The Pennsylvania State University. The Graduate School. The Department of Sociology SOCIAL CAPITAL AND INEQUALITY IN POLITICAL PARTICIPATION

The Pennsylvania State University The Graduate School The Department of Sociology SOCIAL CAPITAL AND INEQUALITY IN POLITICAL PARTICIPATION A Dissertation in Sociology by Michael D. Stout 2008 Michael D. Stout Submitted in Partial Fulfillment of the Requirements for the Degree of Doctor of Philosophy May 2008

ii The dissertation of Michael D. Stout has been reviewed and approved by the following: Duane Alwin McCourtney Professor of Sociology, Demography, and Human Development and Family Studies Thesis Advisor Chair of Committee R.S. Oropesa Professor of Sociology and Demography Alan Sica Professor of Sociology Constance A. Flanagan Professor of Youth Civic Development John McCarthy Professor of Sociology Head, Department of Sociology & Crime, Law and Justice Signatures are on file at the Graduate School

iii ABSTRACT This study seeks to contribute to our understanding of the role that resources play in political participation. The fundamental question asked is, what role do structurally embedded resources, or social capital, play in mobilizing people to become politically active? This question is addressed in a series of steps. In the first step, a measurement model of social capital incorporating trust, social networks, and participation in voluntary associations is created. Next, structural equation modeling is used to determine the extent to which there is inequality in social capital by race/ethnicity and socioeconomic status. In the final step, a structural equation model where social capital acts to mediate the effects of race/ethnicity and SES on five kinds of political participation is tested. The five types of political participation are: (1) general political engagement; (2) attending political meetings and rallies; (3) participating in boycotts, demonstrations, or marches; (4) signing petitions; and (5) membership in political organizations. The results suggest that socioeconomic status is positively associated with all three social capital variables, and that race/ethnicity is negatively related to trust and network diversity. The results suggest that social capital is an important resource that facilitates a range of political activities in the United States. The results also support the conclusion that differences in levels of social capital have implications for (and contribute to) differences in a variety of forms of political participation.

iv TABLE OF CONTENTS LIST OF TABLES...vii LIST OF FIGURES...ix ACKNOWLEDGMENTS...x CHAPTER 1: RESOURCES AND POLITICAL PARTICIPATION...1 Introduction...1 Political Participation...2 Resources and Political Participation...3 Social Capital...5 Chapter Structure...7 CHAPTER 2: A THEORY OF SOCIAL CAPITAL...9 Introduction...9 Overview of Social Capital...11 The Social Capital Concept...12 Individual-Level Social Capital...13 The Homophily Principle...15 Collective Social Capital...16 The Relation of Trust to Social Capital...19 Thick Trust vs. Thin Trust...20 Trust and Action...21 The Importance of Trust...22 The Causal Link between Trust and Social Capital...24 Social Capital as a Resource for Individuals and Collectivities...27 A Definition of Social Capital...27 A Theory of Social Capital...28 Chapter Summary...41 CHAPTER 3: DATA AND MEASURES...43 Introduction...43 Social Capital Community Benchmark Survey 2000...43 Measures...45 Independent Variables...45 Social Capital Variables...46 Dependent Variables...50 Control Variables...51 Analytic Strategy...53 CHAPTER 4: A MEASURE OF SOCIAL CAPITAL...55 Introduction...55 Accessing Social Capital...56 Measuring Social Capital...58 Voluntary Association...59

v Social Networks...60 Social Trust...62 Analytic Strategy...62 Results...63 Measurement Model...63 Chapter Summary...66 CHAPTER 5: INEQUALITY AND SOCIAL CAPITAL...68 Introduction...68 Inequality and Social Capital...68 Analytic Strategy...70 Structural Model...71 Chapter Summary...76 CHAPTER 6: MEASURE OF POLITICAL PARTICIPATION...78 Introduction...78 Definition of Political Participation...78 Types of Political Participation...79 Distinguishing Forms of Political Participation...80 Measures of Political Participation...81 Analytic Strategy...85 Measurement Model...85 Chapter Summary...88 CHAPTER 7: A MODEL OF POLITICAL PARTICIPATION...90 Introduction...90 Socioeconomic Status and Political Participation...90 Hypotheses...92 Race/Ethnic Status and Political Participation...93 Hypotheses...95 Social Capital and Political Participation...98 Voluntary Association and Political Participation...98 Social Networks and Political Participation...101 Social Trust and Political Participation...103 Inequality in Social Trust...105 Hypotheses...106 Theoretical Model of Political Participation and Analytic Strategy...108 Results...112 Electoral Engagement...114 Attend Political Meeting or Rally...118 Signed Petition...123 Political Association Membership...126 Political Protest...130 Discussion...135 Conclusion...140

vi CHAPTER 8: CONCLUSION...141 Introduction...141 Summary of Findings...141 Inequality and Social Capital...141 Inequality in Political Participation...142 Implications for Political Participation in the United States...144 Future Research...146 BIBLIOGRAPHY...150

vii LIST OF TABLES Table 3.1. Descriptive Statistics for Control Variables by Social Status Variables...51 Table 4.1. Results of Measurement Model for Social Capital Variables...65 Table 5.1. Descriptive Statistics for Full Sample, and by Social Status...71 Table 5.2. Standardized SEM Path Coefficients of Social Status and Control Variables on Social Capital...74 Table 5.3. Unstandardized SEM Path Coefficients of Social Status and Control Variables on Social Capital (Standard Error in Parentheses)...75 Table 6.1. Results of Measurement Model for Political Participation Variables...88 Table 7.1. Weighted Sample Statistics of Political Participation by Race/Ethnicity and Socioeconomic Status...112 Table 7.2. Model Fit Statistics for the Seven Structural Equation Models...112 Table 7.3. Standardized Path Coefficients (Direct Effects) for Regression of Electoral Engagement on Social Status, Social Capital, and Control Variables...114 Table 7.4 Unstandardized Path Coefficients (Direct Effects) for Regression of Electoral Engagement on Social Status, Social Capital, and Control Variable...115 Table 7.5. Indirect, Direct, and Total Effects of Social Status and Social Trust on Electoral Engagement...117 Table 7.6. Standardized Path Coefficients (Direct Effects) for Regression of Attended Political Meeting or Rally on Social Status, Social Capital, and Control Variables...119 Table 7.7. Unstandardized Path Coefficients (Direct Effects) for Regression of Attended Political Meeting or Rally on Social Status, Social Capital, and Control Variables...119 Table 7.8. Indirect, Direct, and Total Effects of Social Status and Social Trust on Attended Political Meeting or Rally...122 Table 7.9 Standardized Path Coefficients (Direct Effects) for Regression of Signed Petition on Social Status, Social Capital, and Control Variables...123 Table 7.10 Unstandardized Path Coefficients (Direct Effects) for Regression of Signed Petition on Social Status, Social Capital, and Control Variables...124

viii Table 7.11 Indirect, Direct, and Total Effects of Social Status and Social Trust on Signed a Petition...126 Table 7.12 Standardized Path Coefficients (Direct Effects) for Regression of Membership in Voluntary Political Associations on Social Status, Social Capital, and Control Variables...127 Table 7.13 Standardized Path Coefficients (Direct Effects) for Regression of Membership in Voluntary Political Associations on Social Status, Social Capital, and Control Variables...128 Table 7.14 Indirect, Direct, and Total Effects of Social Status and Social Trust on Membership in Voluntary Political Associations...129 Table 7.15 Standardized Path Coefficients (Direct Effects) for Regression of Participated in Boycott, Demonstration, or March, on Social Status, Social Capital, and Control Variables...131 Table 7.16 Standardized Path Coefficients (Direct Effects) for Regression of Participated in Boycott, Demonstration, or March, on Social Status, Social Capital, and Control Variables...132 Table 7.17 Indirect, Direct, and Total Effects of Social Status and Social Trust on Participation in Political Protest...134

ix LIST OF FIGURES Figure 2.1 Theoretical Model of Social Capital...40 Table 3.1. Descriptive Statistics for Control Variables by Social Status Variables...51 Figure 4.1 Graphical Depiction of the Measurement Model...64 Figure 6.1. Graphical Depiction of Measurement Model of Political Participation...86 Figure 7.1 Theoretical Model of Political Participation...109

x ACKNOWLEDGMENTS Many people contributed to the development of the work presented in this dissertation. First and foremost I would like to thank Duane Alwin. I will forever be indebted to Duane for his patience, understanding, and uncanny ability to provide me with the wise advice at the most appropriate time. With his help and encouragement I was able to slog through some incredibly stressful life-events in order to get this dissertation done. Some of the best memories of my time at Penn State will be of our regular meetings at Otto s. I could not have asked for a better advisor, mentor, or friend. Thank You. The idea for this dissertation has been in development, in one form or another, ever since I took my first sociology class as an undergraduate sociology major at Temple University. There are two individuals who had a particularly strong impact on my understanding of the sociological imagination. I would like to thank Bill Yancey and Josh Freely. My collaboration with the two of them taught me valuable lessons about what it means to be a sociologist and the importance of conducting rigorous sociological research. I would also like to thank the members of my dissertation committee, in alphabetical order, Connie Flanagan, Sal Oropesa, and Alan Sica. All three provided me with valuable advice at each stage of the dissertation process. In addition to being on my committee, Connie Flanagan has contributed greatly to my interest in the link between social capital and civic engagement through our work on the Social Responsibility Project, and has provided me with the opportunity to develop my analytic skills. Serving as a Teaching Assistant for Sal Oropesa s research methods and statistics courses has

xi taught me the importance of learning to explain statistics in a down-to-earth manner, and has served as the basis for many stimulating conversations that have helped to strengthen this dissertation. My discussions with Alan Sica and my time as a student in his social theory classes have instilled within me a desire to take a balanced approach to sociological research, where quantitative methods serve as a means-to-an-end and not as an end-in-itself. I have been incredibly fortunate to have made many good friends in my time at Penn State, and each has, at various times, helped me through challenges I faced in graduate school and throughout the dissertation process. The conversations I ve had with them at bars and parties, and in the Oswald lab, proved to be an invaluable resource that has significantly contributed to my understanding of social capital and politics. In no particular order, I would like to thank Jacob Cheadle, Dan Hawkins, Jason Miller, Julie Pelton, Craig Wiernik, Paul Kendeffy, Jacob Hibel, and Paul-Phillipe Parè. I would also like to thank my family, which has been incredibly supportive of me, emotionally and financially, throughout my entire life. Without them I would not have had the opportunity to obtain an education and I would be a very different person than I am today. I would particularly like to thank my dad, Cos, and my stepmom, Mary. They have always been proud of what I ve accomplished and have always encouraged me to be the best person I can be. I would also like to thank my mom, Kathy, who passed away while I was writing the dissertation. She was always incredibly proud of me, and I wouldn t be where I am today without her. This dissertation is dedicated to her. Last but not least I would like to thank Sue Chu. I couldn t have asked for a more loving, thoughtful, funny, and supportive person to accompany me on this long, strange

xii trip. She s helped to keep me grounded and has always been there for me when I need her.

1 Introduction CHAPTER 1: RESOURCES AND POLITICAL PARTICIPATION If some citizens opt not to take part in politics because they prefer to spend their time and money in other ways or because they do not care about public affairs, then there is less reason for concern if their views do not receive proportionate attention from political elites. But if they do not take part because they lack the resources that facilitate political activity, then these departures from the norm of political equality may pose a more serious challenge to democracy (Verba et al 1993: 456). The above quote raises an important question regarding the relationship between representative democracy and political participation. The key question is, To what extent is political participation voluntary, and to what extent is political participation the result of inequality in the distribution of politically-relevant resources? Naturally, most people assume that individuals have the right to choose whether they will take the time to vote or to participate in an anti-war march, and that individuals can choose whether they will write a check to contribute to a political campaign or a political cause. However, this assumes that people are knowledgeable about opportunities to participate, have the free time and the expendable income that make participation possible, and have the skills that make participation worthwhile. After all, not everybody possesses equal amounts of the knowledge, skills, time, and money that are required for effective political participation, and not everybody is presented with opportunities to use those skills even when they do have them. This is why developing an understanding of political participation is key for understanding and addressing political, social, and economic inequalities in the United States. Such an understanding is important since the extent to which politically-relevant resources are stratified by social class or race/ethnicity plays a part in determining how

2 the benefits of political representation are distributed among the population. So, if particular social groups are less likely to vote because they lack the information and resources required to vote, then their preferences will not be addressed by politicians. It is often assumed that elected officials respond more to the preferences of voters than those of nonvoters. While this assumption often goes untested, at least one study has found that voter preferences predict the aggregate roll-call behavior of Senators while nonvoter preferences do not (Griffin and Newman 2005). It is logical to assume that this would carry over to other types of participation as well, such as participating in political protest, signing petitions, and participating in political organizations. The present study was undertaken in order to understand the role resources play in political participation. It is hypothesized that inequality in participation by race/ethnicity, and by social class are largely attributable to inequality in the resources required to participate. Specifically, this study proposes and tests a theoretical model where access to structurally embedded social resources (social capital) acts to mediate the effects of social class and race/ethnicity on political participation. Political Participation A good deal of scholarly attention has been paid to political participation in America. There have been a number of theoretical and qualitative studies regarding political participation and American democracy that date back at least as far as de Tocqueville s treatise on Democracy in America. Over the last several decades a significant body of empirical knowledge about American political participation has also emerged. Major contributions in the measurement of participation and on the causal

3 association between participation and representation have been made by Sidney Verba and his colleagues, who have dedicated a majority of their professional lives to asking and seeking answers to questions related to the inner-workings of American participatory democracy. The book Voices and Equality (1995) is the culmination of more than 30 years of their work on this subject. The analyses presented in Voices and Equality reveal an unprecedented amount of empirical information about the nature of political participation in the United States. They develop a resource model of political participation that goes beyond traditional SES and rational-choice models of behavior. Contrary to the expectations of rational-choice theories, they report that political activists tend to cite many types of gratification that result from their political behavior, and that the social networks and friendships that develop as a result of participation also play a significant role in the motivation to participate. They found significant differences in certain types of participation by race and social class, and most importantly for the study presented here, they found that resources generated from non-political activity, such as the communication and organizational skills obtained from participation in voluntary associations, are useful for political participation. Resources and Political Participation Verba and his colleagues observed that politically-relevant resources, which vary by social class and race/ethnicity, act to facilitate political participation. The most important of these resources are time, money and civic skills. Civic skills are defined as organizational and communication skills that are useful for most types of political

4 participation. The authors note that people have opportunities to develop their politicallyrelevant civic skills in nonpolitical contexts, through the demands of their jobs, for example, or through participation in nonpolitical voluntary associations. The role played by participation in nonpolitical voluntary associations is particularly relevant for this study. Verba and his colleagues note that participation in nonpolitical voluntary associations facilitates political participation in three ways. First, participation in nonpolitical voluntary associations provides opportunities for people to develop politically-relevant organizational and communication skills. Second, participation in nonpolitical voluntary associations exposes people to political messages and political discussions that lead to political participation. Third, nonpolitical voluntary associations serve as recruiting grounds for attempts at political mobilization. According to Verba et al (1993: 457), Not only do those who lead them make explicit attempts to get people involved in politics but at work, in church and in voluntary associations people develop interpersonal networks that serve to bring them into politics. The role played by nonpolitical voluntary associations for exposing people to political messages and for recruiting them into political activity are both consistent with research on the role social networks play in political participation. Several studies, which will be discussed in more detail in later chapters, have found that the characteristics of people s social networks also play a role in determining their likelihood of participating in politics. Verba and his colleagues, for instance, observed that when recruitment for political activity occurs through interpersonal networks, those being solicited are more likely to say yes than when strangers make the request; while other studies have found

5 that diverse friendship networks increased people s likelihood of participating in one or more politically-oriented activities (e.g., Kotler-Berkowitz 2005). The observation that participation in nonpolitical voluntary association influences political participation, and that social networks influence political participation fit nicely with the logic behind the concept of social capital. The concept of social capital emphasizes the role played by norms of trust and reciprocity, social networks, and hierarchical organizations in the accessibility of structurally embedded resources; and how these resources provide social, political, and economic returns for actors. Social Capital Robert Putnam has linked social capital to civic engagement and political participation (1993; 1995; 2000). He observed a strong empirical relationship between the presence of voluntary associations in communities and the quality of life in those communities. Participation in voluntary associations, he argues, results in individual and collective benefits. He claims that these individual and collective benefits are the direct result of stocks of social capital, which he defines as social networks and the norms of reciprocity and trustworthiness that arise from them (p. 19). He found that communities with high levels of social capital had political structures that functioned efficiently, and political elites that were responsive to the needs of their citizens. He also found that the residents of high social capital communities showed higher levels of civic engagement and participation in politics. Social capital is a relational concept that is a characteristic of modern social structures. It functions at the micro-, meso-, and macro-levels of society (Lin 2001). It

6 provides people with the motivation to act at the micro-level, is a property of social networks at the meso-level, and is a property of hierarchical social organizations at the macro-level (Lin 2001). There have been several problems with the ways that social capital has been used in empirical research. One of these problems is that it is either discussed as a property of individuals, or as a property of collectivities (Portes 1998). Another problem is that social capital has been defined in many ways, and has thus been operationalized in many ways. It has come to represent everything and anything in social life, which has limited its usefulness as a sociological concept. Many of the controversies and problems associated with social capital result from the fact that the theoretical development of the concept has consistently lagged behind its use in empirical research. In this study social capital is defined as the resources embedded in social networks that are accessed and used by actors in social actions; social actions are most likely to occur when norms of trust (mutual expectations that actions taken will be reciprocally beneficial) are present. This definition is based on a network theory of social capital, developed by Nan Lin (2001), which addresses many of the problems associated with prior conceptualizations of social capital. A multidimensional measure of social capital that takes into account trust, networks, and voluntary association is developed and utilized in this study. This measure is then used to examine the extent to which social capital is a useful resource for political participation.

7 Chapter Structure Chapter 2 provides an overview of the literature on social capital and presents a theory of social capital, developed by Nan Lin, which is used as the basis for the construction of the empirical measure used in later chapters. Lin s theory of social capital revolves around social networks and incorporates micro-, meso-, and macro-level processes. The key contribution of this chapter is that it provides a synthesis of the conceptual development of social capital as a characteristic of communities and individuals. Lin s network theory of social capital makes it possible to develop empirical measures of social capital that are relational in nature and that encompass characteristics of multiple levels of structure. As was mentioned above, the theoretical framework for social capital has developed at a slower rate than the rate at which the concept has been used for empirical analysis, which has had deleterious effects on the usefulness of social capital as a relevant concept for social scientific analyses. Chapter 3 presents a description of the data used, information on the measures used, and the analytic strategy used for the analyses throughout the study. The data for the study were derived from the Social Capital Community Benchmark Survey 2000. Chapter 4 presents a measurement model of social capital, based on the theory presented in Chapter 2. The key contribution of this chapter is that it provides a measure of social capital that incorporates the main theoretical propositions of Lin s theory into a multidimensional measure of social capital consisting of social trust, network diversity, and participation in voluntary associations. This measure is used in the empirical analyses of later chapters.

8 Chapter 5 examines inequality in social capital by socioeconomic status and race/ethnic status. This chapter is important since a key argument made throughout the present study is that inequality in social capital leads to inequality in political participation. This chapter thus examines this assumption in order to see the extent to which social capital varies according to race/ethnic characteristics and socioeconomic status. Chapter 6 provides a definition of political participation and discusses how types of political activity are distinguishable from one another based on three criteria: (1) the resources required to participate; (2) the capacity the act has to convey information to elected officials; and (3) whether the volume of a political activity holds the capacity to exert pressure on elected officials. Chapter 6 then constructs a measurement model of political participation in which types of political activity are distinguished according to those criteria. Chapter 7 proposes a theoretical model in which the effects of socioeconomic status and race/ethnic status on political participation are mediated by a multidimensional conceptualization of social capital. Building upon the theoretical developments given Chapter 2 and 6, and the measurement specifications presented in Chapters 3 and 6, Chapter 7 empirically assesses the mediation hypotheses and summarizes the results. Chapter 8 summarizes the results of the study, identifies the key contributions of the study, and provides some ideas for directions of future research.

9 CHAPTER 2- A THEORY OF SOCIAL CAPITAL The distinction between voluntary activity and paid work is not always clear. It is possible to serve private economic purposes through social and political activism many people seek to do well while doing good. They undertake voluntary activity for which they receive no compensation in their churches, in charities, in politics in order to make contacts or otherwise enhance their jobs or careers. Furthermore, for many of those who participate in politics, the policy issues that animate their activity have consequences for their pocketbooks. Conversely, many people get involved in genuinely voluntary activity that is an extension of their paid employment. For example, an accountant may lend his or her professional expertise as part of unpaid service on a hospital or museum board. Those who work for nonprofits or political organizations often extend their commitment with additional volunteer work on behalf of the objectives pursued through their paid employment. In all these cases, the border between voluntary participation and paid employment is blurry (Verba, Schlozman, & Brady 1995: 39). Introduction The above quote describes how economic and social resources provide both the means and the motivation for individuals to voluntarily engage in political activity. In Voices and Equality (1995), Sidney Verba and his colleagues describe the characteristics of politically active American citizens. They report that activists, on average, possess higher levels of politically-useful resources than non-activists do. Activists have better economic resources, including higher incomes, more formal education, and more prestigious occupations than non-activists. Their higher incomes provide them with extra money to contribute to political causes, their formal education provides them with politically-relevant knowledge and civic skills, and their jobs provide them with politically-useful social networks and contribute to the development of their civic skills. Verba and his colleagues also report that activists possess better social resources than non-activists. They are more likely to belong to and to voluntarily participate in

10 nonpolitical organizations, and are more likely to serve in leadership roles within those organizations. Participation in voluntary associations contributes to the development of useful civic skills, establishes politically useful social networks, and increases opportunities for recruitment into political activity. While participation in voluntary nonpolitical associations may not be overtly political, Verba et al argue, it has important implications for the effectiveness of political action, and for the likelihood of engaging in political behavior. The fundamental premise of Voices and Equality is that people invest all of the resources at their disposal, i.e. their economic capital, their human capital, and their social capital (which includes cultural capital), in order to influence the political decisionmaking process. Socioeconomic inequality at both the individual and the group level therefore determine, in part, whose preferences are represented by elected officials. The authors argue, for example, that as political campaigns have become increasingly expensive, the use of economic resources, in the form of campaign contributions, has become an increasingly effective means of political expression. In addition, social resources, such as access to a diverse set of social networks and participation in voluntary associations, increase the likelihood of participating in politics, and increase the effectiveness of political participation. Thus, as the saying goes, it s not what you know, but who you know that is determines the effectiveness of political engagement. The research conducted for the present study seeks to shed light on the process through which economic, human, and social resources increase the likelihood of participating in several types of political behavior: (1) general political engagement; (2)

11 attending political meetings and rallies; (3) participating in boycotts, demonstrations, or marches; (4) signing petitions; and (5) membership in political organizations. It is hypothesized that socioeconomic status (human capital and economic capital), measured as race/ethnicity and social class, directly affect an actor s levels of social resources (their social capital), measured as diversity, or heterogeneity, of social networks, participation in nonpolitical voluntary associations, and social trust. Their social resources, in turn, influence their propensity to participate in politics. In other words, it is hypothesized that social capital mediates the effects of socioeconomic status on political participation. Following the logic of Voices and Equality, a resource-based model for political participation is tested. It is argued that social capital is a resource that plays an important role in shaping how socioeconomic inequality leads to inequality in political participation, and therefore has consequences for inequality in political representation. In this chapter, the concept of social capital is explicated and a theoretical framework is presented, based on the notion of social capital as a resource that motivates action, which is utilized for political participation. Overview of Social Capital Over the past two decades, the term social capital has become a staple of the social scientific lexicon. Perceptions that social capital is essentially a positive thing, and that the underlying causes of a variety of social problems are the result of social capital deficits, have become so ubiquitous that the interest in and debate over social capital has spread beyond the ivory tower of academia into policy circles, and has even grabbed the

12 attention of the mainstream media 1. To scholars and academics who advocate its use in models of human behavior, social capital adds to our understanding of the complex interrelationships between human behavior and a wide range of social phenomena. To scholars and academics critical of social capital, it is a poorly conceptualized, poorly measured, poorly understood, and overemphasized fad, which makes up in publicity and hype what it lacks in substance. For policy analysts and politicians who perceive social capital as a community resource, the debate is focused on whether public policy can effectively increase social capital and, if so, how best to do so in specific communities across the country, and throughout the nation at large. For the purposes of this study, a model of political participation is proposed where social capital mediates, or partially mediates, the effects of social status on the five types of political participation mentioned above. The Social Capital Concept The term social capital has occasionally appeared in scholarly work since at least 1916 2, its popularity in contemporary social science has been due to independent work by scholars such as Pierre Bourdieu, James Coleman, Nan Lin, Alejandro Portes, and Robert Putnam. In its current formulation, the concept of social capital has been elaborated through two distinct, but related, perspectives: (1) as an asset of communities or larger social structures, such as nations, and (2) as an asset of individuals. The former focuses on social capital as a resource that benefits groups, communities, and even nations (e.g., 1 The media attention is due, in large part, to the release of Robert Putnam s book Bowing Alone in 2000, which became a national bestseller in the United States. 2 Putnam (2000) traces the first use of the term back to L.J. Hanifan, state supervisor of rural schools in West Virginia, in 1916.

13 Putnam 2000). The latter focuses on social capital s use by individuals, and how they access and utilize resources embedded in social networks to gain returns in instrumental actions (e.g., finding a higher-status job), or to preserve gains in expressive actions (Lin 2001) 3. Individual-level Social Capital: Networks and Embedded Resources At the individual-level, social capital is like human capital in that it is assumed that its investment can be made by the individual with an expected return (or profit) to the individual 4. The two focal points for analyses of social capital at the individual level are: (1) how individuals invest in social relations, and (2) how individuals capture resources embedded in these relations to generate a return (Lin 2001) 5. Individual-level conceptualizations of social capital have been discussed by a number of scholars whose expertise lies in the study of social networks. In this literature, an individual s location, or position, in a social network is associated with their level of access to embedded resources, and with the quality of resources they have access to. For example, Flap (1991) specifies three types of social capital: (1) the number of people within one s social network who are prepared or obliged to help you when called upon to do so; (2) the strength of the relationship, which indicates a willingness to help; and (3) the resources of the people in the network. To Flap, social capital consists of 3 There has been some controversy surrounding whether social capital is a property of individuals or of communities (see Portes 1998). Generally, scholars have reached a consensus that it is both (Lin 2001). 4 Aggregation of individual returns on social capital can carry external benefits for the collectivity (Putnam 2000). 5 There are two types of such resources that an individual can gain access to and use: personal resources; and social resources (Lin 1982). Personal resources are those that are possessed by an individual, and can include material or symbolic goods. Social resources are those that are accessed through an individual s social ties. The extent to which individuals have access to different social resources is dependent on the extensity and diversity of their social ties.

14 resources provided through strong social relationships with others; it is the product of the availability of social resources and the likelihood that others in the network will offer such resources for help. Burt s (1992) theory of structural holes, follows a similar vein. It draws generally on sociological ideas put forth by Simmel (1955[1922]) and Merton (1968[1957]) on the agency generated through conflicting structural affiliations, and on the economic concepts of monopoly power and oligopoly. Burt combined these general principles with sociological principles drawn from the literature on social networks from the 1970s, particularly the work of Granovetter (1973) on the strength-of-weak-ties, Freeman (1977) on betweenness centrality, and Cook and Emerson (1978) on the benefits of exclusive exchange partners, to produce a network model of competitive advantage. Burt s formulation of structural holes posits that weaker connections between groups represent holes in the social structure of the market. These structural holes create a competitive advantage for individuals whose networks span these holes. Structural holes serve as a buffer, where people on either side possess different resources and information. As a result, nodes 6 that are situated near a hole have the opportunity to control the flow of information and contact between people of different groups. Thus, individuals with contact networks rich in structural holes are the individuals who know about, have a hand in, and exercise control over, more rewarding opportunities (Burt 2001). The Homophily Principle 6 In the social network literature, individuals are generally represented by nodes, or points, in the network, and the ties that link the nodes, their networks, are represented by the lines that connect the nodes.

15 A defining characteristic of social networks is that, generally, birds of a feather flock together. In the network literature, this is referred to as the homophily principle. The homophily principle states that people s personal networks (e.g., their primary groups) are structured in such a way that they are homogeneous regarding socioeconomic, demographic, and intrapersonal characteristics. McPherson, Smith- Lovin, and Cook (2001) have found that homophilous interactions limit the quality and amount of access to information and other embedded resources actors have, and that homophily in race and ethnicity create the strongest divides in our personal environments, followed by homophily in age, religion, education, occupation, and gender. Lin (2001) posits that homophilous interactions are normative, and that heterophilous interactions are abnormal. Additionally, homophilous interactions tend to be expressive, or conducive to satisfying the maintenance of psychological and social resources, while heterophiloius interactions are instrumental, or conducive to obtaining additional social resources. Summary: Individual-Level Social Capital For individuals, then, the benefits of social capital are largely determined by network position (location in a network relative to a structural hole), by the resources accessible through their network position ( embedded resources ), and by the nature of interactions (expressive or instrumental) that take place between actors within (homophilous) and across (heterophilous) networks.

16 Collective Social Capital For collectivities, the social capital discussion generally revolves around two perspectives: (1) how certain groups develop and maintain social capital as a collective resource, and (2) how collective resources enhance group members life chances (Lin 2001). Bourdieu (1986), and Coleman (1988, 1990) have constructed formulations of this perspective in their work, and Putnam (1993, 2000) has tested these formulations in his empirical work on social capital and civic engagement. Bourdieu (1986) distinguishes among three types of capital: (1) economic, (2) social, and (3) cultural. He defines social capital as being made up of social obligations or connections. It is the aggregation of actual or potential resources which are linked to possession of a durable network of institutionalized relationships of mutual acquaintance and recognition or in other words to membership in a group (p.248). Groups provide their members with collectively-owned capital, which grants the members credit. In this form capital is represented by the size of the network and the volume of the capital (economic, cultural, or symbolic) possessed by those to whom a person is connected. Thus, social capital for Bourdieu is a collective resource shared by the members of a defined group that has clear boundaries, and obligations of mutual exchange and commitment. To Bourdieu, the emergence of social capital is produced by the group s members. Repeated exchanges among members of the group establish and reinforce mutual recognition and boundaries (trust) in order to affirm and reaffirm the collectivity

17 of the capital, and each member s claim to the capital. Bourdieu also views social capital as a front for economic capital. He states, for example, that economic capital is at the root of all other types of capital, and that every type of capital is reducible in the last analysis to economic capital (pp. 252-253). For Bourdieu, then, social capital is a form of capital possessed by members of a group. Through the social networks of group members, the capital can be used by members as credits. Thus, social capital is a collective resource that provides credits to group members, and is maintained and reinforced as members continue to invest in their relationships. Coleman (1988; 1990) views social capital as a key to linking the sociological principle that individual action takes place within a social context with the economic principle that wholly self-interested actors act on the basis of utility maximization. His formulation of social capital accepts the economic principle of rational action yet attempts to show how that principle, in conjunction with particular social contexts, can account not only for the actions of individuals in particular contexts but also for the development of social organization (p. S96). Based on Granovetter s (1973) account of the strength-of-weak-ties, Coleman argues that neoclassical and neoinstitutional economic theories have failed to take into account that economic actors are embedded in a web of concrete personal relations and networks of relations that play a significant role in generating trust, in establishing expectations, and in creating and enforcing norms (p. S97). In defining social capital, Coleman (1988) posits that it is a particular kind of resource available to an actor (p. S98). His formal definition of social capital states that:

18 Social capital is defined by its function 7. It is not a single entity, but a variety of different entities, with two elements in common: they all consist of some aspect of social structures, and they facilitate certain actions of actors whether persons or corporate actors within the structure. Like other forms of capital, social capital is productive, making possible the achievement of certain ends that in its absence would not be possible. Like physical and human capital, social capital is not completely fungible but may be specific to certain activities. A given form of social capital that is valuable in facilitating certain actions may be useless or even harmful for others (p. S98; emphasis added). To Coleman, social capital is a resource embedded in the structure of relations between and among individual or corporate actors. Thus, it is social, as it is not a resource belonging to the actors themselves, and it is not represented by the physical tools of production. Putnam (2000) bases his formulation of social capital on Coleman s. He defines social capital as social networks and the norms of reciprocity and trustworthiness that arise from them (p. 19; emphasis added). He notes that social capital benefits both individuals and communities. Individuals benefit from social capital, for instance, when they use their social networks to obtain a job. For communities, social capital has externalities that impact the wider community. Social capital is therefore both a private good and a public good. According to Putnam, for example, service clubs, like Rotary or Lions, mobilize local energies to raise scholarships or fight disease at the same time that they provide members with friendships and business connections that pay off personally (p.20). In Bowing Alone (2000) and in Making Democracy Work (1993), Putnam posits that the number of voluntary social organizations and the prevalence of membership and 7 Some controversy has surrounded the tautological implications of Coleman s contention that social capital should be defined by its function. Lin (2001) suggests that the social capital should be operationalized independently from its effects.

19 participation in these organizations are indicators of levels of social capital in democratic societies. Membership and participation in voluntary associations promote and enhance collective norms of trust and reciprocity, which are essential for the collective well-being (Putnam 1993; 1995; 2000). The collective view of social capital, then, emphasizes how social capital is a unique characteristic of social structures that confer benefits upon groups and their members. The Relation of Trust to Social Capital A recurring theme throughout the literature on social capital is the important role played by trust. Coleman, Putnam, and others explain that trust is a key component of the structure of relations that is so important for generating social capital. Specifically, they argue that social capital is most effectively utilized when social interactions take place in an environment characterized by mutual trust and norms of reciprocity (Coleman 1988; Putnam 2000). For instance, Putnam (2000) defines social capital as social networks and the norms of reciprocity and trustworthiness that arise from them (p. 19). Coleman posits that social relations are most likely to generate desirable outcomes (capital) for individuals when they are embedded in a social context characterized by norms of trust, mutual obligation and expectations of reciprocity. Thus, in the social capital literature, trust is viewed as a resource that promotes sociability among members of a community or social network. Trust also promotes sociability across group boundaries and encourages the establishment of weak ties and the creation of network bridges. As such, trust is a

20 multifaceted concept that is closely linked to the social environment within which social interaction takes place. Trust has been described as a belief that the consequences of someone else s actions will be acceptable from our point of view (Misztal 1996). Luhmann (1979) defines trust as the subjective reduction of complexity or uncertainty. In this sense it is a coping mechanism that allows individuals to adapt to the complex and uncertain conditions that increasingly characterize modern societies (Lewis and Weigert 1985). Thick Trust vs. Thin Trust In The Moral Foundations of Trust, Uslaner (2002) distinguishes between moralistic trust and strategic trust. Moralistic trust is the faith we place in strangers. Strategic trust is confidence in the people we know. Strategic trust is based on experience with particular individuals, while moralistic trust is based on collectively shared values, beliefs, norms, and expectations. Moralistic trust is indicative of a positive outlook, or an optimistic world-view; it is the sense that we can collectively alter the nature of our social world for the better. This type of trust is similar to what most scholars refer to as social trust. Uslaner notes that personal experience has little or no effect on this kind of trust. Social trust reflects the belief that there are underlying values common to all members of a society; that we are connected to people who are different from ourselves; and that we have a moral responsibility for their fate (Uslaner 2002). Following a similar logic, Putnam (2000) distinguishes between thick trust and thin trust. Thick trust is embedded in personal relations that are strong, frequent, and nested in wider networks (136). Thin trust, rests implicitly on some background of

21 shared social networks and expectations of reciprocity (136). He argues that thin (social) trust makes it possible for citizens to work together to solve community-level problems in modern societies, and that it encourages civic engagement. Social trust is a concept that is closely related to social capital. Social capital is defined as the resources embedded in social networks accessed and used by actors for actions (Lin 2001:25). Social trust is defined as the mutually shared expectation, often expressed as confidence, that people will manifest sensible and, when needed, reciprocally beneficial behavior in their interactions with others (Welch et al 2005:497). Trust and Action As a basis for action, individuals choose to act or react toward others depending on the amount of trust they have placed in them or the level of trustworthiness the others are perceived to have generated (Hardin 2002). The extent to which interactions are successful for maintaining or obtaining additional resources (e.g., social capital) is largely determined by the presence of trust, and the extent to which networks are embedded in a trusting social environment. For expressive actions, thick trust helps to maintain social and psychological resources (Putnam 2000). For instrumental actions, those motivated by the desire to obtain different or better resources, thin trust is important (Putnam 2000). Thin trust, or social trust, encourages the formation of weak ties and diverse social networks. Weak ties make it possible for people to interact with others who are located in different social-structural positions and social circles, and who possess different resources (Granovetter 1973). According to Uslaner (2002:9), actors who take