Social Capital Theory

Capacity-centered community change efforts promote the development of social capital in communities. Social capital theory posits that the exchange of resources that occurs through the connections, relationships, and trust among individuals and within communities is vital to community prosperity (Putnam, 2000). In essence, social capital involves the perceived potential for and realized benefits that are achieved through the interchange of resources and assets within a given social network (Bourdieu, 1986; Coleman, 1988). Both individuals and communities can benefit from exchanges through social connections (Lin, 2008; Putnam, 2000), and social capital has been considered a catalyst for individual- and community-level economic mobility (Putnam, 2000; Reim, 2013).

Social capital can generate both positive and negative outcomes (Woolcock & Narayan, 2000), however, and low-wealth, underresourced communities often lack the positive social capital that promotes community advancement. For residents in communities with low levels of community social capital, concentrated poverty and social isolation may be barriers to upward economic mobility (Greenbaum, Hathaway, Rodriguez, Spalding, & Ward, 2008). Similarly, communities characterized by many close social relationships can constrain individual and community outcomes by limiting information and enforcing negative norms (Wilson, 1996). Despite the potentially negative aspects of social capital, interventions aiming to increase the desirability of a low-resource neighborhood and improve quality of life for residents often work to strengthen social capital, which has been described as both the driver and the outcome of positive community change (Pitzer & Streeter, 2015).

At the individual level, social capital has been described as the capacity of one’s relationships to yield access to resources that can result in positive personal outcomes (Lin, 2008; Putnam, 2000). To some extent, every person possesses social capital, which is influenced by one’s community connection, cognitive abilities, financial resources, and overall sense of well-being and satisfaction in life (Brehm & Rahn, 1997).

This paper focuses on social capital because community organizers engage both its individual- and community-level representations. The work of community organizers bridges the micro–macro divide by engaging individuals with variable stores of social capital in change efforts in places with variable amounts of community social capital. Therefore, it is necessary to understand how communities differ in social capital and related resources and how individual social capital profiles differ. Although much research has examined community-level social capital and related constructs, there has been less exploration of the social capital profiles of individuals in low-resource communities. Additionally, there are few resources to guide scholars and community social workers in identifying valid, reliable pathways for building social capital for people living in concentrated disadvantage. Thus, the purpose of this research is to identify the social capital profiles of individuals in low-resource communities and to illustrate how this information can be used to improve resident recruitment and engagement in community change.

Social Capital and Social Ties

Social ties have been described as the relationships people have with friends and neighbors and the resources that individuals gain access to as a result of those connections (Lin, 2001; Portes, 1998). The literature describes many types of social ties—including bonding, bridging, strong ties, and weak ties—which impact social capital within groups and communities, across groups and communities, and according to the degree of closeness those ties imply. Bonding social capital refers to the collective strength and quality of relationships within a group of people, while bridging social capital explains the strength and quality of relationships with others outside the community or network (Banks & Butcher, 2013; Putnam, 2000). Weak social ties are relationships that mainly consist of the exchange of information, whereas strong social ties describe tight-knit networks with close emotional ties (Ellison, Steinfield, & Lampe, 2007; Putnam, 2000).

Social ties have been measured as the presence of cohesive and resourceful relationships and connections within a neighborhood network (Moore et al., 2010). Research has found social ties to be related to mental well-being and life satisfaction (Helliwell & Putnam, 2004); however, studies have shown that the psychological benefits of social ties are more prevalent in males and people with higher socioeconomic standing (Kawachi & Berkman, 2001). Social ties can also connect residents in underresourced communities to opportunities, but the type of social ties one has is important. For example, Tegegne (2015) found that U.S. immigrants relied heavily on their close social ties when searching for employment, which resulted in limited opportunities and low-wage jobs. Tegegne’s study suggested that the type of social ties one has matters, and connections beyond one’s core group—bridging social capital ties with others outside of one’s neighborhood or socioeconomic group—can open doors for new and a wider range of opportunities. Although strong social ties may lead to improved mental well-being and life satisfaction, weak social ties can be a way to attain knowledge and access new opportunities for positive life outcomes, as was demonstrated in early social network research that illustrated how weak ties facilitated access to employment opportunities (Granovetter, 1973). Bridging social capital—which connects diverse groups of people (e.g., different racial and cultural backgrounds)—can be an opportunity to widen and increase an individual’s access to and exchange of resources, resulting in positive life gains.

Social Capital and Civic Engagement

Civic engagement has been described as the level of involvement in social participation and formalized group memberships practiced by members of a community (Shah, 1998). Commonly recognized civic engagement behaviors include how active community members are in volunteerism, politics, community organizations to address common concerns, religious organizations, and in social and interpersonal activities (Putnam, 1995). Research has demonstrated that having close social ties within a network can predict higher levels of civic engagement, both within and outside of a given network (Nisanci, 2017).

According to Putnam (1995), civic engagement is a critical driver for building community, and more people being engaged in civic institutions indicates greater levels of community social capital. Indeed, communities with high levels of civic engagement are most likely to be successful in addressing complex community-level social issues (Putnam, 1995). Research has established the connection between civic engagement and collective efficacy, whereby the higher the level of civic engagement in a community, the higher the presence of bonding social capital, and subsequently, the higher the degree of perceived collective capacity to address community concerns (Collins, Neal, & Neal, 2014). In another study, Mangum (2011) found a connection between trust and civic engagement and that many of the factors (e.g., age, social class, religious activity) that led to increases in civic engagement also led to increases in trust.

Social Capital and Trust

Trust can facilitate positive social relationships in pursuit of mutually beneficial goals (Coleman, 1988; Ross, Mirowsky, & Pribesh, 2001). As both a key component of and a necessary precursor to establishing social capital, trust is critical to the creation, support, and maintenance of social capital in order to strengthen neighborhoods (Usher, 2007). Generalized trust is concerned with general trust in others, including those not personally known; personalized trust describes trust toward those in one’s immediate inner circle (Marschall & Stolle, 2004; Stolle, 2002; Uslaner, 2002; Yamagishi & Yamagishi, 1994). Putnam’s (2002a) conceptualization of social trust includes both personalized trust (e.g., kinship and neighboring) and generalized trust (e.g., general trust in others and institutions, such as local government and law enforcement). Generalized trust is “less intensive but more extensive” than personalized trust (Stolle, 2002, p. 399), and research has found that generalized trust can be a stronger predictor of financial prosperity than other dimensions of social capital (Knack & Keefer, 1997; Stolle, 2002). Higher levels of civic engagement are correlated with higher levels of generalized trust, and communities benefit when members are highly engaged and working cooperatively (Stolle, 2002).

Mistrust, on the other hand, has been defined as “an absence of faith in other people based on a belief that others are out for their own good and will exploit or victimize you in pursuit of their goals” (Ross et al., 2001, p. 569). Mistrust exists when the perceived potential cost for negative outcomes of trusting behavior outweighs the perceived potential benefit for trusting behavior (Coleman, 1988; Smith, 2010). Mistrust can be detrimental to communities, stifling progress and weakening the community social fabric. High levels of mistrust—both personalized and generalized—are present in many vulnerable and marginalized communities, as living in distressed neighborhoods marked by chronic crime and concentrated poverty prohibits a culture of openness and cooperation. Both individual characteristics and experiences and characteristics of the communities in which people live influence trust (Alesina & Ferrara, 2002). In the absence of trust, reciprocal relationships and civic engagement may weaken, thereby impacting capacity for community improvements (Pyles & Cross, 2008).

Social Capital and Reciprocal Relationships

Reciprocity is another dimension of social capital (Menzel, Buchecker, & Schulz, 2013; Putnam, 1993), and norms of reciprocity represent an influential resource embedded within social relationships. A relationship is considered reciprocal when resources are shared in an equitable and interchangeable process for the mutual benefit between and among members of a given community (Sánchez-Franco & Roldán, 2015; Tamjidyamcholo, Bin Baba, Tamjid, & Gholipour, 2013). Reciprocity in relationships has been described as how willing community members are to help one another and receive help in kind (Kawachi & Berkman, 2000). Norms of reciprocity involve the customary expectations among community members of helping behaviors within a network (Sánchez-Franco & Roldán, 2015), and when examining the exchange of resources in a community, reciprocity has been noted as a critical social capital construct to investigate (Chan & Li, 2010).

Reciprocal relationships and helping behaviors have been found to directly impact trust (Sánchez-Franco & Roldán, 2015). Concerning reciprocity, trust has been described in the literature as the “willingness to take risks” based on the confidence that there will be a return on investment of the helping behavior in the long term (Onyx & Bullen, 2000, p. 24). Reciprocity can exist in relationships both within close-knit groups and among those with looser ties. Further, it has been suggested that communities with high levels of reciprocal relationships are more invested in one another, and that reciprocity is evidence of prosocial community behaviors (Onyx & Bullen, 2000).

At the community level, the trust and reciprocal relationship dimensions of social capital are related to the concept of neighborhood collective efficacy. Criminologists have found the community social capital concept inadequate for explaining differences in crime rates across neighborhoods. They sought to identify the mechanisms through which social factors were related to crime. Collective efficacy measures the amount of social control and social cohesion perceived by residents to be present in a neighborhood. Multivariate research findings revealed that lower neighborhood crime rates were associated with higher levels of neighborhood collective efficacy (Morenoff, Sampson, & Raudenbush, 2001; Sampson, 2012).

Together, neighborhood social capital and collective efficacy research can inform neighborhood assessment efforts. Having insight into the amount and type of social ties, civic engagement, trust, and norms of reciprocity present in a neighborhood helps organizers to understand neighborhood social context and how this may be associated with socioeconomic conditions. Furthermore, identifying the amount of collective efficacy present can indicate how amenable residents may be to participating in interventions. However, this research does not explain how differences in social capital dimensions lead to collective efficacy, nor does it tell organizers how to engage individual residents to help develop the social control and social cohesion constructs that comprise collective efficacy. In the following section, we review community interventions that are pertinent to enhancing social capital and collective efficacy.

Types of Community Interventions

Community-level social capital interventions can be separated into two broad categories: social network interventions and interventions that build trust, reciprocity, and civic engagement. Focusing on the social ties, public health social network interventions seek to promote the adoption of positive health behaviors (Valente, 2012). Opinion leaders are identified by survey respondent nominations and targeted to disseminate information throughout the community (Valente & Pumpuang, 2007; Kelly et al., 2006; Starkey, Audrey, Holliday, Moore, & Campbell, 2009). However, some researchers have argued that formal community leaders are not always the best people to target to promote change because they have a vested interest in the current social order (Rogers, 2003; Valente, 1995). Thus, other network interventions identify individuals who are useful for promoting information diffusion based on their position in the community’s social network (Borgatti, 2006), specifically, individuals with bridging ties or other indicators of network centrality. These approaches require data on the entire social network within a community (global network data) and use mathematical algorithms to identify the most useful people to target.

Other community social capital interventions target the trust, reciprocity, and civic engagement dimensions of social capital. For example, the Saguaro Seminar’s Social Capital Building Toolkit Version 1.2 (Sander & Lowney, 2006) focused on the types of events and groups that are more likely to generate trust among small and large groups of participants, instructing organizers to host celebrations or events where people can enjoy shared interests, encourage residents to do favors for one another, discuss community issues, work on projects to achieve a common goal, and build one-on-one relationships. Other work has argued that community practitioners can cultivate reciprocity across groups when bringing groups together in a community-change initiative (Harris et al., 2013). This work is limited by a focus on groups and organizations as primary actors; when individual agents are discussed, the value of their social capital is defined only in terms of the quantity present.

Alone, these types of interventions provide insufficient direction for community organizers to target participants to affect broad community change. In the case of social network interventions, the goal is typically information dissemination and not broad community change, such as increases in trust, community capacity, health and well-being, collective efficacy, or reductions in crime.

Current Study

The current study builds on prior research by focusing on types and levels of social capital held by individuals residing in three low-resource communities. Specifically, we sought to identify whether participants demonstrated different social capital profiles, and if so, how those profiles differed across social capital dimensions. Understanding community members’ social capital profiles may help organizers to plan more effective community engagement strategies.

Research Design and Procedures

We used a community-engaged, mixed-methods approach to examine our two research questions. Dimensions of social capital used in the survey questions, measures, and methods of data collection were discussed and agreed upon in conversations between community leadership and research partners, which included residents from the target neighborhoods. To inform development of the survey instrument, we conducted six focus groups with neighborhood residents during the initial stages of data collection. Qualitative data were collected from focus groups to capture residents’ perceptions of the constructs of social capital, and the findings were used in the development of a household-level survey instrument, which the research team devised and finalized in collaboration with community leadership and residents.

The four summary social capital variables resulting from the focus group findings and collaborative survey development process included measures of trust, civic engagement, social ties, and reciprocal relationships. The survey instrument included adapted versions of scales containing social capital constructs measuring social ties, trust, and civic engagement from the Harvard Social Capital Community Benchmark Survey–Short Form (Putnam, 2002b) and measuring reciprocal relationships as originally used in the National Survey of Black Americans (Jackson & Neighbors, 1997), later adapted for use in the Making Connections Initiative (Annie E. Casey Foundation, 2013).

Surveys were administered through in-person interviews at residents’ homes in fall 2014 and spring 2015. Participants were informed that their participation was voluntary, surveyors shared informed-consent documents, and residents were told about their rights regarding confidentiality and the right to refuse participation or opt out of the study at any time.

Participants

Participants in the household-level survey included one person per household—self-identified as the “head of household”—from occupied residences located in the target neighborhoods, which were comprised of two census tracts. At the time of this study, census data indicated that of the 4,697 residents in the neighborhoods, 57.7% were living in poverty, 97% were minority (92.9% Black), and 27% had less than a high school degree; the median income was $24,298 (U.S. Census Bureau, 2013). In comparison, 25.4% of all residents were living in poverty in Baton Rouge, where the citywide median income was $38,953 (U.S. Census Bureau, 2013). The sample population included the total number of households (2,401 in 2013) with a vacancy rate of 20.95%; surveyors identified an additional 65 housing units as vacant or unsafe.

The face-to-face household-level survey used a random sampling strategy, with a completion goal of 200 households. The random sample was generated from a list of known addresses for occupied households in an Excel spreadsheet, using the RAND() method. Of the households approached, 54 refused to participate, and 47 were not home. After three attempts, households not available were removed from the list and new addresses were randomly selected as replacements. A total of 121 households completed the survey; 392 residents (adults and children) were reported to reside in those households.

Measures

Data Analysis

To identify a typology of individual social capital, we used Ward’s method of hierarchical cluster analysis using squared Euclidian distances to identify initial clusters—a method that is appropriate for use with continuous variables (Everitt et al., 2011, p. 258). This top-down method starts with all individuals in one cluster and separates the sample into increasing numbers of clusters based on a mathematical algorithm. In contrast to factor analysis, which aims to combine individual items in a measurement scale into fewer constructs, the purpose of cluster analysis is to place objects into groups.

To determine the optimal number of clusters, we examined the dendrogram and means of each summary variable. A dendrogram is a tree diagram that depicts the process of classifying objects into clusters. Nodes in the dendrogram depict clusters, and these are represented by horizontal lines that connect multiple vertical lines (Everitt et al., 2011, p. 88). In our study, the terminal nodes (the end points on the left side of the dendrogram; Figure S1, online) represent survey respondents. The entire dendrogram provides a picture of all cluster solutions. Analysis of variance (ANOVA) determined whether the variable means differed across clusters. The total within-group variance parameter from the F test provided evidence that cluster groupings were not due to chance. This and the number of residents in each cluster were used to identify the best cut or optimal number of clusters. No consensus exists regarding the correct procedure for determining the best cut, and subjective judgement based on subject-area expertise is expected to be used (Baxter, 1994, in Everitt et al., 2011). Thus, a subjective discussion of the selection of cluster number is provided, including a discussion of substantive differences between alternate cluster solutions.

Descriptive Statistics

Table 1 depicts the characteristics of survey participants. More than three quarters of respondents were female (77%), and most were African American (96%). Nearly half of respondents were single (48%), and more than half had attained a high school education or GED, with an additional 31% reporting some college, having a college degree, or more than a college degree. Of those responding to the question regarding household income, 58% reported earning less than $15,000 in the year prior to the survey; 23% reported owning their own home. The average age of respondents was 47.69. On average, respondents lived in their neighborhood around 10 years and in households with a mean of 1.95 adults and 1.32 children. With regard to the social capital dimensions, descriptive findings revealed the sample mean level of civic engagement was 9.04 (SD = 3.23) with a slightly higher mean reciprocity score (10.72; SD = 3.04). The mean scores for trust and social ties were 6.52 (SD = 2.38) and 5.88 (SD = 2.29), respectively.

Cluster Analysis

The dendrogram revealed potential cluster groupings of two, three, four, or five. (Figure S1, online). Using ANOVA, we determined that the means of each dimension of social capital differed significantly across all groups in each of these cluster groupings (Table 2). The two- and five-cluster solutions were not selected. The two-cluster solution provided very little substantive information other than differentiating between one group with higher means on all dimensions of social capital and another group with lower means on all social capital dimensions. The five-cluster solution had fewer than 10 cases in three or more clusters. Because the three- and four-cluster solutions were identical except for the classification of six individuals in a separate group, we present detailed information about the four-cluster solution, a substantively unique cluster that provides more insight into variations in social capital profiles. We first present the four-cluster solution, then compare the four-cluster solution to the three-cluster solution.

The four-cluster solution resulted in four groups whose means on each of the four dimensions of social capital were significantly different from one another, comprising distinct social capital profiles. Table 3 presents descriptive statistics for social capital dimensions across clusters.

Individuals in Cluster 2 demonstrated the highest overall level of social capital. This cluster had the highest mean scores for civic engagement (12.6) and social ties scores (8.7); its mean levels of trust ranked third (6.3) and tied for lowest mean reciprocal relationship score (9.7). We labeled this social capital profile top civic engagement and social ties, low trust and reciprocity (TCS_LTR; n = 17). This profile indicates the highest levels of participation in civic activities such as volunteering and attending public meetings, religious services, and community events; serving as a leader in an organization; talking to leaders about or working to improve neighborhood problems; and voting. At the same time, this group reported the most close connections to friends, those who are different than them, and community leaders. The group’s lower trust score revealed less generalized trust in neighbors, police, and the local government. Finally, individuals with this profile provide and receive help from friends and family less than all others.

Cluster 1 ranked second in overall social capital and civic engagement with slightly lower levels of civic engagement (10.0) than Cluster 2. This group also had the second highest level of trust (8.0) and reciprocity (10.2) but ranked third in social ties (5.0). This group was classified as high civic engagement, trust, and reciprocity, low social ties (HCTR_LS; n = 21). Thus, this group’s social capital profile is characterized by participating in the neighborhood’s civic life to a large extent; articulating trust in neighbors, police, and leaders; and providing and receiving help from family and friends. Group members invited friends, people of different racial backgrounds, and neighbors to their homes less often than two other groups.

Cluster 3 ranked third in overall social capital. Labeled high reciprocity and social ties, low civic engagement and trust (HRS_LCT; n = 31), this group had the highest mean reciprocity score (12.4) but had the lowest level of trust (5.1). It ranked second for social ties (6.2) and third for civic engagement (8.6). This group’s social capital profile is dominated by providing and receiving help from family and friends and visiting with others in their homes. Group members have relatively low levels of civic engagement and the lowest levels of generalized trust.

Cluster 4 ranked fourth in overall social capital. Although this group had the highest mean social trust score (9.0), it had the lowest mean scores for civic engagement (6.0) and social ties score (4.3), and it tied for the lowest reciprocity score (9.7). Individuals with this social capital profile, classified as high trust (HT; n = 6), expressed the highest trust level of all the groups and the lowest scores for all other social capital dimensions. Figure 1 compares social capital dimension scores across the four clusters.

The three-cluster solution combined the HT group from Cluster 4 with Cluster 1 in the four-cluster solution. This did not change the overall rank of Cluster 1 on the dimensions of social capital, nor did it substantively change the means. However, it did obscure the existence of a group of residents whose social capital profile was defined by the highest levels of trust but the lowest levels of the other social capital dimensions.

Prior research has discussed only the correlations among social capital measures and with other predictors and outcomes. No known research has attempted to classify individuals into groups based on their social capital profile. Thus, we present the correlation results for all social capital dimensions to enable a comparative discussion of findings. In this study, all social capital measures are positively and significantly correlated with one another, ranging from a moderate association between civic engagement and social ties (r =.65, p < .001) to a weak association between trust and reciprocity (r = .19, p < .05). Civic engagement is positively correlated with trust and reciprocity (r = .29, p < .01; r = .46, p < .001, respectively). Social ties are positively associated with trust and reciprocity (r = .24, p < .05; r = .35, p < .001).

Top Civic Engagement and Social Ties, Low Trust and Reciprocity

The TCS_LTR group includes individuals who are highly involved in civic groups and volunteering and who invite dissimilar individuals and community leaders to their homes, indicating ties that bridge social networks. A prior history of civic engagement indicates that members of this group are likely to participate in future community engagement efforts (Putnam, 1995) and possess increased collective capacity to address community concern (Collins et al., 2014). Similarly, the high bridging social ties scores indicate that this group may be useful in disseminating information to others in the community (Borgatti, 2006).

The pattern of high levels of civic engagement and low levels of trust present in this group seems to contradict previous findings that civic engagement is positively correlated with generalized trust (Stolle, 2002). However, this finding illustrates the difference between our analysis and prior work. Whereas past work has identified correlates of various dimensions of social capital in communities, cluster analysis clearly illustrates that positive correlations do not indicate that all individuals with high levels of civic engagement also trust others. The social profiles derived from our cluster analysis reveal individual social capital complexity that warrants organizers’ consideration in terms of community engagement efforts.

For this group, lower levels of social trust may require community organizers to put extra effort into demonstrating credibility and establishing the relevance of their initiative before attempting to engage group members, who are likely already community leaders. The reasons this group is less trusting than others are unknown. Perhaps the low levels of trust are related to personal characteristics, or perhaps group members’ prior experiences engaging in community-change efforts have demonstrated to them that trust is not always warranted. For example, during one research team meeting, a community resident discussed the importance of building trust and credibility. He observed that two types of organizations come into their community: those that make promises and walk away, and those that actually make something happen. Future research should explore the impact of organizational efforts on trust.

This group also is less involved in reciprocal relationships than the HCTR_LS group. Scholars have suggested that communities with high levels of reciprocity have a high level of care for their neighbors (Onyx & Bullen, 2000). The measure of reciprocal relationships used in our study depicts the extent to which residents provide and receive help from family and friends. This may mean they do not need or want to give or receive help, or it may indicate an unidentified need or potential resource. These speculations should be the basis of future research. In any case, this group would be important to engage because its high level of social ties would help elicit the participation of others through their diverse networks—particularly important in efforts to reduce the negative effects of place-based economic inequality.

High Civic Engagement, Trust, and Reciprocity, Low Social Ties

Members of the HCTR_LS group are involved in civic institutions, have high levels of trust, and are involved in reciprocal relationships but less frequently invite unlike individuals into their own homes or visit them in their homes. It is possible that this group is so busy with community-engaged activities that they are not often home in the evenings or on the weekends to socialize. It is also possible that they spend their spare time with strong ties who then engage in reciprocal relationships. It may also be that they have demanding jobs that leave little time for social engagement outside of work. Whatever the reason for their low level of reported social ties, this group is a natural target for organizers to recruit for involvement in community engagement efforts and would likely be the easiest to engage in the early stages of a community-change initiative. Care should be taken to help this group feel comfortable when exposing them to unlike others and preparing them to address or confront community leaders, since frequent contact with unlike others is limited.

High Reciprocity and Social Ties, Low Civic Engagement and Trust

The HRS_LCT group is characterized by high levels of reciprocal relationships and high social ties, but the group has the lowest levels of generalized trust and lower levels of civic engagement than all but one group. Members of this group are engaged in mutual aid with individuals they know and may not have the time or resources to participate in community engagement activities. Their exceptionally low trust levels in the presence of reciprocal relationships is unexpected given prior research, which has found that norms of reciprocal relationships and helping behaviors directly and positively impact trust (Sánchez-Franco & Roldán, 2015). Although the type of community (i.e., virtual vs. neighborhood) and measures of reciprocal relationships and trust differed in this study from the measures used by Sánchez-Franco and Roldán, this incongruent finding illuminates the need for further research to understand the motivations of members of this group. Clearly, gaining their trust to elicit participation in community efforts will require careful, targeted communication and an understanding of their time demands. Future research should also explore the extent to which this group suffers more negative effects of place-based economic inequality. Group members’ higher levels of reciprocal relationships and low levels of trust may result from witnessing these negative effects among friends and relatives and needing to intervene when other institutions fail to help.

High Trust

Perhaps the most interesting and unexpected group that emerged from this study is the HT group. These individuals expressed the highest level of generalized trust and lowest levels of all other dimensions of social capital. This type of generalized trust has been found to be highly correlated with other dimensions of social capital (Knack & Keefer, 1997; Stolle, 2002). Thus, because members of this group trust social institutions, they may be more willing to engage in community improvement efforts. However, since they are not already in the habit of participating in civic activities or exchanging assistance and resources, it may be difficult to elicit and sustain their engagement. A community organizer might provide opportunities for capacity building and networking for this group, as its members might engage with a community-change initiative but not know where to start. Organizers should be mindful that this group represented a small number of individuals; future research with larger samples should assess factors predicting this social capital profile.

Limitations

Several limitations of this research should be considered. First, although the sample was randomly selected, it came from a contiguous group of neighborhoods in one midsized southern city, and some respondents did not respond to all survey items. Additionally, 77% of respondents identified as female, resulting in an oversampling of females in the community. Because the sample size was too small to incorporate nonresponse rates, the study sample was not fully representative of the neighborhood. Finally, the total number of respondents and the low number of respondents in one of the clusters prevented more complex analysis.

Implications for Research and Practice

Future research should determine if social capital profiles vary by geography and as compared to national rates. Studies are also needed to identify factors that predict each social capital profile and assess the extent to which social capital profiles change over time. Additionally, future research could examine the impact of organizing efforts on trust, reciprocal relationships, social ties, and civic engagement, and how this subsequently influences social capital profiles. To advance social work practice in neighborhoods, the Putnam social capital short form and reciprocity measures might be administered to community members prior to an organizing effort to help identify members’ social capital profile. Interventions might also be targeted to different social capital groups, with outcomes measured to determine the effectiveness of the various strategies used based on social capital profiles. Additionally, future practice research might include the development and testing of a brief screening instrument to assess the social capital profile status of individual community members. Such an instrument may allow social workers to determine the best method for engaging individuals based on their profile and also help them to understand which profile groups are missing from the engagement effort.

Conclusion

Advocacy in social work practice is essential to promoting social justice (Brown, Livermore, & Ball, 2015), and engaging residents of low-resource communities in advocacy and community development efforts is necessary to ensure that community- and policy-level changes reflect their desires and promote their interests. Understanding individual social capital profiles and designing strategies to maximize their engagement is essential to expanding these efforts and reducing place-based economic inequality and related socioeconomic conditions.