Social Capital Theory and Quantitative Approaches in Measurements: Disaster Literature Focus

Abstract

Disaster recovery depends heavily on the support of social networks and the resources they can generate. Yet such support is difficult to measure and assess. This paper reviews existing quantitative approaches to measure social capital within a disaster context. The article addresses (1) how is social capital conceptualized in the disaster literature? and (2) what social capital measures have been used based on existing conceptual frameworks (e.g., bonding, bridging, and linking)? We review how social capital has been defined and what properties of social capital make it important in the disaster planning contexts. Then we explore and assess existing approaches used to measure social capital while offering suggestions for potential improvements. These potential improvements to social capital indices will capture and measure social capital in a more comprehensive way and have a more solid theoretical basis based on the social capital and disaster resilience literature. Each type of social capital is expected to have different consequences and effects on disaster planning processes and phases; therefore, it is important to distinguish and measure social capital variables in a holistic way for policymakers and practitioners to make use of social capital information in preparing for disasters.

Introduction

A growing body of literature advocates for the incorporation of social capital into efforts to enhance community resilience and foster sustainable, locally driven disaster management (Berke and Campanella 2006; Rydin and Pennington 2000; Kumari and Frazier 2021). Social capital refers to the “features of social organizations such as networks, norms, and social trust that facilitate coordination and cooperation for mutual benefit” (Putnam 2000, p. 225). It has been shown to play a significant role in enhancing resilience across multiple phases of disaster (Meyer 2018). Despite ample evidence demonstrating the contributions of social capital to community resilience, it has yet to become a central component of disaster planning (Kumari and Frazier 2021). One reason for this underutilization could be attributed to the challenges associated with quantifying social capital in a systematic manner.

Indeed, while social capital measurement has traditionally relied on qualitative approaches (i.e., surveys and focus group interviews) focusing on limited geographical areas, these approaches not only hinder broader empirical studies across varying contexts but are also less accessible for local authorities and planners to gauge the level of social capital in their related communities (Meyer 2018; Engbers et al. 2017). To tackle these challenges, scholars suggest quantitative approaches to measure social capital using secondary data sets.

This paper reviews the existing quantitative approaches to measure social capital from a perspective of disaster planning through two primary research questions: (1) how is social capital conceptualized in the disaster planning literature? and (2) what social capital measures have been used based on existing conceptual frameworks (e.g., bonding, bridging, and linking)? We review how social capital has been defined and what properties of social capital make it important in the disaster planning contexts to answer the first question. To address the second question, we compile, explore, and assess existing approaches used to measure social capital. We conclude by critiquing the existing social capital measures and provide suggestions for improvements and refinements for better conceptualization of social capital with quantitative approaches.

Emergence of the Social Capital Concept

The concept of social capital has a long intellectual history that can be traced back to the work of Hanifan (1916). Hanifan (1916) defined social capital as accumulated assets of social groups (i.e., family and community) wherein people share some degree of similarity (i.e., interest or background). He noted that social capital could benefit each group member in satisfying their needs. A few decades later, Jane Jacobs brought the term social capital into the community planning literature through her work. In her seminal book, The Death and Life of Great American Cities (1961), Jacobs described social capital as networks of relationships between neighbors that are slowly built up through everyday activities and interactions in a neighborhood. She believed that social capital provided a foundation for mutual trust, shared efforts, and resilience. Jacobs (1961) focused on the community as a social unit, but her conceptualization of social capital is in line with Hanifan’s definition in terms of recognizing social capital as an accumulated asset within a social group based on reciprocal relationships.

Bourdieu (1986) redefined the concept of social capital by expanding its focus to more informal mutual social networks at the individual level. He defined social capital as “the aggregate of the actual or potential resources which are linked to possession of a durable network of less institutionalized relationships of mutual acquaintance and recognition” (Bourdieu 1986, p. 248). Building on Bourdieu’s works, Coleman (1988) specified the following elements of social capital: obligations and expectations, information channels, and social norms. The study examined how the outcomes from social capital change under social structural conditions. Coleman (1988, p. S98) defined social capital as a capital that “inheres in the structure of relations between actors and among actors.”

While Bourdieu and Coleman understood social capital as a private good, Woolcock and Narayan (2000) and Putnam (2000) focused on the collective outcome of social capital (Monteil et al. 2020). Woolcock and Narayan (2000, p. 226) referred to social capital as “the norms and networks that enable people to act collectively,” with an emphasis on the source of social capital rather than consequences. Putnam (2000, p. 225) further expanded and publicized the notion of social capital, highlighting its role in civic engagements in modern society, in his book Bowling Alone: America’s Declining Social Capital. He defined social capital as “features of social organization such as networks, norms, and social trust that facilitate coordination and cooperation for mutual benefit.”

As summarized in Table 1, in scholarly works over the decades, the concept of social capital has been expanded and developed into a more complex concept involving different levels and dimensions that can generate both positive and negative outcomes. The evolution of social capital research has inspired several different social science disciplines, including disaster planning.

Table 1.

Definitions of social capital through time

Study	Social capital definition	Note
Hanifan (1916)	Accumulated assets of social groups (i.e., family and community) wherein people share some degree of similarity (i.e., interest or background)	The original concept of social capital
Jacobs (1961)	Networks of relationships between neighbors that are slowly built up through everyday activities and interactions in a neighborhood. She believed that social capital provides a foundation for mutual trust, shared efforts, and resilience	Incorporated community as a social unit, in which social assets can be accumulated and stored
Bourdieu (1986)	The aggregate of the actual or potential resources which are linked to possession of a durable network of less institutionalized relationships of mutual acquaintance and recognition	Expanded the social capital concept to informal mutual social network
Coleman (1988)	Inheres in the structure of relations between actors and among actors	Specified the elements of social capital: obligations and expectations, information channels, and social norms
Woolcock and Narayan (2000)	The norms and networks that enable people to act collectively	Highlighted source of social capital
Putnam (2000)	Features of social organization such as networks, norms, and social trust that facilitate coordination and cooperation for mutual benefit	Highlighted collective perspective of social capital in facilitating coordination and cooperation

Distinct Properties of Collective Social Capital

Social capital has several distinct properties that set it apart from other forms of capital, such as economic capital and physical capital. Firstly, social capital functions primarily as a public good (Aldrich 2011). Public goods are characterized by nonexcludability and nonrivalry. Nonexcludability refers to the difficulty of excluding nonpaying consumers from consumption. The nonexcludable trait of social capital helps to create a more inclusive and equitable community environment. Collective resources in social capital, such as norms and trust, are available to anyone joining the community. This often also happens regardless of whether they have contributed to the formulation of social capital or have other types of capital (i.e., human and financial capital) to exchange for social capital. For example, equal access to social capital allows anyone, especially marginalized populations, to participate in community activities and decision-making processes. Therefore, the gaps between the advantaged and disadvantaged populations can potentially be narrowed.

Nonrivalry, the other trait of public goods, means one’s consumption does not depreciate the value of capital or affect its availability for others. Typically, the use of traditional modes of capital involves a decrease in its value or availability. For instance, the use of an emergency disaster shelter can reduce the availability of the shelter to others in need. However, the use of collective social capital does not necessarily diminish its value or availability rather, it often enhances its effectiveness (Adler and Kwon 2002). Lee and Fraser (2019) found that individuals who have directly or indirectly experienced the value of social ties are more likely to actively participate in civic activities than others. Social networks with expanded memberships are more likely to achieve practical benefits through collective actions, such as receiving recovery funds (Lee and Fraser 2019). Further, social capital is the only form of capital that can be enhanced during an emergency period (Lee and Fraser 2019). Damage from disasters on the built and natural environments and their resultant human losses can, consequently, bring enormous economic burdens. While the scarcity of resources due to disaster impacts may last for long periods of time, social capital continues to be available and can even be enhanced during such difficulties and events. For example, the Vietnamese American community in New Orleans rapidly recovered from the damage following Hurricane Katrina although they had scarce financial and human resources compared to the nearby communities (Aldrichi et al. 2018). With the leadership of the local Vietnamese church, they capitalized on strong social ties, efficiently addressed impediments including insufficient resources, and eventually restored their community in a shorter period of time compared to the neighboring communities having better access to resources.

The Theoretical Roles of Social Capital in Disaster Management

Disaster management refers to a series of activities and endeavors to reduce the disaster impacts. It typically comprises of plans, structures, arrangements, and resource mobilization established to return to normal routines in a comprehensive and coordinated way to respond to the whole spectrum of emergency needs. A commonly used model of disaster management consists of four phases: preparedness, response, recovery, and mitigation. These four phases are a continuous cycle with some overlaps between each phase.

Kapucu (2008) summarizes the theoretical explanations of how the coordination of social capital and social networks improves the effectiveness of overall disaster planning processes. One explanation discussed is that the inclusion of diverse social networks and stakeholders helps local authorities make better decisions that are pertinent to the local context (Lyles et al. 2014; Burby 2003). Local authorities have to make choices and take actions based on the information received from organizations and individuals in the community (Kapucu 2008). As the information can often be ambiguous or conflicting, a given message has more than one possible interpretation. Interpretation of disaster-related information entails unique challenges that are often time-sensitive, need engineering knowledge, and even require a sense of ethics. Through active communication, participants can collectively interpret and make sense of the information in their environment (Kapucu 2008). Timely, transparent, and comprehensive information dissemination allows for informed decision-making.

Disaster events involve time compression, which creates a time of intense information flows and a high level of uncertainty due to the disruption under abnormal conditions (Olshansky et al. 2012). Intensive information flows require a high capacity to analyze and interpret information using knowledge, experiences, and intellectual resources. Stakeholders of each organization bring different sets of capacity and social capital, including social networks, local knowledge (i.e., information on local capacity), skills, experiences, and authority to administer rules, all of which are critical assets to interpret and analyze disaster-related information reflecting local context. On the other hand, a lack of local knowledge and limited incorporation of existing social networks can bring negative consequences, such as accentuation of preexisting inequalities (Ganapati and Ganapati 2008).

Organizational learning is another theory with which to understand how community coordination benefits the disaster response and recovery. The scope and complexity of emergency response operations necessitate a flexible learning approach. Organizations and individuals learn through processes of knowledge acquisition, information dissemination, information interpretation, and organizational memory (Kapucu 2008). Informative opportunities, such as public workshops that take place during planning processes facilitate dialogue on disaster-related issues among community individuals and organizations and allow them to adjust their goals and performance in accordance with changing conditions and demands in each disaster phase (Kapucu 2008). Such dialogues and adjustments potentially strengthen social networks around disaster goals, formulate collective interests, enhance the level of disaster awareness, and further lead to collective action toward hazard mitigation and resilience building (Brooks 2019; Berke et al. 2012; Nelson and French 2002; Sabet and Khaksar 2020). Conversely, evidence shows that a lack of understanding between authorities and community members can lead to ineffectiveness, insult, damage, and limited coordination of emergency management processes (Tierney and Oliver-Smith 2012).

Frameworks of Measuring Social Capital

Social capital can be further distinguished and assessed by the strength of relationship ties and the attributes of social networks. The three types of social capital studied most include bonding, bridging, and linking, and each generates different effects for individuals and communities during disasters (Aldrich 2012; Aldrich and Meyer 2014; Kyne and Aldrich 2020; Trump et al. 2018; Hawkins and Maurer 2010; Kawamoto and Kim 2019; Kim et al. 2006; Larsen et al. 2004; Szreter and Woolcock 2004).

Bonding social capital describes the strong solidarity between individuals within family groups, small groups, or local communities that typically share similarities in terms of race, ethnicity, religion, and cultural backgrounds (Granovetter 1973). These relationships among homogeneous individuals—such as family members and close friends—provide mutual emotional support, and personal empowerment and assistance. Bonding social ties are often referred to as a sociological superglue that holds individuals together based on deep trust and makes them work as one social entity (Rydin and Holman 2004; Putnam 2000, p. 21). Research on bonding social capital has generally found that a community with a higher level of homogeneity tends to show a higher level of civic participation. Thus, bonding social capital provides stable nourishment for public participation (Aldrich 2011). This strong relationship can provide physical assistance and emotional support, especially in traumatic disaster situations (Hawkins and Maurer 2010). Additionally, a higher level of bonding social capital increases the likelihood of receiving warnings, locating shelters, and obtaining recovery assistance (Aldrich and Meyer 2014). Strong bonds, however, can also have negative effects for those not within the group, creating forms of not-in-my-back-yard (NIMBY)ism in postdisaster scenarios (Aldrich and Crook 2008).

Bridging social capital, in contrast, refers to weak ties that loosely connect heterogeneous groups of people across varying cultural distinctions such as religious traditions, national origins, races, ethnicities, and sexual orientations (Granovetter 1973; Wuthnow 2002). This type of social capital is known for “bridging within a community among people with diverse interests and points of view, and building links between the disinvested community and the mainstream” (Crawford et al. 2008, p. 539). These ties are often described as a lubricant that alleviates conflicts that may exist between individuals or social groups with different backgrounds. Studies show that bridging social capital nurtures a healthy environment for democracy of the larger society as it goes beyond the immediate social groups, overcomes divisiveness and insularity, and encourages not only tolerance but also cooperation that may be useful for addressing large-scale social problems, such as crime, poverty, and disaster disruptions (Wuthnow 2002). Also, bridging ties allow individuals to create more far-reaching connections outside their coherent groups (Holman and Rydin 2013; Putnam 2000, p. 21). While each bridging tie can be considered weak in isolation; ironically, these ties can bring large and far-reaching effects. Scholars have emphasized that these weak ties are more likely to spread new information, transfer crucial resources, enable people to assess external assets, and spread innovations compared to bonding ties (Granovetter 1973). Bridging social capital is known for providing the most important assistance to flood-affected households, particularly in the response and recovery phases (Babcicky and Seebauer 2017). Bridging social capital allows the disaster victims without close relatives or friends in the region to be supported by weak ties outside the affected area (Elliot et al. 2010).

Lastly, linking social capital describes the connections between different individuals or social groups across explicit, formal organizations or institutionalized authority in power, influence, wealth, and prestige (Szreter and Woolcock 2004). These ties are often referred to as vertical or scaffold ties as they connect individuals and groups with key actors at a higher level of power who often have the authority to distribute scarce resources (Rydin and Holman 2004, p. 123). Individuals and social organizations with linking social capital may be beneficial for addressing their crisis as it allows access to information about assistance and help (Wuthnow 2002). Linking social capital in disaster research is often used to refer to the vertical ties between communities and local authorities, such as local government or first responder agencies. These local authorities are crucial sources of workforces and economic assistance for the households that suffer severe physical damage (Babcicky and Seebauer 2017). Communities with strong linking ties to government officials, for example, can speed their receipt of postdisaster aid (Aldrich 2011). The key features of bonding, bridging, and linking social capital are summarized in Table 2.

Table 2.

Summary of the characteristics of bonding, bridging, and linking social capital

Characteristics	Bonding social capital	Bridging social capital	Linking social capital
Structural form	Strong ties, usually between family and close friends	Weak ties, usually between members of different social groups	Weak ties usually involve formal agents (i.e., local government agencies)
Key metaphor	Superglue that holds people together make them work as one entity	Lubricant that alleviate conflict	Scaffold that helps people access key actors at a higher tier
Density of ties	Dense and thick	Sparse and thin	Sparse and thin
Range of resources available from the network	Bounded and limited, similar type of resources	Extended and far-reaching, diverse type of resources	Extended and far-reaching, diverse and unlimited resources
Direction	Horizontal	Horizontal	Vertical
Entities in network	Homogeneous	Heterogeneous	Heterogeneous
Formal/informal	Informal relationships	Informal or formal relationships	Informal or formal relationships

Sources: Data from Rydin and Holman (2004); Wilkin et al. (2019).

Review of the Existing Quantitative Indices of Social Capital at the Community Level

Many social capital frameworks at the community level include subjective dimensions, such as individuals’ trust and altruism toward others in the community (Engbers et al. 2017). To measure such subjective dimensions, survey and interview methods have been widely used. However, because such approaches inevitably involve cost and time constraints, large-scale systematic attempts that entail social capital measurement have been limited (Engbers et al. 2017, p. 550). Further, relative to disaster studies, measuring social capital with such methods can limit scholars’ ability to address gaps in knowledge on social capital in the disaster contexts. Considering the unpredictable nature of disasters, surveys can normally only be implemented postdisaster; therefore, it is nearly impossible to assess changes in social capital or impacts of predisaster attributes on disaster losses (Meyer 2018).

To overcome the critical limitations in survey and interview approaches, there are a few attempts to measure social capital using secondary data sets. This paper focuses on the indices based on the framework of bonding, bridging, and linking social capital, which is now a widely accepted, tested, and established social capital framework (Woolcock and Narayan 2000; Kyne and Aldrich 2020). This framework preserves different types of connections, and it allows this study to explore the contributions of every kind of social capital in disaster management.

One of the attempts is Fraser (2021), who constructed a social capital index using the Japanese Census data set (see Table 3). The paper compiled an annual data set of 1,741 Japanese municipalities from 2000 to 2017. Seven variables for bonding, five variables for bridging, and four variables for linking were used for each component. The index was compared with the social vulnerability index (SoVI) using the SoVI approach proposed in Cutter et al. (2003) and confirmed to be theoretically valid and robust.

Table 3.

Comparison of social capital indicators between studies

Concept	Kyne and Aldrich (2020) (US context)	Fraser (2021) (Japanese context)	Suggested indicator set (US context)	Note
Bonding
Demographic similarity	(Race) Race fractionalization	(Nationality) Nationality fractionalization	(Race) Race fractionalization	—
	(Ethnicity) Ethnicity fractionalization		(Ethnicity) Ethnicity fractionalization
	(Education) Negative absolute difference between % of total population with college education and % of total population with less than high school education	(Education) Negative absolute difference between percentage of total population with college education and percentage of elementary school graduates	(Education) Negative absolute difference between % of total population with college education and % of total population with less than high school education	—
	(Race-Income) Gini coefficient	—	(Race-Income) Gini coefficient	—
	(Employment) Absolute difference between % of total employed and % of total unemployed labor force	(Employment) Absolute difference between % of employed and % of unemployed labor force	(Employment) Absolute difference between % of total employed and % of total unemployed labor force	—
	(Gender-income) Gender income fractionalization	(Gender-employment) Fractionalization of employment equality by gender	(Gender-income) Gender income fractionalization	—
	—	(Religion) Fractionalization by Religious Minority	(Religion) Fractionalization by Religious Minority	Proposed to move from bridging social capital
			(Religion) Number of religious organizations per capita
Language competency	% of total population proficient English speakers	—	% of total population proficient English speakers	—
Communication capacity	% of total households with a telephone	Television broadcast reception contracts per capital	% of total households with a telephone	—
Non-elder population	% of total population below 65 years of age	% of total population below 65 years of age	% of total population below 65 years of age	—
Bridging
Religious organizations	Religious organizations per 10,000 persons	Religious organizations per capita	—	Proposed to move to bonding social capital
Civic organizations	Civic organizations per 10,000 persons	Nonprofit organizations per capita	Number of civic organizations per 1,000 persons	—
Social embeddedness - charitable ties	Member of charitable organization (% of total)	Unions per capita	Number of grantmaking/giving services per 1,000 persons	—
	Member of fraternal order (% of total)		Number of unions per 1,000 persons
	Member of union (% of total)
Social embeddedness-Neighborhood ties	—	Community centers per capita	Community facilities per 1,000 persons (i.e., libraries, recreation centers)	—
		Libraries per capita
Social embeddedness-Civil society participation and norm adaptation	—	Volunteer participation rate	—	—
		Voter turnouts in prefectural/ lower hose elections
Social attachment	—	—	% veteran population	Newly proposed in this paper
Linking
Political linkage	% of total voting-age population who are eligible for vote	—	Voting rate in presidential election	Alternative variable proposed
Government linkage	% of total local government employees working for local governments	Local government employees per capita	% of total employees working for public administration sector	—
	% of total state employees working for the state governments	Prefectural government employees per capita
	% of total federal employees working for the federal agencies	Prefectural police per capita
		% of vote for ruling party in house of Reps elections
Political linkage-political activities	Attended political rally/speech/organized protest activities (% of total)	Prefectural assembly members per capita	Number of peaceful protest events per capita	—
		% population of vote for ruling party in prefectural election	% vote supporting the ruling party of the state/federal
Institutional acceptance	—	—	Census response rate	Newly proposed in this paper
Political plurality	—	—	Political fractionalization	Newly proposed in this paper

Note: Bold type indicates the subcomponents that describe demographic similarity.

Similarly, Kyne and Aldrich (2020) also constructed a social capital index using a publicly available US data set of 3,134 counties across the contiguous US (Table 3). Nine variables for bonding and 10 variables each for bridging and linking were used to calculate the index. They also validated the index with two established indices, including the SoVI and the Baseline Resilience Index (BRIC)—both developed by Susan Cutter and her colleagues (Cutter et al. 2003, 2010). These two sets of indicators capture the social capital based on the bonding, bridging, and linking typology. While those indices are validated using established and widely accepted indices, there is still room for improvement as follows.

First of all, the indicators of religious involvement need to be reclassified. Both Kyne and Aldrich (2020) and Fraser (2021) used indicators that represent quantity of religious organizations in a community, using the number of religious organizations per capita. While these indicators are classified as bridging social capital in both studies, there is plenty of evidence that suggests religious involvement is more appropriate to be classified as bonding social capital. This is not merely a matter of how it is technically classified; as it is described in the previous section, bonding and bridging have different consequences and effects on disaster planning processes and phases, it is important to distinguish between them more properly.

As Putnam (2000) himself admits, bonding and bridging are not either/or categories that can be exclusively separated but rather more-or-less dimensions that can be used to compare different forms of social capital. Still, he clarifies that one of the contrasting features between them is exclusiveness versus inclusiveness. Bonding social capital is inward-looking and tends to reinforce conformity and solidarity, which can often be exclusive to others who do not share this conformity (Monteil et al. 2020; Putnam 2000). On the other hand, bridging social capital is outward-looking and inclusive, encompassing people across diverse social cleavages and enabling linkage to external assets and information diffusion (Monteil et al. 2020; Putnam 2000). In summary, the key factor that determines whether religious variables are closer to bonding or bridging is whether religion makes the overall community more inclusive or exclusive.

In this regard, several studies suggest that churches and faith-based organizations provide opportunities to reach out to people outside of their religious communities, forming inclusive social ties. These organizations offer not only places of worship but also opportunities for education, recreation, and civic services (Heuser 2005). Empirical evidence shows that individuals’ religious attendance is positively associated with broader civic involvement, such as volunteering, participating in civic activities, and joining small support groups (Park and Smith 2000).

However, a majority of studies indicate that religious attendance does not necessarily indicate inclusiveness. Park and Smith (2000) observed community insularity among the committed and religiously active group of churchgoing Protestants who tend to attend church-related volunteering but are less likely to attend non-church-related volunteering (Park and Smith 2000, p. 283). Further, studies show that the social network derived from religious involvement often divides the community as a whole. Putnam (2000) reported that religious tradition and norms shared among religious communities can lead to intolerance or ignorance toward differences or outsiders (Putnam 2000). As a more extreme example, Heuser (2005) mentions that some ultraconservative and/or fundamentalist religious communities seek isolation and social dominance. The author explains that they typically focus their concerns and resources inward, detaching themselves from participation in other organizations or civic society as a whole.

Secondly, we suggest to consider including an additional indicator to measure bridging social capital: The percentage of the veteran population using American Community Survey data set from the US Census. There are studies that implies the association between military service experiences and bridging social capital. For example, Patulny et al. (2015) explored how military service experiences affect bonding and bridging social capital, and found a significantly positive relationship between war service and bridging social capital. It reveals that while many veterans may have mental health issues [i.e., post-traumatic stress disorder (PTSD)] that hinder the formation of bonding social capital, they showed surprisingly robust social bonds via symbolic attachment to nationalism and the broader society, increasing bridging social capital (Patulny et al. 2015). Nesbit and Reingold (2011) suggested similar results from their investigation on the effect of military service on volunteering by demographic groups. The study shows that veterans are more likely to have volunteered in the past year than nonveterans, although there are some variations depending on their marital status or race (Nesbit and Reingold 2011). These studies explain that military socialization creates more opportunities for broader citizen engagement.

Thirdly, while political linkage is an important part of linking social capital, the variables for political linkage are insufficient, especially in the study of US context. For example, Kyne and Aldrich (2020) used the variable “% of total voting-age who are eligible for voting,” which serves as a proxy for the potential quantity of linking ties. As a more enhanced indicator that quantifies the linking ties, we propose the “% of total population that cast a vote in the presidential election,” which reflects the quantity of actual functioning linking ties, aiming to employ a more nuanced indicator. This will effectively reflect how well the community members understand and take advantage of social institution to express their needs through institutionalized channels.

Fourthly, linking social capital is not based on one-way networks. Existing indicators heavily focus on the capacity to reach out to community members, using variables such as the percentage of governmental employees per capita. However, for the governmental linkage to work properly, community members should interact with governmental employees or react to the governmental needs with favor. In this sense, variables such as the response rate of the Census survey from the US Census could be a good proxy that reflect the extent of institutional acceptance.

Finally, we suggest including an indicator of political plurality to more comprehensively capture linking social capital. Current linking social capital indicators focus on the number of personnel in governmental bodies, such as the percentage of local government employees in a community. These indicators assume that communities with more personnel in governmental bodies would have more vertical networks with community entities. While we agree that local authorities are important elements of linking social capital as they usually have access to variety resources, it is doubtful whether those linking ties are necessarily responsive regardless of the context. Rubin (2016) theorizes that a lack of institutionalized process of accountability produces unresponsive linking ties because local authorities operate in the context of complex political dynamics. Particularly during times of urgent need caused by natural disasters, resources are often allocated based on political logic or power relations, rather than actual need, as it is difficult for local authorities to meticulously review and distribute available resources according to established protocols in such a short time. A study on relief expenditure across US counties found that politically supportive constituencies are more likely to receive disaster relief from the governor (Healy and Malhtra 2009; Rubin 2016). Similarly, Reeves (2011) found that during the US presidential elections from 1981 to 2004, states that were highly competitive in terms of elections received twice as many disaster declarations from the president, even after accounting for the real impact of natural disasters (Rubin 2016).

On the other hand, a political environment that embraces pluralism and diversity ensures that local authorities are held accountable. Politics literature found that government responsiveness during times of crisis is stronger in more pluralistic political settings (Rubin 2016). In this sense, incorporating a proxy indicator of political pluralism based on political fractionalization within the linking social capital is needed to reflect the effectiveness of the existing linking ties.

To operationalize the political pluralism, we suggest computing the political fractionalization using a data set from the Election Data Science Lab, which provides county-level total votes to Democrats, Republicans, Green, and other parties for presidential election.

Conclusion

This article reviews quantitative measures of social capital at the community level, with the aim of promoting its use in disaster management. Historically, disaster management decisions have relied on physical infrastructure, overlooking the significant role social capital plays across all phases of disaster management (Meyer 2018). While physical infrastructure is effective, it often requires tremendous resources for construction and maintenance, making it unavailable for marginalized communities. However, social capital can be available regardless of resource availability once it is formed well enough to function effectively. Moreover, unlike other forms of capital, social capital primarily serves as a public good, creating a more equitable environment even for marginalized groups and possibly narrowing the gaps between advantaged and disadvantaged populations. Despite its unique roles that cannot be expected from other forms of capital, social capital has been underutilized in enhancing community disaster resilience due to the lack of standardized measures quantifying it. To address this issue, there have been a few attempts suggesting quantitative measures based on secondary data sets. This study suggests several points for improving those early attempts.

The first suggested improvement is to reclassify the current variables utilized to measure social capital related to religious activities. While previous studies classified such variables as bridging social capital, we argue that it is more appropriate to classify them as bonding social capital. The second improvement is for the current variables related to the political linkage of the federal government. We suggest using variables that reflect the number of individuals who participated in an important political event, such as presidential elections. Third, we suggest incorporating additional indicators, which are % veteran population, Census response rate, and political fractionalization.

These improvements to social capital indices will better capture and measure social capital in a more comprehensive way and have a more solid theoretical basis based on the social capital and disaster resilience literature. It is not merely a matter of how social capital is technically classified and measured, as each type has different consequences and effects on disaster planning processes and phases. Therefore, it is important to distinguish and measure social capital variables in a holistic way for policymakers and practitioners to make use of social capital information in preparing for disasters. Nevertheless, this study comes with several caveats. First, the social capital index heavily relies on aggregated proxy data, potentially overlooking many of the nuances at the individual level and neglecting vital interlocal interactions play a role in shaping social capital the local environment. Second, the suggested proxy indicators in this study lack empirical testing, rendering the study unable to ensure the effectiveness of these suggestions. Hence, further study is required to test the suggested proxy indicators using real-world data set. The tests may include internal consistency testing, incorporating validated sets of indicators as delineated by Kyne and Aldrich (2020) and Fraser (2021). Additionally, validation efforts should involve actual disaster damage amount data sets along with the established sets of indicators such as Cutter et al. (2003) Social Vulnerability Index (SoVI).