Skip to main content

How Housing Mobility Affects Education Outcomes for Low-Income Children

Evidence Matters Banner Image

Fall 2014   


        Housing’s and Neighborhoods’ Role in Shaping Children’s Future
        How Housing Mobility Affects Education Outcomes for Low-Income Children
        Protecting Children From Unhealthy Homes and Housing Instability

How Housing Mobility Affects Education Outcomes for Low- Income Children


      • The Annie E. Casey Foundation’s Making Connections initiative is a prominent example of a philanthropically directed, place-based intervention. Through tailored social services, technical assistance, grants, and capacity-building resources, the Making Connections initiative sought to improve conditions in targeted low-income communities. Between 2002 and 2010, the Casey Foundation worked in 10 U.S. cities with the goals of improving employment, asset development, education, health, and civic participation.
      • A study of survey data from Making Connections by Theodos, Coulton, and Budde examined housing mobility and school mobility patterns among residents of these neighborhoods, assessing whether these moves led children to better- or worse-performing schools.
      • Practitioners of place-based initiatives can derive important lessons from the research about how changing residences and schools affects the academic achievement of low-income students.
      • Children in Making Connections neighborhoods experienced the greatest improvements in school quality after changing school districts, confirming previous research that found housing mobility positively affects low-income children when mobility leads them to higher-performing schools.

A group of young school children being helped to cross a busy street by a school crossing guard.
Decades of research demonstrate that neighborhood conditions influence children’s educational outcomes.
Improving education outcomes for low-income children is a topic of pressing concern for researchers, policymakers, and educators, especially in light of evidence that over time, the widening gap in test scores between children from rich and poor families and the growing divide between these groups in completed schooling hinder the socioeconomic mobility of low-income children.1 Policy efforts to promote academic achievement among low-income children often focus on school-based investments such as increased teacher training, smaller class sizes for early grades, and curriculum development.2 Researchers and educators also recognize that improving poorly performing schools requires comprehensive community-building activities designed to strengthen the neighborhoods these schools serve. This awareness, which stems from decades of research demonstrating that neighborhood conditions, including racial segregation, influence children’s education outcomes, has led to a growth in place-based initiatives that target specific low-income neighborhoods with comprehensive economic, social, and educational resources from the public and private sectors.3

Place-based interventions, often referred to as “community-change initiatives,” are led by philanthropies, nonprofits, or governments seeking to improve conditions in targeted low-income communities.4 By supplying tailored social services, technical assistance, grants, and capacity-building resources in a specific geographic area, place-based initiatives intend to benefit residents directly through improved services and indirectly through strengthened social networks.5 The Annie E. Casey Foundation’s Making Connections initiative was such a place-based approach, with the goals of improving employment, asset development, education, health, and civic participation.6 Between 2002 and 2010, the Casey Foundation invested $500 million in the program, which took place in low-income neighborhoods across 10 U.S. cities. In addition to on-the-ground interventions, Making Connections pursued public policy advocacy at the city and state levels in support of community development priorities in each neighborhood.7

The initiative, which collected broad data on neighborhood families through a longitudinal survey, inspired a range of research studies examining various aspects of how housing and neighborhoods affect outcomes for children in communities targeted for a place-based initiative.8 The effect of housing mobility on education outcomes for low-income children is one such research area that bears important implications for place-based policy and practice. Researchers Theodos, Coulton, and Budde studied housing mobility and school mobility patterns among residents in Making Connections neighborhoods.

The Theodos, Coulton, and Budde research study offers important lessons about how families moving residences and moving schools can negatively and positively impact improved academic achievement for low-income students. The research base is varied. On the one hand, housing mobility can enable low-income children to switch into higher-performing schools, potentially leading to important educational gains. On the other hand, changing schools as the result of a move can disrupt or interfere with children’s educational attainment, especially if the change is to a lower-quality school.9

Although the Making Connections data spurred many program evaluations and research articles, this study is unique for two reasons. First, the study analyzes the interplay between residential moves and school changes in low-income neighborhoods, which is important because most studies focusing on educational outcomes analyze residential and school mobility changes in isolation.10 Second, the sites in the study occur in the context of a place-based initiative, offering present-day practitioners a critical opportunity to understand how housing mobility and school mobility affect these efforts. In particular, the study potentially could inform the federal government’s ongoing place-based initiatives, such as Promise Zones, Promise Neighborhoods, and Choice Neighborhoods, among others, whose program designs incorporated many lessons from previous community-change initiatives that foundations have sponsored (see “Housing’s and Neighborhoods’ Role in Shaping Children’s Future,” p. 1).11

This article highlights findings from the Theodos, Coulton, and Budde study on the relationship between housing mobility and education outcomes for low-income children; considers the relevance of these findings to place-based policy and practice, especially ongoing federal efforts; and situates this study within the research base on housing mobility and education outcomes.

Making Connections Initiative

The Casey Foundation’s Making Connections initiative was a comprehensive community change demonstration project targeting low-income neighborhoods in cities selected based on data indicators of child and family need and proven leadership capacity at the community level.12 The Casey Foundation narrowed its initial selection of 22 cities down to the 10 cities that had previously demonstrated the local capacity and institutional support necessary to successfully implement the initiative.13 The 10 cities were Denver, Des Moines, Hartford, Indianapolis, Louisville, Milwaukee, Oakland, Providence, San Antonio, and White Center (a suburb of Seattle).14 Through a mix of grants, technical assistance, and social investment, the Casey Foundation sought to strengthen the community, and thereby strengthen families, by improving participants’ connections to “economic opportunity, positive social networks, and effective services and supports.”15 Addressing the needs and challenges of both children and their parents simultaneously is referred to as a “two-generation” strategy.16

A family loading boxes onto a moving truck.
Residential mobility can have either a positive or a negative impact on families and children.
The Casey Foundation created local site teams in each city, composed of foundation staff; consultants; and partners from community foundations, nonprofits, and government agencies.17 Each local team worked with the Casey Foundation to determine neighborhood boundaries, delineating a Making Connections neighborhood with a median size of 4.9 square miles and median population of 30,598.18 The teams were responsible for improving outcomes for children and families and employed various strategies to achieve cross-site goals such as increasing employment and earnings, improving the reading proficiency of students completing third grade, and developing the leadership capacity of residents so they can better participate in civic processes. For example, some sites created one-stop job training and employment-related service centers called Centers for Working Families. The Casey Foundation also contributed social investments in each location, including certificates of deposits in local community financial institutions. The team’s staffing structure evolved over the course of implementation, with local coordinators replacing foundation staff as leads for each site team.19

Although the designers of the initiative acknowledge that it “did not achieve the desired population-level change in its neighborhoods,” Making Connections did have cross-site success in program goals related to employment, asset-development, and children’s school readiness.20 For example, the enrollment of preschool children in schools across sites increased from 1,558 to 1,999 between 2005 and 2007. In addition, the percentage of students able to read proficiently by the end of third grade improved in seven of the eight schools focused on by the initiative.21 Activities to improve education outcomes in Making Connections sites inspired two ongoing Casey Foundation programs: the Campaign for Grade-Level Reading, with the goal of improving the reading proficiency of low-income children by the end of the third grade, and Attendance Works, which promotes better policy and practice around school attendance. In addition to instructing work within the Casey Foundation, the strategies and lessons learned from Making Connections inform other foundation and government place-based initiatives.22

The Casey Foundation and researchers used a longitudinal cross-site survey as the primary data source for evaluating the Making Connections initiative’s impact on children and families.23 Researchers at the National Opinion Research Center at the University of Chicago and the Urban Institute designed the survey, which was to be completed by an adult resident aged 18 or older in each home, and collected the responses through in-person and phone interviews in three waves: from 2002 to 2004, 2005 to 2007, and 2008 to 2010.24 Researchers designed the survey to obtain a representative sample of children and families in the targeted neighborhoods.25 If families with children moved between waves, then the surveyors contacted and interviewed the resident at his or her new address.26 The survey included questions on various topics of interest, including employment; income; level of hardship; community engagement; satisfaction with neighborhood services; and perceptions of neighborhood quality, safety, and social cohesion.27 The survey also contained a separate section about each child living in the home, including questions related to schools children attended and school readiness.28 Theodos, Coulton, and Budde analyzed the data from this survey to determine whether moving residences and moving schools influenced education attainment for children in the Making Connections neighborhoods.

Housing Mobility and School Mobility in Making Connections

In their study, Theodos, Coulton, and Budde sought to “examine the relationship between residential and school mobility in these sites and to determine the circumstances that are associated with children switching to better or worse performing schools as a result.”29 The study relies on a hypothesis, borne out in the evidence base, that low-income students should attend high-performing schools because “the overall performance of a school’s student body influences individual achievement.”30 This study uses reading and math test scores for each grade to define school performance, yielding a composite rank score for each school relative to all other schools in the state.31 The researchers examined the three waves of Making Connections, classifying changes in data between waves 1 and 2 as “period 1” and between waves 2 and 3 as “period 2.”32

Theodos, Coulton, and Budde analyzed whether a child’s change in school rank between periods corresponded with a change of residence or school. Did housing mobility cause families in Making Connections neighborhoods to switch to higher-, lower-, or comparable-quality schools? Although some school changes are natural (such as moving from elementary to middle school or from middle school to high school), other school changes are not the result of advancing to the next grade level. The authors classified the former school changes as “promotional” and the latter as “nonpromotional.”33 Nonpromotional school mobility can be attributed to a number of voluntary and involuntary reasons, such as moving to a new residence or transferring to another school (either because students request transfers or schools ask students to transfer because of disciplinary or academic issues).34 They also calculated the distance that students moved from their original homes and schools, determining whether each move crossed school district and neighborhood boundaries. In addition, the researchers compared students on a range of economic, education, and housing characteristics, including household income level, parental education level, race and ethnicity, and family’s ability to afford food.35 Some major highlights from this Making Connections research study included the following:

  • Little variation existed in the overall quality of the schools neighborhood children attended. In period 1, the schools children attended had an average state rank in the 27th percentile. In period 2, the average state rank of these schools was in the 26th percentile. Segments of the student population, however, experienced improvements and declines in school quality across periods; 38 percent of students experienced the greatest variation in school performance between periods, with 19 percent experiencing school rank improvements of two or more deciles and 19 percent studying at schools ranked two or more deciles below their period 1 school.36
  • Of the school changers, 43 percent experienced changes of two or more deciles in school rank compared with 19 percent of school stayers. This finding confirms that the greatest changes in school quality — whether improvements or declines — occurred as a result of changing schools.37
  • Most students (51%) attended schools outside of the neighborhoods defined by the initiative. Students in the initiative neighborhoods exhibited a high rate of school mobility. Between periods 1 and 2, 78 percent of students changed schools, and 22 percent of students remained in the same school. Of those changing schools, 56 percent did so for promotional reasons and 22 percent did so for nonpromotional reasons.38
  • More than half of the children (55%) changed residences between periods. This finding tracks with national trends for housing mobility rates among low-income families.39 Fifty-nine percent of families changing schools also changed residences compared with 41 percent of those remaining in the same school between periods.40

Additional variables that led to measurable changes in the rank of schools that children attended were parental education levels and the household’s economic security. According to Theodos, Coulton, and Budde, each “additional level of parental education is associated with an increase in [average percentile] state rank of 0.8 [points] by period 2” and “households that experienced worsened food security between the two periods or that experienced food insecurity at both periods were associated with declining school performance ranks (–2.7 and –2.5 [points], respectively).”41

The finding related to worsened food security is especially noteworthy given the relationship between financial distress and housing mobility among low-income families. Cohen and Wardrip point out that poor and near-poor families move the most frequently, which reflects a “range of often complex forces,” including residential instability related to housing cost burden, loss of employment, and the lack of a safety net.42 Movers who frequently relocate short distances in response to financial stress or housing problems are known as “churners.” In an earlier analysis of housing mobility data from Making Connections, Coulton, Theodos, and Turner find that 46 percent of movers were churners, who had a median annual income of $14,000 and relocated a median distance of 1.7 miles.43 In addition, 24 percent were “nearby-attached,” middle-aged movers who relocated close by but did so more because of life-cycle factors than a desire to leave their neighborhood, and 30 percent were “up-and-out movers,” who relocated greater distances to improve their housing and neighborhood satisfaction.44 Evidence suggests that housing mobility triggered by economic distress hinders children’s academic achievement.45

Although Theodos, Coulton, and Budde find that children in Making Connections neighborhoods demonstrate high rates of both housing and school mobility, those mobility rates did not lead to drastic improvements in the quality of the schools they attended.46 According to the authors, children who changed school districts, which requires a change in residence, tended to move to higher-ranked schools, resulting in “an average improvement in percentile state rank of 8.9 points.”47 Moreover, families’ degree of financial distress, as measured by difficulty affording food, contributed to switching to lower-performing schools.48 Although the authors note that “it is not the case that no children who remained within the same school district saw improvement (or that all children leaving their school district did),” the data reveal that children had to move to schools outside of the target neighborhood to experience improvements in school rank.49 These mobility dynamics reveal important implications about place-based policy and practice.

Research Limitations in Making Connections

A young girl and her teacher playing musical instruments together.
Policymaking efforts to promote academic achievement among low-income children often focus on teacher training, small classes, and curriculum development.
The Making Connections data are limited in certain critical aspects. First, the families participating in the longitudinal survey do not constitute a representative sample of U.S. neighborhoods because they were “deliberately selected for a community-change initiative and may differ from other low-income neighborhoods in important ways.”50 Second, the initiative did not capture data on where childless households moved, excluding potentially relevant data on how mobility patterns differed between these groups.51 Third, and most relevant to the topic of educational attainment, the research study relied on student test scores in math and reading proficiency as a measure of school performance.52 Researchers acknowledge that test scores are a limited measure of school quality, pointing to the need for collecting more comprehensive measures such as social and behavioral outcomes and college readiness.53 As Theodos, Coulton, and Budde note, their study’s reliance on test score data is likely masking “some differences in quality.”54 In addition, although the study examines the role of children’s race and ethnicity on housing and school mobility rates, the authors do not specifically investigate how segregation or integration by race or ethnicity influences children’s educational outcomes.55 In particular, the study does not consider whether moving from a segregated to a nonsegregated neighborhood results in better educational outcomes for children and for minority children in particular. Nevertheless, these researchers highlight important issues about the relationship between housing and neighborhood context and the well-being of children.

Relevance for Place-Based Policy and Practice

High rates of housing and school moves outside of neighborhood boundaries can impede the implementation of place-based initiatives, such as Making Connections, that focus their activities to improve school and neighborhood quality within defined geographic areas.56 The success of a place-based initiative depends in part on residential stability in the target area; to benefit from an initiative’s services and capacity-building efforts, families need to have “access to these programs for some minimum amount of time,” and demonstrable improvements in neighborhood capacity depend on “stability in emerging leaders and networks.”57 As Coulton, Theodos, and Turner make clear, community-based initiatives and local practitioners must heed the reality of high rates of residential mobility:

Efforts to improve the well-being of families and children by strengthening conditions in poor neighborhoods cannot simply assume that families will remain in one place long enough to fully benefit. Many of the Making Connections movers remained nearby, however. These nearby movers may retain social connections from their original residential location and may still participate in activities and services there. This finding highlights an opportunity for community-based initiatives to continue serving families who move but remain nearby.58

A diagram showing that housing characteristics, stability, and educational outcomes are interrelated.
Note: Recreated with permission from the Urban Institute.
Source: Mary Cunningham and Graham MacDonald. 2012. “Housing as a Platform for Improving Education Outcomes among Low-Income Children,” WhatWorks Collaborative, 5.
Kingsley, Jordan, and Traynor posit that the high rates of housing mobility in Making Connections neighborhoods demonstrate the complexity of residential mobility, which requires new thinking about place-based policy.59 In particular, the authors note that practitioners and policymakers of community-change initiatives must tailor their responses to meet the circumstances of different types of mobility.60 The appropriate policy options are different, for example, for “up-and-out movers” — those with higher incomes who relocate to better neighborhoods — than for “churning movers” — those with lower incomes who tend to move shorter distances in response to complications with housing arrangements and financial stress.61 Kingsley, Jordan, and Traynor offer community-based organizations a range of recommended practices tailored to the circumstances of each type of residential mobility.62 The authors suggest that although housing and other supportive counseling potentially benefit all movers, up-and-out movers in particular would benefit from housing counseling that enables them to make sensible housing choices in lower-poverty neighborhoods.63 Reducing some of the barriers to mobility into lower-poverty neighborhoods, such as racial discrimination, is an effective response for this group of movers.64 Churners, on the other hand, need to reduce the harmful effects of residential instability associated with insufficient income.65 Because low-income families who frequently move short distances because of economic stress are at higher risk of becoming homeless, Kingsley, Jordan, and Traynor suggest that strategies and programmatic approaches should focus on homelessness prevention services such as providing legal services to prevent evictions and “network organizing.”66 Network organizing is a strategy that grassroots neighborhood organizations use to strengthen both family capacity and social networks, allowing churning movers to connect to needed services. Lawrence CommunityWorks, for example, is a community development corporation that uses network organizing strategies to produce affordable housing in Lawrence, Massachusetts.67

Housing mobility patterns within targeted neighborhoods also pose a challenge to the evaluation of place-based initiatives.68 Determining whether a program has improved neighborhood economic outcomes is difficult, for example, if the families benefiting from the program’s success become up-and-out movers. As Coulton, Theodos, and Turner assert, place-based neighborhood interventions “may improve services for neighborhood residents or create employment and other opportunities, but needy families might not remain in the same neighborhood long enough to benefit.”69

Place-based initiatives with the goal of improving educational outcomes in low-income communities endeavor to strengthen the quality of neighborhood schools. The U.S. Department of Education’s Promise Neighborhoods program, for example, aims to “significantly improve the educational and developmental outcomes of children and youth in our most distressed communities” (see “Housing’s and Neighborhoods’ Role in Shaping Children’s Future,” p. 1).70 The goal of these efforts is to create high-performing neighborhood schools that function as an “anchor point for numerous partnerships that strengthen programs for children and promote parent and community engagement.”71 As the Making Connections data demonstrate, 83 percent of children at baseline were attending schools ranked below the 50th percentile in the state.72 Considering the number of low-performing schools located in low-income neighborhoods, it is perhaps unsurprising that families are willing to move long distances to access higher-ranked schools, and the research base on housing mobility and education outcomes confirms this.73 Why families move influences whether education outcomes for low-income students will be better or worse after the move.

Housing Mobility and Educational Attainment

The literature on the relationship between housing mobility and educational outcomes is mixed.74 For low-income children, the effect of housing mobility on school attainment depends on a range of factors. The transmission of the effect is either direct, as in a disruption in the children’s instruction and curriculum, or indirect, as seen through the move’s effect on the children’s parents or peer network.75 The effect also depends on the frequency of moves and whether the move is in response to distress factors such as poverty, low-quality housing, or domestic violence.76 Frequent moves, which are often referred to as hypermobility, present “special challenges to children’s well-being.”77

A young boy reading aloud from a book.
Place-based initiatives that bring children into higher-performing schools potentially lead to important educational gains.U.S. Department of Education
Researchers use conceptual models to illustrate how housing mobility contributes to education outcomes for children. Cunningham and MacDonald developed a model demonstrating the relationship between housing instability and outcomes such as school changes, absenteeism, behavioral problems, test scores, years of schooling completed, and other students’ scores (see fig. 1).78 This model situates various research findings on housing mobility and education outcomes, establishing whether mobility creates a positive or negative pathway to education success.79

As demonstrated by the large number of churning movers in Making Connections neighborhoods, lower-income families are more likely to move, creating possible adverse impacts for children’s schooling such as disrupted instruction and excessive absenteeism.80 These moves are often unplanned or involuntary, caused by foreclosure, eviction, or cost burden, among other reasons.81 The negative effects of housing mobility extend beyond the moving family, affecting both the old and new schools as well as the neighborhood to which the family relocates. Kerbow, for example, finds that Chicago schools with highly mobile student populations had a negative effect on teachers’ instruction and ability to keep progress on curriculum.82 In such highly mobile schools, all students suffer as “review and catch-up work become the norm.”83 Frequent housing mobility also affects children’s educational achievement. A 1994 study by the U.S. Government Accountability Office (then the U.S. General Accounting Office) finds, for example, that children who changed schools three or more times since first grade had a greater likelihood of repeating a grade or achieving lower reading scores by the third grade.84 A Minnesota study finds that first- through sixth-grade students who moved three or more times over a 6.5-month period between 1994 and 1995 scored an average of 20 points lower on standardized reading tests than did their peers who did not move.85

Although a number of studies confirm negative short-term consequences associated with housing mobility and school outcomes, some researchers suggest that the long-term benefit of moves that lead to higher-quality schools may outweigh these short-term costs.86 In a study of Canadian children, for example, Hango discovered a long-term positive relationship between housing mobility and the completion of high school.87 Other research indicates that, with the help of parents, children can recover from the disruption to their schooling that moving causes.88 A body of research also confirms that residential mobility can lead to positive educational outcomes, especially when these moves give children access to high-quality schools and neighborhoods.89

Children using playground equipment in a neighborhood park.
The relationship between housing and its neighborhood context influences the well-being of children.
Affordable housing strategies, including housing mobility programs and inclusionary zoning policies, have the explicit intention of providing low-income families with the opportunity to move to communities with strong school systems.90 For example, Heather Schwartz’s research into Montgomery County, Maryland’s inclusionary zoning policies, which mandate that a portion of all new residential development in the county must be set aside as affordable housing, finds that over a five- to seven- year period, students in public housing attending low-poverty schools outperformed their public housing peers in moderate-poverty schools in both math and reading (see “Housing’s and Neighborhoods’ Role in Shaping Children’s Future,” p. 1).91 A key takeaway from Schwartz’s research is that low-income children who attend economically integrated schools created by an inclusionary zoning program experience positive school effects that accrue with continued exposure to higher-income peers.92

HUD’s Moving to Opportunity for Fair Housing (MTO) demonstration, an experimental program that provided a treatment group with housing vouchers and mobility counseling to help them move from high-poverty to low-poverty neighborhoods, surprisingly did not lead to improved educational outcomes for low-income children (see “Housing’s and Neighborhoods’ Role in Shaping Children’s Future,” p. 1).93 These results run counter to the educational outcomes associated with the Gautreaux Assisted Housing Program, Chicago’s residential desegregation relocation program. The Gautreaux program, which arose out of a legal settlement in the Hills v. Gautreaux lawsuit regarding racial segregation in Chicago’s public housing, eventually moved 7,000 African-American public housing families to the suburbs, mostly during the 1980s.94 Certain findings on children’s educational experiences illustrate the difference in results between the Gautreaux program and MTO. Eighty-eight percent of children who moved to the suburbs as part of the Gautreaux program went to schools with average ACT scores at the national average or above. By contrast, less than 10 percent of students from MTO’s experimental group attended schools ranked at the 50th percentile or higher in their state.95 Researchers posit that MTO’s lack of educational benefit might be attributed to the short period of time that some families spent in lower-poverty neighborhoods and parental decisions to keep children in their previous schools.96 In a three-city study of MTO, Ferryman, Briggs, Popkin, and Rendón find that many children in the experimental group did not access higher-performing schools because their parents were “information poor” and did not make use of formal sources of information about schooling options, such as teachers or school staff, or receive formal counseling. Rather, parents relied “heavily on referrals provided by their networks of relatives and friends.”97 An additional factor that may explain the discrepancy in educational outcomes is that MTO movers in the experimental group, unlike the families participating in the Gautreaux program, moved to neighborhoods that were not “substantially more affluent or less segregated than their original neighborhoods.”98

The foregoing research summary demonstrates that housing mobility has different consequences on the educational outcomes of low-income children depending on the circumstances surrounding a move; the level of racial segregation in the destination neighborhood is a crucial variable. As Guy maintains, residential mobility has the potential to be “either a positive or a negative phenomenon for families and neighborhoods.”99


Theodos, Coulton, and Budde’s analysis of housing and school mobility patterns in Making Connections neighborhoods is an important contribution to the research base. Whereas previous research on housing mobility and school mobility examined these changes in isolation, Theodos, Coulton, and Budde study the interplay between residential moves and school moves. As Making Connections demonstrated, some moves can have a positive effect on educational outcomes, especially when mobility leads students to higher-performing schools. Very few children experienced sizeable gains in the quality of schools attended, however, and those improvements tended to accompany moves outside of the student’s school district. Most changes of schools and residences were not associated with school improvements, and some moves actually led students to lower-quality schools. Although this finding is not surprising in the context of the low-income neighborhoods studied under the Making Connections initiative, it provides an opportunity to inform ongoing place-based efforts.

The study’s finding that high rates of mobility did not correspond with increases in educational attainment suggests that place-based initiatives targeted in neighborhoods should help children from low-income families either switch into higher-ranked schools or minimize mobility into lower-ranked schools. As the findings demonstrate, many children in neighborhoods identified for a place-based initiative were attending schools outside of the target area and, therefore, not benefiting directly from targeted investments in a school within the Making Connections boundaries. In addition to improving conditions in schools within a small geographic area, place-based community-change practitioners should also address some of the structural impediments to accessing higher-quality schools.100 Furthermore, this research study confirms the importance of complementary strategies: both improving conditions in low-income neighborhoods and helping low-income residents move to neighborhoods of opportunity. Housing mobility programs have a particular role to play in this regard.

  1. Greg Duncan and Richard Murnane. 2011. “Executive Summary: Whither Opportunity? Rising Inequality, Schools, and Children’s Life Chances,” Russell Sage Foundation. “Between 1978 and 2008, the gap between the average mathematics test scores of children from high- and low-income families grew by a third (from 96 points on an SAT-type scale to 131 points in 2008).…By the 1990s, more than 20 percent of men and almost as large a fraction of women had less education than their parents. Since education has been the dominant pathway to upward socio-economic mobility in the United States, the growing gap in educational attainment between children from rich and poor families is likely to result in increased income inequality in future generations and hinder the intergenerational socio-economic mobility that has been a source of pride for Americans.”
  2. Heather Schwartz. 2010. “Housing Policy Is School Policy: Economically Integrative Housing Promotes Academic Success in Montgomery County, Maryland,” Century Foundation, 13.
  3. Claudia Coulton, Brett Theodos, and Margery A. Turner. 2012. “Residential Mobility and Neighborhood Change: Real Neighborhoods Under the Microscope,” Cityscape 14:3, 56; Maarten van Ham, David Manley, Nick Bailey, Ludi Simpson, and Duncan Maclennan. 2012. “Neighbourhood Effects Research: New Perspectives,” in Neighbourhood Effects Research: New Perspectives, eds. Maarten van Ham, David Manley, Nick Bailey, Ludi Simpson, and Duncan Maclennan, London: Springer Dordrecht Heidelberg, 3; Patrick Sharkey. 2012. “Residential Mobility and the Reproduction of Unequal Neighborhoods,” Cityscape 14:3, 9–31; Ingrid Gould Ellen, Keren Horn, and Katherine O’Regan. 2012. “Pathways to Integration: Examining Changes in the Prevalence of Racially Integrated Neighborhoods,” Cityscape 14:3, 33–54.
  4. For a helpful overview of the accomplishments of other place-based community change initiatives, please see: Anne Kubisch, Patricia Auspos, Prudence Brown, and Thomas Dewar. 2010. “Community Change Initiatives from 1990–2010: Accomplishments and Implications for Future Work,” The Aspen Institute, Roundtable on Community Change.
  5. Claudia Coulton, Brett Theodos, and Margery Turner. 2009. “Family Mobility and Neighborhood Change: New Evidence and Implications for Community Initiatives,” Urban Institute; “Place-Based Initiatives,” Center for the Study of Social Policy website ( Accessed 20 August 2014.
  6. Annie E. Casey Foundation. 2013. “Community Change: Lessons from Making Connections,” 2.
  7. Ibid.
  8. For more information about the Annie E. Casey Foundation’s Making Connections initiative, please visit For a list of major research studies conducted with Making Connections data, please visit
  9. Brett Theodos, Claudia Coulton, and Amos Budde. 2014. “Getting to Better Performing Schools: The Role of Residential Mobility in School Attainment in Low-Income Neighborhoods,” Cityscape 16:1, 62.
  10. Ibid.
  11. Ibid., 62: “Moreover, for a decade or more, foundation-sponsored community change initiatives have invested in building the social fabric, institutional capacity, and civic engagement in distressed neighborhoods, with the expectation that doing so would lead to improved educational outcomes for children;” Annie E. Casey Foundation, 21.
  12. Annie E. Casey Foundation, 1. 
  13. For more information about site selection, please see: “Frequently Asked Questions,” National Opinion Research Center at the University of Chicago website ( Accessed 11 September 2014; Annie E. Casey Foundation, 1.
  14. “Frequently Asked Questions.”
  15. Coulton, Theodos, and Turner, vii; Annie E. Casey Foundation, 1.
  16. Annie E. Casey Foundation, 3.
  17. Ibid., 1.
  18. “Frequently Asked Questions.”
  19. Annie E. Casey Foundation, 2, 9.
  20. Ibid., 2, 16–7; Anne Kubisch. 2010. “Structures, Strategies, Actions, and Results of Community Change Efforts,” in Voices From the Field III: Lessons and Challenges From Two Decades of Community Change Efforts, eds. Anne C. Kubisch, Patricia Auspos, Prudence Brown, and Tom Dewar, Washington, DC: Aspen Institute, 15: “Most [comprehensive community-change initiatives] can show improvements in the well-being of individual residents who participated in programs in their target neighborhoods. Some produced physical change in their neighborhoods through housing production and rehabilitation, some reduced crime, and a few also sparked commercial development.…[F]ew (if any) have been able to demonstrate population-level changes in child and family well-being or rates of poverty.”
  21. Annie E. Casey Foundation, 4.
  22. Ibid., 3.
  23. Theodos, Coulton, and Budde, 65: “Of the 10 sites, 7 had three survey waves; in Hartford, Milwaukee, and Oakland, only the first two waves were administered.”
  24. “Frequently Asked Questions.”
  25. Theodos, Coulton, and Budde, 66.
  26. Ibid.; Authors also note that the survey did not reinterview childless households that moved between survey waves because the initiative “focuses on the well-being of families with children.”
  27. Coulton, Theodos, and Turner, 60.
  28. “Frequently Asked Questions;” “Study Design,” National Opinion Research Center at the University of Chicago website ( Accessed 11 September 2014.
  29. Theodos, Coulton, and Budde, 62.
  30. Ibid., 64.
  31. Ibid., 67: “Because proficiency levels are state defined and, therefore, vary across states and over time, we first ranked reading and math scores for each grade (where tests were administered) for every school in each of the 10 states in this study, based on the percentage of students who tested at or above the state’s proficiency level in that subject. This methodology provided a ranking of each school’s relative performance on each test administered for each grade at that school in that year. We then averaged the schools’ math and reading rankings for all tested grades in a given year to get a yearly composite rank score for each school in our analysis.”
  32. Ibid., 66.
  33. Ibid., 67.
  34. Rebecca Cohen and Keith Wardrip. 2011. “Should I Stay Or Should I Go? Exploring the Effects of Housing Instability and Mobility on Children,” Center for Housing Policy, 9; Russell Rumberger. 2003. “The Causes and Consequences of Student Mobility,” The Journal of Negro Education 72:1, 12.
  35. Theodos, Coulton, and Budde, 69.
  36. Ibid.
  37. Ibid., 75.
  38. Ibid.
  39. Ibid., 74.
  40. Ibid.; Coulton, Theodos, and Turner, 69–70.
  41. Ibid., 77.
  42. Cohen and Wardrip, 3, 6.
  43. Coulton, Theodos, and Turner, 69.
  44. Ibid.
  45. Maya Brennan. 2011. “The Impacts of Affordable Housing on Education: A Research Summary,” Insights from Housing Policy Research (May), 3.
  46. Ibid., 78.
  47. Ibid.
  48. Ibid., 79.
  49. Ibid., 77.
  50. Coulton, Theodos, and Turner, 84.
  51. Ibid.
  52. Theodos, Coulton, and Budde, 79.
  53. Mary Cunningham and Graham MacDonald. 2012. “Housing as a Platform for Improving Education Outcomes among Low-Income Children,” What Works Collaborative, 10; For more comprehensive definitions of school quality used by researchers, parents, and government officials, see: Linda Darling-Hammond. 1996. “Restructuring Schools for High Performance,” in Rewards and Reform: Creating Educational Incentives That Work, eds. Susan H. Fuhrman and Jennifer O’Day, New York: Jossey-Bass, 144–94, as cited in Cunningham and MacDonald, 6.
  54. Theodos, Coulton, and Budde, 79.
  55. Ibid., 77–8: “Racial differences emerged in the multivariate analysis, confirming findings from previous research on school performance and mobility. Relative to non-Hispanic White children and controlling for other factors, non-Hispanic African-American children had more negative school performance change measures. Regression results showed that non-Hispanic African-American children experienced a decline in the percentile of school state rank on the order of 4.4 compared with that of non-Hispanic White children. Hispanic children also fared worse than non-Hispanic White children controlling for other factors, with a decline of 3.4 in the percentile of school state rank relative to non-Hispanic Whites.”
  56. Ibid., 62.
  57. Coulton, Theodos, and Turner, 56.
  58. Ibid., 84.
  59. G. Thomas Kingsley, Audrey Jordan, and William Traynor. 2012. “Addressing Residential Instability: Options for Cities and Community Initiatives,” Cityscape 14:3, 168.
  60. Ibid.
  61. Ibid., 164.
  62. Ibid., 177. Recommended efforts to provide residential stability to churners, who move short distances because of financial duress, include the following: (1) establish or broaden outreach; (2) establish or strengthen referral functions; (3) establish or strengthen workforce development, financial management, and housing and mobility counseling; (4) provide affordable housing in the community, with a portion earmarked for residentially unstable families; (5) maintain links and services to outmovers; and (6) collaborate with and become stronger advocates for the reform and strengthening of relevant citywide programs.
  63. For more information about how housing counseling affects the housing decisions of residents who move to lower poverty neighborhoods, see: Mary Cunningham and Noah Sawyer. 2005. “Moving to Better Neighborhoods With Mobility Counseling,” Metropolitan Housing and Communities Center, Urban Institute.
  64. Kingsley, Jordan, and Traynor, 168.
  65. Ibid., 169.
  66. Ibid., 170–5.
  67. Ibid., 175, 176.
  68. Cynthia Guy. 2012. “Guest Editor’s Introduction,” Cityscape 14:3, 3; Theodos, Coulton, and Turner, xii.
  69. Coulton, Theodos, and Turner, 59.
  70. “Promise Neighborhoods,” U.S. Department of Education website ( Accessed 24 August 2014.
  71. Theodos, Coulton, and Budde, 80.
  72. Ibid., 70.
  73. Cunningham and MacDonald, 3.
  74. Brennan, 1–3; Cohen and Wardrip, 7; Tama Leventhal and Sandra Newman. 2010. “Housing and Child Development,” Children and Youth Services Review 32, 1169; Tama Leventhal and Jeanne Brooks-Gunn. 2000. “The Neighborhoods They Live In: The Effects of Neighborhood Residence on Child and Adolescent Outcomes,” Psychological Bulletin 126:2, 315–8; Ingrid Ellen and Margery Turner. 1997. “Does Neighborhood Matter? Assessing Recent Evidence,” Housing Policy Debate 8:4, 848–50; Arthur Reynolds, Chin-Chih Chen, and Janette Herbers. 2009. “School Mobility and Educational Success: A Research Synthesis and Evidence on Prevention,” paper presented at the Workshop on the Impact of Mobility and Change on the Lives of Young Children, Schools, and Neighborhoods, Board on Children, Youth, and Families, National Research Council, June 29–30, 2009, Washington, DC.
  75. Cohen and Wardrip, 7.
  76. Brennan, 2.
  77. Cohen and Wardrip, 4.
  78. Cunningham and MacDonald, 4.
  79. Ibid.
  80. Brennan, 1.
  81. Cohen and Wardrip, 3, 9.
  82. David Kerbow. 1996. “Patterns of Urban Student Mobility and Local School Reform Technical Report,” Center for Research on the Education of Students Placed at Risk.
  83. Cunningham and MacDonald, 8.
  84. U.S. General Accounting Office. 1994. “Elementary School Children: Many Change Schools Frequently, Harming Their Education.”
  85. Kids Mobility Project. 1998. “Kids Mobility Project Report,” Hennepin County Office of Planning and Development and Community Connections, Hennepin County, MN.
  86. Leventhal and Newman, 1169.
  87. Darcy Hango. 2006. “The Long-Term Effect of Childhood Residential Mobility on Educational Attainment,” Sociological Quarterly 47:4, 631–64.
  88. Anne Hendershott. 1989. “Residential Mobility, Social Support, and Adolescent Self-Concept,” Adolescence 24:93, 217–32; as cited by Brennan, 2.
  89. Cunningham and MacDonald, 8.
  90. Brennan, 2.
  91. Schwartz, 6.
  92. Ibid., 32.
  93. Lisa Gennetian, Lisa Sanbonmatsu, Lawrence Katz, Jeffrey Kling, Matthew Sciandra, Jens Ludwig, Greg Duncan, and Ronald Kessler. 2012. “The Long-Term Effects of Moving to Opportunity on Youth Outcomes,” Cityscape 14:2, 137–68.
  94. Brennan, 5.
  95. Larry Orr, Judith Feins, Jacob Robin, Erik Beecroft, Lisa Sanbonmatsu, Lawrence Katz, Jeffrey Leibman, and Jeffrey Kling. 2003. “Moving to Opportunity for Fair Housing Demonstration: Interim Impacts Evaluation,” U.S. Department of Housing and Urban Development, 111.
  96. Margery Austin Turner, Austin Nichols, and Jennifer Comey with Kaitlin Franks and David Price. 2012. “Benefits of Living in High-Opportunity Neighborhoods: Insights from the Moving to Opportunity Demonstration,” Metropolitan Housing and Communities Center, Urban Institute; Kadija Ferryman, Xavier de Souza Briggs, Susan Popkin, and Maria Rendón. 2008. “Do Better Neighborhoods for MTO Families Mean Better Schools?” Metropolitan Housing and Communities Center, Urban Institute, 3.
  97. Ferryman et al., 2, 3.
  98. Brennan, 5. See also: Greg Duncan and Anita Zuberi. 2006. “Mobility Lessons from Gautreaux and Moving to Opportunity,” Northwestern Journal of Law & Social Policy 1, 110–26: “One important way in which the Gautreaux program differs from MTO’s is that the Gautreaux programs were part of a legal settlement involving racial discrimination and designed to provide families living in highly segregated neighborhoods of concentrated poverty in Chicago the opportunity to move to much better neighborhoods, where ‘much better’ was defined as more racially integrated. In contrast, Moving to Opportunity targeted only class. It provided families with opportunities to move to more affluent neighborhoods, defined as those with poverty rates under 10 percent, but attached no racial criteria whatsoever to the destination neighborhoods. In fact, most MTO families moved to highly segregated, if more affluent, neighborhoods.”
  99. Guy, 1.
  100. Theodos, Coulton, and Budde, 81.


Previous Article               Evidence Matters Home              Next Article

The contents of this article are the views of the author(s) and do not necessarily reflect the views or policies of the U.S. Department of Housing and Urban Development or the U.S. Government.