Home Articles Reader Opinion Editorial Book Reviews Discussion Writers Guide About TCRecord
transparent 13

Homeless Education and Social Capital: An Examination of School and Community Leaders

by Peter M. Miller - 2011

Background/Context: This study contributes to the literature on the schooling of homeless and highly mobile students. Although previous work has detailed the demographics of homelessness, the effects of homelessness on academic progress, and particular legal issues in homeless education, this research focused on how individual and institutional relationships influence homeless education.

Purpose/Objectives: The purpose of the study was to develop deeper understanding of how schools and shelters helped create educational social capital for students and families who were experiencing homelessness. The guiding research questions for the study were: (1) How do school and shelter leaders perceive social capital as influencing the education of students who are homeless? and (2) How do school and shelter leaders’ relational networks influence the education of students who are homeless?

Setting: Data were collected from three homeless shelters and three public schools that are located in a large city in the Mid-Atlantic region of the United States.

Participants: A total of 31 interviews were conducted with shelter-based administrators, case workers, and child development specialists, and school-based principals and central office administrators.

Research Design: A qualitative collective case study research design was employed.

Findings: Homeless students and families appeared to have insufficient stores of productive social capital, and although schools and homeless shelters provided them with some important relationships and resources, school and shelter leaders’ own shortages of bridging social capital limited the extent to which efficient educative active could occur.

Conclusions/Recommendations: Based on the findings, it is suggested that schools and shelters prioritize social capital development and improve interorganizational networking. Specifically, purposeful efforts should be made to develop school-shelter-family networks that are heterogeneous in composition.


Educating students who are experiencing homelessness is a complex and difficult task that beckons the service of multiple individuals. Key actors from school districts, shelters, and other social service agencies are called to join the efforts of families to address the students’ needs—needs that are often considerably affected by the students’ extremely difficult “outside of school” lives. Predictably, as these diversely positioned school and community leaders work toward a common end—academic and social growth for the students—the obstacles and challenges that they face are abundant. The purpose of this qualitative study was to learn more about how these challenges were navigated in one particular urban setting. Specifically, guided by the work of James Coleman (1988) and other social capital theorists, I examined urban administrators’ perceptions of how social capital issues influenced the educational experiences of students and families who are experiencing homelessness.


It is notoriously difficult to identify how many people are homeless in the United States. Most homeless counts provide broad ranges of numbers (i.e., between 2.3 and 3.5 million people; Bowman & Barksdale, 2004) over specific periods of time (1 day, 1 month, 1 year, and so on) and, depending on the counting criteria, these estimates tend to fluctuate greatly.1 Regardless of estimating techniques and criteria, however, there is widespread recognition that recent spikes in home foreclosures, coupled with broader recession-related dilemmas, have escalated the breadth and depth of the homelessness crisis in the United States. Although many of those who, for years, struggled to maintain stable housing continue to languish, others who never before encountered such challenges to meet daily living needs are now straining to avoid conditions of homelessness. Local indicators of the escalating severity of the homeless crisis are evident even before macro-level estimates (HUD data, and so on) become available. For example, the Allegheny, Pennsylvania, County Department of Human Services reported a 35% increase in the number of users of emergency shelter services in the Pittsburgh Metro area during the last quarter of 2008, with over a third of these individuals being first-time users of county services (i.e., the “new homeless”). Michael Lindsey, director of homeless services for Allegheny County, suggested that these numbers are not peculiar to the Pittsburgh region, but can seen with striking clarity in diverse areas across the United States (personal communication, January 7, 2009). In fact, a recent (2008) National Coalition for the Homeless survey supported this assertion, indicating that 72% of state and local homeless coalitions noted significant increases in homeless populations in their areas since 2007.

The surge in homelessness is especially troubling in that it is not relegated to adults; in fact, for several years, the growth rate of school-age children who are homeless—10% percent—has far surpassed the overall 2% growth rate in the United States (Finkelstein, 2005). Some recent examples illuminate this trend:

During the first few months of the 2008–2009 school year, the Clark County, Nevada, School District reported 1,500 homeless students—twice the number of students that was reported the entire previous year. (Gewertz, 2008)

The Macomb Intermediate School District in suburban Michigan reported its highest ever number of homeless students (514) early in the 2008–2009 academic year—a 33% increase over the previous year. (Johnson, 2008)

Data collected in 2008 from over 1,000 school districts by the National Association for the Education of Homeless Children and Youth indicated that with 2 months remaining in the school year, most districts had already identified and served more homeless students than ever before in a full year.

Accordingly, previous point-in-time estimates that there are between 900,000 and 1.4 million (Capuzzi & Gross, 2004) homeless children in the United States are only a fraction of the numbers to come. For example, the Washington, D.C.-based family advocacy group First Focus projects that 2.2 million new home foreclosures are likely to affect more than 2 million children within the next 2 years. Not all these students will experience homelessness, but many certainly will. Homelessness, then, affects not just the poorest of the poor in the United States; it has touched (and will continue to touch at increasing rates) diverse students from urban, suburban, and rural backgrounds throughout the country (Cauce, 2000; Jencks, 1994).


Not surprisingly, homelessness has a vicious impact on children’s growth and development (Hagan & McCarthy, 1997; Nunez, 1996; Stormont, 2004). The multitude of conditions associated with homelessness among youths include institutional neglect, hunger, splintered families, inadequate health conditions, social stigma, and physical, psychological, and substance abuse (Gibel, 1996; Gross & Capuzzi, 2004; MacLean, Embry, & Cauce, 1999; Rafferty, 1997). Additionally, Cauce (2000) found extremely high rates of suicide among children who experience homelessness. These assorted troubling conditions have been described as both preconditions to homelessness and effects of homelessness. This range of “homeless devastation” among families and children has been documented in several noteworthy books, including those by Jonathan Kozol (1988), Susan Quint (1994), and Christopher Jencks (1994).

Considering the plethora of social problems associated with homelessness, it is not surprising that children in such conditions tend struggle in school.2 Perhaps the most immediately evident indicator of these struggles can be seen when looking at school attendance statistics. Students who are experiencing homelessness miss an exceptionally high number of class days. It was estimated that in 1998, 45% of homeless students were not attending school (Stronge, 2000). Unfortunately, even when children who are homeless do attend class, they tend to be highly “school-mobile.”3 That is, as their families move from one temporary location to the next, the children change schools (Taylor & Adelman, 2000). Situated within the daunting social struggles that contribute to and sustain homelessness are serious implications for such mobility, because “with each change in schools, a student is set back academically by an average of four to six months” (Moore, 2005, p. 2). School mobility, in fact, has been identified as an especially important variable to examine among homeless students in that it separates them from many other at-risk students, whose otherwise difficult situations (often characterized by poverty, failing schools, and troubled neighborhoods) are not as consistently impacted by the accompaniments of mobility (Miller, 2009b). In response, The McKinney-Vento Homeless Assistance Act of 1987 (since reauthorized in 2001 as Title X, Part C of the No Child Left Behind Act–• 725) was adopted to give students who are homeless the option to remain in their schools of origin or to enroll in the school nearest to their new shelter (or other place of temporary residence). McKinney-Vento has benefitted many students by giving them opportunities to maintain stable school experiences, but because of practitioners’ limited understandings of the policy, their limited capacities to effectively implement its costly transportation requirements, and/or, in many instances, their blatant disregard for its stipulations, the policy’s broad potential to lessen student mobility has not been actualized (refer to Miller & Hafner, 2008, for full analysis of McKinney-Vento’s impact on the homeless education system).

In addition to (and, undeniably, in relation to) attendance-related school dilemmas, research indicates that students who experience homelessness are, in comparison with other students, affected to a much greater extent by social/behavioral dilemmas and are much more likely to be placed at risk of academic failure (Gibel, 1996; Gross & Capuzzi, 2004; Vissing, 2000). Specifically, experiences of homelessness correlate with decreases in test scores and increases in disability identification, dropout rates, and violent behavior (Bowman & Barksdale, 2004; Lively & Kleine, 1996; Stronge & Reed-Victor, 2000). Also, in comparison with others, it is more than two times as likely that homeless children will repeat a grade (Tucker, 1999).

Despite these seemingly overwhelming educational difficulties that face homeless children, research indicates that school remains one of the few places where homeless children can potentially maintain a degree of normalcy in their lives (Tucker, 1999). Schools can provide children with physical care, space, stability, security, autonomy, positive social interactions, competency, and creative outlets (Eddowes & Butcher, 2000). Noguera (2001) explained,

Children who are homeless, undocumented, sick or disabled, hungry or abused, all have a right to public education. Given the harsh realities confronting the poorest people in this country, schools are often the only place where children can be guaranteed at least one meal, a warm building, and relative safety under adult supervision. Public schools are, in effect, the most significant remnant of the social safety net available to poor people in the United States. (p. 197)

In turn, Rafferty (1997) wrote that when administrators and teachers provide the students with special resources and services to meet their particular needs, schools can be particularly adept at “cushioning the blow” (p. 48) of homelessness. When considering issues of homelessness among youth, then, schooling is a pivotal area of concern.


The leadership practice4 that guides schools and shelters is regularly identified as a factor that is of central importance in the education of students who are homeless (Stronge & Reed-Victor, 2000). Case workers, child development specialists, and youth coordinators at shelters, and principals, social workers, and district administrators from schools all have fundamental roles in getting students enrolled in school, transported to school, and socially and academically integrated within their school (Miller, 2007b).5 Their work is collaborative, in that they must communicate about students’ situations of homelessness and academics, providing each other with information and support that is necessary for students’ needs to be met. Although almost all interagency collaboration is challenging, the collaborative leadership work of schools and shelters seems to be especially difficult because the problems that accompany homelessness are so complex and pervasive. Stronge (2000) emphasized that in such conditions, “The quality of leadership of the people who are part of interagency partnerships is critical and their efforts should build on their collective vision, commitment, and competence” (p. 15). Unfortunately, these leaders—regardless of how exceptional they may be in terms of their commitment and competence as inner-building administrators—are often limited in their capacities to collaborate by their paucity of experience with and understanding of each other (Miller, 2007a). Although it is certainly clear that the leadership practice that influences the education of students who are homeless is affected by these collaborative limitations, we need to learn more about how school and shelter workers attempt to work together and, in turn, why they often struggle to forge productive relationships. Accordingly, this study examines school-shelter relationships that are founded in service of students who are homeless. I then describe the construct of social capital, which serves as the theoretical guide for the study.


With roots in sociological theory, social capital is an empirical concept that has been defined in a multitude of ways and investigated in diverse organizational and disciplinary settings (Portes, 1998). The multifarious definitions of the term include Putnam’s (1993, 1995) well-known collective/community-level construction of social capital as the “features of social organization, such as networks, norms, and trust, that facilitate coordination and cooperation for mutual benefit” (p. 66); Coleman’s (1988) description of social capital as individuals’ relationally situated guides and benefits that undergird human capital development; and Lin’s (1999, 2000) flexibly applied construct of social capital as the advantages that individual and collective actors accrue from their locations within social networks. In these and other iterations, social capital has been used effectively in the field of education to examine students’ and families’ environments, expectations, opportunities, and achievements (e.g., Coleman; Goddard, 2003; Greeley, 1997; Lee & Croninger, 1994, 2001; Stanton-Salazar, 1997). Broadly summarized, this research tends to suggest that certain kinds of social capital facilitate productive educational experiences for students, families, and/or communities. Accordingly, when used with clarity and precision, social capital is a concept that can lend valuable guidance to individual and collective-level analyses6 of leadership, collaboration, and the schooling of children who are homeless.


Situated within the field of educational research and social capital, this study draws particularly from James Coleman’s (1988) perspective. Coleman’s seminal article entitled “Social Capital and the Creation of Human Capital” was of foundational importance to the discussion of social capital in education. Here (and in his later works), he highlighted the central relevance of social capital in individuals’ social, educational, and corporate lives and examined how various organizational structures and cultures build on and/or create social capital within and among diverse actors. Coleman contrasted social capital with the more easily identifiable and commonly addressed constructs of physical and human capital7:

If physical capital is wholly tangible, being embodied in observable material form, and human capital is less tangible, social capital is less tangible yet, for it exists in the relations among persons. Just as physical capital and human capital facilitate productive activity, social capital does as well. For example, a group within which there is extensive trustworthiness and extensive trust is able to accomplish much more than a comparable group without that trustworthiness and trust. (p. 100–101)

As implied here, Coleman and others (such as Greely, 1997; Putnam, 1995; and Warren, 2005) have suggested that extensive reserves of mobile social capital are often invaluable in the attainment of desired social, educational, and professional ends. Coleman, whose work guides much of this study’s analysis, described social capital as being most clearly manifested in relations that provide (1) obligations, expectations, and trustworthiness; (2) information channels; and (3) norms and effective sanctions. Additionally, he explained that certain kinds of social structure are especially important in facilitating these relational characteristics.

Obligations, Expectations, and Trustworthiness

Illustrating how social capital is gained from relations that provide obligations, expectations, and trustworthiness, Coleman wrote,

If A does something for B and trusts B to reciprocate in the future, this establishes an expectation in A and an obligation on the part of B. This obligation can be conceived as a credit slip held by A for performance by B. . . . This form of social capital depends on two elements: trustworthiness of the social environment, which means that obligations will be repaid, and the actual extent of obligations held. (p. 102)

Trust, as described here, is often forged through family, religious, cultural, and/or community bonds. As “members” of such social networks, individuals can operate freely and openly without fear of deceit or disingenuousness. They hold mutually understood obligations to be truthful to each other and to help each other in times of need. This type of social capital is highly pragmatic—individuals help each other “get things done.” In these particularly tight webs of relationships, individuals who violate their group’s ethical norms and expectations (i.e., those who do not honor their “social/relational debts”) face multiply manifested repercussions in social, educational, professional, and other arenas. So, just as much can be derived from “credit slip” relationships, much can be lost for those who violate the trust that underlies such commitments.

Information Channels

Coleman (1988) described another important form of social capital as the “potential for information that inheres in social relations” (p. 104). “Information” is depicted as important in that it provides a basis for action; unfortunately, the acquisition of relevant “life” information can be a complex and time-consuming task. For example, it would seem that the more information a mother has about her son’s schooling experiences, the better she would be able to support him academically and socially. However, most mothers are limited in the amount of time that they have to read school handbooks, curriculum guides, and the like. In this scenario, the parent could gain vital information in such areas quickly and accurately if she had relationships with teachers, administrators, or other parents—any of whom would be fonts of school information. Relationships, then, not only can provide information that can assist individuals in their daily lives, but they can do so in a way that is efficient. Coleman summarized, “The relations in this case are not valuable for the ‘credit slips’ they provide in the form of obligations that one holds for others’ performances or for the trustworthiness of the other party, but merely for the information that they provide” (p. 104).

Norms and Effective Sanctions

Finally, presenting social capital in the form of behavioral norms and effective sanctions, Coleman (1988) described how tightly knit relationships with others often provide individuals with influential cues about how they should act. Oftentimes these cues suggest that individuals should forgo their own self-interests for the good of the collective group of which they are a part. Relations here provide norms and moral guidance for social and professional behavior.8 Coleman wrote that, “In some cases, norms are internalized; in others, they are largely supported through external rewards for selfless actions and disapproval for selfish actions. . . . This social capital, however, like the forms described earlier, not only facilitates certain actions; it constrains others” (p. 105). A “positive” example of how relationships can facilitate productive action could be a teenage girl who is part of a group of friends who are very socially conscious. As a part of this group, the girl would be influenced to act responsibly and to treat others with respect. Her much-valued bonds with these friends would be strengthened to the extent that her actions paralleled their collective values. These “behavioral cues” are a noteworthy source of social capital.

Structural Foundations of Social Capital

Importantly, Coleman (1988) asserted that certain kinds of social structure are especially efficacious at facilitating the aforementioned forms of social capital. Most notably, social capital seems to especially flourish in networks where there are multiplex relations. Broadly speaking, this refers to situations in which the profits derived from relationships are not limited to those that are most immediately apparent; rather, they provide social support, expectations, obligations, and/or information that facilitate productive action in multiple venues. For example, if your neighbor is a banker who has a child in the same class as your own child, the benefits of your relationship with her would likely exceed those that might typically be associated with a “good neighbor” (such as keeping an eye on your house when you are away and lending you a cup of sugar when you need it). From this neighbor, you might also gain valuable information about what is happening at school and how your children are progressing, and because she is a banker, you might even learn from her about how to refinance your home. This single relationship could help you become a more effective community resident, parent, and consumer. Coleman (1988) explained,

In (multiplex relations), persons are linked in more than one context (neighbor, fellow worker, fellow parent, coreligionist, etc.), while in simplex relations, persons are linked through only one of those relations. The central property of a multiplex relation is that it allows the resources of one relationship to be appropriated for use in others. Sometimes the resource is merely information, as when two parents who see each other as neighbors exchange information about their teenagers’ activities; sometimes, it is the obligations that one person owes a second in relationship X, which the second person can use to constrain the actions of the first in relationship Y. Often, it is resources in the form of other persons who have obligations in one context that can be called on to aid when one has problems in another context. (pp. 107–108)

Especially when discussing issues related to parenting and schooling, a particularly important form of multiplex relation occurs when parents are close acquaintances or friends with the parents of their children’s friends, as well as with teachers, principals, and other influential adults in the community. Parents do not need to operate independently in these conditions; their children can be raised with a unified set of expectations and behaviors, and their development can be addressed holistically (Warren, 2005). Such “intergenerational closure” promotes positive social and educational experiences. Coleman explained,

The social capital that has value for a young person’s development does not reside solely within the family. It can be found outside as well in the community consisting of the social relationships that exist among parents, in the closure exhibited by this structure of relations, and in the parents’ relations with the institutions of the community. (p. 113)

It is clearly evident, then, that value can be derived from structures and relationships that facilitate the development of large “social capital accounts.”

Value of Social Capital

Interestingly, Coleman (1988) suggested that the extent of value that social capital has varies from individual to individual. For example, in areas where there are extensive government services (such as public transportation and childcare services), people are less dependent on the social capital that inheres in interpersonal relations. Further, those who have extensive reserves of physical (material objects) and/or human capital (like skills and intelligence) might be less in need of social capital. On the other hand, those who have a dearth of personal resources or who live in areas where public services are provided on a very limited basis are often extremely dependent on social capital. For such individuals—often those who occupy the lowest levels of socioeconomic status—social capital provides information, support, trust, and closure that is essential to their productivity. Implied here is the fact that social capital is most important to those who are poor and disenfranchised in society.

Although there are many ways to analyze leadership and collaboration that serves students who are homeless, this study is informed by Coleman’s (1988) conceptualization of social capital. Specifically, I sought to learn about how the relationships and social networks of leaders and the homeless families they serve influenced the educational process. The guiding research questions for the study were: (1) How do school and shelter leaders perceive social capital as influencing the education of students who are homeless? and (2) How do school and shelter leaders’ relational networks influence the education of students who are homeless?


Qualitative interviews, observations, and document analyses were used to investigate participants’ perceptions of school-shelter leadership and collaboration in Middleton,9 a large city in the Mid-Atlantic region of the United States. Data were collected from three shelters and three schools in Middleton’s urban downtown corridor.10


Each of the three shelters that was chosen for this study serves a high number of families from downtown Middleton and is well regarded for its long-standing presence in the local community. The Family Space (TFS) is a long-term shelter for women and their children who are experiencing homelessness for any reason; Mothers’ Center and Support (MCS) is a 60-day shelter for women and their children who are without permanent housing because of issues of abuse and/or domestic violence; and Hope Society (HS) is an emergency shelter for women and children who experiencing homelessness for any reason. Although the shelters differ in their specific missions (focuses on long-term residence, emergency residence, domestic violence, and so on), the sizes of their staffs (MCS has over 50 staff members, whereas the others are smaller), and other key areas, they are each relevant for this study because they engage issues of homeless education in downtown Middleton on a daily basis. Specifically, because each of these shelters is located in the Middleton School District and the children from their shelters typically attend Middleton Schools, the insights from staff members at TFS, MCS, and HS are particularly meaningful in this focused examination of leadership and collaboration for homeless education.

An array of professionals carries out key duties at each of these shelters, but I was particularly interested in gaining the insights of those who work directly with students and families. Accordingly, rather than interviewing the executive directors of the shelters—leaders who play key administrative roles in the shelters but who have little day-to-day interaction with students, families, or schools—I interviewed case managers (those who work with mothers and families on a daily basis), child development specialists (those who support students’ wider social, emotional, and cognitive growth), and after-school programming specialists (who guide students’ in-shelter tutoring and academic activities). These “frontline” leaders’ perspectives of students, families, schools, and the educational process were rich and experientially informed.


The schools that were chosen for the study, Macon Elementary School, Lane Middle School, and Williams Elementary School, are each located within a few blocks of at least one of the previously described shelters. Of these, Macon, which is located just two blocks from TFS, usually has the highest enrollment of shelter-based homeless students (estimated by school leaders to be at least 25–30 students, out of 360 total, at any given point in time). Staff members at Macon, which serves predominantly African American students, openly acknowledge homelessness as a pressing issue in their school community. Like Macon, many of Lane’s students are highly mobile, but not all because of homelessness. The racially and socioeconomically diverse school is located in close in proximity to two major universities where many Lane students’ parents/guardians are students, faculty, or staff. Students cycle in and out of Lane at significantly higher rates than at other neighborhood schools in the region. Although Lane does not tend to have as many homeless students as Macon (estimated by school leaders to be 15–20 students at any given time out of 402 total), the school was chosen for the study because it was identified by staff members at TFS as one that “works well” with homeless students. The principal at Williams Elementary was unable to offer an estimate of how many homeless students tend to be enrolled at his school (which has 197 total students) but suggested that his students—most of whom are African American—are the “poorest of the poor” and that “most” of them experience regular episodes of housing instability. He claimed that even though many of his students do not reside in neighborhood shelters, they regularly bounce between houses and therefore, based on McKinney-Vento’s inclusion of those who “double-up” with others as being eligible for homeless services, are homeless.

The schools are all part of the Middleton School District, which, like many large urban school districts on the eastern seaboard, has faced significant struggles in recent years. The district has been plagued by dwindling enrollment, faltering achievement marks (as measured by No Child Left Behind standards), low graduation rates, and financial problems. The district served over 300 students who experienced homelessness during the 2006–2007 school year—a number that reflects only a small portion of the total number of children who were without permanent residence in the community (many more go unidentified).11 In addition to interviewing leaders at Macon, Lane, and Williams, I also interviewed relevant administrators from the district’s central office, including the district homeless liaison who plays a central role in the enrollment and transportation of homeless students (refer to Table 1).

Table 1. Middleton Area Research Sites and Participants



Interview Participants

The Family Space (TFS)

Long-term site (up to 2 years) for women and families experiencing homelessness for any reason; small staff (15).

Case managers (2), child development specialists (4).

Mothers’ Center and Support (MCS)

60-day (maximum) site for women and children experiencing homelessness due to domestic abuse or violence; large staff (60).

Child specialists (4), after-school program staff (6).

Hope Society (HS)

Emergency shelter for those experiencing homelessness for any reason; highly mobile clientele.

Case manager, family specialist; liaison.

Macon Elementary

Urban; 360 students; 99% African American students; shares neighborhood with TFS.

Principal (2), central administrator, liaison (2).

Lane Middle

Urban; 402 students; racially and socioeconomically diverse; shares neighborhood with MCS.

Principal, central administrator, liaison.

Williams Elementary

Urban; 197 students; low socioeconomic status; 99% African American students.

Principal, central administrator, liaison.


In all, 31 interviews were conducted with shelter-based administrators, case workers, and child development specialists, and school-based principals and central office administrators over a 12-month period. Each of the participants was identified as a key actor in the leadership of school-shelter collaborative arrangements in Middleton. They were information-rich (Patton, 1990) participants in that they were willing to share openly about their personal experiences, beliefs, expectations, and concerns relating to the education of students who are homeless. The interviews were semistructured (Creswell, 1998) and loosely structured around issues of social capital (refer to Table 2 for the semistructured interview protocol). All the interview sessions were tape recorded, held at the participants’ places of employment, and lasted between 45 minutes and 2 hours. Importantly, documental data provided vital sources of support. These data took the shape of hundreds of pages of shelter literature on family in-take procedures, school contact information, and resident policies and was used to question and confirm interview data.

Table 2. Semistructured Interview Protocol


What is the mission of this organization (purposes, population served, etc.)?


What do you do? What’s your typical day like? Etc.


What is your background (education, etc.) and training like?


Why do you do this work? (what led you to it and why do you still do it?)


What are your short and long-term professional aspirations?


What are the biggest challenges/rewards of this work?


Do you have a standardized protocol that you follow with mothers/children when they check in? Do all staff follow it? Do you identify student assets/strengths?


How long do mothers/children typically stay here? Do the kids usually stay at their previous schools?


What is the nature of your interactions with students and their mothers?


What types of challenges are the mothers/children typically facing when they check in here? What are the “biggest” ones?


What types of supportive networks do the mothers/families tend to have?


What other relationships are important to the mothers?


What school partners do you work with the most in attempting to meet the needs of students?


How often do you speak with them? What is the nature of your relationship with them?


To what extent do you understand the work and responsibilities of teachers and principals and/or homeless agency staff? What do you think their biggest challenges and responsibilities are?


Do educators help you (and/or other staff) learn about how to best assist students with their work?


What are your impressions of the school administrators/teachers with whom you’ve worked in terms of their commitment and capacity to meet the needs of students who are homeless? Any district-level people?


Who else have you collaborated with in order to meet the needs of students who are homeless?


What influences do local, state, and/or federal policies have on your work? To whom or what do you feel most accountable in your work with students and their families?


 Are you familiar with the McKinney-Vento Act? If so, what implications does it have upon your work?


What creative strategies have you employed to meet student needs?


What are the best things being done here to help students?


What specific types of assistance would you like as you attempt to meet the needs of students who are homeless?


To identify convergence in the data (Patton, 1990), all interview transcripts, field notes, and documents were coded by the researcher. First, an inductive, open coding process was utilized to identify emergent themes. Although the reflexive nature of qualitative research accepts that the position of the researchers does influence the themes that surface, this thorough open coding process allowed themes that were truly reflective of the participants' responses to emerge from the data.

To discern the relationships between these major emergent themes, an axial coding process was utilized next, whereby the researcher engaged in “clustering themes into conceptual groupings, making metaphors for the integration of diverse pieces of data, subsuming particulars into the general, shuttling back and forth between first-level data and more general categories, noting relations between variables, building a chain of evidence, and making conceptual/ theoretical coherence” (Huberman & Miles, 1998, p.26). After the data were open and axial coded and emergent themes were identified and relationally described, they were examined in relation to social capital theory (refer to Table 3 for examples of how codes were developed and how these codes led to the findings).

Before moving on to the findings, it is important to note how the researcher’s identity played a role in the data collection and analysis process and what steps were taken to ensure the validity of the study. Having past professional experience as an employee at a homeless shelter and, later, as a high school teacher, I entered both school and shelter sites with a degree of tacit understanding about the work that is done in such organizations. Although no two schools or shelters face identical issues or have preceisely the same governing rules and processes, there are certain habits, rhythms, and concerns that seem common to these settings. My familiarity in this regard helped me to “speak and understand” the language of schooling when I was interviewing school participants, and speak and understand the language of homeless work when I was talking with shelter participants. Although my comfort and fluency in both types of settings were helpful in collecting and making sense of the data, it was also crucial to ensure that my interpretations of the data were accurate by conducting member checks with participants. Finally, to help verify the conceptual and theoretical logic and clarity of the manuscript’s higher level interpretations, two faculty colleagues and one community-based social activist served as cross-readers. There were few instances in which these reviewers perceived data significantly different than I did, and in these instances, we entered into further dialogue and ultimately came to mutual conclusions.

Table 3: Examples of How Raw Data Led to Codes and Findings



Not surprisingly, the participants in the study uniformly indicated that inadequacies in productive relationships and social networks regularly inhibited educational possibilities for homeless students and families. Families at TFS, MCS, and HS were described as having virtually no social capital that could help instigate effective education action. In turn, they were seen to be fundamentally dependent on school and shelter leaders’ capacities to help them learn about and navigate basic education processes. It was especially troubling, then, to learn that the school and shelter leaders themselves (and the systems in which they operated) also had limited stores of educationally oriented social capital—capital that undergirded their abilities to serve homeless students and families. These widespread productive social capital deficits appeared to present major obstacles to students’ possibilities for success in schooling. Although the findings ultimately unfolded as being most noteworthy as they related to the school and shelter leaders’ social capital, I begin by describing these leaders’ perceptions of social capital deficits among the homeless.12


Although participants spoke somewhat differently about the wide-ranging “problems” in the lives of homeless families (abuse, addiction, mental illness, and so on), they found agreement in explaining that these families suffer from deficits of productive and/or strategic relationships. Shelter and school leaders suggested that families’ broken (and, in some cases, never-existent) social networks contribute to their conditions of homelessness and make it difficult for them to accomplish daily tasks that lead to education and upward social mobility. They most commonly described these relational voids as coming about as a result of destructive personal choices and transient lifestyles. Davidra Rivers, a case worker from TFS, shared her strong sentiments:

We have a lot of women who have burned their bridges with their supportive networks. The difference between me and them in that situation is that my family would never have me here. I would never be made to spend one iota of time in an emergency shelter. It just wouldn’t happen. I have so many friends and so much family. And I think for a lot of the people here, that’s not the case. . . . They have poor, I don’t want to say poor social skills, but poor interpersonal relationships. They don’t understand what it is to be a friend. . . . They burn bonds by borrowing and not giving back. There’s no guilt. The survival skills are very raw. I want to say, very animalistic, very primitive, very, you know, it’s a fight for myself. Forget everybody else. They will take from each other and they will not look back. They are out for whatever it is that they need.

Davidra was joined in this general perspective—that many of the homeless are “bridge-burners”—by several other shelter leaders. However, the most commonly voiced “barrier” to relationship development was the highly mobile lifestyles of homeless families. Supporting the literature on homelessness that documents transience as a typical accompaniment to homelessness, the families living in HS, MCS, and TFS were described as being highly mobile. For example, the following statement by Kendra Barlow, a child development specialist at MCS, was representative of the words of many others: “A lot of the kids are going from school to school to school and they are not able to make friends. Because they’re constantly going from one place to the next.”

Leaders described this lack of supportive relationships—whether brought on by bridge-burning, mobility, or other factors—as greatly reducing the “social capital stores” of the homeless. They described this as having serious implications for the educational opportunities of these families’ children in that they had little day-to-day support, they had no insider perspectives into the schooling process, and they had no models for appropriate behavior.

Little Day-to-Day Support

Viewed through Coleman’s (1988) social capital lens, the dearth of relationships rooted in obligation, expectation, and trust left the homeless families without much day-to-day support in facilitating their children’s school success. They did not have relational “credit slips” to cash in during times of need. They couldn’t call on friendly neighbors, coworkers, or family members for support. Seemingly simple tasks like getting children to school and helping them with homework were often difficult and sometimes altogether unfeasible. HS caseworker Tommie Sluby explained,

Some of the hassle is if, for instance, you have a mother with four children here and she has an infant and she’s got to get another child to school in the morning. She’s got to take three or four children out of here (the shelter) just to get one child to school. On a city bus. So it’s a real challenge.

Davidra Rivers agreed that such circumstances are both common and debilitating, especially considering the many daily tasks that await homeless mothers who are attempting to get their own personal lives straightened out. She claimed that “there are a lot of moms who sacrifice their kids going to school so that they can get some of their needs met.”

No “Insider” Perspective

Also noted by school and shelter leaders was the absence in homeless mothers’ lives of “insider” perspectives of the schooling process. Using Coleman’s (1988) language, they lacked multiplex relationships and/or those that served as “information channels” in this area. Parents were depicted as very rarely having any meaningful relationships with school personnel—relationships that undoubtedly would have notable potential for providing pragmatic school information. TFS child supervisor Lannie Swann noted that their lack of strategic relationships and understandings of the education system contribute to the condition that “the parents are often detached from what their kids are doing.” Homeless families’ scarcities of school-based information channels were seen as being exacerbated by their high rates of mobility. Knowing that they would be “on the move” in short matters of time, they were depicted as being reticent to invest themselves socially or emotionally in any particular relationships—including ones at their children’s schools. Macon School principal Gabby Voce explained, “Parents often see this [their stays in school] as a temporary thing so they don’t get involved. When they’re in transition homes for two weeks they don’t see a purpose [for developing relationships].”

No Behavioral Models

Relationship-derived social capital in the shape of “norms and effective sanctions” was also perceived to be in short supply among homeless students and families. School leaders specifically emphasized that some homeless students’ periodic behavioral problems in school were partially attributed to their lack of friendships. Such networks, it was asserted, would have provided them with social support and models for acceptable behavior. Without such social anchors, schools leaders implied that there were no peer-enforced sanctions for homeless students who display disruptive or inappropriate behavior. Principal Blount said, “We see a lot of acting out. A lot of insecurities. A lot of distrust. And the way it plays out is that there is total confusion and misbehavior.” Norms for productive educational behavior were also portrayed as being virtually nonexistent for parents. Shelter leaders commonly described homeless mothers as being “clueless” about how to behave as a supportive parent in school-related issues—information that is regularly modeled by other parents who are immersed in normative networks.


It became evident that despite their overall shortage of productive relationships and social linkages (as described by shelter and school leaders), homeless families were not altogether devoid of positive social capital. Most notably, they did appear to benefit significantly from their contacts with shelter personnel, and they were well served by the work of several building and district-level administrators in the Middleton School District whom they came to know. The relationships formed between shelter workers and homeless mothers appeared to be especially critical because they provided the mothers with some basic social supports (like child care and after-school programming; see Table 4) that were otherwise very difficult for them to attain. Further, from these relationships (which were admittedly limited in depth and duration), the homeless gained key advocates and information channels. Tommie Sluby explained how she helped her clients navigate seemingly complex school systems by informing them of their rights as stipulated by the McKinney-Vento Act:

If they have a concern or if there’s a policy issue and they say, “Well I won’t be able to do this.” I say, “Oh yes you can [stay in your school of origin]! Legislation provides for your child to be enrolled in school and for transportation to be provided on an expedited basis.”

Tommie’s comment was representative of those provided by other shelter leaders. Similarly, the data indicated that relationships with school principals and social workers also provided homeless families with a degree of educationally relevant social capital in that they made attempts to help students become socialized to their new schools (for example, Macon School had a buddy system that paired new students with classmates who could “teach them the ropes”), and they tried to get parents involved (Lane School had an assortment of parent engagement programs). It was, in fact, quite evident that both school and shelter leaders, cognizant of homeless families’ deficits in relationship networks, often extended themselves beyond their official calls of duty to provide mothers and students with social support. Shelter leaders Davidra Rivers (case worker at TFS), Tommie Sluby (case worker at HS), Aussie Carr (family advocate at MCS), Patricia Garrity (program director at TFS), Wilma Laimbeer (after-school director at TFS), and Tina Kempton (child support specialist at MCS), and school leaders Gabby Voce (principal at Macon School), Adrienne Dantley (principal at Lane School), Nikki Wills (regional homeless liaison), and Elma Bennett (Middleton School District homeless liaison) were especially impressive in their respective positions as advocates for the homeless. They each demonstrated remarkable passion, empathy, skill, determination, and commitment in their relationships with students and families in need.

Table 4. Direct Resources Provided by Schools and Shelters That Can Help Create Social Capital

School-Based Resources

Shelter-Based Resources

“Buddy system” for new students (Lane)

After-school academic support (MCS, TFS)

“Engaged parents” program (Macon)

Childcare services (HS, MCS, TFS)

Community partner readership program (Williams)

Personal counseling services (MCS)

Parent consultancy during times of transition (regional homeless education office)

Community partnerships for the arts (MCS)

Parent curriculum training (Middleton Schools central office)

Job-finding support programs (MCS, TFS)


Transitional housing assistance (TFS)

Importantly, it appeared that the social capital provided by each of these leaders flowed directly from the leaders’ own reserves of social capital. That is, the well-established organizational connections and associations of shelter and school workers were mobilized in service of homeless families who needed help with the schooling process. For example, Adrienne Dantley, the longtime principal at Lane School, explained that when trying to help homeless students get situated in her school, she regularly calls on Elma Bennett, the Middleton School District homeless liaison, to get key information about student records and enrollment. Having known Elma for many years, Adrienne established a level of trust and understanding with her that greatly facilitates her work and in turn assists homeless families. Adrienne claimed to have similar relationships with numerous other Middleton District employees. Such inner-system networks of relationships—which were also described by several shelter leaders—appeared to undergird leaders’ capacities for helping the homeless learn about and adapt to the schools and shelters. It was of note, however, that leaders’ social capital inhered predominantly in relationships with those who were “professionally like” them. School leaders talked of strong collaborative relationships as being with other school leaders in the district. Shelter leaders spoke about strong relationships with colleagues in their buildings (and in a few instances with other shelter-based leaders). Rarely noted were comments about strong relational bonds between school and shelter personnel.13 So, in their work with clients who lacked any substantial amounts of social capital, the leaders themselves were limited. Their social networks were seen as being largely homogeneous, resulting in personal and organizational deficits in what Granovetter (1973) referred to as “structural holes” and what others similarly describe as “bridging capital”— connections across communities and ties among those who are different along a relevant dimension of social life (Wood & Warren, 2002). In the next section, then, I change from my focus on perceived productive social capital deficits among the homeless to the “bridging” challenges described by the leaders.


Issues of personal and organizational unfamiliarity between school and shelter workers appeared to inhibit relational network development and productive collaboration and, ultimately, to limit leaders’ capacities for providing social closure in the educational arena for students and families who were experiencing homelessness. The findings here were pivotal to the implications of this study.

Personal and Organizational Unfamiliarity

When asked about her relationship with the leaders (principals, social workers, and so on) of MCS’s neighboring schools, child support specialist Tina Kempton said, “I’m actually not really familiar with the schools” and admitted that there was virtually no relationship there at all. Her sentiments were echoed by MCS child development specialist Kendra Barlow, who claimed, “I’ve never even met any of the social workers in the Middleton School District”; HS case worker Tommie Sluby, who said, “I don’t have a relationship with or talk with a lot of school personnel”; and MCS director of after-school services Dana Duff, who was unable to recall names of any principals or social workers from nearby schools. Similarly, when asked about who they work with at the homeless shelters, Williams School principal Donald Royal was unsure, and Macon School principal Gabby Voce said, “It’s really hard to know who’s in charge there [at TFS].” Clearly evident here was the relational disconnect between school and shelter leaders. Even though leaders of the schools and shelters are only separated by two or three city blocks and they work with many of the same children each day (Macon School enrolled 50–60 homeless students each year, many of whom were staying at nearby TFS, and Lane enrolled about 15 homeless students each year, most of whom were from MCS), they knew very little about each other. Such personal unfamiliarity clearly restricted the leaders’ capacities for working together efficiently.

Not only did this unfamiliarity with one another affect general efficiency levels, but it also precluded school and shelter leaders from knowing each other’s broader values and “ways of thinking” about education and schooling. For example, although every shelter-based leader portrayed homeless students’ education as being of paramount importance and worthy of special effort and consideration on the part of the schools, the principals almost uniformly viewed homeless students the same way they viewed other students. They expressed great value and care for these students, but with the exception of transportation and immediate placement accommodations (mandated by the McKinney-Vento Act), they did not rush to provide services and responses that were specific to homeless students’ unique psychosocial needs. Such perspectives were demonstrated by principals Adrienne Dantley, Donald Royal, and Gabby Voce. Adrienne, an experienced and highly successful administrator, said that her school treats “highly mobile” homeless children the same way they treat other highly mobile students. She explained,

Where our school is located in proximity to all the local universities, we have a lot of families in transition inasmuch as their families are affiliated with research or going to school or doing an internship or residency—because we’re close to [five universities]. So we have a lot of culturally diverse families in this area who are from all over. We’ve had families from Australia, or Slovenia, or Colombia. . . . You know, children pop in and pop out. You know, they’re here from Iceland one semester and gone the next. So it’s kind of what we’re all used to. The children are familiar with it.

Her policy of treating homeless students the same as other students in transition is witnessed in her policy of not telling the teachers of the students that the children are living in a shelter and her general tendency to not involve shelter personnel in students’ school matters: “We try to make sure that the kids are not identified whatsoever.” Although her intentions here are noble—to ensure that children are not stigmatized by the homeless label—the “sameness” of treatment policy appears to contradict the perspectives of shelter leaders who adamantly contended that the children need special attention and adaptations (while still being held accountable). It is important to emphasize here that Adrienne’s commitment to serving students is solid, and her overall record of care and success is impressive (her school is recognized as one of the best in the district). Her philosophy of service, however, is highly different from those espoused by the shelters.

Donald, a first-year principal at Williams School, described a similar perspective as Adrienne. He had no special programs or policies in place to serve students’ needs that are specifically affiliated with their conditions of homelessness. Surprisingly, Donald claimed that the shelter-based students at his school are often better adjusted socially and emotionally than others at the school:

The kids here don’t see it as anything different if someone is staying at a shelter. In this school, there is such a thin line between those who are homeless and those who are not. This is a very low SES community. Actually, some of the kids who are at shelters have better situations. At least they usually show up at school cleaned and fed. We get the poorest of the poor in the area here. Some of the kids from the shelters are a bit more well-adjusted than those who aren’t.

Like Adrienne, Donald expressed concern for homeless students and families and appeared to be genuinely interested in helping them, but as the leader of a school where almost every student comes from a very poor and unstable household, his immediate concerns were broader in scope than just those aligned with homelessness.

Gabby also admitted to not implementing many policies, procedures, or trainings at Macon School that were specific to homelessness. Her explanation, however, was slightly different from those of Adrienne and Donald. She claimed that she was simply too overburdened with school achievement-related concerns to spend much time discussing and considering issues related to homeless students’ social and emotional needs. These standardized-test-related concerns escalated considerably over the past two years, since a new superintendent was hired to “clean up” the Middleton District. Whereas she would have considered homeless-specific programs and policies in the past, Gabby admitted that since his arrival, principals and teachers have been strongly encouraged to allocate their time and resources in areas that would be of potential benefit in the area of testing: “Now the focus has shifted and it’s all about testing and getting the kids proficient.”

In addition to this unfamiliarity about “ways of thinking” about schooling, school and shelter leaders also appeared to be divided by their unfamiliarity with each other’s organizational structures and “ways of working.” Consequently, they claimed to have a difficult time figuring out how they could work together—a clear indicator of school and shelter leaders’ limited amounts of bridging capital as they sought to educate students who are homeless. For example, multiple shelter-based workers provided stories about how difficult it was to figure out whom to speak with (and how to get in touch with them) in the Middleton School District for any given issue that arose. A frustrated Davidra Rivers explained,

If I need to call the school and talk to the assistant principal, the attendance record keeper, whoever, I definitely would say that there is a slow response. Period. . . . If I’m having to talk to the school social worker regarding a certain child or a certain activity, it may be several days to hear from that person. And I don’t know why that is. I don’t know if that person’s not in the school 24 hours a day. I don’t know what their assignments are . . . I think that the problem definitely becomes the structure of the district, of their administration.

When pressed to further explain her sentiments, Davidra identified particular frustrations with principals, complex school structures, and schools’ overall lack of responsiveness:

I can’t tell you how many times I have tried [calling] them [principals]. Because when you get the district printout [of school contact information], you don’t get the social worker’s phone number. You don’t get the secretary’s phone number. You get the principal’s phone number. So you’re assuming that they have elected themselves to be the contact person if you need to contact someone there at the school. Well, why else would their names be there, right? You call the principal and I mean, you can leave 75 messages and that person is not calling you back! . . . Who do we call? Who do we talk to to deal with it? There’s not a lot of communication when a child is failing, especially if it’s one of our children. There are all kind of excuses. . . .  “I don’t know the mailing address. I didn’t have the phone number. This wasn’t right. That wasn’t right.’”

Most of the leaders at HS and TFS described similar difficulties in reaching and understanding schools. They tended to situate these difficulties within their general unawareness of “the [school] system.” They did not know whom to contact about specific issues that arose, nor did they understand the way the large district “fit together.”

Clearly, then, the disconnect between school and shelter leaders was sustained by a lack of personal, philosophical, and organizational understanding between parties. Although none of the data were indicative of individual leaders’ or organizations’ shortage of commitment or desire to help homeless students and families, many of the leaders perceived such a shortage among their school/shelter counterparts. Personal and organizational unfamiliarity with leaders who operate in different contexts became problematic here in that tension and other-blaming perspectives were bred (and social capital was squelched). Examples of this interorganizational tension and other-blaming were numerous, including the following statements by shelter and school leaders:

Gabby Voce of Macon School claiming that shelter leaders need to make a better effort to learn about her school:

I think that they really should know the expectations of the school. Because they’re right there and the parents have to get their kids in school and then the parents have to go out and seek employment. Really I would think that the shelter should be an advocate for the kids who have to go through several transitions at the same time. . . . Shelter staff should visit here. At least they could then know what we’re all about.

Dana Duff of HS claiming that school officials are biased against and/or indifferent about students and families who are homeless:

I certainly pick up on their attitudes and their views of the homeless. . . . The attitude of certain people in the school district is “Oh,well, she’s in a shelter—we’ll get to her. She’s in a shelter.” A real negative attitude. . . . Believe me, when I’m in here doing the one on ones [conversations with school district personnel], I hear it! There’s an element of indifference that’s an undercurrent to this. All it takes is one or two people who come across with that mentality—it could be a receptionist or any other staff—and when it does, look out! It’s really a challenge . . . I have to yell at them. . . . They don’t want their boats rocked!

Although Dana was the only participant to allude to outright confrontation, more subtle signs of other-blaming, as implied by Gabby, were pervasive. Contrary to the comfort and mutual respect that typically characterize productive interorganizational collaborative endeavors (Miller & Hafner, 2008), a palpable sense of tension and uncertainty marked participants’ words about their “partners down the street.” Again, the negative perceptions that leaders voiced were interesting, even surprising, considering that when analyzed within their particular work setting, all the leaders demonstrated impressive commitment, passion, and skill in their service of students and/or the homeless.


It is important to note that there were four leaders who provided noteworthy exceptions to this theme of insufficient bridging social capital. These leaders, Nikki Wills (a regional homeless coordinator), Elma Bennett (the Middleton District homeless liaison), Patricia Garrity (TFS family specialist), and Aussie Carr (MCS family advocate), were able to tap into broad interorganizational networks in service of homeless students’ educations. In the case of Nikki and Elma, their very job descriptions are dependent on developing and/or capitalizing on broad interorganizational networks. As a regional homeless coordinator, Nikki works with hundreds of schools and shelters in nine counties. Included in her wide-ranging responsibilities are educating schools and shelters about policies regarding the education of homeless students, providing educationally assistive material resources to students and families, and directing supplementary educative programming for homeless students. Elma, the Middleton School District homeless liaison, is deeply involved in coordinating transportation arrangements for students. In this capacity, she works with both school and shelter representatives on a daily basis. All the participants in this study spoke glowingly about Elma and Nikki. They were described as responsive and conscientious in helping to get children enrolled as quickly as possible and as key informants about the McKinney-Vento Act. However, despite the apparently excellent networking being done by Nikki and Elma, their work was seen to be predominantly focused on issues of transportation and policy implementation. In that they were each responsible for working with many different schools and shelters and they were not actually housed in or representing any one particular school or shelter (Nikki works in a large office building a few miles away from downtown, and Elma works in the Middleton District’s central office building), their capacities for engaging the previously described unfamiliarity and tension between specific schools and shelters were limited. They simply did not have the time to work on relationship development that might have ultimately engendered fruitful collaboration on more localized levels.

The only two “frontline” participants (those who were actually housed in a school or shelter) in the study who appeared to have what Lin (2000) referred to as heterogeneous networks were Patricia and Aussie. With diverse personal and professional acquaintances, they both appeared to possess extensive reserves of bridging social capital. Their impressive connections with leaders from schools, shelters, and other community agencies greatly facilitated their work with students and families who were homeless. Patricia’s bridging capacity was rooted in her past position as an employee in the Middleton School District. She was armed with experiential understanding of who people were and how the system worked and was therefore able to make efficient inroads into the schools in ways that other shelter-based leaders were limited. She explained,

Fortunately I used to work there [in the Middleton School District], so I’m really close to a lot of the staff members. I’m able to communicate with them and I know what a lot of them do, what their responsibilities are. I was with the district for 21 years. I was what you call a “floater,” spending time at all the different schools in the district. So I kind of have a relationship with at least two or three people in each building. I’m familiar with their programs.

Davidra Rivers, Patricia’s supervisor at TFS, emphasized the value of Patricia’s bridging capital: “Patricia’s relationship with the schools is an invaluable resource. It’s never been better for us.”

In contrast, Aussie’s vast network of resourceful relationships in the community seemed to be a result of her exceptional personal commitment to networking. During the course of our interview, she repeatedly harkened back to the value of networking with those in the schools. In fact, Aussie claimed that this was the most important facet of her job at MCS. She explained, “Through the networking we get all these relationships going. And it allows us to get on the phone and say, ‘Hey, we have a child who might be interested in this,’ and they’ll say, ‘Oh, you know what, there’s a scholarship open for them.’” Aussie’s job description as a family advocate did not require her to forge relationships with school secretaries, teachers, and social workers, but she took it upon herself to do it because, as she said, “I love this job and these kids.” Her extra time and efforts in this regard appeared to pay off for the students in Aussie’s program at MCS, as Aussie acted almost as a second mother for them—she gathered important information about what was going on at school (in terms of students’ behavior, assignments, and so on), she assisted the children in getting them enrolled in supplementary social and educational programming, and she provided them with much-needed nurturing and guidance during difficult times.

The bridging capital of Patricia and Aussie, although extremely beneficial in the short term, was limited in that it was fundamentally tied to their individual characteristics rather than to the larger systems of practice in which they worked. As a result, the productive networks they established were, from organizational perspectives, not very sustainable. In fact, after interviewing Patricia during the early phases of this study, I returned to TFS 2 months later to interview more participants, only to find that she was no longer employed at the shelter. This illuminated the problem facing organizations that are overly reliant on specific bridging characteristics possessed by individuals rather than developing systems of practice that institutionalize collaboration and boundary spanning. Having not done this, TFS found itself back at square one with its school-based networks after Patricia’s departure.


The findings from my investigation into the leadership practice in the Middleton School District support indications in the literature that homeless students and their families are hindered by shortages of social capital that can facilitate productive education experiences. Specifically, participants in the study described these shortages as problematic because homeless families, who are already faced with so many daunting challenges (poverty, addiction, mental illness, physical abuse, and so on), are in desperate need of supportive relationships and networks to help with their children’s schooling experiences. With awareness of this critical gap in the lives of the homeless, the shelter workers in particular attempted to fill some of the major service needs of their clients. They provided them with childcare services, after-school academic programs, and advice on schooling processes. Although these appeared to be valuable services, the shelter and school workers themselves were seen as having limited bridging social capital—which made it difficult for them to lead and collaborate with one another and, ultimately, to serve the homeless students with great efficacy. These deficits in bridging capital appeared to be rooted in personal and organizational unfamiliarity. Although there were four leaders who did indeed appear to have valuable bridging capacities, their abilities to help develop educationally relevant social capital among the homeless was limited in that they focused predominantly on issues of transportation, and/or their bridging work was not institutionalized and therefore unsustainable.


After engaging questions of how school and shelter leaders perceive social capital as influencing the education of homeless students, it became evident that it was not just homeless students and families’ productive social capital that was in short supply, but also that of the systems of leadership practice. For shelter and school leaders to help create diverse forms of social capital among the homeless in Middleton, it became quite clear leaders need to make conscious efforts to increase and diversify their own stores of social capital. These findings have important implications for the Middleton-based leaders, as well as leaders in similar urban contexts and researchers of issues related to the schooling of homeless children. In the next section, I discuss several specific implications for those who work in this arena (both practitioners and researchers). This discussion is informed by some key elements of social capital theory.


It is evident that systems of leadership practice that engage homeless education issues in Middleton and other similar areas should focus purposefully on developing and activating social capital that can facilitate productive schooling experiences for all children. Schools’ natural predispositions to prioritize the development of human capital (math skills, science skills, reading levels, and so on) can, if too narrowly pursued, discourage attempts to cultivate strategic family and community relationships in the lives of students. Examples of human capital “tunnel vision” include hyperintensive efforts to raise standardized test scores at all costs. Fostered by the current high-stakes testing pressure-cooker environment, these very policies that promise to “leave no child behind” seem to do just that to those who are in most need. Specifically, Coleman’s (1988) claim that human capital can be useless without the presence of social capital heeds attention in the context of homeless education; those students who have the least material resources (such as the homeless) are in dire need of supportive social networks if they are to succeed academically, socially, and professionally. Especially for students who are homeless, social capital undergirds human capital. The findings here are largely in concert with those of Lee and Croninger (1994), Stanton-Salazar (1997), and Goddard (2003)—each of which indicated that social capital can help contribute to academic success. Accordingly, when thinking of homeless education, leaders do need to think about issues relating to transportation and test taking, but it appears to be of fundamental importance to think about how to cultivate productive relationships in the lives of students and families. Schools and shelters should explicitly consider how they can engender obligations, expectations, and trustworthiness for the homeless, how they can provide information channels for the homeless, and how they can present norms and effective sanctions to the homeless.


To help foster social capital in their students/clients, schools and shelters need to develop their own interorganizational networking capacities. This assertion is made with support from both the wider body of research on homeless education, which describes effective school-shelter collaboration as being critical (Altshuler, 2003; Duffield, 2000; Eddowes & Butcher, 2000), and the data from this study, which indicate that school administrators and shelter case workers, child development specialists, and family advocates could serve vital roles as “closers” of students’ and families’ social networks. Here I hearken back to Coleman’s (1988) claim that “for families that have moved often, the social relations that constitute social capital are broken at each move. Whatever the degree of intergenerational closure available to others in the community, it is not available to parents in mobile families” (p. 113). In systems of practice like the one in Middleton, the fragmented lives of homeless families are likely to persist at least until school-shelter networks are strengthened (see Figure 1). A move toward social closure might occur, for example, when TFS leaders and Macon School administrators (who are located only two blocks apart and in turn work with many of the same families) become familiar enough with each other personally (knowing each other’s names and having working respect for one another) and organizationally (knowing who should be contacted and how they can be best reached in various situations) that they can provide each other with constructive information and updates on students’ lives. This could entail a classroom teacher calling the after-school programming staff at the shelter to let them know what a student’s homework assignment is or to inform them about any behavior issues the student has. It might also entail the case worker from the shelter calling the principal to let her know that a family is encountering a particularly difficult problem on a given day and to ask her to keep an extra close eye on the student. The provision of social closure here does not require extensive resources or new programming—just a concerted effort among shelter and school leaders to get to know one another and to share pertinent student information on a regular and ongoing basis.

Figure 1. Fragmented and closed social networks among homeless families


In this discussion of social closure, three considerations are especially important. First, more so than for many other students living high-poverty situations, there are tangible opportunities to develop meaningful social networks for sheltered homeless students. Here I revisit the statement made by Williams School principal Donald Royal in which he indicated that his students who are living in shelters can actually be better off than the other poor children in his school. He suggested that shelter students, although usually going through very turbulent times, have some degree of stability in that they have meals, a place to sleep, and after-school programming—things that some of his other students do not have. To extend this line of reasoning, shelters that are staffed by caring and intelligent adults (such as all of those interviewed at HS, MCS, and TFS) could be seen as having potential to provide social closure that has never previously existed in students’ lives. Rather than having one (or sometimes two) overburdened parent as their advocate, students who are homeless can benefit from multiple advocates who, if working collaboratively within and across organizational boundaries, can help them live up to their academic and social potential. This reinforces Warren’s (2005) claim about the critical role of community organizations as builders of social capital: “With their roots in the community, these organizations serve as mediators between families and schools. In this role, they can help schools understand families better and families understand schools better” (p. 28).

Inherently tied to these specific “opportunities” that shelters have to facilitate social closure, a second noteworthy concern here relates to the time frames that specific shelters have to help create closure. Depending on the nature of services that are provided by each of the shelters, the amount of time that they have to catalyze meaningful relationships for families varies greatly. Leaders at shelters like TFS that provide long-term bridge services for up to 2 years, for instance, have noteworthy opportunities to develop enduring “shelter-based” connections with families and schools. Such instances fit the example described earlier—closed networks are created among parents, shelter leaders, and school leaders, and in turn, students benefit. Shelters like HS that provide emergency shelter for very short periods of time, however, have only brief windows of time to work with parents and students (generally, anywhere from 1 to 30 days). It would be unrealistic and beyond the scope of institutional missions and capacities for such agencies to attempt to forge long-lasting shelter-based relationships with parents and children. Indeed, the services (and, accordingly, relationships) provided at these agencies are short-term in nature. However, such agencies might still work effectively toward closing families’ longer term “education networks” by quickly connecting parents and students with relevant community organizations and services. In these settings (Boys & Girls Clubs, YMCAs, after school programs, churches, and so on), meaningful and lasting relationships—ones that outlast families’ shelter stays—can be formed. Examples of such “community linking work” have been witnessed recently in inner-city Pittsburgh, where several well-connected emergency shelters have effectively connected highly mobile homeless families with stable community partners who can help students and parents for longer periods of time (refer to Miller, 2009a).

A third and final consideration in this discussion of social closure pertains to the notion of “bridging capital,” to which I referred earlier. Loosely related to Granovetter’s (1973) conceptualization of “weak ties” and Burt’s (1992) description of “structural holes,” bridging capital is constituted in the profitable relationships one has with those who have notable differences in backgrounds and/or situations. In that bridging capital exposes individuals to new and previously nonexistent life possibilities, my findings support indications that bridging capital is vital in the homeless education context—for students, parents, and leaders alike. 14 Although some social capital theorists—most notably Coleman (1988)—suggest that dense “closed” networks are most beneficial, and other theorists suggest that bridging networks or “weaker” ties are most fruitful (e.g., Granovetter; Burt), my findings suggest that both are critical in homeless education contexts like Middleton. The (previously described) information, norms, and sanctions that families and leaders derive from dense, closed networks (again, refer to Figure 1) need to be joined by possibilities for educational, professional, and social advancement that are often brought about by relationships with diversely positioned individuals and institutions. Accordingly, it should be emphasized that the closed networks that schools and shelters help create should be heterogeneous in nature. Such networks help foster stability and information sharing (within and between schools, shelters, and families) while concurrently exposing individuals to new ideas, insights, and opportunities. Lin (2000) explained this need for network heterogeneity:

People in lower socioeconomic status tend to use local ties, strong ties, and family and kin ties. Since these ties are usually homogeneous in resources, this networking tendency reinforces poor social capital. . . . For the disadvantaged to gain a better status, strategic behaviors require accessing resources beyond the usual social circles and routine exchanges. (p. 789)

As suggested by Lin (2000), such diversely composed networks—those that move beyond usual school and shelter circles—are especially important for those who have traditionally been placed on the fringes of society, for, although men and individuals from middle- and upper-level socioeconomic backgrounds tend to regularly interact with wide-ranging, well-resourced individuals and institutions, women, people of color, and the poor tend to relate more exclusively with those who are like them, thereby limiting their access to academic, social, and professional resources and opportunities. The homogeneity of these relationships can provide a valuable source of solidarity and trust, but it can also mitigate their capacities to avoid and/or more effectively engage difficult life situations like homelessness.

Importantly, the correlation of shelter/school leaders’ social capital and families’ social capital stores is seen clearly here, given than diverse leadership networks that are tied to schools, shelters, and other resourceful community institutions are foundational to families’ development of heterogeneous networks.


Finally, this study has implications for those who conduct research that focuses on issues of homelessness and schooling. A clear limitation of this study is that it draws entirely from the understandings and perspectives of school and shelter leaders. Although these leaders offer valuable, well-informed insights into leadership and other key aspects of homeless education, future work would certainly benefit from data gathered from homeless parents and students themselves. Additionally, the use of quantitative and/or mixed-method research designs in school and shelter settings (such as those employed by Miller, 2009a) could provide rich findings that might generalize–and even trigger productive action—across wider spectrums of practice. Finally, with skyrocketing rates of homelessness overwhelming school districts across the United States and billions of dollars in federal stimulus money earmarked for homeless and other high-poverty students, countless homeless education action plans are being hastily developed or revised. Research is needed that longitudinally tracks the efficacy of these plans and of the wider school/community human and financial resource distribution. Homeless education, then, truly is an area that is ripe for inquiry and desperate for answers.


1. For example, whereas the HUD definition of homelessness narrowly includes only those on the streets or in shelters, public schools (and many other social services) also include other populations, such as those who are forced to double up with other families and those who live in motels/hotels.  

2. The McKinney-Vento Homeless Assistance Act of 1987 (since reauthorized in 2001 as Title X, Part C of the No Child Left Behind Act–• 725) describes “homeless children and youth” as “individuals who lack a fixed, regular, and adequate nighttime residence.”

3. Moore (2005) explained that “within a year, 41% of homeless children will attend two different schools and 28% of homeless children will attend three or more different schools” (p. 2).

4. We refer here to Spillane’s (2006) conceptualization of leadership as a practice that is distributed among leaders, followers, and their situation over time.

5. The McKinney-Vento Act gives students who are homeless the option to remain in their schools of origin or to enroll in the school nearest to their new shelter (or other place of temporary residence). In all cases, schools are required to allow students to enroll (with or without their personal records) and to provide them with immediate transportation.

6. Lin (1999) wrote, “Most scholars agree that (social capital) is both individual and collective goods; that is, institutionalized social relations with embedded resources are expected to be beneficial to both the collective and the individuals in the collective” (p. 33).

7. Coleman described physical capital as being embodied in “tools, machines, and other productive equipment” (p. 100), and human capital as being “created by changes in persons that bring about skills and capabilities that make them able to act in new ways” (p. 100).

8. It is important to note that social capital, as described by Coleman, is a neutral concept that is not always mobilized for “moral” purposes. For example, members of illegal crime circles commonly tap into considerable pools of social capital in their illicit activities.

9. Pseudonyms are used for all proper names in this study.

10. This study focuses on issues relating to the education of sheltered homeless students. There are many other “types” of homeless students – those who are doubled up, those on the streets, those awaiting foster care, etc. – all of whom face the daunting challenges of homelessness in different ways.

11. Data provided by the Middleton homeless liaison.

12. It is important to note that although leaders perceived their clients to have relational “deficits,” the homeless were predominantly described as lacking social capital in the educational arena. That is to say, they were not described as being altogether devoid of positive relationships—just devoid of those that would help them to thrive in schooling.

13. Shelter leaders did claim to have productive working relationships with central district administrators Elma Bennett and Nikki Wills, who assisted them with getting students enrolled in school. They did not, however, describe having close relationships with other school-based leaders.

14. Refer to Lin (2000) for a further description of structural holes and weak ties.


Altshuler, S.J. (2003). From barriers to successful collaboration: Public schools and child welfare working together. Social Work, 48(1), 52-63.

Miller, P. (2007a). Examining-boundary spanning leadership in university-school-community partnerships. Journal of School Public Relations, 28(2), 189–211.

Miller, P. (2007b, April). Who’s in charge? Examining the levels of leadership that affect university-school-community partnerships. Poster presented at the American Educational Research Association annual meeting, Chicago, IL.

Miller, P. (2009a). Boundary spanning in the homeless education situation: Notes from an emergent faculty role in Pittsburgh. Educational Administration Quarterly, 45, 616–630.

Miller, P. (2009b). Leadership and collaboration for homeless students: Examining community perceptions. Urban Review, 41, 222–250.

Miller, P., & Hafner, M. (2008). Moving toward dialogical collaboration: A critical examination of a university-school-community partnership. Educational Administration Quarterly, 44(1), 66–110.
Bowman, D., & Barksdale, K. (2004). Increasing school stability for students experiencing homelessness. In Overcoming Challenges to Providing Transportation to the School of Origin (pp. 2–22). Greensboro, NC: National Center for Homeless Education.

Burt, R. S. (1992). Structural holes: The social structure of competition. Cambridge, MA: Harvard University Press.

Capuzzi, D., & Gross, D. R. (2004). Youth at risk: A prevention resource for counselors, teachers, and parents (4th ed.). Alexandria, VA: American Counseling Association.

Cauce, A. M. (2000). The characteristics and mental health of homeless adolescents: Age and gender differences. Journal of Emotional and Behavioral Disorders, 8, 230–240.

Coleman, J. S. (1988). Social capital in the creation of human capital. American Journal of Sociology, 94, 95–120.

Creswell, J. (1998). Qualitative inquiry and research design: Choosing among the five approaches. Thousand Oaks, CA: Sage.

Duffield, B. (2000). Advocating for homeless students. In J. H. Stronge & E. Reed-Victor (Eds.), Educating homeless students: Promising practices (pp. 203–245). Larchmont, NY: Eye on Education.

Eddowes, E. A., & Butcher, T. (2000). Meeting the developmental and educational needs of homeless infants and young children. In J. H. Stronge & E. Reed-Victor (Eds.), Educating homeless students: Promising practices (pp. 21–43). Larchmont, NY: Eye on Education.

Finkelstein, M. (2005). With no direction home: Homeless youth on the road and in the streets. Belmont, CA: Thomson Wadsworth.

Gewertz, C. (2008, November 5). Districts see rising numbers of homeless students. Education Week. Retrieved from http://www.edweek.org

Gibel, L. C. (1996). Attitudes of children toward their homeless peers. New York: Garland.

Goddard, R.D. (2003). Relational networks, social trust, and norms: A social capital perspective on students’ chances of academic success. Education Evaluation and Policy Analysis, 25(1), 59–74.

Granovetter, M. (1973). The strength of weak ties. American Journal of Sociology, 91, 481–510.

Greeley, A. (1997). Coleman revisited: Religious structures as a source of social capital. American Behavioral Scientist, 40, 587–594.

Gross, D. R., & Capuzzi, D. (2004). Defining youth at risk. In D. Capuzzi & D.R. Gross (Eds.), Youth at risk: A prevention resource for counselors, teachers, and parents (4th ed., pp. 3–23). Alexandria, VA: American Counseling Association.

Hagan, J., & McCarthy, B. (1997). Mean streets: Youth crime and homelessness. Cambridge, England: Cambridge University Press.

Huberman, A. M., & Miles, M. B. (1998). Data management and analysis methods. In N. K. Denzin & Y. S. Lincoln (Eds.), Strategies of qualitative inquiry (pp. 179–210). Thousand Oaks, CA: Sage.

Jencks, C. (1994). The homeless. Cambridge, MA: Harvard University Press.

Johnson, C. G. (2008, July 11). Mortgage crisis is leaving children homeless. OneWorld.net, July 11. Retrieved from http://www.alternet.org/story/91122/

Kozol, J. (1988). Rachel and her children: Homeless families in America. New York: Ballantine Books.

Lee, V. E., & Croninger, R. C. (1994). The relative importance of home and school in the development of literacy skills for middle-grade students. American Journal of Education, 102, 286–329.

Lee, V. E., & Croninger,  R. G. (2001). The elements of social capital in the context of six high schools. Journal of Socio-Economics, 30, 165–167.

Lin, N. (1999). Social networks and status attainment. Annual Review of Sociology, 25, 28–51.

Lin, N. (2000). Social capital: A theory of social structure and action. Cambridge, England: Cambridge University Press.

Lively, K. L., & Kleine, P. F. (1996). The school as a tool for survival of homeless children. New York: Garland.

MacLean, M., Embry, L., & Cauce, A. (1999). Homeless adolescents’ paths to separation from family: Comparison of family characteristics, psychological adjustment, and victimization. Journal of Community Psychology, 27, 179–187.

McKinney-Vento Homeless Assistance Act of 2001–Title X, Part C of the No Child Left Behind Act–• 725.

Moore, J. (2005). Collaborations of schools and social service agencies (Research report). Greensboro, NC: National Center for Homeless Education.

National Association for the Education of Homeless Children and Youth. (2008). The economic crisis hits home: The unfolding increase in child and youth homelessness. Washington, DC: Author.

Noguera, P. (2001). Transforming urban schools through investment in the social capital of families. In S. Saegert, J. Thompson, & M. Warren (Eds.), Social capital and poor communities (pp. 189–214). New York: Ford Foundation.

Nunez, R. (1996). The new poverty: Homeless families in America. New York: Insight Books.

Patton, M. Q. (1990). Qualitative evaluation and research methods (2nd ed.). Newbury Park, CA: Sage.

Portes, A. (1998). Social capital: Its origins and applications in modern sociology. Annual Review of Sociology, 24, 1–24.

Putnam, R. D. (1993, Spring). The prosperous community: Social capital and public life. The American Prospect, 35–41.

Putnam, R. D. (1995). Bowling alone: America’s declining social capital. Journal of Democracy, 6, 65–78.

Quint, S. (1994). Schooling homeless children: A working model for America's public schools. San Francisco: Jossey-Bass.

Rafferty, Y. (1997, December/January). Meeting the educational needs of homeless children. Educational Leadership, 48–52.

Spillane, J. P. (2006). Distributed Leadership. San Francisco: Jossey-Bass.

Stanton-Salazar, R. (1997). A social capital framework for understanding the socialization of racial minority children and youths. Harvard Educational Review, 67(1), 1–39.

Stormont, M. (2004). Nowhere to turn: Homeless youth. In D. Capuzzi & D.R. Gross (Eds.), Youth at risk: A prevention resource for counselors, teachers, and parents (4th ed., pp. 341–371). Alexandria, VA: American Counseling Association.

Stronge, J. H. (2000). Educating homeless children and youth: An introduction. In J. H. Stronge & E. Reed-Victor (Eds.), Educating homeless students: Promising practices (pp. 183–202). Larchmont, NY: Eye on Education.

Stronge, J. H., & Reed-Victor, E. (2000). Educating homeless students: Promising practices. Larchmont, NY: Eye on Education.

Taylor, L., & Adelman, H. (2000). Connecting schools, families, and communities. Professional School Counseling, 3, 298–308.

Tucker, P. (1999). Providing educational services to homeless students. Journal for a Just and Caring Education, 5(1), 88–107.

Vissing, Y. (2000). Meeting the educational needs of intermediate and middle school homeless students. In J. H. Stronge & E. Reed-Victor (Eds.), Educating homeless students: Promising practices (pp. 45–66). Larchmont, NY: Eye on Education.

Warren, M. (2005). Communities and schools: A new view of urban education reform. Harvard Educational Review, 75(2), 133–173.

Wood, R., & Warren, M. (2002). A different face of faith-based politics: Social capital and community organizing in the public arena. International Journal of Sociology and Social Policy, 22(9/10), 6–54.

Cite This Article as: Teachers College Record Volume 113 Number 5, 2011, p. 1067-1104
https://www.tcrecord.org ID Number: 16089, Date Accessed: 12/6/2021 4:28:43 PM

Purchase Reprint Rights for this article or review
Article Tools

Related Media

Related Articles

Related Discussion
Post a Comment | Read All

About the Author
  • Peter Miller
    University of Wisconsin–Madison
    E-mail Author
    PETER M. MILLER is an assistant professor in the Department of Educational Leadership and Policy Analysis at the University of Wisconsin–Madison. His research focuses on issues of educational leadership, homeless education, and social capital theory. He has recently published work in venues such as Educational Administration Quarterly and The Urban Review.
Member Center
In Print
This Month's Issue