then they have to go right back to their camps. Mistrust Despite the strong sense of mu tual respect among homeless pe rsons, a degree of mistrust was present towards homeless persons, or inside rs, and outsiders. The most mistrust between homeless persons was present among those particip ants who were living in tent city. Many described violence and drug abuse, and reported it was a very fright ening way to live. Joe: You always have to watch your back. Louise: I started drinking again because Im scar ed. Yeah I talk to people and I have friends, but its scary. A lot of people want to take adva ntage of you. Its especially dangerous out here for a woman. Outsider mistrust was present and was conveye d by most respondents as a defense mechanism. Joe: Homeless people dont trust anybody, basically. 75

PAGE 76

Despite this, most homeless persons approached and asked to participate in the study were friendly even when declining to participate in the interviews. Outsider mistrust was also portrayed in a disdain for the system. Two par ticipants stated that they had worked with homeless advocacy groups in the past. In both inst ances, the participants had worked with groups who were attempting to create shelter spaces fo r homeless persons, one exclusively for women in Gainesville, and one for all homeless persons in Orlando. Joe: I used to know some people in Orlando w ho tried to get the uh, naval base turned into a homeless center and the government just laughed at us. They sa id theyd research our proposals. They had medical facilities there, housing, schoolrooms, textbooksshut it down and built new houses on it. [] if you have a bad habit, say you smoke a cigarette, five years from now youll probably be smoking a pack a day. If youre homeless, and somebody doesnt, at first, try to put you in th e right direction, or help you, youre gonna stay homeless. Goal Orientation All of the participants were able to articulate specific goa ls about their future, but the goals described were overwhelmingly related to ending their homelessness. This is expected because the most pervasive facet s of their lives while homeless are the struggles of survival while homeless, in addition to making attempts to become housed and self-sufficient. The attitudes and beliefs expressed by participants lend themselves to goal development. These included positive outlook on the future, self-effica cy, motivation, frustration at their current life situation, and faith. The subthemes that were pr esent in the narrative on goal orientation and development are centered on outlook including the fear and uncertainty related to living on the streets, goal setting, and self-efficacy, which includes aspects of motivation. Outlook Respondents expressed mixed feelings regarding their futures. While many of them were hopeful and positive when asked directly about thei r goals and perspective, tones of fear and uncertainty were pervasive in the discourse. Pa rticipants repeatedly mentioned that being 76

PAGE 77

homeless is scary and hard. Despite this their outlooks quickly turned positive when discussing their goals and hopes for the future. Goals Respondents were asked to discuss persona l short-term and long-term goals. All respondents listed goals pertaini ng to ending their homelessness. The most common goals were related to obtaining housing, obtaining or ma intaining employment, and staying positive. When probed to differentiate between short-term and long-term goals, par ticipants stated that their goals were the same, because not being homeless anymore was both a short-term hope and a long-term maintenance. Joe: I dont plan on being homeless for too l ong. When I do something I put a full effort into it, you know. When I go to wo rk, I put a full effort in tryi ng to go to work. I try to put full effort in a conversation. Joe: Im more concerned about my health ri ght now than anything else. And looking for basic housing, which Ill find. Peter: To do better. Ive got a part time job now [] well, I got put on child support so trying to make those payments. The tone of outward hopefulness was maintain ed in the discussions of goals. Respondents discussed their personal strengths and seemed like they were, at times, trying to convince themselves that their goals were achievable. Self-efficacy & motivation A strong sense of self-dependence and indi viduality was present among interviewees. When discussing the circumstances of their b ecoming homeless and when talking about other homeless persons, they were adamant that homeless people are not bad people, but are victims of circumstance and bad luck. Despite the setbacks and often traumatic experiences relayed by interviewees, they were all certai n that they could get themselves out of the situation, with or without outside help. 77

PAGE 78

Joe: [the solution is] doing it for yourself. Ev erybody is different. I guess work, too, it gives you self-esteem. This attitude persisted when prompted to consider forces out of their control that may contribute to their homelessness, particularly lack of jobs and affordable housing. Initially, respondents expressed frustration at the apparent decline in oppor tunities and perceived lack of cooperation and understanding from governments and institutions, but often turned this frustration around by stating they just have to keep trying. Faith was also mentioned as a personal motivator. Leadership Identification of leaders Participants were unable to identify specific leaders within the homeless community. There was an overall attitude that homeless persons did not have time to act as leaders, because they needed to fend for themselves. One participant st ated he did not know of any leaders because he does not get personal with other homeless pers ons. When prompted further, none of the participants were able to identify leaders among the homeless population in Gainesville, but often mentioned leaders within the community and continued to discuss what are important characteristics in leaders. When asked to identify whom they viewed as leaders, eight of the nine participants identified themselves. Ralph: Myself. I dont go to nobody else. I dont hang out with but 2 or 3 of those homeless guys, and thats about it. When asked to identify leaders in the Gain esville community in relation to homeless advocacy or services, Arupa Freeman, founder a nd operator of the HomeVan, was repeatedly identified as a key figure. Participants also mentioned organizations and agencies as leaders because they help you out a lot. The agencies mentioned most often were St. Francis House, 78

PAGE 79

Fire of God Ministries, and the Salvati on Army, although participant also expressed dissatisfaction with the way the Salvation Army is run. Disdain towards the city government, including the housing authority, a nd the police were repeatedly e xpressed. Four participants reported having been arrested for trespassing or public intoxicati on within the past year. These participants were frustrated w ith the way the city government wa s dealing with homeless persons and stated that they have no choice but to loiter as they have nowhere to stay. Louise: I aint got nowhere to sleep, so I have to trespass. I aint got nowhere to drink, so I have to drink in public. Definitions of leadership When prompted to define what makes a person a leader, participants consistently mentioned being trustworthy and reliable as essential characteris tics for leadership. In addition, participants asserted that leadership is often an element of personality, and that a willingness to help, listen, and work with others is essential. Joe: Anybody who isnt presidentyou mean a l eader in the homeless community? What you need is honesty. You get somebody you can trus t, thats leader material. Especially when youre on the street. Louise: Personality is really important. Ada: A leader has to know people, and has to want to work with people and help them. Personal Experience Participants were eager to talk about their experiences as homeless persons. In doing so, personal characteristics and values were relayed. A strong sense of disdain and lack of belief in the political system emerged from this discourse which may contribute to the strong sense of self and personal responsibil ity discussed previously in the section on group identity. Additionally, the circumstances surrounding their homelessness, coping methods, and support systems were discussed. 79

PAGE 80

Homeless experience The first question in the interview, How would you describe your experience as a homeless person elicited various negative comments. Ada: It sucks. Peter: Its scary, being homeless. Its scary living from day to day and wondering where youre gonna lay your head at. Wondering wher e your next meal is gonna be coming from. Thats scary. Youre unemployed, ai nt got a job, cant get a job. Joe: Awful. Unstable. And its th e worst thing you could be. Reasons mentioned by participants for becoming homeless varied from inability to find work, divorce, domestic abuse, substance abuse, poor physical and/or mental health, and having been released from prison. In most cases, more than one factor contribute d to an individuals becoming homeless. Peter: When I got out of prison, I wasnt trying to be homeless. Its just something that came up. Coping Interviewees persiste ntly mentioned coping methods. Many mentioned the allure of drugs and alcohol as coping mechanisms when living on the street. The ease of access and influence of other homeless persons contributes to dependence. Jenny: Theres lots of drugs ar ound. Hard drugs, and alcohol. Especially if you go to the tent cities, its everywhere. Peter: Yeah they be doing drugs all over the pl ace. Even right here [outside the St. Francis House] at night. If you co me out here late at night you see all kinds of crack pipes and stuff around. One participant mentioned that she had been sober for 9 years prior to becoming homeless. As soon as she was on the street again, she started drinking to d eal with the fear and uncertainty. 80

PAGE 81

Smoking cigarettes is also pervasive in the homeless community. Four interviewees were smoking during interviews, and four others were smoking either before or after the interviews. When prompted about smoking, one interviewee menti oned it was just a way to pass the time. Support systems When asked about family, all participants stat ed that they had no connections with family. Two participants explicitly stated that they are considered black sheep in their families and are not welcome. One participant lost his wife and da ughter in a car accident several years earlier, which drove him to become an alcoholic and suffer a psychiatric decompensation. Louise: I dont talk to any family. I got a son, but he says Im worthless and a failure so I dont need anything from him. The rest of my family is the same way. Im the black sheep. Ada: Were all black sheep here. Thats probably the reason our families dont talk to us no more. Many participants stated that partners and hus bands or wives who were also homeless were the only people they were close to. As previ ously mentioned in the discussion on friendships with homeless persons, participants did not men tion any need to seek to support from other homeless persons, but did recognize that they could seek help if they wanted to. Participants reported learning about services and progr ams through other homeless persons, but the interactions were usually limited to sharing information about meeting personal needs and superficial chat. The most commonly reported s upport was faith. Many participants maintained hope that God would bring them out of their situation. A New Model As mentioned in the literature review, no model exists to pr edict the course of social movement emergence. Several concepts, most notably group identity and goal formation, have been discussed as being present in social movements, but the effect of these concepts on social movements, and their place over the course of action has not b een examined. In the literature 81

PAGE 82

review, I proposed a model (Figure 2-1) for new so cial movements based on the literature in both sociology and social psychology relating to group dynamics and collective action. This study was limited to examining the pres ence of three precursors to action, group identity, group cohesion, and goal orientation in a so cial group that may or may not be able to act collectively. Because the group being studied has not engaged in any mode of protest to date, it is impossible to measure action, however, action capaci ty is defined as the ability to act based on the presence of the thr ee related variables. Based on the fi ndings of this study, the above model may not be accurate in predicting goal orienta tion. The regression analysis suggests that group cohesion is the predictor of goal orientation for the study sample. The correlation suggests that group identity and group cohesion are related. Th is suggests that a model where group cohesion alone precedes goal orientation would be more accurate. Although the causative relationship between group identity and group cohesion cannot be discerned based on the findings in this study, it can be inferred that group identity would precede group cohesion if the final step before action were goal orientation. This differs from the originally proposed model in that group identity and group cohesion do not equally determine the presence of goal orientation and formation. The revised model is shown in Figure 5-1. Limitations Research on fringe groups is notably difficult and this study was no exception. The referral sampling method did not carry out as exp ected. Many of the study participants did not refer other homeless persons because they did not know how to locate these individuals. They were also hesitant because they did not want thos e that they were referring to feel as though they had been called out for their homelessness. Beca use of this, it is impossible to determine how many tiers of referrals were achieved. The findings of this study are therefore not generalizable to the overall homeless population or to homeless persons in general. 82

PAGE 83

Only nine persons of 63 total in the sample ag reed to the qualitative interview. This is a small number compared to the overall study grou p and a very small fraction of the 952 totally homeless persons in Alachua County. The informa tion gathered from the interviews cannot be generalized to the study sample, th e general population of homeless pe rsons in Gainesville, or to homeless persons in general. Additionally, the locations of data collection limited the representativeness of the sample. Homeless persons spend their time in various locations throughout Alachua County. Data were collected primarily at the St. Francis House shelter and soup kitchen and at the Downtown Community Plaza during daylight hours and during three food handouts arranged by the HomeVan. As previously explained, these site s were chosen primarily for safety of the researcher and because these were the most accessible sites. Despite the above sampling limitations, the demographic makeup of the quantit ative sample was closely reflective of the demographic makeup of all homel ess persons in Gainesville. Negative stigmas associated with being a hom eless person may have rendered some items for measuring group identity useless. The most nota ble example from data collection is the item I am glad to be a homeless person from the gr oup identity scale. The issue here stems not from negative attitudes towards being a group member, but negative attitudes towards being homeless, disenfranchised, stigmatized persons with limited resources. Pre-testing of the instruments did not reveal this issue. This negative stigma also limited the willingne ss of participation in either the quantitative or qualitative portion of the study. This was more present in the qualitative portion, as only nine of 63 persons asked to participate agreed to do so. The social sensitivity of the factors often leading to homelessness such as substance abuse, domestic violence, and mental illness, make it 83

PAGE 84

difficult for people to feel comfortable discussi ng their situation, even if the study does not involve those issues. Additionally, the quality of data are difficult to determine due to the above-mentioned stigma and negative social impli cations of homelessness. While th e concepts being studied were not closely related to the circumstances of homelessness, homeless persons develop negative attitudes towards outsiders becaus e of the perception that nothi ng is being done to help. I had many people decline to participate by commenti ng on the vast number of studies having been performed and a belief that nothing has come of them. The homeless population in Gainesville has not participated in a social movement as of the data collection for this st udy. Because of this, it was impossi ble to measure action capacity based on action. This limits the inferential power of the proposed model because I was unable to measure action capacity as an outcome variable. Goal orientation was chosen as the outcome, as in the model it is the final step before action capacity is achieved. This model has never been tested and is based solely on the conceptual frameworks laid out by previous scholars on new social movements. Because it has never been teste d, there is a possibility that goal orientation is not the final step before action capacity can be achieved. Implications for Research The findings of this study sugge st several implications for re search on social movements: Group cohesion is related to the presence of gr oup identity, and thus should be examined as a variable in social movements. The causal relationship between group ident ity and group cohesion is unknown and should be further investigated. Quantitative instrumentation on social movement variables needs to be further developed. The importance of group cohesion in social movement dynamics needs to be further explored. Group cohesion has been studied for de cades in the social psychology research for 84

PAGE 85

groups as small as work groups and as large as communities. Whether it has been excluded from social movement research because it is not an entirely sociolog ical concept is unclear, but the importance of its presence in social groups is un deniable. Exploratory st udies are necessary to verify the presence of group cohesion in establ ished social movement groups. Comparisons of group cohesion levels for social movement gr oups and non-social movement groups should be carried out in addition to qualita tive examinations of movement participants perceptions of cohesiveness among groups. This study suggests that group cohe sion is closely tied to group id entity. The nature of this relationship needs to be determined to better understand how group dynamics influence collective action. At this time it is difficult to propose an effective model for social movements. New social movement literature is heavily focused on the role of identity in movement activities; now that it has been suggested that group cohesion is closely re lated to group identity, it is necessary to examine this relationship, as it may play an equal role establishing a movement. The causal direction of this relationship must be esta blished quantitatively to develop a viable model. Qualitative explorations focused on group cohe sion of new social movements with strong identity components will resolve whether group cohesion plays an equa l role in movement emergence as group identity, or if its merely a byproduct. Research on new social movements tends to be highly qualitative. However, the use of quantitative and qualitative measures increase s the validity of findi ngs, and therefore the theoretical implications. Although the role of identity has been thoroughly researched from a new social movement perspective, no viable quantitative instruments are available from the literature. The instruments used in this st udy were taken from social psychology research. Researchers must develop and test valid and reliable instruments for measuring the presence of 85

PAGE 86

social movement components. Additionally, th e instrumentation available should be made appropriate for groups where membership may be socially shameful. While being homeless does not necessarily mean that a person is part of a homeless movement or shares common identity with other homeless persons, most participants in this study indicated high levels of group identity despite not being ha ppy to be a homeless person. In addition to the above implications for so cial movement research, the findings of this study can be used to improve the methods used for research on fringe groups. Despite sampling difficulties, consultation with experts and homele ss persons allowed me to decide which sites would get the most representative sampling of homeless persons to as k to participate. I was able to achieve a representative sample by accessing only two sites where homeless people congregate in all of Gainesville. Use of this kind of insider knowledge allows researchers to make the best of conducting studies on hard to reach persons by allo wing representativeness without necessarily achieving generalizability. Figure 5-1. Revised model for new social movements. 86

PAGE 87

APPENDIX A ITEMS IN GROUP IDENTITY SCALE I have a lot in common with other homeless people. I feel strong ties to other homeless people. I find it difficult to form a bond with other homeless people.* I dont feel a sense of being conne cted with other homeless people. I often think about the fact that I am a homeless person. Being a homeless person is not rela ted to how I feel about myself.* Being a homeless person is an impor tant part of my self-image. The fact that I am a homeless pe rson does not really cross my mind. Im glad to be a homeless person. I often regret that I am a homeless person. I dont feel good about bei ng a homeless person. I feel good when I think about myself being a homeless person. Reverse scored items 87

PAGE 88

APPENDIX B ITEMS IN GROUP COHESION INDEX I hang out with other homeless people. I travel with other homeless people. Most of my daily interaction is with homeless people. Homeless people work together. I like hanging out with other homeless people. I enjoy working together w ith other homeless people. Homeless people help each other. When I need help, I turn to other homeless people. Homeless people share common beliefs. I learn about services through other homeless people. I seek comfort from homeless people. Homeless people feel that they can ma ke a difference by working together. All of my homeless friends ha ve been here a long time. Most of my friends are homeless people. When I get sick, I turn to other homeless people for help. I eat with other homeless people. I sleep in the same place as other homeless people. When I seek help, I usually do so with other homeless people. I seek advice from other homeless people. When homeless people have a plan, they stick to it. Homeless people believe that they can improve their own situation. Homeless people believe that polic ymakers are not on their side. 88

PAGE 89

APPENDIX C ITEMS IN GOAL ORIENTATION SCALE I just try to avoid being unskilled at doi ng the things I need to do to survive. When I am doing something, I find myself thinking a lot about not messing up. When I am doing something, I focus on not doing worse than I have done in the past. I just hope I am able to main tain enough skills to survive. I am just trying to avoid performing the things I need to do poorly. I am willing to choose a challenging ta sk that I can learn a lot from. For me, development of my personal ability is important enough to take risks. I often look for opportunities to develop new skills and knowledge. I enjoy challenging and difficult task s where Ill learn new skills. I like to show that I can perform better than others. I prefer to work on tasks that allow me to prove my ability to others. I try to figure out what it takes to prove my ability to others. I enjoy it when others are aw are of how well I am doing. I would avoid taking on a new task if there were a chance that I would appear very unskilled to others. Avoiding a show of low ability is more impor tant to me than learning something new. I prefer to avoid certain situati ons where I might perform poorly. Im concerned about taking on a task if ot hers could tell that I had low ability. 89

PAGE 90

APPENDIX D INTERVIEW SCHEDULE Homeless Experience How would you describe your experience as a homeless person? Thoughts about Homeless People How would define a homeless person? Follow up probe: Some people make distincti ons between different types of homeless people, ie: long-term, short-term, or one-time. Do you think this is tr ue that differences between these groups actually exist? Follow up to the follow up: do you iden tify with any of these groups? Social Activities Who do you spend the most time with? Why? Follow up probe: why or why not do you spend time with other homeless people? Everyday Activities Where do you sleep at night? Are there usually the same people there night after night? Follow up: Who are these people? Would you describe them as being part of your group? Where do you normally have your meals? Do you usually eat with the same people? Follow up: who are these people? Would you desc ribe them as being part of your group? Leadership To you, what makes a person a leader? Do you know any people within the homeless community that you would identify as leaders? Who? Follow up: Do you see yourself as being a leader? Why/why not? Are there any people you identify as leaders for homeless issues who are not necessarily part of the homeless community? Who? Goal Orientation What are some of your goa ls related to your situa iton as a homeless person? Follow up: Do you feel that you can achieve these goals? Why/why not? List 2-3 personal goals fo r the next 3 months. Follow up: Why did you mention these? Do you feel that these are achievable? List 1-2 personal goals for the next year or more. Follow up: Why did you mention these? Do you feel that these are achievable? If different from above: why do you consider th ese goals to be more long-term than the previous ones? 90

PAGE 91

APPENDIX E INTERVIEW PROCESSING FORM Site:_________________ Date:________________ Participant ID:_________ 1. Main issues or impor tant points to note 2. Information I got (or failed to get) on target questions Subject Details Goals Group ID Group Cohesion Leadership 3. Other important or interesting points 4. Nonverbal cues which may provide information Cue Context & Details Silence Body Language Refusal to answer 91

PAGE 92

LIST OF REFERENCES Adelson, J. (2007, September 13). Police find tent city abandoned[ Electronic version] The Gainesville Sun Retrieved January 1, 2008, from http://www.gainesville.co m/article/20070913/NEWS/709130315 Adelson, J. (2007, September 6). Plaza surr ounded by controversy[ Electronic version] The Gainesville Sun Retrieved January 1, 2008, from http://www.gainesville.co m/article/30070906/NEWS/709060320 Ames, C. (1992). Classrooms: Goals, structures, and student motivation. Journal of Educational Psychology, 84, 261-271. Ames, C. & Archer, J. (1988). Achievement goals in the classroom: Students learning strategies and motivation processes. Journal of Educational Psychology, 80 260-267. Ansoff, H.I. (1965). Corporate strategy New York: McGraw-Hill. Austin, J.T., & Bobko, P. (1985). Goal-setting theory: Unexplored areas and future research needs. Journal of Occupational Psychology, 58 289-308. Baranik, L.E., Barron, K.E., & Finney, S.J. (20 07). Measuring goal orientation in a work domain: Construct validity evidence for the 2x2 framework. Educational and Psychological Measurement, 67 697-718. Barron, K.E. & Harackiewicz, J.M.( 2001). Achiev ement goals and optimal motivation: Testing multiple goal models. Journal of Personality and Social Psychology, 80( 5), 706-722. Bergdall, T. (2003) Reflections on the catalytic role of an outsider in asset-based community development (ABCD). Retrieved January 15, 2008, from http://www.sesp.northwestern.edu/ docs/ABCD-outside-catalyst.pdf Bernstein, M. (2005). Identity politics. Annual Review of Sociology, 31 41-74. Boatswain, S.J. & Lalonde, R.N. (2000). Social identity and preferred et hnic/racial labels for blacks in Canada. Journal of Black Psychology, 26 216-234. Bollen K.A. & Hoyle, R.H. (1990). Perceived cohesion: A conceptual and empirical examination. Social Forces, 69 (2), 479-504. Broward Outreach Centers. (2007). Broward outreach centers: Hollywood and Pompano Beach: Comprehensive homeless assistan ce centers in south Florida Retrieved February 1, 2008, from http://www.miamirescuemission.com/browardoutreach.htm Brown, E.E., & Brooks, F. (2006). African Ameri can and Latino perceptions of cohesion in a multiethnic neighborhood. American Behavioral Scientist, 50 (2), 258-275. 92

PAGE 93

Brown, R., Condor, S., Matthews, A., Wade, G ., & Williams, J. (1986). Explaining intergroup differentiation in an industrial organization. Journal of Occupational Psychology, 59 273286. Bucker, J.C. (1988). The development of an instrument to measure neighborhood cohesion. American Journal of Community Psychology, 16 (6), 771-791. Bryman, A. (2004). Social research methods Oxford: Oxford University Press. Buechler, S.M. (1995). New so cial movement theories. The Sociological Quarterly, 36 (3), 441464. Cameron, J.E. (2001). Social identity, modern sexism and perceptions of personal and group discrimination by women and men Sex Roles, 45 743-766. Cameron, J.E. (2004). A three-factor model of social identity. Self and Identity, 3 239-262. Cameron, J.E. & Lalonde, R.N. (2001). Social identification and gender-related ideology. British Journal of Social Psychology, 40 59-77. Cameron, J.E., Salto, T., Lalonde, R.N., & Lay, C. (1997). A multidimensional measure of ethnic identification Poster session presented at the me eting of the Canadian Psychological Association, Toronto, Canada. Carmines, E.G. & Zeller, R.A. (1979). Reliability and validity assessment Thousand Oaks, CA: Sage Publications. Caton, C.L., Dominguez, B., Schnazer, B., Hasin, D.S., Shrout, P.E., Felix, A., McQuistion, H., Opler, L.A., & Hsu, E. (2005). Risk factors for long-term homelessn ess: Findings from a longitudinal study of firsttime homeless single adults American Journal of Public Health, 95(10), 1753-1759. Cattell, V. (2001). Poor people, poor places, and poor health: the mediating role of social networks and social capital. Social Science & Medicine 52, 1501-1516. Chapman, E. (2002). Effects of social cohesiveness and cooperative incentives on small group learning outcomes. Current Research in Social Psychology, 7 (17), 293-310. Cherry, E. (2006). Veganism as a cultural movement: A relational approach. Social Movement Studies, 5 (2), 155-170. Chin, W.W., Salisbury, W.D., Pearson, A.W., & Stollak, M.J. (1999). Perceived cohesion in small groups: Adapting and testing the percei ved cohesion scale in a small-group setting. Small Group Research, 30 751766. Clark, J. (2004). Social movement theory and patr on clientelism: Islamic social institutions and the middle class in Egypt, Jordan, and Yemen. Comparative Political Studies, 37(8), 941968. 93

PAGE 94

Committee to End Homelessnes s in Kings County. (2005). 10 year plan to end homelessness in Kings County: Highlights Seattle, WA. Retrieved January 8, 2008, from http://www.cehkc.org/highlights-3-9.shtml Committee to End Homelessn ess in Kings County. (2005). Frequently asked questions about a roof over every head in Kings county: Our community's 10-year plan to end homelessness. Seattle, WA. Retrieved January 8, 2008, from http://www.cehkc.org/10YPlanFAQs.pdf Coy, P. & Woerhle, L. (2000). Social conflicts and collective identities Lanham: Rowman & Littlefield Publishers, Inc. Crook, W.P., Mullis, R.L., Cormille, T.A., & Mu llis, A.K. (2005). Outcome measurement in homeless systems of care. Evaluation and Program Planning, 28 379-390. Dale, A. (2004, April 1). No place like home: Womens housing advocacy rebirth in Toronto. Women and Environments International Magazine. Daskal, J. (1998). In search of shelter: The growing shortage of affordable rental housing. Washington, DC: Center on B udget and Policy Priorities. Deaux, K. (1996). Social identification. In E.T. Higgins & A.W. Kruglanski (Eds.), Social psychology: Handbook of basic principles (pp. 777-798). New York: Guilford. DeCarmine, J. (2007, January 25). Executive summary: GRACE marketplace one-stop homeless center proposal Gainesville, FL. DeVaus, D. (2001). Research design in social research London: Sage Publications. DeVellis, R.F. (2003). Scale development: Theory and applications Thousand Oaks, CA: Sage Publications. Dweck, C.S. & Leggett, E.L. (1988) A social-cognitive approach to motivation and personality. Psychological Review, 95, 256-273. Einwohner, R. (2006). Identity work and collec tive action in a repressi ve context: Jewish resistance on the Aryan side of the Warsaw ghetto. Social Problems, 53 (1), 38-56. Ellemers, N., Kortekaas, P., & Ouwerkerk, J.W. (1999). Self categorization, commitment to the group, and group self-esteem as related but distinct aspects of social identity. European Journal of Social Psychology, 29 371-389. Elliot, A.J. & McGregor, H.A. (2001). A 2x2 achievement goal framework. Journal of Personality and Social Psychology, 80 501-519. Elliot, A.J. & Church, M. (1997) A hierarchical model of appr oach and avoidance achievement motivation. Journal of Personality and Social Psychology, 72 218-232. 94

PAGE 95

Elliot, E.S. & Dweck, C.S. (1988). Goals: An approach to motivation and achievement. Journal of Personality and Social Psychology, 54 5-12. Finney, S.J., Pieper, S.L., & Barron, K.E. (2004). Examining the psychometric properties of the Achievement Goal Questionnaire in a general academic context. Educational and Psychological Measurement, 64 (2), 365-382. Fisk, D., Rakfeldt, J., & McCormack, E. (2006). Assertive outreach: An effective strategy for engaging homeless persons with substa nce use disorders into treatment. The American Journal of Drug and Alcohol Abuse, 32 469-486. Fitzpatrick, C. (2004, October 25). Housing firs t becoming the standard model for homeless populations. Alcohol & Drug Abuse Weekly, 16 (41), 1-3. Future Generations. (n.d.) SEED-SCALE process for community change. Retrieved March 1, 2008, from http://www.future.org/app lied-research/process-change. Gainesville City Commission. (2006, October 30). Meeting minutes: Joint city/county commission special meeting. Gainesville, FL. Goodwin, J. & Jasper, J. (Eds.).(2003) Rethinking social movement s: Structure, meaning, and emotion. New York: Rowman & Littlefield Publishers. Gurin, P. & Markus, H. (1989). Cognitive conseque nces of gender identity. In S. Skevington & D. baker (Eds.) The social identity of women (pp.152-172). London: Sage Publications. Haynes, S.N., Richard, D.C.S., & Kubany, E.S. (1995). Content validity and psychological assessment: A functional approach to concepts and methods. Psychological Assessment, 7(3), 238-247. Herzog, H. (2005). On home turf: Interv iew location and its social meaning. Qualitative Sociology, 28 (1), 25-46. Heywood, A. (1997). Politics. London: Macmillin Press, Ltd. Hinkle, S., Taylor, L.A., Fox-Cardamone, D.L., & Crook, K.F. (1989). Intragroup identification and intergroup differentiation: A multicomponent approach. British Journal of Social Psychology, 28 305-317. Homelessness: A solvable problem. (2007) America, 196 (8), 5. Ilgen, D.R., Shapiro, J., Sala s, E., & Weiss, H. (1987). Functions of group goals: Possible generalizations from individuals and inferences for training Unpublished manuscript, Michigan State University. Jasper, J.M. (1997). The art of moral protest: Culture, biography, and crea tivity in social movements. Chicago: University of Chicago Press. 95

PAGE 96

Jensen, B. (2005). Program pioneering Housi ng First is catching on across the country. Chronicle of Philanthropy, 17 (21). Kauffman, L.A. (1990). The anti-politics of identity. Sociological Review, 90(1), 67 Kipnes, D. & Joyce, A. (1999). A brie f review of group cohesion research. The Group Circle. Klein, H.J. & Mulvey, P.W. (1995). Two investig ations of the relationships among group goals, goal commitment, cohesion, and performance. Organizational Behavior and Human Decision Processes, 61 (1), 44-53. Kriesi, H., Koopmans, R., Duyve ndak, J.W., & Giugni, M. G. (1992). New social movements and political opportunities in Western Europe. European Journal of Political Research, 22, 219-244. Langman, L. (2005). From virtual public sphere s to global justice: A critical theory of internetworked social movements. Sociological Theory, 23 (1), 42-74. Legander, S. (2006). Housing firs t: A program to help people m ove off the streets and into treatment. Behavioral Healthcare, 26 (5), 38. Marvasti, A.B. (1999). Being homeless: Textual and narrative constructions Unpublished doctoral dissertation, University of Florida, Gainesville. McAdam, D. & Snow, D.A. (2001). Social movements: Readings on their emergence, mobilization, and dynamics. Los Angeles: Roxbury Publishing Company. Merton, R. (1983). Introduction to sociology. New York: Harcourt Brace Jovanovich, Inc. Miles, M.B. & Huberman, A.M. (1994). Qualitative data analysis: An expanded sourcebook. Thousand Oaks, CA: Sage Publications. Miller, D.C. & Salkind, N.J. (2002). Handbook of research design & social measurement (6th Ed.). Thousand Oaks, CA: Sage Publications. Miller, H. (1991). On the fringe: The dispossessed in America. Lexington, Massachusetts: Lexington Books. Morris, A. (1981). Black southern student sitin movements: An analysis of internal organization. American Sociological Review, 45 744-67. Mowbray, C.T., Cohen, E., & Bybee, D. (1993). The challenge of outcome evaluation in homeless services: Engagement as an intermediate outcome measure. Evaluation and Program Planning, 16, 337-346. Mueller, C.M. (1994). Conflict networks and the origins of womens liberation. In D. McAdam & D.A Snow (Eds.), Social movements: Readings on their emergence, mobilization, and dynamics (pp. 158-171). Los Angeles: Roxbury Publishing Company. 96

PAGE 97

National Alliance to End Homelessness. (2006). How much does the Federal Government spend on homelessness? Explainer: Questions and answers on homeless policy and research National Coalition for the Homeless. (2006). Why are people homeless?. Retrieved April 15, 2007, from http://www.nationalhomeless. org/publications/facts.html National Low Income Hous ing Coalition. (2005). The crisis in Americas housing Retrieved March 1, 2008, from www.n lihc.org/doc/072106chart.pdf New York City Department of Homeless Services. (2007). NYC department of homeless services. Retrieved March 15, 2008, from http://www.nyc.gov/html/dhs/html/home/home.shtml Nicholls, J.G. (1984). Achievement motivation: Conceptions of ability, subjective experience, task choice, and performance. Psychological Review, 91 328-346 Nickelson, I. (2004). The district should use its upcomi ng TANF bonus to increase cash assistance and remove barriers to work Washington, D.C.: D.C. Fiscal Policy Institute. Nolen, S.B. (1988). Reasons for studying: Moti vation orientations a nd study strategies. Cognition and Instruction, 5, 269-287. Noonan, R.K. (1995). Women against the state: Political opportunities and collective action frames in Chiles transition to democracy. Sociological Forum, 10 81-111. OLeary-Kelly, A.M., Martocchio, J.J., & Frink, D.D. (1994). A revi ew of the influence of group goals on group performance. Academy of Management Journal, 37(5), 1285-1301. Obst, P., Zinkiewicz, L., & Smit h, S.G. (2002). Sense of community in science fiction fandom: Part 1: Understanding sense of community in an international community of interest. Journal of Community Psychology, 30 87-103. Padgett, D.K., Gulcure, L., & Tsemberis, S. (2 006). Housing first serv ices for people who are homeless with co-occurring serious mental illness and substance abuse. Research on Social Work Practice, 16 (1), 74-83. Phelan, J., Link, B.G., Moore, R.E., Stueve, A. (1997). The stigma of homelessness: The impact of the label homeless on attitudes towards poor persons Social Psychology Quarterly, 60(4), 323-337. Poland, B. & Pederson, A. (1998). Reading between the lines: Interpreting silence in qualitative research. Qualitative Inquiry, 42 (2), 293-313. Poletta, F. & Jasper, J. (2001). Collec tive identity and social movements. Annual Review of Sociology, 27 283-305. Reicher, S. (2004). The contex t of social identity: Domina tion, resistance, and change. Political Psychology, 25(6), 921-945. 97

PAGE 98

Riordan, C.M. & Weatherly, E.W. (1999). Defi ning and measuring employees identification with their work groups. Educational and Psychological Measurement, 59 310-324. Rollinson, P. (2007). A rural problem t oo: Homelessness beyond the big cities. Planning. Ronan, W.W., Latham, G.P., & Kinne, S.B. (1973) Effects of goal setting and supervision on worker behavior in an industrial setting. Journal of Applied Psychology, 58 302-307. Rubin, H.J. & Rubin, I.S. (1995). Qualitative interviewing: The art of hearing data Thousand Oaks, CA: Sage Publications. Russell, A. (2005). Myth and the Zapatista movement: exploring a network identity. New Media & Society, 7 (4), 559-577. Ryen, A. (2001). Cross-cultural interviewing. In J.F. Gubrium & J.A. Holstein (Eds.). Handbook of interview research: Context & method (pp. 335-354). London: Sage Publications. Seashore, S.E. (1954). Group cohesiveness in the industrial work group Ann Arbor: University of Michigan, Institute for Social Re search, Survey Research Center. Sheskin, D.J. (2004). Handbook of parametric and nonparametric statistical procedures. Boca Raton, FL: Chapman, & Hall. Simsek, S. (2004). New social movements in Turkey since 1980. Turkish Studies, 5 (2), 111-139. Sommer, R. & Sommer, B. (2002). A practical guide to behavi oral research: Tools and techniques. New York: Oxford University Press. Snow, D., Soule, S., & Krieski, H. (Eds.) (2004). The Blackwell companion to social movements. Malden, MA: Blackwell Publishing. Speer, P.W., Jackson, C.B., & Peterson, N.A. ( 2001). The relationship between social cohesion and empowerment: support and new implications for theory. Health Education & Behavior, 28 716-732. Swithinbank, T. (1997). The street press: hom elessness, self-help, and social identity. Gender and Development, 5 (3), 45-51 Tajfel, H. (1978). Differentiation betw een social groups London: Academic Press. Tajfel, H. & Turner, J.C. (1979). An integrative th eory of intergroup conflict. In W.G. Austin & S. Worchel (Eds.), The Social Psychology of intergroup relations (pp. 33-47). Monterey, CA: Brooks-Cole. Tarrow, S. (1988). National politics and collective action: Recent theory and research in Western Europe and the United States. Annual Review of Sociology, 14 421-440. The Economic Policy Institute. (2005). Minimum wage: Frequently asked questions. Retrieved April 12, 2008, from http://www.epi.org/conten t.cfm/issueguides_minwage_minwagefaq. 98

PAGE 99

Thompson, S.J., McManus, H., Lantry, J., Windsor L., & Flynne, P. (2006). Insights from the street: Perceptions of services a nd providers by homeless young adults. Evaluation and Program Planning, 29, 34-43. Tinker, C. (2007, September 6). Reaction varies on homeless center site [Electronic version]. The Gainesville Sun. Retrieved January 1, 2008, from http://www.gainesville.co m/article/20070906/GUARDIAN/70906015 Tortu, S., Goldsamdt, L.A., Hamid, R. (2002). A practical guide to research and service with hidden populations. Boston: Allyn & Bacon. Touraine, A. (1988). Return of the actor: Social th eory in post-industrial society Minneapolis: University of Minnesota Press. Touraine, A. (1985). An introduction to the study of social movements. Social Movements, 52(4), 749-787. Tsemberis, S., Gulcur, L., & Nakae, M. (2004) Housing first, consumer choice, and harm reduction for homeless individua ls with a dual diagnosis. American Journal of Public Health, 94(4), 651-656. U.S. Department of Housing and Urban Development. (2007). Homelessness: Programs and the People They Serve. Washington, DC. Retrieved from January 1, 2008, http://www.huduser.org/publications/ homeless/homelessness/ch_1c.html U.S. Department of Housing and Urban Development. (2007). McKinney-Vento act. Washington, DC. Retrieved January 1, 2008 from http://www.hud.gov/offices/cpd/hom eless/lawsandregs/laws/ U.S. Department of Housing and Urban Development. (2007). The homeless in America: A profile based on the first annual hom eless assessment report (AHAR) Washington, DC. Retrieved January 1, 2008, from http://www.hud.gov/offices/cpd/homeless/library/homelessnessfactsheet.pdf U.S. Department of Housing and Urban Development. (2007). Homeless assistance programs Washington, DC. Retrieved January 1, 2008, from http://www.hud.gov/offices/cpd/homeless/programs/index.cfm Useem, B. (1985). Disorganization and the New Mexico prison riot of 1980. American Sociological Review, 50 667-688. Vahabzadeh, P. (2001). A critique of the ultimate referentiality in the new social movement theory of Alberto Melucci. Canadian Journal of Sociology, 26 (4), 611-633. Walgrave, S. & Verlhust, J. (2006). Towards new emotional movements? A comparative exploration into a specific movement type. Social Movement Studies, 5 (3), 275-304. 99

PAGE 100

Walsh, E.J. & Warland, R. (1983). Social movement involvement in the wake of a nuclear accident: Activists and free riders in the TMI area. American Sociological Review, 48 764780. Warren, C.A.B. (2001). Qualitative interviewing. In J.F. Gubrium & J.A. Holstein (Eds.). handbook of interview research: Context & method (pp. 83-101). Thousand Oaks, CA: Sage Publications. Weber, R. P. (1985). Basic content analysis Thousand Oaks, CA: Sage Publications. Weems, G.H. & Onwuegbuzie, A.J. (2001). The impact of midpoint responses and reverse coding on survey data. Measurement and Evaluation in Counseling and Development, 34, 166-176. Weinreb, L.F., Buckner, J.C., Williams, V., & Ni cholson, J. (2006). A comparison of the health and mental status of homeless moth ers in Worcester, Mass: 1993 and 2003. American Journal of Public Health, 96 (8), 1444-1448. Weldon, E. & Weingart, L.R. (1993). Group goals and group performance. British Journal of Social Psychology, 32 307-334. Wievorka, M. (2005). After new social movements. Social Movement Studies, 4 (1), 1-19. Williams, J.C. (2005). The politics of homelessness: Shelter Now and political protest. Political Research Quarterly, 58 (3), 497-509. Willis, G.B. (2005). Cognitive interviewing: A tool fo r improving questionnaire design Thousand Oaks, CA: Sage Publications. Zander, A., Forward, J., & Albert, A. (1969). Adap tation of board members to repeated success or failure by their organization. Organizational Behavior and Human Performance, 4, 5676. Zugman, K. (2003). Political cons ciousness and new social movement theory: the case of fuerza unida. Social Justice, 30 (1), 153-176. 100

PAGE 101

BIOGRAPHICAL SKETCH Norma Cadavieco was born in 1984, in Miami, Florida. She grew up mostly in Miami and Fort Lauderdale, graduating from Western High School in Davie, Florida in June 2002. She earned her Bachelor of Science degree in fa mily, youth, and community sciences from the University of Florida in August 2006. Norma was accepted into the joint B.S./M.S. program in Family, Youth, and Community Sciences before he r final year of undergraduate study. In this program, she was able to begin gradua te courses during he r senior year. Upon completion of her masters program, Norma hopes to continue researching homelessness and social movements to improve programming and outreach for homeless persons and the poor in the U.S. Norma currently resides in Brooklyn, New York with her black cat, Radon. She is working as a case manager in a transitional housing facilit y in Manhattan for formerly homeless adults. 101