«AmericAn neighborhoods: inclusion And exclusion Volume 16, Number 3 • 2014 U.S. Department of Housing and Urban Development | Office of Policy ...»
As exhibit 4 shows, focusing generally on local or neighborhood ties fails to capture important differences between relationships with ties who live within the same residential building and those who live elsewhere in the neighborhood.18 People interact less frequently with ties from within the same building—on average, 78 times a year compared with 155 times a year with ties who live elsewhere. A lesser proportion of the average building network are kin, the same race or ethnicity, the same gender, or better off, whereas a greater proportion has coresident children. Building networks have a substantially greater average density than the networks of those who live elsewhere in the neighborhood—34 compared with 10 percent.
Although the proportion of within-building ties in the average core network is similar to the proportion of ties who live elsewhere in the neighborhood (16 percent for both groups), fewer of the total within-building ties named are primary ties or those named in the core network. On average, Using the General Social Survey social network data, Marsden (1987) reported an average density of 0.61 compared with averages of 0.44 in Fisher’s (1982) study and 0.33 in Wellman and Wortley’s (1989) study of Toronto residents. In these studies, density is defined as the proportion of ties who are especially close to one another, rather than our more liberal measurement of the proportion of ties who interact regularly (regardless of emotional connection).
Too few respondents named no building ties in their personal network to analyze separately (N = 15); however, exploratory analysis suggests that these individuals differed in key ways from those who had at least one building relationship. Those with no building ties were more likely to have moved from within the community, to have more kin ties, and to have a lower household income at followup. Although they had, on average, smaller networks (on average they named 2.8 total ties), their networks generally included a greater share of relationships with ties who are better off, and they reported receiving expressive, instrumental, and informational support from a greater proportion of their network than those who named one or more affordable-housing residents in their network.
60 American Neighborhoods: Inclusion and Exclusion Building Ties: The Social Networks of Affordable-Housing Residents
0.6 out of 2.4 within-building ties were named in the core network versus 0.6 ties out of 0.8 ties who live elsewhere in the neighborhood—25 versus 75 percent, respectively. When asked how many people in the building the respondent considered a close friend, the typical response was none (the average was 1.2). When asked about their reliance on neighbors in the building, most respondents did not perceive people they knew in the affordable-housing building to be essential resources; 61 percent of respondents disagreed or strongly disagreed with the statement, “I would have a hard time getting by without the help or assistance my neighbors provide,” and 68 percent disagreed or strongly disagreed with the statement, “I rely on the people I know in my building a lot.”19 At face value, these findings appear to mirror previous work on relocated public housing residents, which found limited interaction among neighbors; however, the fact that residents are less wellconnected to other building residents than to those who live in the surrounding neighborhood or beyond does not necessarily mean that they do not convey resources or help to support the daily lives of residents. To the extent that building ties augment other relationships, they may represent a unique source of support, resources, or (new) information not otherwise available to low-income individuals and also help to connect residents with others.
Receipt and Provision of Support Among Neighbors To investigate the content of these relationships, we calculated whether the respondent provided or received one or more instances of expressive, instrumental, or informational support with each local tie. Exhibit 5 presents summary statistics for the proportion of local ties who were activated for specific ends, including the subsets of ties who live in the same building and who live elsewhere in the neighborhood. Overall, we find that most relationships with neighbors include instances of one or more types of support but that the specific utility and directionality of the relationship varies according to proximity.
We see lower rates of social exchange with neighbors from the same building than with those who live elsewhere. Respondents exchanged (received and provided) support of one or more types with 81 percent of within-building ties compared with 96 percent of ties who lived elsewhere in the community. In general, affordable-housing residents provided support to a greater share of withinbuilding ties than the share of those from whom they received support or assistance. This pattern is seen across all three types of support and shows the potential for within-building ties to tax the limited resources of residents; however, 85 percent of respondents disagreed or strongly disagreed with the statement, “Sometimes I feel overwhelmed by the help or assistance I provide to my neighbors.” Affordable-housing residents exchange instrumental support with the smallest share of local ties, which is particularly true of within-building networks. Residents exchange informational support with the greatest share of ties; on average, they exchange information with 70 percent of their withinbuilding network and with 93 percent of their network ties who live elsewhere in the neighborhood.
Although residents exchange all three types of support with a smaller share of their within-building network than with ties who live elsewhere, it is clear that residents of affordable housing do interact with one another and that these relationships convey varied types of support and sharing of resources.
Respondents were read six statements about the people they knew in their building and asked how strongly they agreed or disagreed with each using a five-point Likert scale: strongly disagree, disagree, neither agree nor disagree, agree, or strongly agree.
Neighbor Characteristics and the Likelihood of Interaction To assess what individual characteristics are associated with certain types of support, we used a complementary dataset of each building tie20 named by one of the 105 affordable-housing residents in the study who nominated one or more unique individuals who lived at the same address.
Exhibit 6 presents a series of logistic regression models that estimate the odds of providing or receiving each type of support.
Overall, being the same gender and same race or ethnicity as the respondent is significantly associated with greater odds of expressive support (received or provided) but not of instrumental or informational support. Household composition—specifically, both the respondent and the local tie having one or more coresident children—is significantly associated with greater odds of receiving and providing all three types of support, particularly instrumental support. Frequency of interaction is significantly associated with greater odds of receiving emotional and instrumental support and of providing instrumental support, but not with the odds of receiving informational support or of providing emotional or informational support. Because sustained attention has focused on the proposed benefits of more affluent neighbors, particularly for access to information and job contacts, we include a binary variable for whether the respondent indicated the tie was generally better off. Relative status and the odds of providing or receiving any of the three types of support exhibit no statistically significant association.
Individual ties remained anonymous; therefore, more than one respondent may have nominated the same individual.
Average proportion of building ties activiated for specific type and direction on support.
b We further examined what characteristics are associated with specific forms of information received by the respondent, both because informational support was the most prevalent form of exchange between residents and because research and policy have focused heavily on the potential benefits of neighbors who may act as informational resources. Exhibit 7 presents a series of models that examine the association between the characteristics of building ties and the odds of the respondent receiving information or advice about five different topics. Both the respondent and tie having one or more children is positively associated with receiving information about school or childcare and housing, but not with receiving work and job information or discussing a neighborhood issue. Although respondents are more likely to seek information about childcare or finding a school or tutor for their children from a neighbor who is better off (odds ratio = 3.0, p.05), they are no more likely to receive advice about their job, work, or finding a new job, nor about neighborhood issues, housing issues, or a housing search.
responses to informational resources questions.
Discussion Affordable-housing residents in our case study are socially connected to a range of individuals who include family, friends, and neighbors. Although the average network is relatively homogenous regarding race and ethnicity and, particularly, gender, greater diversity exists in terms of relative status—on average, 30 percent of ties are better off and 13 percent are worse off. Slightly less than one-half of the average overall network consists of ties to individuals who live outside of the residents’ community. Affordable-housing residents do not appear to have the kind of dense, potentially redundant, and locally bound networks that are often ascribed to lower income households.
Ties with those who live in the neighborhood but not in the same building are similar to an individual’s overall network in terms of homophily, frequency of interaction, and relative status.
For the average resident, 75 percent of these neighborhood ties were named in the core personal network. By contrast, ties to other affordable-housing residents in the same building differ in key ways from the characteristics of the rest of the network. Relationships in the building are somewhat more diverse, ties interact less frequently, and a smaller share of relationships are to others who are better off. Although within-building networks have a greater average density than those that comprise individuals living elsewhere in the neighborhood, the proportion that interacts regularly remains low.
For the average network, only 25 percent of building ties were named in the core personal network.
Certain characteristics were associated with a greater likelihood of receipt or provision of support between affordable-housing residents. Expressive support is more likely between similar individuals; however, homophily is not associated with instrumental or informational support. Both ties 64 American Neighborhoods: Inclusion and Exclusion Building Ties: The Social Networks of Affordable-Housing Residents having one or more children is the factor that is mostly consistently associated with provision and receipt of support between building residents. This finding is consistent with other qualitative research that finds that children facilitate interaction across socioeconomic groups (Chaskin and Joseph,
2011) and improve access to other resources, such as childcare centers that act as brokers to other institutions and services (Small, 2009). Residents are significantly more likely to receive information about childcare or finding a school or tutor for their child from a building tie who is better off.
Although affordable-housing residents do not perceive a great degree of reliance on building ties and report few or no close friends in the building, we find meaningful interaction and exchange of multiple types of support that may help residents to both get by and get ahead. Taken together, this finding suggests that residents of affordable housing access a broad range of social resources, with relationships to neighbors in the building acting as supplemental or secondary ties. The value of these ties depends partly on the direct resources and knowledge of the individual and partly on the resources of others in the broader network. For this reason, it is less important that most building relationships are with ties who are doing about the same than it is that these ties, in turn, are connected to a range of others outside the building, many of whom are better off. Because residents exchange informational resources with a substantial proportion of their building network, it is important to consider if and how building networks facilitate access to new or different knowledge, rather than the mere exchange of information. In our case study, we see the potential for affordable-housing residents to benefit directly from exchanges with other building residents and indirectly by becoming connected to neighbors who have access to social resources. Whether these same processes would work in residential developments with a broader mix of incomes is unclear. More research is needed that examines the social lives of affordable-housing residents in different contexts. This research is particularly important if we are to understand how this population functions in complexes with a broader income mix and how policies can support greater social integration across income levels.
Acknowledgments The authors thank their dedicated project team, particularly their field interviewers who contributed to this project. They also thank their anonymous reviewer for providing thoughtful feedback. This project is supported by U.S. Department of Housing and Urban Development grant H-21613CA.
Authors Elyzabeth Gaumer is the Director of Housing Policy Research and Program Evaluation at the Department of Housing Preservation and Development, City of New York.
Ahuva Jacobowitz is the Deputy Director of Housing Policy Research and Program Evaluation at the Department of Housing Preservation and Development, City of New York.
Jeanne Brooks-Gunn is the Virginia and Leonard Marx Professor of Child Development and Education at Teachers College and College of Physicians and Surgeons, Columbia University.