UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

From participation to social cohesion : an analysis of variation in the development of social capital… Legun, Katharine 2008

Your browser doesn't seem to have a PDF viewer, please download the PDF to view this item.

Item Metadata


24-ubc_2008_fall_legun_katharine.pdf [ 2.07MB ]
JSON: 24-1.0066516.json
JSON-LD: 24-1.0066516-ld.json
RDF/XML (Pretty): 24-1.0066516-rdf.xml
RDF/JSON: 24-1.0066516-rdf.json
Turtle: 24-1.0066516-turtle.txt
N-Triples: 24-1.0066516-rdf-ntriples.txt
Original Record: 24-1.0066516-source.json
Full Text

Full Text

 FROM PARTICIPATION TO SOCIAL COHESION: AN ANALYSIS OF VARIATION IN THE DEVELOPMENT OF SOCIAL CAPITAL IN COASTAL BRITISH COLUMBIA by KATHARINE LEGUN  A THESIS SUBMITTED IN PARTICAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF MASTER OF ARTS in THE FACULTY OF GRADUATE STUDIES (Sociology)  The University of British Columbia (Vancouver)  August 2008  ©Katharine Legun 2008   ii ABSTRACT  Social capital refers to the relationships between people that are productive: it can provide people with access to resources, ease transactions, and facilitate social and economic development at the community level. It has been conceptualized as both associations between people and attitudes of trust and cooperation that enable ties to be productive. Within communities, these attitudes underlie social cohesion, which can be defined as social integration and a propensity to cooperate and contribute to the community. Moreover, it is interaction and social engagement that develops social capital by creating and maintaining relationships and fostering social cohesion. This thesis presents an analysis of the development of social capital in coastal British Columbia by considering how the social participation of community members generates socially cohesive attitudes. Moreover, I empirically consider how this relationship varies for different people in different places and across two different types of participation. Formal participation refers to engagement in structured and organized group activities, such as rotary clubs or sports teams, while informal activities are casual irregular and often spontaneous, such as visiting with friends. Using a series of multiple linear regressions on survey data from rural coastal communities in British Columbia, I test how the relationship between these two types of participation and social cohesion varies according to people’s socio-demographic characteristics or the communities in which they live. Not only does this research consider who develops social capital in this way, but also whether the relationship between participation and social cohesion differ along these social lines. The results show that processes of social capital development reflect the characteristics and social environments of community members in coastal British Columbia.  The variability shows that social capital development is embedded within particular contexts in ways can lead to inequalities in social capital.  iii TABLE OF CONTENTS ABSTRACT.................................................................................................................................... ii LIST OF TABLES.......................................................................................................................... v LIST OF FIGURES ....................................................................................................................... vi LIST OF BOXES.......................................................................................................................... vii ACKNOWLEDGEMENTS......................................................................................................... viii CHAPTER 1: INTRODUCTION................................................................................................... 1 Introduction................................................................................................................................. 1 Literature Review........................................................................................................................ 5 Defining Social Capital: From Participation to Social Cohesion ........................................... 6 Social Cohesion – Definition and Operationalization .......................................................... 11 Participation – Definition and Operationalization ................................................................ 14 Relationships between Participation and Social Cohesion ................................................... 16 Socio-demographic Factors and Social Inequality ............................................................... 20 Community ........................................................................................................................... 22 Summary ................................................................................................................................... 23 CHAPTER 2: DATA AND METHODS ...................................................................................... 24 Introduction............................................................................................................................... 24 Context...................................................................................................................................... 25 The Sample and Distributions................................................................................................... 25 Participation .......................................................................................................................... 25 Socio-Demographic Factors.................................................................................................. 28 Place...................................................................................................................................... 28 Analytic Strategy ...................................................................................................................... 29 Social Cohesion: Factor Analysis ......................................................................................... 29 Binary Relationships and Missing Data................................................................................ 31 Multiple Linear Regression................................................................................................... 32 Interaction Effects................................................................................................................. 33 Summary ................................................................................................................................... 35 CHAPTER 3: RESULTS.............................................................................................................. 36 Social Cohesion: Factor Analysis ............................................................................................. 36 Exploratory Work and Missing Cases ...................................................................................... 37 Regression Results .................................................................................................................... 40 Informal Participation and Leadership.................................................................................. 40 Formal Participation and Leadership .................................................................................... 41 Informal Participation and Social Cohesion ......................................................................... 41 Formal Participation and Social Cohesion............................................................................ 42 Interaction effects...................................................................................................................... 47 Socio-Demographic Factors.................................................................................................. 47 Community Factors............................................................................................................... 51 Summary ................................................................................................................................... 54 CHAPTER 4: ANALYSIS AND DISCUSSION ......................................................................... 57 Introduction............................................................................................................................... 57 From Participation to Social Cohesion: How can Social Capital be Developed? .................... 57  iv Informal Participation ........................................................................................................... 58 Formal Participation.............................................................................................................. 61 Who is Developing Social Capital? Socio-Demographic Factors ............................................ 66 Age........................................................................................................................................ 66 Gender................................................................................................................................... 69 Household Income and Education ........................................................................................ 70 Ethnicity................................................................................................................................ 74 The Importance of Community................................................................................................. 76 Interactions: The Effects of Participation Vary ........................................................................ 79 Socio-Demographic Factors.................................................................................................. 80 What Does this Tell Us About Social Capital in Coastal British Columbia?........................... 87 How to Build a Cohesive Community .................................................................................. 87 Capital and Return Deficits................................................................................................... 89 Place and Social Capital........................................................................................................ 90 CHAPTER 5: CONCLUSION ..................................................................................................... 92 Insights, Applications and Directions for Future Research ...................................................... 92 Schmoozers and Machers ..................................................................................................... 94 Broader Contexts: Restructuring........................................................................................... 95 Aboriginal People in Coastal British Columbia.................................................................... 98 Inequality in Social Capital................................................................................................... 99 Applications: Building foundations for Social Capital Development .................................... 100 Encouraging the Conditions for Social Capital Development............................................ 100 Education as an Avenue to Social Assets ........................................................................... 100 Community as a Conductor of Schmoozing ....................................................................... 101 Targeting Missing Links (Not Everyone is Going to Get a Degree) ...................................... 102 Future Research ...................................................................................................................... 103 Summary ................................................................................................................................. 104 BIBLIOGRAPHY....................................................................................................................... 106 APPENDIX 1: SURVEY QUESTIONS FOR SOCIO-DEMOGRAPHIC VARIABLES ........ 111 APPENDIX 2: ETHICS CERTIFICATE ................................................................................... 114   v LIST OF TABLES  Table 1: Distributions……………………………………………………………………….. .28 Table 2: Missing Cases…………………………………………………………………….... .32 Table 3: Social Cohesion Scales……………………………………………………………...37 Table 4: Social Cohesion Distributions……………………………………………….…..…. 37 Table 5: Bivariate Relationships………………………………………………………….…..39 Table 6: Informal Participation and Attitudes Towards Leaders……………………….….....43 Table 7: Formal Participation and Attitudes Towards Leaders………………………… ........44 Table 8: Informal Participation and General Social Cohesion……………………….……… 45 Table 9: Formal Participation and General Social Cohesion…………………………………46  vi LIST OF FIGURES  Figure 1: Development of Social Capital from Micro to Macro………………………….…. 10 Figure 2: Social Capital in Operation: Participation and Social Cohesion…………………...10 Figure 3: Informal Participation and Attitudes towards Leaders by Education…………........48 Figure 4: Formal Participation and Attitudes toward leaders by Marital Status………….......49 Figure 5: Formal Participation and Attitudes toward leaders by Workstatus…………….......49 Figure 6:  Informal Participation and Social Cohesion by Ethnicity……………………….... 50 Figure 7: Formal Participation and Social Cohesion by Ethnicity…………………………... 50 Figure 8: Formal Participation and Attitudes towards Leaders by Community…………… ...52 Figure 9: Informal Participation and Social Cohesion by Community………………….........53 Figure 10: Informal Participation and Social Cohesion by Community……………………...54 Figure 11: Informal Participation and Social Cohesion by Community……………………...54 Figure 12: Age and Informal Activities……………………………………………………....67 Figure 13: Age and Social Cohesion……………………………………….………………... 67   vii LIST OF BOXES Box 1: Formal Participation Survey Questions…….……………………………………… ...26 Box 2: Informal Participation Survey Questions…........……………………………………..27 Box 3: Sense of Isolation…….……………………………………………………………. ....29 Box 4: Social Cohesion Likert Scale Questions…..………………………………….…… ....30  viii ACKNOWLEDGEMENTS   I would like to thank Dr. Ralph Matthews, whose comments and insights have been invaluable to me while I formulated the ideas for this research and developed them into a thesis. I am also extremely grateful for his mentorship throughout graduate school, and am truly indebted to him for the skills, knowledge and expertise his supervision has provided.  I would also like to thank Dr. Gerry Veenstra for inspiring my interest in statistics and coaching me through their application in this research. His clear and perceptive comments on drafts and in conversation have been so helpful in the development of this thesis and have made the experience so much more educational and enjoyable.   1 CHAPTER 1: INTRODUCTION Introduction  The more we get together, the better our community. Recent literature on social capital has repeated this basic idea many times over, tying civic participation and social ties to a variety of community assets including trust, cooperation, political efficacy and economic development (Putnam, Leonardi, and Nanetti 1993; Putnam 2000). Such research has initiated an onslaught of academic and governmental interest. The World Bank (Social Capital Initiative 1998) has referred to social capital as the “missing link” in development, while other authors have referred to social capital as a “master key” to development (Kliksberg 1999).   The basic premise for these arguments is that social capital is the product of participation and social ties at the individual level embedded in broader systems of social cohesion that engender cooperation and exchange. This thesis will add depth to the simple notion that participation generates social cohesion by delineating the relationships between different types of participation and social cohesion, and by considering how those relationships vary for different groups of people.  I will consider whether participation invariably coincides with social cohesion, or if they represent a more dynamic process of social capital development that reflects the circumstances in which they are embedded. Social cohesion, in this study, refers to a propensity to cooperate in a community, and includes concepts of trust, sense of community, and reciprocity. This definition bridges across literatures on social cohesion (Stanley 2003a, 2003b) and social capital (Putnam 2000: 134-147; Sacks 2002). Participation will be categorized as formal or informal to generally align with instrumental and expressive actions as described by Lin (2001: 43). Social cohesion  2 will also be distinguished based on factors that emerge within the data. However, as authors have distinguished between general and institutional trust (Matthews and Côté 2005; Putnam 2000: 137; Veenstra 2002), I expect to see a distinction between general cohesion that applies to cooperative attitudes towards the general population and institutional cohesion that is reflected in cooperative attitudes towards institutions and leaders. The relationship between participation and social cohesion may only be salient across particular types of participation and cohesion. Furthermore, the relationships that do emerge may be contingent on factors such as socio- demographic characteristics and community characteristics that may influence both whether people participate and the degree to which that participation produces cohesion. In other words, these factors may antecede the relationship between participation and cohesion, such that they influence the degree to which a community member participates, and they may influence the strength of the relationship between participation and social cohesion, such that people with particular socio-demographic factors or in particular communities generate more social cohesion through the same levels of participation. In essence, this research will attempt to demonstrate that social capital can produce social cohesion, but that it is contingent on the types of social cohesion and participation in question, and also on the characteristics of participants. The following questions will be empirically addressed:  1. Is the relationship between participation and social cohesion salient across both formal and informal participation and different types of social cohesion, such as general and institutional? If so, does the relationship differ in strength between these various dimensions?  2. Do socio-demographic factors influence whether people participate in community life, and the extent to which their participation produces cohesion?  3. Does the community in which a person lives influence whether they participate in community life and does the nature of the relationship between participation and cohesion vary across communities?   3  Answers to these questions will help to establish the social contingencies of social capital in the generation of social cohesion, illuminating a more nuanced interpretation of civic participation for the purpose of community resilience and development. While the research findings will contribute to the academic discourses around social capital, they may also have particular relevance for the communities in which the study is situated. This research is embedded in the context of coastal British Columbia and is part of a much larger, multi-year and multi-stage endeavour called the Resilient Communities Project. The twenty-four communities that comprise the basis for this study traditionally have resource- based economies involving some combination of fishing, logging and mining. Not only do these industries provide an economic foundation for the development of coastal life, they are also enmeshed with the cultural and social practices of community members. Over the past decade, these foundations have shifted dramatically and in many cases, disappeared from beneath coastal populations (Young and Matthews 2007), leaving the economic, cultural and social structures of communities unstable at best. In Masset, for example, the population went from 1,529 in 1991 to just 921 in 20011. Other communities have not suffered such a drastic decrease in size but have certainly experienced a significant economic decline. From 1991 to 2001, the average employment income decreased in sixteen of the twenty-four communities: in Prince Rupert, average income went from $39,214 to $30,438; in Alert Bay, from $36,252 to $24,311; and in Gold River, from $45,410 to $30,520. In many cases, the drop in earnings is particularly poignant for men. As mines close, mills disappear from the local landscape, and marinas empty, so too disappear important sources of income and infrastructure. The Resilient Communities Project aims to explore the ways that social capital can support communities through social and  1 Statistics Canada data compiled in Community Profiles on the Resilient Communities Project website, www.resilientcommunitiesproject.com. Extracted April 10, 2008.  4 economic turbulence by aiding adaptation, decreasing vulnerability, and facilitating economic growth and diversification.  A recent experience I had in Port Hardy exemplifies the importance of social resources in sustaining coastal communities. Every July, Port Hardy holds FILOMI Days to celebrate the fishing (FI), logging (LO) and mining (MI) industries in the community. It seems somewhat ironic to celebrate three resource industries that are no longer a source of stability, but instead are commonly cited among community members as a source of misfortune. For some, surely the celebration of a declining set of industries must cause cynicism or grief, and certainly some community members reflected such inclinations. Yet, the FILOMI festival lures quite a crowd and is often mentioned in casual sidewalk conversation, even with the occasional contentious undertone. These observations led me to believe that participation in the festival may build cohesion within the community, through a celebration of nostalgia, common history and familiarity. My hunch was substantiated as a local woodworker mentioned that, after years of missing out, he had reluctantly attended the festivals the previous year only to find that it reignited business relationships with other community members. Through face-to-face interaction, community members were reminded of his skills and perhaps the value of familiarity, and commissioned work from him.  Social resources in coastal communities can provide a safety net for residents and also aid economic development. This thesis considers social cohesion to be such a safety net, and the links between people, or social capital, the threads that constitute its fabric. This project will empirically test the dimensions and dependencies of the relationship between participation and social cohesion through the application of multiple linear regression models to survey data. The results will be relevant to both literature on social capital and the particular communities from  5 which the data were generated. Establishing whether there is a difference between general cohesion and cohesion with institutions and their association with participation will add to the social capital literature. By establishing whether participation and cohesion are contingent on socio-demographic and community factors, the study will contribute to the discussion on social inequality and social capital (Lin 1999a). By comparing formal and informal participation and the factors that influence whether people participate and become more socially cohesive, the study will also aid in a discussion of the comparative values of different types of participation in community development. Each of these findings will provide insights for decision-makers in coastal British Columbia, firstly, by identifying populations who do not participate and generate cohesion, or whose cohesive returns on participation are less than others, secondly, by illuminating the role of place in shaping social capital, and thereby identifying particular communities that provide fertile ground for the development of cohesion through participation, and thirdly, by identifying the strengths and weaknesses of different forms of participation as strategies to develop different types of social cohesion.  Literature Review The sociological story of participation and cohesion in coastal B.C. will unfold over the following pages. The literature review will begin by situating participation and social cohesion as they will be employed in this study within the academic discourses of social capital.  These literary locations will also elucidate the potential value in distinguishing institutional cohesion from general cohesion and formal participation from informal participation. The different possible combinations will be outlined, discussed and theorized.  6 After placing relationships between participation and cohesion within a theoretical framework, the various social contingencies that are likely tied to those relationships will be articulates in terms of socio-demographic factors and community. To begin with, the potential socio-demographic contingencies of participation and cohesion will be embedded within the discussion of social inequality and social capital, largely developed by Nan Lin (2000). In line with that work, the contingent factors will be discussed in terms of capital deficit, relating to access to and investment in social capital, and return deficit, or the varying degree to which participation generates cohesion. The community in which an actor lives will be considered through a discussion of place. Place can impact whether community members participate formally or informally by providing different structural resources and also the degree to which they generate social cohesion through those forms of participation.  Defining Social Capital: From Participation to Social Cohesion Social capital has taken a variety of definitions in academic discourses. Bourdieu (1985: 248) provides one of the earliest accounts of social capital, describing it as “the aggregate of the actual or potential resources which are linked to possession of a durable network of more or less institutionalized relationships of mutual acquaintance or recognition.” This definition tying durable social relationships, or social ties, to resources has remained a pivotal concept to social capital. Boudieu (1985: 249) goes on to suggest that “the profits which accrue from membership in a group are the basis of the solidarity which makes them possible.” In this statement, investment and participation in group relations, ties and resources are explicitly reciprocally linked to group solidarity and social cohesion. Not only is the link between participation and social cohesion somewhat intuitive, but with the rise in interest in social capital, a number of  7 studies have supported the connection empirically (Carpiano 2007/9; Stolle 1998; Veenstra 2002). The most notable and probably most widely read example is Putnam’s Bowling Alone (2000), which ties a decline in civic participation to the fragmentation of American communities. Civic participation, symbolized by bowling leagues, is being replaced by individualism and the ‘lone bowler’, creating a decline in trust, reciprocity and cooperation, or social cohesion, to the detriment of community politics, economy and health. In an earlier publication, Making Democracy Work (1993), Putnam similarly explaines economic discrepancies between Northern and Southern Italy as a consequence of different levels of participation and the resulting differences in social cohesion between community members and within economic and political institutions. Through this literature and the work of Putnam in particular, a basic model has been developed in which participation and networks, as micro forms of social capital, aggregate into more collective social assets such as norms of trust and cooperation, or what I refer to as social cohesion (Figure 1). Participation reinforces ties and associations, building intra-personal trust and reciprocity, developing into more general forms of trust, sense of community, and a propensity to cooperate. As these attitudes become more common among individuals in a community, they can develop into collective norms that are considered a macro form of social capital. This macro form of social capital is comprised of individuals who share socially cohesive attitudes. Hence, the movement between participation and individual attitudes provides an indicator of this process underway. In order to be productive, social capital must include aspects of both micro and macro components. While some authors argue that network ties and embedded resources provide the most cogent measureable form of social capital (Lin 1999a), ties are not useful nor their  8 embedded resources accessible if associations are not paired with a degree of shared trust, reciprocity and cooperative attitudes. Sandefur and Laumann (1998: 484) refer to networks as “potential social capital,” recognizing that ties and resources are only valuable if they can be mobilized. An individual can have a plethora of ties without being able to capitalize on a single one if those ties lack the mutual trust and cooperative imperative to make them productive. I may know my neighbor and see her frequently. However, I may not lend her my electric drill upon request, were I lacking a general trust in people and a sense of obligation to my local community. Moreover, I would not ask her to lend me anything if I did not think reciprocity was a reasonable and normal practice.  Similarly, one can live in a very cohesive community, but without having connections to people, it would be difficult to access their resources. In isolation, ties and social cohesion may not be particularly productive. In tandem, these associations embedded within shared attitudes of cooperation, trust and social cohesion, function as social capital and produce an array of individual and social benefits. At the individual level, intrapersonal ties, operating with a degree of trust and cooperation, can provide people with jobs (Lin and Dumin 1986; Green, Tigges, and Diaz 1999), information (Burt 1992), and access to resources (Van Der Gaag and Snijders 2005; Enns, Malinick, and Matthews 2008), leading to individual outcomes such as increased socio-economic status (Lin 1999b). At a societal level, social capital relates to economic development (Knack and Keefer 1997; Narayan and Pritchett 1999; Putnam, Leonardi, and Nanetti 1993; Putnam 2000), health (Kawachi et al. 1997; Lavis and Stoddart 2003), political efficacy, (Putnam, Leonardi, and Nanetti 1993; Wilkinson 1996), and adaptation to environmental change (Adger 2003). This paper focuses specifically at the process in which participation, and the exercise of social networks, generates or reinforces social cohesion, indicated in this research at the individual level as a propensity to cooperate. This  9 process that is constantly reinforcing links between the individual and the collective is the foundation of social capital. This research is focused on the individual, and therefore cannot consider macro social capital as societal norms or social cohesion as a property of a collective. It will, however, consider how individuals develop the attitudes that aggregate into the macro forms of social capital as an indicator of those collective values being developed. As a consequence, this thesis will address how community members develop the cohesive sentiments, or propensity to cooperate, that are components of social cohesion, as a macro form of social capital (Figure 1). As Figure 2 indicates, there is also an aspect of this relationship that is reciprocal. As people participate socially in their communities, developing more ties and social cohesion, they will also likely feel a stronger impetus to participate. In part, this reciprocal relationship will be encapsulated by the community variable in the regression models. As participation develops social cohesion, it becomes a property of the community, and any effects it would have on participation will consequently be controlled by the community variable. As a consequence, one can expect community to explain a significant amount of the variation in social cohesion, while also encapsulating other important properties of community such as infrastructure. The implications of this approach on the interpretation of the regression models will be further discussed in the interpretation and analysis portion of this thesis.  10   11 Social Cohesion – Definition and Operationalization The concept of social cohesion has a significant history in sociological literature, running though Hobbes’ Leviathan (1991), De Toqueville’s Democracy in America (2000), Tönnies’ Community and Society (1957), and Durkheim’s Division of Labour in Society (1984). These texts are all concerned with the ways in which populations become contributing members of a greater social unit. More nuanced definitions of social cohesion, however, have emerged in contemporary literature that still retain this theme of social order. Maxwell (1996) relates social cohesion to shared values and communities of interpretation, commitments to equality, and a sense that community members are engaged in a shared enterprise. More methodically, Jenson (1998) defines social cohesion as a location along five different social scales: belonging- isolation, inclusion-exclusion, participation-non-involvement, recognition-rejection, and legitimacy-illegitimacy. The various approaches to social cohesion can all be united under a general theme of cooperation. This thesis will subsequently adopt a rather simple definition proffered by Stanley (2003b: 8), who observes that,  “social cohesion appears to be based on the willingness of people in society to cooperate with each other in the diversity of collective enterprises that members of society must do in order to survive and prosper.” The willingness to cooperate is the crux of social cohesion as it will be used in this thesis, and is a productive framework with which to approach resilience in coastal British Columbia. As Sacks (2002: 147) aptly notes, “survival turns out not so much to depend on individual strength as on habits of co- operation”. A willingness to cooperate, and the underlying concepts of trust, reciprocity and sense of community are precisely the qualities required to provide a social safety net and a foundation for growth and development during periods of change.  12 Sense of community, trust and reciprocity will be operationalized to encapsulate social cohesion in this project.  Sense of community is invaluable to the rural coast, as it binds community members to place. Place refers to the embedding of social relationships in time and space, in contrast to the emptying of space Giddens (1990: 18) refers to, so that communities are meaningful spaces where people feel at home. Similarly, Sacks (2002: 150) suggests that informal relationships grounded in honesty, respect, trust and reciprocity embed people in “larger groupings in and through which we develop our identities,” embodying what Jenson (1998) terms inclusion and belonging in her construction of cohesion. Hence, while labour shifts and socio-demographics along the coast may be changing, sense of community can allow people to still retain somewhat of a stable social identity and sense of belonging. Without a sense of place there is little to tie people to rural communities while they undergo the economic turbulence recently experienced in British Columbia. Sense of community also refers to how people feel about their communities as places and people. This type of sense of community, therefore, involves attitudes towards the community as a whole and towards community members. Two measurements thus will be applied in this study to encapsulate “sense of community”: sense of embeddedness in the community and sentiments towards the community as a whole. Trust is also an element of social cohesion as a foundation of reciprocity and exchange and thus collaboration and cooperation. Trust and reciprocity are also aspects of social cohesion that have been emphasized throughout the literature on social capital (Coleman 1990; Hardin 2002; Putnam 1993, 2000).  According to Putnam (2000: 135), “honesty and trust lubricate the inevitable frictions of social life.” Hobbes (1991) broaches the issue of trust in noting that two parties are paralyzed from engaging in any sort of exchange without a degree of mutual trust. This state of paralysis has re-emerged as the Prisoner’s Dilemma (Sacks 2002: 147-148; Putnam  13 2000: 288) whereby two prisoners are each confronted with the opportunity to disclose the details of their crime, reducing their own punishment while increasing the other’s, or trusting that the other will uphold his or her silence so that they both suffer no punishment. If their relationship is strong, and they trust one another, they can both acquire the most optimal outcome to their dilemma. Trust is an essential element of productive social ties as it enables community members to mobilize ties and access resources. Hence, trust plays a valuable role in keeping communities together and cooperative by allowing members to share resources without feeling vulnerable to opportunism. In other words, trust reduces the transaction costs of exchange. Honesty and trust are tied to what Jenson (1998) terms recognition and legitimacy. Trust for one another, and developing that trust into a more general faith about people, communities, and their systems, is a fundamental component of a cohesive community. Authors also identify a type of social cohesion that relates to leaders and institutions. If people believe that their community’s institutions are sound and that leaders are representing a common interest and effectively working towards those goals, there will be a greater incentive to participate in decision-making, and also a greater incentive to support and adhere to decisions made. As a consequence, political initiatives for collective services would be more cogent and elicit greater public support (Stanley 2003: 8,12). Furthermore, if business and political leaders feel they have greater support and legitimacy among community members, they will be more likely to stay committed to their communities, invest time and energy, and spearhead projects and programs. If there is greater faith in institutions and their representatives and leaders, cooperation is more predictable and less risky, reducing transaction costs and engendering greater creativity and innovation (North 1990; Stanley 2003b: 12).  14 These sentiments towards leaders and institutions are likely quite different from general forms of cohesion, although some authors tend to group them together (Jenson 1998; Maxwell 1996; Stanley 2003c). The social relationships that emerge between community members, and between community members and institutions, are quite different and involve different processes. As Putnam (2000: 137) suggests “Trust in other people is logically quite different from trust in institutions and political authorities.” Similarly, Sacks (2002: 149) differentiates contractual and covenantal ties. Not only does Veenstra (2002) use different categories relating to institutional and general trust, he also finds that they were related to different forms of participation. Hence, when developing the concept of social cohesion into a measurement, I expect to find two related but distinct forms of cohesion, general and institutional, that may have different relationships with participation.  Participation – Definition and Operationalization Putnam (2000), Sacks (2002), Stanley (2003b) and Veenstra (2002) suggest that one way to develop social cohesion is through participation.  By interacting face-to-face and developing social ties and linkages, people become more familiar with each other and develop the trust, reciprocity and sense of community that underlies social cohesion. From this perspective, participation captures the action foundation of social capital emphasized by Lin (2001: 48-53) in his writings on networks and ties, while also being embedded within the literature on macro social capital that stresses the importance of civic participation. Indeed, Lin (1999a: 32) suggests that authors who focus on micro and macro social capital agree that “it is the interacting members who make the maintenance and reproduction of this social asset possible.” Furthermore, this connection between participation and ties is supported by Enns et al. (2008)  15 who used the RCP dataset to find a correlation between social participation and social ties. Hence, I will be using participation as a reflection of social networks in action. The research on a social capital from a network perspective suggests that ties can be differentiated as weak or strong, bonding or bridging. Granovetter (1973: 1361) quantifies the strength of a tie as a “combination of the amount of time, the emotional intensity, the intimacy and the reciprocal services which characterize the tie.” Bridging ties stretch across social networks, linking people from distant social arenas, whereas bonding ties are those that occur between members of a close, and often closed, social group (Burt 1992). Lin (2000: 45) notes people have a tendency to homophily, to building relationships with people with similar socio- demographic backgrounds and interests, such that close social bonding networks among kin groups and people sharing neighbourhoods or occupations would likely be often stronger than heterophilious relationships that occur between socially disparate individuals. In general, then, bonding ties would tend to be stronger, and bridging ties, weaker. Bonding and bridging ties are associated with different actions through which they are developed and maintained. Bridging ties that reach beyond an actors’ network are developed through instrumental social action practiced for the purpose of developing ties and expanding networks (Lin 2001: 46). Expressive social action, on the other hand, is practiced to reinforce intimate ties between actors, and is thus associated with strong and bonding ties (Lin 2001: 46). Instrumental and expressive actions are captured in what I have termed formal participation and informal participation. Formal participation encapsulates actions that occur in a group setting organized around a particular endeavour. Examples include a participation in a quilt guild, rotary club or political party. Because the group is formed around an action, it can be expected to be a somewhat heterophilious and socially geared toward expanding networks, aligning it with  16 instrumental social actions. Informal participation is often spontaneous or irregular, and may occur between a small and selective group of people. Examples include dinner with friends, going to a movie or playing cards. The action is formed around a particular group of people and is intended to enhance and solidify existing relationships, and thus would likely be more homophilious.  Informal participation is much more angled toward expressive social actions, and strong, bonding ties. There is some overlap between the two categories of participation, as formal participation can also be a forum for expressive social action and reinforcing bonding social ties. Similarly, one can have dinner with a business associate for instrumental purposes. In general, however, if someone wants to meet new people, and thus engage in an instrumental social action, it would be most appropriate to join a formal organization or club. Similarly, if someone is losing touch with a friend, and wants to build a stronger bond with them, it would be logical to visit in a more personal and intimate social setting. Hence, while these categories are ideal types (Weber 2004: 314) and not absolute, they do capture general social patterns of purposeful actions aimed at the development of social capital. They are used here as analytic tools to distinguish between formal and informal participation within the context of networks and social capital. Relationships between Participation and Social Cohesion The basic relationship between participation and cohesion will be tested for the two types of participation, formal and informal, and potentially two types of social cohesion, as I expect institutional and general cohesion to emerge as two distinct types. The relationships will be tested with the suspicion that a significant correlation will exist between some matches and not for others. In particular, one would expect formal participation to be tied to a more institutional  17 cohesion. Both formal and informal participation may be tied to general cohesion, but for different reasons and potentially different applications. The leaders involved in community institutions may also be involved with formal groups as a way to build their own networks, leading community members to build relationships with them and gain institutional cohesion. Furthermore, given that civic groups are institutions, they may engender a greater comfort with institutional structures overall. Indeed, this link between formal participation and institutional cohesion has led scholars to connect social capital to democratic ideals, resonating with the 19th century work of de Tocqueville. As Stolle (1998: 498) notes, this connection is “based on the assumption that membership in voluntary associations facilitates mobilization and socialization effects, as well as democratic learning processes within these associations.” Yet in reflecting on the work of Sacks (2002: 151) and Giddens  (1990: 88), it may also be the case that institutional cohesion, particularly in the context of modernity, is not built through experience within institutions or contact with leaders, which are largely relationships defined by contract. Instead, institutional cohesion may be grounded in pre-formed perceptions that establish a type of faith in leaders and institutionalized systems. Formal participation may produce cohesion by providing a forum through which community members can develop diverse ties and expand networks with people from a broader set of social backgrounds, potentially generating general trust as unfamiliar community members become familiar to participants. These ties are diverse in the sense that they are heterophilious – they are between people with different backgrounds and interests. These ties also have an economic value, as weak, bridging ties help actors “get ahead” and provide actors with more diverse sources of information and access to more diverse resources (Burt 1992; Granovetter 1973; Lin 1999b), potentially inspiring reciprocal relationships. Hence, the relationship between  18 formal participation and cohesion would relate to the economic growth and development of rural communities on coastal British Columbia. On the other hand, the more formal social structure of participation may also detract community members who lack social confidence, limiting the extent of heterogeneity that can be achieved. Furthermore, intra-group tendencies towards homophily may actually implicitly exclude some community members, furthering social fractions in the community. Informal participation has received less attention from scholars with regards to both network development and social cohesion. Expressive social actions, and the related bonding form of social capital, can reinforce boundaries around particular social groups and potentially isolate individuals from a community. Along these lines, de Tocqueville (2000: 78) condemns informal participation when he claimed that it was a manifestation of individualism: Individualism is a calm and considered feeling which disposes each citizen to isolate himself from the mass of his fellows and withdraw into the circle of family and friends; with this little society formed to his taste and leaves the greater society to look after himself.  As a consequence, these exclusive groups can actually conflict with community interests and reduce social cohesion, creating what Rubio (1997) terms perverse social capital. Informal participation certainly has the potential to produce negative consequences for communities. However, rather than discounting the value of informal participation altogether, this literature illuminates the need to consider its value within broader social contexts. Informal participation, is seems, may have a particularly important role to play in the development of social cohesion. In his description of covenantal relationships, Sacks (2002: 149) suggests that the bonds of belonging that emerge in strong relationships foster the honesty, trust and reciprocity that underlie social cohesion. Furthermore, bonding social capital and strong social ties may provide  19 people with reliable resources in times of crisis.  Woolcock (2002: 232) consequentially associates bonding social capital with destitution and survival rather than progress, which may overlook a more enduring purpose for bonding social capital. Devine et al (2003) find that bonding ties between family and friends influence the middle-class’ ability to migrate in pursuit of new opportunities and adapt to new environments. For vulnerable groups, the bonding ties that informal participation produces may generate the means for survival, but can also provide a sense of security and confidence for a broader spectrum of the social scale that enables community members to innovate, adapt and aid others during turbulent times such as those currently experienced in rural BC. The relationship between participation and cohesion may not be salient throughout the different potential dimensions of participation and cohesion. Instead, the correlations between participation and cohesion may be contingent on the types of participation being practiced and the types of cohesion under consideration. Illuminating how these relationships differ will provide insights into the various ways social capital can operate and greater depth into its character and potential role in coastal British Columbia. Even this fuller picture, however, would be shallow were it to ignore the more extensive social contingencies which influence the development and utility of social resources, such as socio-demographic factors and the effects of place. These factors may influence whether people participate and the extent to which participation generates cohesion. They may also spuriously create the appearance of a relationship when there is none, or conceal a relationship by independently influencing both participation and cohesion in opposing directions.   20 Socio-demographic Factors and Social Inequality  The contingency of the relationship between social capital and cohesion on socio- demographic factors illuminates social inequality in social capital. Capital often accrues within particular groups of people in society, reinforcing social inequality. Research has shown that social capital follows this pattern, as people occupying more privileged positions in society will have a better stock of social ties, and a more advantageous network position (Campbell, Marsden, and Hurlbert 1986; Green, Tigges, and Diaz 1999; Lin and Dumin 1986). Alternatively, people who occupy more disadvantaged social positions have a capital deficit (Lin 2000: 790-791).  A significant body of literature has emerged that delineates the factors that determine a capital deficit, including economic or socio-economic status  (Kawachi et al 1997; Wilkinson 1996), gender, employment categories, educational levels (Enns, Malinick and Matthews 2008), ethnicity and race (Green, Tigges and Diaz 1999; Elliott 1999), and religiosity (Veenstra 2002).  This project will explicate the factors that lead to a capital deficit in coastal British Columbia. In addition to income, gender, employment, education, ethnicity and religiosity, additional factors of age and marital status will be included. As scholars have been concerned with a decline in social capital (see Putnam 2000; Castells 2000), age is an important indicator of generational changes. People of different ages may also engage more in different types of participation. Similarly, people who are married or in a common-law relationship may engage socially with others in a way that differs from someone who is single or divorced.   Less research has explored what Lin (2000) terms a return deficit, or the ways in which the same level of social investment may produce different returns depending on socio- demographic factors. While research has suggested that people who occupy lower levels along a  21 social hierarchy are more likely to invest in poorer social resources, less research has considered whether equal levels of investment in equally valuable resources will provide more or less benefits for people who occupy different levels along a social hierarchy. For this project, a return deficit would indicate that some social groups develop more social cohesion than others when they participate. For example, people with a lower household income may engage in fewer activities, and thus have a type of a capital deficit. It may also be the case that when people with a lower income do engage in social activities, they do not build as much trust as someone with a higher household income, because they are more financially vulnerable. Moreover, people who have a notably low household income may not feel as positively about their communities as others when they engage in social activities because they feel economically excluded and relatively deprived. A capital deficit will explicate the factors that determine whether people invest in social capital by participating, and thus produce social cohesion, whereas a return deficit will explicate the contingent value of that participation, whereby the varying productivity of social capital will be shown across different socio-demographic factors. A return deficit will be elucidated by testing interaction terms that will capture whether the relationship between participation and cohesion varies across different socio-demographic factors. The details of this process will be more fully outline in the following chapter on methodology.  This thesis will illuminate the social factors that lead to a social capital deficit in rural communities along coastal BC, by explicating which social groups do not participate, and consequentially have a reduced level of cohesion.  Variables that antecede the relationships between participation and social cohesion will be discussed within this context. As a unique contribution to an emerging discussion of social capital and inequality, this project will also show those social positions for which social capital generates produces less social cohesion than  22 others, indicating that these groups are experiencing a return deficit. A capital deficit will establish the socio-demographic factors that predict whether people participate in the first place, demonstrating the contingency of the relationship, and a return deficit will show the contingent value of that participation.  Community Research has shown that different characteristics of place can influence the development of social capital. In recognition of the role locality can play, some scholars advocate the consideration of locality in research on social capital (Beggs, Haines, and Hurlbert 1996; Enns, Malinick, and Matthews 2008; Veenstra 2005). For example, rural actors have networks characterized by more “bonding” social capital founded on dense ties than their urban counterparts (Beggs, Haines, and Hurlbert 1996; Onyx and Bullen 2000). The characteristics of places, in other words, may determine whether community members develop social capital, and also the ways that social capital is developed and applied. Along these lines, Bradbury (2006) argues that the physical routes through which social capital is accessed are underemphasized in the literature. He argues that transport services play a fundamental role as a physical route through which rural people access and generate social capital.  These experiences of place and particularly the sense of mobility and accessibility of place may not be experienced the same way by everyone. Places, and the sense of isolation some individuals feel in their communities, are factors that may provide a background to the practices of social capital. This thesis will consider whether the relationship between participation and cohesion emerges only in particular places, and furthermore, whether places influence the extent to which participation generates social cohesion. In other words, this thesis will investigate whether there is an interaction  23 between participation and place, such that participation generates more social cohesion in some communities than others. Hence, I will essentially establish whether the relationship between participation and cohesion varies between different places.  Summary This thesis will consider how social capital is developed on coastal British Columbia by exploring how community members build social cohesion through participation. This link between individual ties and shared attitudes of trust and reciprocity are foundational to the concept of social capital. By comparing the relationships between formal and informal participation and potentially two different types of social cohesion, general and institutional, I will explicate whether the relationship is salient across different possible dimensions of participation and cohesion. Furthermore, the influence of socio-demographic and community factors on those relationships will be explored in two ways: firstly, by delineating whether the factors influence whether people participate, and secondly, by testing whether the relationship varies in strength across the factors, suggesting that some generate more cohesion than others through their participation. Altogether, this research will establish the contingencies of social capital development and practice via the process through which participation generates social cohesion.   24 CHAPTER 2: DATA AND METHODS Introduction  This thesis will explore the ways in which different people develop social cohesion by participating in the social life of their communities as a representation of social capital development processes. The relationship between both formal and informal participation and social cohesion will be tested by using multiple linear regression models on a data set generated through survey responses. The research will establish who is more likely to engage in these types of social capital development, and whether some people who are engaged reap fewer social benefits than others. These individuals will be differentiated according to socio-demographic factors and the community in which they reside, which will be considered as antecedent variables that influence whether people participate, or factors that interact with participation such that participation generates more social cohesion for some than others.  This chapter will outline the methodological approach aimed at delineating how community members along coastal British Columbia develop social capital. I will begin by putting the data in context by describing the methodological aspects of the broader research project in which the research is embedded. The chapter will continue by describing the variables to be used in the study and their distributions. I will also describe the process through which the social cohesion variables will be created. The analytic strategy will be described, including a discussion of exploratory work, missing cases, the development of regression models, and the use of interaction terms. The chapter will conclude with a brief summary of the chapter.  25 Context The data for this project were produced through the Resilient Communities Project (RCP), a multi-year project, three phase study. The first phase identified and researched the social and economic characteristics of coastal communities in British Columbia through BC Statistics, and identified 131 coastal communities with a population over 50 and less than 30,000. The second phase of the project generated the data that will be used for this project. Of the 131 communities, 24 were selected using stratified a stratified random sample, and a random sample of 4,386 households was mailed surveys. A 60% response rate generated a final sample size of 2,881.   The survey largely focused around social capital which was operationalized through an array of indicators, and included sections on demographics, employment, community, trust, social activities, networks, media use and health. The data generated through the second phase of the project will be used for this thesis. The third phase of the project included in-depth interviews of leaders and community members in six of the twenty-four communities, and built qualitatively on many of the themes quantified in the second phase. As a consequence, many of the findings from this phase of the study could provide a foundation for a follow-up study of the interview data. The Sample and Distributions Participation  In conjunction with instrumental and expressive social actions, the research will include formal and informal participation. Each participation variable refers to a range of activities, rather than the frequency of activities, or commitment to those activities. However, the range of activities is intended to measure the level of social engagement and participation in the  26 community, rather than commitment to a particular activity or social group.  The number of different activities people do is more likely to measure the extent to which a respondent is expanding and maintaining his or her social networks than would the frequency of activities or degree of commitment to particular activities. The number of different activities is intended to give an approximate measurement of the amount of time spent participating, as someone involved with more activities will likely be committing more time overall, all else being equal. More importantly, individuals who engage in a diversity of activities can be assumed to be engaging with different people with a variety of interests, expanding or maintaining a broader and perhaps more diverse social network. The formal participation variable reflects the number of distinct types of activities, from a list of 23 including an ‘other’ category” (See Box 1). Measuring the variety of different groups would indirectly assess the variety of different people, with different interests, with whom the respondent was engaged.           Box 1: Formal Participation Survey Questions  Are you currently involved with any of the following (Please select Yes for all that apply.)  a. Artistic or craft/hobby group (eg, Pottery Guild, crafter’s association, quilting club) b. Business (eg, Chamber of Commerce, Band Economic Development Committee, Community Economic Development Committee) c. Church-related activities (eg, choir, bible study or care group, coffee or social committee) d. Community Service Group (eg, Crisis Centre, Food Bank or Community Kitchen) e. Cultural or ethnic associations (eg, Heritage Association, First Nations cultural groups) f. Educational (eg, Parent Advisory Council, School Board/Trustee, Curriculum/Language groups) g. Environmental (eg, resource conservation/management/action groups) h. Health (eg, Cancer Society, Health Auxiliary, alternative or traditional healing groups) i. Neighbourhood (eg, Resident’s Association, crime prevention groups) j. Political (eg, political party, Band or Tribal Council, town council) k. Self-help or support l. Service Club (eg, Lions Club, Rotary Club) m. Social Club (eg, card playing, music, book club) n. Spiritual/religious group o. Sports or recreation (eg, soccer, karate, Little League, Curling Club, weekly pick-up games) p. Work-related (eg, union, cooperative, professional association) q. Youth (eg. Girl Guides, Boy Scouts, 4H) r. Any other activities not listed above. (Please specify)    27 Respondents were asked to indicate whether they participated in the activity, and whether they did so inside the community, outside the community or both. As this project was concerned with engagement in community life and activities that involved face-to-face interaction with other community members, participation was only counted if it is undertaken either in the community or both inside and outside the community. Informal participation similarly denotes the total number of casual activities in which a respondent participates derived from a list of 15 (See Box 2). Once again, each item on the list asked respondents to indicate whether the activity occurred inside or outside the community, or both, and responses were only added into the informal participation variable if the activity occurred inside the community or both inside and outside.           In general, respondents engage in a greater variety of informal social activities than formal (Table 1). For each measure, responses tend to cluster at the lower levels of participation, while a few respondents participate in much higher levels of participation, pulling the mean up above the median. Box 2: Informal Participation Survey Questions  In the past 30 days, have you done any of the following activities with people who do not live in your household? (Please select Yes for ALL that apply)  a. Had a meal at a restaurant with others? b. Had a meal with others in your home or their home? c. Watched a movie or television program with others in your home or their home? d. Played cards or games with others in your home or their home? e. Gone with others to a movie at a theatre? f. Gone to a church social function g. Gone to a religious service? h. Attended a sports event? i. Participated with others in a sports or recreational activity? j. Gone with others to a bar or tavern? k. Gone shopping with others? l. Gone sports fishing or hunting with others? m. Participated with others in a craft or artistic activity? n. Participated in a First Nation ceremonial or spiritual activity? o. Participated in a First Nation cultural or social activity?  28 Table 1: Distributions Interval Variables Mean Median Standard Deviation IQR Min Max Informal Participation 4.38 4 2.448 3 0 15 Formal Participation 2.70 2 2.162 3 0 17 Age 54.83 54 15.51 23 19 104 Sense of Isolation 3.17 3 1.346 2 1 5  Household Income  6.55 6  2.898  5  1  12 Categorical Variables 1 Valid % Gender (Male) 1437 51.3 Religiosity (Religious) 2057 73.9 Ethnicity (North American Indian) 206 7.5 Marital Status (Single) 887 31.1 Education (University or Prof. Degree) 578 20.7 Workstatus (Unemployed) 1180 41.6 Socio-Demographic Factors Socio-demographic variables in the study include age, gender, religiosity, ethnicity (North American Indian), household income, marital status, education and work status2. The distribution of these variables is included in Table 1. Household income will be treated as a continuous variable in the regression models, although the values each represent an income range that begins at “less than $5,000” (1) and “$5,000 to $9,999” (2) and increases by $10,000 per unit up to “100,00 and over” (12). Marital status, education and workstatus have all been recoded into dummy variables with scores of 1 for being single (separated, divorced, widowed or never married), holding a university or professional degree (BA, MA, Ph.D. or Professional Degree) and being unemployed. Each of these variables was taken directly from respondents’ answers to straightforward survey questions that asked respondents to identify themselves from a range of options. Place Two variables refer to place. The first simply represents the community in which the respondent lives, and was procured in much the same way as the socio-demographic variables.  2 The questions used to generate the socio-demographic factors is included in Appendix 1  29 The sample is distributed evenly over the twenty-four communities, with approximately 120 (4%) respondents in each. The community with the smallest sample is Alert Bay, with 71 respondents, and the largest is Comox, with 155. The second variable, sense of isolation, relates to a respondents’ perceptions regarding the isolation of their community and is generated from a survey item that asked respondents to rate how they feel about their community on a five point scale along six dimensions (Box 3). The distribution for “sense of isolation” is included in Table 1.  Analytic Strategy Social Cohesion: Factor Analysis The first step of this project will be aimed at producing the social cohesion variables. They will be developed by conducting a factor analysis of 27 Likert Scale questions that relate to the concept of social cohesion (Box 4). Box 3: Sense of Isolation  For each pair of items listed below, please darken the circle along the scale which comes closest to how you feel about your community.  5    4    3    2    1 32. Good Place to raise kids                      O   O   O   O   O Poor place to raise kids 33. Good employment opportunities         O   O   O   O   O      Poor employment opportunities 34. Safe                                                      O   O   O   O   O Dangerous 35. Isolated                                                O   O   O   O   O Not isolated 36. Good schools                                       O   O   O   O   O Poor schools 37. High crime                                           O   O   O   O   O Low crime 38. Good health care                                 O   O   O   O   O Poor health care  30                     Box 4: Social Cohesion Likert Scale Questions  Trust The next set of statements deal with issues of trust. Please indicate your level of agreement by filling in the appropriate circle.  1. Most people can be trusted. 2. You can’t be too careful in dealing with people. 3. Most people in this community can be trusted. 4. Generally, I don’t trust people from outside this community as much as I trust people who live here. 5. There are groups in this community that I do not trust 6. Young people in this community can be trusted. 7. In this community, you should never leave your home without locking it. 8. Most people in this community are likely to try to take advantage of you if they get the change. 9. The police responsible for this community can be trusted to do a good job. 10. I trust the leaders in this community to respond to community needs. 11. Business leaders in this community can be trusted 12. First Nation leaders in this community can be trusted. 13. The politicians who represent this community can be trusted to do a good job.   Your Community This section asks questions about your community and the people who live there. Please indicate your level of agreement by filling in the appropriate circle.  1. I am very satisfied with this community as a place to live. 2. People in this community have a weak sense of community. 3. It is hard for people to make close friends in this community. 4. I would move away from this community if a good job came up somewhere else. 5. The future of this community depends more on what happens outside the community than inside it. 6. People like me don’t have a say about what the political leaders in this community do. 7. The political leaders in this community are good, capable people. 8. The political leaders in this community generally represent the interests of a few powerful groups. 9. Compared to other communities on the B.C. coast, this community is doing pretty well economically. 10. This community doesn’t have enough services for the elderly. 11. Business leaders in this community are creating new economic opportunities here. 12. The First Nation leaders in this community play an important role in its social life. 13. First Nation groups are important to this community’s economic wellbeing. 14. Using First Nation culture and history is a good way to promote economic development in this community.   31 Putnam (2000: 137) notes that institutional and generalized trust differ in society, and are generated through different processes. Similar distinctions are made by Sacks (2002: 149), and have been empirically distinguished by Veenstra (2002). Trust is an important aspects of social cohesion. Hence, while authors tend to group the types of social cohesion involving leaders and institutions with those aimed at people more generally, this study will not assume they are the same. Scales will be formed from the primary factors emerging from the factor analysis. To ensure that the scales are internally consistent, scale reliability analysis will be performed, after which point the scales will be constructed by taking the mean of response values across relevant items. Binary Relationships and Missing Data  The second analytic stage of the study is an exploration of the relationships among all the variables. There are two purposes to this process. The first is to ensure that the variables have a role to play in the study. If they have absolutely no relationship to any of the other variables, it is most unlikely that their inclusion is going to reap any benefits, and I would consider excluding them from the remainder of the study. Secondly, the relationships that emerge will allow me to anticipate what may happen when the variables are included in the regression models, allowing me to better predict and recognize noteworthy results.  The bivariate relationships between the different variables will be tested before and after selecting cases. Only cases that contain values for all of the independent variables for a given set of models will be included, and in particular, missing cases for both informal and formal participation will be excluded to ensure that their effects can be compared. Given the number of missing cases for informal and formal participation (Table 2), it can be expected that the sample size for the regression models will be significantly reduced. The missing cases for formal  32 participation in particular, but informal participation as well, are largely because respondents did not indicate whether they participated inside or outside of the community. By testing the bivariate relationships before and after cases are selected, it will enable me to determine how the reduced sample may impact the regression results. In particular, I will look at whether the strength of the relationships between the independent and dependent variables, represented by Spearman’s rho or eta², changes between the entire sample and after the sample has been reduced by excluding missing cases. I will also consider whether the significance levels of the relationships change between the sample sizes. This comparison will help to determine whether the smaller samples will create sample bias that can reduce the validity of the study. Table 2: Missing Cases Valid Missing Informal Participation 2616 265 Formal Participation 2214 667 Age 2823 58 Gender 2801 80 Marital Status 2856 25 Household Income 2696 185 Education 2787 94 Workstatus 2836 45 Community 2881 0 Sense of Isolation 2805 76 Multiple Linear Regression Following the exploratory work and the analysis of missing cases, I will address my research questions through a series of multiple linear regression models. The research project is interested primarily in the relationship between the two types of participation and the two cohesion variables, and the impact that different factors have on those relationships. The best way to test these relationships will be through multiple linear regressions that add variables in a series of steps. Linear regression is appropriate for scale dependent variables, and steps, beginning solely with the participation variable, will illuminate the changes that occur in the  33 relationship between participation and cohesion when other variables are added. Hence, four linear regression models will be constructed to compare each of the two forms of participation with each of the cohesion scales. This will allow me to make inferences about the different impact of factors on the relationships between participation and cohesion.  The first model will simply include the participation variable and cohesion variable. The socio-demographic variables will be included in three steps because they are the largest group of factors, and this will help to distinguish their various impacts. The first step will include age, and gender; the second set will include religiosity, ethnicity and marital status; and the third will include household income, education, and work-status.  The next step will include the effects of community and sense of isolation. Literature on social capital suggests that these variables may antecede the relationship between participation and cohesion, leading to a capital deficit, or potentially cause the relationship to be spurious. They are also variables for which the strength of the relationship between participation and cohesion may change. In other words, participation may produce more cohesion for some than others along these socio-demographic lines, potentially leading to a return deficit on participation for some social groups.  This part of the investigation requires consideration of interaction effects. Interaction Effects  Following the inclusion of socio-demographic and community factors, interactions terms will be used to test the combined effects of participation and each of the other independent factors. Interaction effects capture what Jaccard and Turrisi (2003: 1) call a moderated relationship, a relationship between two variables that changes depending on the value of a third variable. This thesis is concerned with the way the relationship between participation and cohesion is stronger or weaker as a consequence of socio-demographic or community factors and  34 thus moderated by those factors.  The results would suggest that processes of social capital generation differ for different social groups. One question that will be addressed, for example, is whether the relationship between formal participation and social cohesion is weaker for people who are unemployed. Perhaps people who do not have a job have less resources to contribute to social ties, and consequently do not reap the same networking benefits from formal participation and develop less social cohesion than those who are employed. As a consequence, people who are unemployed would have a return deficit on participation, and may be less able to mobilize ties as a result. This exploration will establish whether the movement from a micro level of social capital in the form of associations and ties to macro social capital, as attitudes and a propensity for cooperation, is less salient across particular populations leading to difficulties in the development of social capital.  An interaction term will be generated by multiplying participation, either formal or informal, with a socio-demographic or community factor. The resulting term will provide a value that captures the unique combined effects of participation and that factor. Take, for example, an interaction between age and formal participation. When age is multiplied by formal participation for each respondent, it will produce a unique value that represents the product of, or interaction between, a respondent’s age and level of formal participation. When the interaction term is added to the regression model while age and participation are held constant, it will show whether the combined effects of age and participation have a unique relationship with social cohesion above and beyond the independent effects of age and formal participation.  If these prove significant, then there is something significant about participating when older or younger. In other words, the effects of participation on social cohesion vary across different ages.  35 Summary The results of the statistical analysis will produce implications for the relationship between participation and social cohesion. They will demonstrate whether different forms of participation do indeed have different effects on cohesion. They will also show who is engaged in these processes, by delineating which of the socio-demographic and community factors are antecedent to the relationships. They will indicate whether the relationships between the forms of participation and aspects of cohesion do exist, or whether there is reason to suspect they are spurious. Lastly, they will show whether these forms of participation generate cohesion to the same degree for everyone, or whether socio-demographic or community factors predispose some people to generate more cohesive sentiments through participation than others. Altogether, through the analysis of participation and social cohesion, this research will examine the ways in which different people are engaged in the development of social capital in coastal British Columbia.    36   CHAPTER 3: RESULTS  Social Cohesion: Factor Analysis A factor analysis was run for 27 Likert scale questions that relate to a respondent’s trust and how they feel about their community. Two factors emerged from the data relating to social cohesion. The first included nine questions that captured sentiments towards people and community members generally (Table 3).  The second theme included questions focused on a respondent’s attitudes towards leaders in the community, including business leaders, political leaders, and First Nations leaders. A scale reliability analysis showed that the two sets of questions are internally consistent and reliable, and two scales were created for general social cohesion and attitudes toward leadership by calculating the means of each set of questions.  These two scales will be used as the dependent variables for this research, representing two aspects of social cohesion. Each of these scales form fairly normal distributions, although each are skewed by small clusters of lower values that pull the mean below the median (Table 4). The scale capturing attitudes toward leadership relates to an institutional dimension of social cohesion. Like the institutional trust identified by Putnam (2000: 137), Veenstra (2002) and Matthews and Côté (2005), attitudes toward leadership reflect how community members perceive the representatives of political, economic and cultural structures in their communities. As discussed in Chapter 1, these attitudes can facilitate community involvement in decision-making processes enabling leaders to make more cogent and innovative decisions, and can also encourage community members to be more receptive to decisions made. These positive relationships are thus a fundamental component of a community’s social cohesion because they provide the foundation for cooperation between community members and the individuals who have more power to shape communities.  37  Table 3: Social Cohesion Scales General Social Cohesion; Chronbach's alpha = 0.752 Most people can be trusted You can't be too careful in dealing with people* Most people in this community are likely to try to take advantage of you if they get the chance* Most people in this community can be trusted I am very satisfied with this community as a place to live People in the community have a weak sense of community* It is hard for people to make close friends in this community* I would move away from this community if a good job came up somewhere else* There are groups in this community that I do not trust* Attitudes Toward Leadership; Chronbach's alpha = 0.827 People like me don't have a say about what the political leaders in this community do* The political leaders in this community are good, capable people The political leaders in this community generally represent the interests of a few powerful groups* I trust the leaders of this community to respond to community needs Business leaders in this community can be trusted First Nation Leaders in this community can be trusted The politicians who represent this community can be trusted to do a good job *Values reversed for consistency   Table 4: Social Cohesion Distributions Mean  Median Std. Dev. Range General Social Cohesion 3.547 3.667 0.548 1.22 to 5 Attitudes Towards Leadership 3.215 3.286 0.642 1 to 5  Exploratory Work and Missing Cases  To ensure comparability across regression models for each dependent variable, cases were only included if they had values for all of the variables to be used in the analysis. As a consequence, the sample sizes for the models were significantly reduced from the original 2881 sample population. For the models predicting attitudes toward leadership, the sample is reduced to 1702. The sample is slightly smaller for the general social cohesion models at 1650.  An exploration of key bivariate relationships was performed for all cases and for the smaller samples that would be tested in the regression models (Table 5). In general, differences between the samples are not particularly important, but do exist in several notable cases. The relationship between  38 ethnicity and formal participation is less statistically significant for both of the reduced samples although in each case the differences between formal participation means across the categories are very similar. For each sample that will be used to test attitudes toward leadership, the relationship between household income and formal participation is weaker, as is the relationship between a respondents’ sense of isolation and formal participation. For the reduced sample for the general social cohesion models, the relationship between household income and social cohesion is weaker in the reduced sample size and insignificant whereas it is significant in the large sample. Other slight differences in the strength of relationships exist between the entire sample and reduced sample sizes, but at a level that is unlikely to influence the models.  The reduced strengths of these variables are meaningful in the interpretation of the regression models, as the significance tests for these may under-represent the effects in the broader population. Overall, however, while the reduced sample size is not perfectly representative of the broader sample, the differences are small and there is little reason to suspect that they will reduce the validity of the study.  Of key importance are the relationships between the participation variables and the dependent variables, general social cohesion and attitudes toward leadership. These relationships remain the same for the largest sample and the reduced sample sizes. These tests suggest that there is a relationship between formal participation and social cohesion, while there is no direct relationship between informal participation and either dependent variable. However, a relationship may be concealed by other factors, as most of the independent variables are related to informal participation.  In particular, age, ethnicity and community may conceal the relationship between informal participation and the two dependent variables, attitudes towards leaders and general social cohesion. Respondents who are younger or have a North American Indian ethnic origin are more likely to engage in informal activities, but are less likely to be cohesively inclined or have positive attitudes toward leaders. Similarly, some communities where informal participation is high, respondents are  39 likely to be less socially cohesive. As a consequence, it can be expected that controlling for these variables may reveal an underlying relationship between informal participation and social cohesion. Table 5: Bivariate Relationships All Cases Soc. Coh. Leadership F. Part I. Part. Social Cohesion 1 0.516*** 0.113*** 0.011 Leadership 0.516*** 1 0.111*** 0.11 Formal Participation 0.113*** 0.111*** 1 0.326*** Informal Participation 0.011 0.011 0.326*** 1 Age  0.278*** 0.231*** 0.077*** -0.178*** Household Income 0.050* 0.063** -0.028 -0.024 Sense of isolation -0.194*** -0.128*** 0.056** 0.053** Gender 0.002* 0.001 0.006*** 0.006*** Religiosity 0.006*** 0.012*** 0.010*** 0.003** Ethnicity (NAI) 0.012*** 0.004** 0.006*** 0.020*** Marital Status (Never Married) 0.003** 0.001 0.002* 0.013*** Education ( Degree) 0.014*** 0.009*** 0.009*** 0.002* Work-status (Unemployed) 0.006*** 0.007*** <0.001 0.008*** Community 0.070*** 0.039*** 0.036*** 0.052*** Attitudes Toward Leadership (N=1702) Soc. Coh. Leadership F. Part I. Part. Leadership ---- 1 0.111*** 0.001 Formal Participation ---- 0.111*** 1 0.340*** Informal Participation ---- 0.002 0.340*** 1 Age  ---- 0.257*** 0.096*** -0.154*** Household Income ---- 0.053* -0.008 -0.038 Sense of isolation ---- -0.126*** 0.050* 0.063** Gender ---- <0.001 0.006** 0.005** Religiosity ---- 0.010*** 0.013*** 0.003* Ethnicity (NAI) ---- 0.006** 0.003* 0.021*** Marital Status (Never Married) ---- <0.001 <0.001 0.016*** Education ( Degree) ---- 0.015*** 0.009*** 0.002* Work-status (Unemployed) ---- 0.009*** <0.001 0.004** Community ---- 0.051*** 0.033*** 0.059*** General Social Cohesion (N=1650) Soc. Coh. Leadership F. Part I. Part. Social Cohesion 1 ---- 0.113*** -0.001 Formal Participation 0.113*** ---- 1 0.322*** Informal Participation -0.001 ---- 0.322*** 1 Age  0.311*** ---- 0.105*** -0.141*** Household Income 0.022 ---- -0.018 -0.039 Sense of isolation -0.186*** ---- 0.055* 0.054* Gender <0.001 ---- 0.006** 0.004** Religiosity 0.005** ---- 0.011*** 0.004* Ethnicity (NAI) 0.017*** ---- 0.003* 0.020*** Marital Status (Never Married) 0.016*** ---- <0.001 0.010*** Education (Degree) 0.015*** ---- 0.013*** 0.001 Work-status (Unemployed) 0.010*** ---- <0.001 0.004* Community 0.084*** ---- 0.031*** 0.056*** Relationships measured using Spearman's rho or eta² *p <0.05; **p<0.01; ***p<0.001  40 Alternatively, the relationship between formal participation and the social cohesion variables may be anteceded by age, education and religiosity. Age is positively correlated with formal participation, attitudes towards leaders and general social cohesion.  Similarly, respondents who hold a degree and are religious are more likely to have these socially cohesive attitudes. Regression Results Informal Participation and Leadership  The most important trend to emerge regarding the relationship between informal participation and attitudes toward leaders is the increased magnitude of the effects of informal participation with the inclusion of other variables (Table 6). The initial relationship, small and insignificant, grows with the inclusion of each set of variables, suggesting that the relationship is concealed by these factors. In particular, age, ethnicity, household income and education all influence peoples informal participation and how they perceive community leaders in contrasting directions. Younger people participate the most, but have a less positive attitude toward leaders. People who are native, have a lower household income and do not have a degree follow a similar pattern of high levels of participation paired with less positive attitudes toward leaders. It is also important to note that gender is significantly related to attitudes towards leaders once other factors are controlled. Given that gender is also related to informal participation, it is likely that, all else being equal, gender is antecedent to the relationship. Women participate informally more and also have more positive attitudes towards leaders. Community causes a similar dissonance between informal participation and attitudes toward leaders. In many communities where levels of informal participation are high, members have less positive attitudes toward their leaders. Respondents participate more when they think their community is isolated, but are similarly likely to be more poorly disposed toward leaders. Once the  41 effects of these factors are controlled, an underlying, significant relationship between informal participation and attitudes toward leadership exists, such that more participation corresponds with more positive attitudes towards leaders.  The implications of these findings will be discussed in the following chapter. Formal Participation and Leadership  Unlike the effects of informal participation, there is an initial relationship between formal participation and social cohesion that is reduced with the introduction of socio-demographic and community variables (Table 7). The relationship is largely the consequence of these other independent factors, as controlling for age, household income and education reduces the relationship below any significant levels. Similar to the relationship between informal participation and attitudes towards leaders, gender is significant once other factors are controlled, and likely also has an antecedent role to play in the relationship. These factors either influence whether people participate and generate these positive attitudes toward leaders, or alternatively, influence formal participation and these positive attitudes independently, causing the relationship to be spurious. Informal Participation and Social Cohesion  The effects of informal participation on social cohesion parallel its effects on attitudes toward leaders (Table 8). The initial weak relationship increases with the addition of other variables that conceal the underlying relationship. When socio-demographic and community factors are controlled, the relationship is significant. In particular, marital status, ethnicity, household income and community conceal the relationship. Each of these variables influences the informal participation of respondents and their level of social cohesion in opposing directions, somewhat anticipated by the preliminary tests, such that the underlying positive linear relationship is masked. People who have  42 never been married participate informally the most, but feel less socially cohesive. Similarly, people From North American Indian descent have higher levels of informal participation, but lower levels of social cohesion, on average, in their communities. Socio-economic status also conceals the relationship, to a lesser degree, and the addition of community and sense of isolation to the models reveals a somewhat weak but significant relationship between informal participation and social cohesion. In contrast, gender is likely to antecede the relationship, although its effects may be somewhat masked by the concealing impacts of the other variables. Women participate more and are more likely to feel socially cohesive within their communities. Formal Participation and Social Cohesion  The relationship between formal participation and social cohesion was non-linear, and was consequently transformed by taking the square root of the participation variable. The results show that the relationship differs from that of informal participation and social cohesion, as there is an initial significant relationship that weakens when socio-demographic factors are controlled (Table 9). Both age and gender influence whether people participate and develop social cohesion, as does income and education. Like informal participation models, ethnicity likely conceals some of the relationship as the preliminary tests show that people with North American Indian ethnic origin participate formally more but have lower levels of social cohesion. The relationship between formal participation and social cohesion is slightly concealed by community and sense of isolation.  43  Table 6: Informal Participation and Attitudes towards Leaders   DV: Leadership (N=1702) Model 1 Model 2 Model 2 Model 3 Model 4 R(R²) 0.009(<0.001) 0.244(0.060) 0.255(0.065) 0.288(0.083) 0.376(0.142) F-test 0.129 35.865*** 19.617*** 17.068*** 8.336*** Independent Variables B B B B B Informal participation -0.002 0.007 0.009 0.011 0.012* Age (as of 2004)   0.011*** 0.010*** 0.010*** 0.010*** Gender (Male)   -0.047 -0.045 -0.061* -0.062* Religiosity     0.045 0.040 0.063 Ethnicity (North American Indian)    -0.124* -0.093 -0.124* Marital Status (Single)     -0.086 -0.062 -0.048 Household Income      0.018** 0.020*** Education (Degree)       0.142*** 0.125** Workstatus (Unemployed)      0.003 0.030 Community° Alert Bay        0.482*** Bella Coola         0.282** Bowser        0.186 Chemainus         0.288** Comox        0.170 Courtenay         0.168 Galiano Island        0.314*** Gold River         0.377*** Hagensborg        0.303** Kitimat         0.287** Masset        0.428*** Mayne Island         0.371** Mill Bay        0.156 Parksville         0.010 Port Alberni        0.102 Port Hardy         0.339** Prince Rupert        0.013 Quadra Island         0.480*** Queen Charolette        0.408*** Sechelt         0.022 Sointula        0.503*** Squamish         0.115 Tofino        0.285** Sense of Isolation         -0.063*** °Reference category Powell River *p <0.05; **p<0.01; ***p<0.001    44   Table 7: Formal Participation and Attitudes towards Leaders DV: Leadership  (N=1702) Model 1 Model 2 Model 3 Model 3 Model 4 R(R²) 0.069(0.005) 0.247(0.061) 0.257(0.066) 0.289(0.083) 0.375(0.141) F-test 8.180** 36.921*** 20.054*** 17.084*** 8.288*** Independent Variables B B B B B Formal participation  0.020** 0.015* 0.015* 0.012 0.011 Age (as of 2004)   0.010*** 0.009*** 0.010*** 0.009 Gender (Male)   -0.043 -0.042 -0.059 -0.061* Religiosity     0.043 0.040 0.064 Ethnicity (North American Indian)    -0.122* -0.088 -0.116 Marital Status (Single)     -0.083 -0.058 -0.043 Household Income      0.018** 0.020** Education (Degree)       0.134*** 0.118** Workstatus (unemployed)      0.005 0.031 Community° Alert Bay        0.480*** Bella Coola         0.275** Bowser        0.174 Chemainus         0.284** Comox        0.169 Courtenay         0.163 Galiano Island        0.297** Gold River         0.367*** Hagensborg        0.292** Kitimat         0.284** Masset        0.421*** Mayne Island         0.363** Mill Bay        0.153 Parksville         -0.003 Port Alberni        0.094 Port Hardy         0.342** Prince Rupert        0.009 Quadra Island         0.470*** Queen Charolette        0.403*** Sechelt         0.018 Sointula        0.481*** Squamish         0.107 Tofino        0.281** Sense of Isolation         -0.063*** °Reference category Powell River *p <0.05; **p<0.01; ***p<0.001   45    Table 8: Informal Participation and General Social Cohesion DV: Cohesion (N=1650) Model 1 Model 2 Model 3 Model 4 Model 5 R(R²) 0.021(<0.001) 0.298(0.089) 0.319(0.102) 0.343(0.117) 0.434(0.188) F-test 0.707 53.364*** 31.081*** 24.257*** 11.343*** Independent Variables B B B B B Informal Participation -0.005 0.004 0.008 0.009 0.013* Age (as of 2004)   0.011*** 0.011*** 0.011*** 0.009*** Gender (Male)   -0.062* -0.063* -0.073** -0.065** Religiosity     -0.011 -0.014 0.014 Ethnicity (North American Indian)     -0.213*** -0.188*** -0.156** Marital Status (Single)     -0.098* -0.081 -0.076 Household Income       0.012* 0.014** Education (University or Professional Degree)       0.118*** 0.093** Workstatus (Unemployed)       -0.022 -0.020 Community Alert Bay         0.270** Bella Coola         0.202* Bowser         0.344*** Chemainus         0.310*** Comox         0.172* Courtenay         0.149 Galiano Island         0.328*** Gold River         0.223** Hagensborg         0.409*** Kitimat         0.193* Masset         0.183* Mayne Island         0.487*** Mill Bay         0.221* Parksville         0.055 Port Hardy         0.043 Powell River         0.241** Prince Rupert         0.114 Quadra Island         0.458*** Queen Charolette         0.257** Sechelt         0.064 Sointula         0.359*** Squamish         -0.031 Tofino         0.194* Sense of Isolation         -0.070*** °Reference category Port Alberni *p <0.05; **p<0.01; ***p<0.001     46   Table 9: Formal Participation and General Social Cohesion DV: Cohesion (N=1650) Model 1 Model 2 Model 3 Model 4 Model 5 R(R²) 0.068(0.005) 0.300(0.090) 0.320(0.103) 0.342(0.117) 0.433(0.187) F 7.685** 54.142*** 31.314*** 24.134*** 11.296*** Independent Variables B B B B B Formal participation (Sqrt) 0.059** 0.034 0.039 0.030 0.046* Age (as of 2004)   0.011*** 0.010*** 0.011*** 0.009*** Gender (Male)   -0.060* -0.061* -0.072** -0.064** Religiosity     -0.010 -0.012 0.015 Ethnicity (North American Indian)     -0.207*** -0.180*** -0.145** Marital Status (Single)     -0.095* -0.078 -0.070 Household Incomeº       0.012* 0.014** Education       0.112*** 0.085** Workstatus (Unemployed)       -0.022 -0.019 Community° Alert Bay         0.275*** Bella Coola         0.205* Bowser         0.348*** Chemainus         0.321*** Comox         0.180* Courtenay         0.154 Galiano Island         0.315*** Gold River         0.222** Hagensborg         0.407*** Kitimat         0.200* Masset         0.185* Mayne Island         0.487*** Mill Bay         0.230** Parksville         0.048 Port Hardy         0.054 Powell River         0.250** Prince Rupert         0.118 Quadra Island         0.457*** Queen Charolette         0.259** Sechelt         0.067 Sointula         0.347*** Squamish         -0.033 Tofino         0.198* Sense of Isolation         -0.072*** °Reference category Port Alberni *p <0.05; **p<0.01; ***p<0.001   47  Interaction effects   Interaction terms for each factor were added individually to Model 5 of each regression set, and showed that the relationship between participation and social cohesion varies across socio- demographic factors and between communities. A diverse set of factors interacted with informal and formal participation to significantly impact each of the dependent variables. These tests reveal that the relationship between participation and social cohesion are moderated by a range of factors. The patterns that emerge with regard to these interactions will be discussed in the following chapter. Socio-Demographic Factors Attitudes towards Leaders   Three socio-demographic factors interacted with informal or formal participation to significantly influence a community member’s attitude toward community leaders. The effects of informal participation are moderated by a respondent’s education. The slope of the linear relationship between informal participation and attitudes toward leadership increases by 0.027 (p = 0.061) for people who hold a university degree, denoted by the magnitude of the interaction term. For those that do not hold a degree, the slope is 0.006, and is insignificant (p=0.352), whereas it increases to 0.033 and is significant (p=0.010) for those who have obtained a university or professional degree (Figure 3). In other words, for those that hold a degree, their attitudes towards leadership can be expected to positively increase by 0.033 for each additional informal activity in which they participate, while informal participation cannot be expected to generate any of these attitudes for those who do not hold a degree.   48   The effects of formal participation are moderated by both marital status and work-status. Respondents who have never been married or who are employed are less likely to generate positive attitudes toward community leaders when they participate in formal activities than those who are married, living in common-law, separated, divorced or widowed or those who are unemployed.  For those community members who have never been in a marriage or marriage-like relationship, the relationship between formal participation and attitudes toward leadership decreases significantly by 0.057 (p=0.028) from 0.015 (p=0.037) to  -0.042 (p=0.096) as illustrated in Figure 4. Interestingly, the relationship increases by about 0.024 (p = 0.087) from 0.002 (p = 0.844) to 0.025 (p = 0.020) for people who are unemployed (Figure 5).  49      General Social Cohesion  The tests showed that respondents of a North American Indian descent generate less social cohesion when they participate both informally and formally than the rest of the population. The interaction term for informal participation and ethnicity produced a B value of -0.043 (p=0.010), such  50 that the slope of the relationship between informal participation and social cohesion for respondents who are not North American Indian is 0.018 (p=0.001), but drops to -0.025 for  respondents who are North American Indian and is not significant. The interaction terms for formal participation and ethnicity changes the slope of the relationship between participation and social cohesion by -0.162 (p=0.011), from 0.064 (p=0.003) to -0.098 (p=0.103). When the interaction effects for both informal and formal participation and ethnicity are included in the model, the effects of ethnicity no longer have a significant relationship with social cohesion.     51  The interaction term for household income and informal participation also has a significant impact on social cohesion with a B value of 0.003 (p=0.039), suggesting that for every unit increase in income range, the slope of the regression line for participation and social cohesion increases by approximately 0.003, indicating that participating matters more for wealthier people than for people with more limited resources. Community Factors Attitudes towards Leaders   None of the interaction terms for community and informal participation showed a significant relationship with positive attitudes toward leaders. However, several of the interaction terms for formal participation and community significantly influence attitudes towards leaders, suggesting that formal participation is a more effective way to generate these attitudes in some places more than others. Using Powell River as a reference category, Alert Bay (B = 0.087; p = 0.057), Bella Coola (B = 0.071; p = 0.095), Bowser (B = 0.094, p = 0.064), Courtenay (B = 0.086; p = 0.063), Kitimat (B = 0.098; p = 0.040), Mill Bay (B = 0.143; p = 0.039), Port Alberni (B = 0.069; p = 0.084), Prince Rupert (B = 0.081; p = 0.069), Queen Charlotte (B = 0.094; p = 0.038) and Sechelt (B = 0.084; p = 0.092) all interacted with formal participation to have a significant impact on a community member’s attitudes toward leaders. The differences in the relationship between formal participation and attitudes toward leaders in communities, comparative to Powell River, are illustrated in Figure 8. These results show that formal participation is more likely to generate positive attitudes toward leadership in the ten communities listed than in Powell River.  When the interaction tests are re-run as binaries without a Powell River as a reference category, measuring the effects of participating in one particular community as compared with every other, none show any significant effects.  52  General Social Cohesion   The interaction terms for formal participation and community showed no significant effects, while several of the interaction terms for informal participation and community were significant and positive. The interaction effects for Gold River (B = 0.056; p = 0.094) Hagensborg (B = 0.075; p = 0.030), Kitimat (B = 0.071; p = 0.024), Mayne Island (B = 0.121; p = 0.002), Queen Charlotte (B = 0.064; p = 0.054) and Squamish (B = 0.083; p = 0.046) relative to Port Alberni show that informal participation generates more social cohesion in some communities than others. When the tests were rerun with the community as a binary without a reference category, they showed that the strength of the relationship between informal participation and social cohesion is very much contingent on the community in which it occurs. The slope of the relationship between informal participation and social cohesion is 0.101 (p = 0.001) for respondents on Mayne Island, but decreases by 0.093 to 0.008 (p = 0.128) for respondents who do not live on Mayne Island (Figure 10). Similarly, the relationship between informal participation and social cohesion is positive and significant for respondents who live in Kitimat (B = 0.048; p = 0.015) and weaker and insignificant for respondents  53 in other communities (B = 0.007; p = 0.174, Figure 11). Some communities produce significantly more social cohesion through informal participation than others, demonstrated by the relative significance of interaction terms, and, as Mayne Island and Kitimat exemplify, the relationship only exists in particular places.    54    Summary  While the analytic focal point of this project is the results from the regression models, some analytic work had to be undertaken to facilitate the regression strategy. A factor analysis of data from social cohesion survey questions showed that there are two distinct facets of social cohesion: attitudes towards leaders and a more general form of social cohesion relating more broadly to trust  55 and sense of community.  These two factors became the dependent variables for the regression models. Exploratory tests were performed between the variables to provide some indication of what to expect in the regression models and to provide a background of relationships on which to draw when interpreting the impact of the variables in the regression models. The results of these tests showed many of the socio-demographic variables, including age, income and education, that were correlated with lower levels of informal participation were also correlated with higher levels of both types of social cohesion. These tests also showed that the people who are older, better educated and who have a higher household income are likely to participate formally more and are also those with higher levels of both types of social cohesion, suggesting that these factors may spuriously cause the relationship or antecede it. Similarly, some of the communities with the highest levels of formal and informal participation have the lowest levels of social cohesion, and vice versa, suggesting that community may also conceal relationships of interest. In short, there are a lot of social contexts in which participation and social cohesion seem to have a dissonance.      The results of the regression models presented only one surprise: while the effects of gender on the two social cohesion variables in the exploratory tests were minor, they showed to have a significant impact in all of the regression models once other factors were controlled, consequently placing them as the only factors to be antecedent to all four relationships tested between participation and social cohesion. More predictably, the effects of informal participation on both types of social cohesion are concealed by these independent variables so that once they are controlled a relationship is revealed. Socio-demographic factors antecede the relationship between formal participation and both types of social cohesion, but the effects of community on the relationship vary for attitudes towards leaders and general social cohesion. Community has very little impact on the effects of formal participation on attitudes towards leaders, while it conceals it the effects of formal participation on social cohesion. While there is a relationship between participation and social cohesion that is somewhat salient across different types of participation and social cohesion, the ways  56 that the relationship exists seems to vary according to different social characteristics and environments.  The variances between participation and social cohesion are empirically substantiated by the results from the interaction tests.  While the regression models alone show that different people are more likely to participate in ways that do or do not lead to higher levels of social cohesion, the interaction tests show that some people can be expected to have higher levels of social cohesion than others when they participate equal amounts. The effects of informal participation on attitudes towards leaders differ depending on a community member’s education, while the effects of formal participation differ according to marital status and work-status.  Alternatively, the effects of informal participation on more general forms of social cohesion is stronger for people with a higher household income and who live in Kitimat or on Mayne Island. The effects of both informal and formal participation on general social cohesion are significantly lower for people with a North American Indian ethnic origin.  Overall, the results of these tests show that the way people participate, informally or formally, reflect socio-demographic factors. Moreover, that participation as an investment in social capital has variable social returns. The following chapter will discuss these differences in social capital development in greater detail, and consider how these variances reflect the different contexts of people’s lives.  57  CHAPTER 4: ANALYSIS AND DISCUSSION  Introduction   As a process of social capital development, how do the participatory activities of community members generate social cohesion? This chapter will try to address this basic question by analyzing the results of Chapter 3 within the context of the study developed in Chapter 1. In particular, the three areas of inquiry posed at the introduction of this thesis will be tackled. First, the salience of the relationship between participation and social cohesion will be considered for both types of participation and social cohesion. This discussion will lead to a consideration of the contingencies of those relationships, or the various contexts that foreground the development of social capital. In the second section of this chapter, I will consider who is developing social capital in these ways, followed by where they are. The third section will consider how the nature of the relationship between participation and social cohesion changes for different people and in different places. After the factors that shape social capital have been elucidated, the chapter will conclude with a summary. From Participation to Social Cohesion: How can Social Capital be Developed?   There are a variety of ways that people can develop the ties and bonds that facilitate exchange and reciprocity. For five years, I have lived in the same neighbourhood and about once a month I hear the same man stroll around the neighbourhood and sing operatically. I imagine  58 this man is known by many people around Vancouver and their familiarity with him may ease his daily transactions and lubricate any social exchange. Surely this vocal character must have a surplus of weak ties, although their efficacy may be limited by what some may perceive as odd behaviour. This project did not exhaustively consider the subtle ways in which people can build social capital, but did look at two particular forms of participation and can suggest that they both generate the attitudes that underlie socially cohesive communities. These two types of participation have been categorized as informal and formal participation and respectively represent expressive actions aimed to solidify existing relationships and instrumental social actions that create relationships and expand networks (Lin 2001: 46). The regression models show that both of these types of participation are related to attitudes towards leaders and general social cohesion although the effects differ slightly between types of participation. The influence of informal participation on social cohesion is more subtle, and only becomes apparent by looking beyond the effects of other factors. Formal participation, on the other hand, has a stronger relationship with both forms of social cohesion, but these effects are largely embedded in particular populations. Informal Participation   People who participate a lot informally are what Putnam (2000: 93) calls schmoozers. Like formal types of civic engagement, Putnam (2000: 108) notes that “informal social connectedness has declined in all parts of American society.” This presents a paradox, as the social conditions that espouse more schmoozing, including higher levels of education, being single and childlessness, have increased in America. These educated childless singles are not the people who schmooze the most on coastal British Columbia, and while this study does not  59 consider changes in levels of informal participation over time, the results do illuminate a new piece of the participatory puzzle and present a new paradox. Populations with higher levels of participation do not also have higher levels of social cohesion. The effects of informal participation are similar for both forms of social cohesion: attitudes towards leaders and general social cohesion. In each case, the relationship is concealed by other factors such that no relationship is evident unless they are held constant. Particular social characteristics dispose people to participate informally a lot and to also have less positive perceptions of community leaders and lower levels of general social cohesion. People that are younger, for example, are more likely to participate informally more, but are also more likely to have more negative perceptions of community leaders and a lower sense of trust and sense of community. Several factors have similar effects so that seemingly, as people participate in more informal activities, they seemingly have less socially cohesive attitudes. Once these other characteristics are taken out of the statistical picture, a small significant relationship becomes apparent. The slope of the relationship between informal participation and attitudes towards leaders is almost horizontal when factors are not controlled (Model 1, Table 6), but increases when socio-demographic and community factors are held constant. High levels of participation and less positive attitudes towards leaders are embedded in these factors, pulling the slope of the relationship down.  The results of the interaction tests further the suspicion that the effects of schmoozing are contextually conditioned. People that have more money, are not of a North American Indian Ethnic origin, and live on Mayne Island or in Kitimat can be expected to gain significantly more  60 social cohesion when they participate. That is, the societal social capital generated by schmoozing is significant, but appears to be highly contingent on who is the schmoozer. Why might the schmoozers occupy such a tenuous role in social capital development? It may be that the bonding ties that more closely correspond to informal activities have a tendency to be more exclusive and remain at the individual level without aggregating to broader social norms and attitudes. Nan Lin (2000: 45) suggests that expressive social actions aimed to enhance existing ties are often more socially homophilious: occurring between similar people with shared interests creating closed social networks.  While these ties can certainly be assets to community members, such tight networks, grounded in sameness, can support exclusive social groups that obstruct social integration. These qualities can lead to perverse social capital (Rubio 1997) and enable groups such as the Mafia or Colombian cartels to operate with more efficiency and security. While it seems implausible that organized crime is going to emerge from close bonds and plague coastal communities of British Columbia, it is certainly possible that particular social groups may become insular, making it more difficult for them to contribute to the social and economic development of these coastal communities. In this particular case, the fragmentation is a significant risk because of the populations that seem to potentially be developing this insularity: younger people and those with a lower socio-economic status. Yet informal participation and the bonding social ties that emerge also correlate to social cohesion once factors are controlled, suggesting that there is a significant underlying relationship that could be enhanced in particular social arenas if the circumstances that obstruct the relationship are resolved. There may be experiences and circumstances that are shared by particular populations that lead them to seek the solace of close friends and family while also leading them to feel also  61 less in synch with their communities. The following sections on socio-demographic factors, community and interaction effects will illuminate some of the characteristics that could be causing people to engage in a greater number of informal activities, while going against the general pattern by feeling less socially cohesive. Formal Participation   Unlike the informal participation models, socio-demographic variables are largely antecedent to the effects of formal participation on the two forms of social cohesion, attitudes towards leaders and general social cohesion. In the regression models, the initial relationships that exist between participation and the dependent variables are reduced below any significant level when socio-demographic factors are controlled. These results suggest that particular socio- demographic groups participate in group activities and feel better about their communities as places with good leaders, more trustworthy people and where they are more socially embedded. In this regard, the effects of formal participation on both forms of social cohesion are similar, and some general patterns about the role of formal types of participation can be deduced. However, the ways in which the role of formal participation unfolds when community is controlled differs between attitudes towards leaders and general social cohesion, opening up some sites of comparison regarding more specific relationships between group participation and social cohesion.  People that are more engaged in a variety of formal activities are what Putnam (2000: 93) calls Machers in reference to the Yiddish term. Machers are people who participate in a variety of organized activities and tend to be more engaged in their communities overall and “those more engaged in community life are both more trusting and trustworthy” (Putnam: 137).  The  62 correlation found between group activities and social cohesion in this study follows the work of Stolle (1998), Veenstra (2002) and Putnam (1993; 2000). These activities whereby community members are bowling together rather than alone, to relay Putnam’s (2000) analogy, foster the types of attitudes that become social assets to the community. The positive attitudes towards leaders, general sense of trust and sense of community facilitate exchange and cooperation, aiding community development and resilience.  These activities are practiced by some more than others, and the cohesive attitudes follow a similar uneven distribution. People that are older, have a higher household income and hold a university or professional degree are the machers of coastal British Columbia. In fact, their influence is so significant, that combined they explain the relationship between formal participation and both social cohesion variables, and individually each are better predictors of social cohesion than levels of participation. In other words, these characteristics are highly correlated with levels of formal participation and attitudes of social cohesion, while others are lacking in both areas. At a micro level, the lack of participation among some socio-demographic populations will lead to a capital deficit, whereby less social investment will lead to fewer ties and fewer accessible resources for the individual. At the macro level, the absence of particular populations may prevent the development of a truly cohesive community. If some members of the population are less likely to cooperate with others, it will be difficult to develop norms of reciprocity and exchange overall, and may stagnate community development. Moreover, the absence of particular groups of people may render certain resources more inaccessible to the network of cooperative community members, undermining what Durkheim (1984) calls organic solidarity.  63   Arriving in a community where half the people are cheerful, talkative and helpful, while the other half are quiet, introverted and hostile would likely lead one to feel somewhat on guard. When those divisions follow a marked socio-demographic distinction, those guarded feelings might become a bit more disturbing, if not somewhat intriguing to the wandering sociologist. I have certainly felt a bit off-put in this way touring rural parts of British Columbia. Communities so divided would not be considered truly socially cohesive, lacking the overall spirit of camaraderie and cooperation. If only some people feel good about their leaders, trusting of each other and embedded in a positive environment, it is unlikely these individual attitudes would develop into broader social norms and characteristics of social cohesion in the overall community. In other words, the capital deficit at a micro level for some may lead to a macro capital deficit in the community more generally, so that the positive impacts of participation remain individualized with limited benefits for community development. The concentration of this type of social capital among some members of the population can also have negative consequences for governance and economic development. Ideas, information and the productive qualities of dialogue are resources that may be lost from community networks that exclude segments of society. The networks of exchange in a community, whereby resources including specialized skills and expertise are accessed through social capital, will also be less efficient if some socio-demographic groups are less engaged and feel less cohesive. For example, the skills of the woodworker who was introduced at the beginning of this thesis may go largely unutilized by local community members if he is not part of a social network characterized by trust and cooperation. As an ex-forestry worker, it is unlikely that he would have gained a degree, and could exemplify one of the people who would be less likely to participate in group activities and have a lower sense of social cohesion.  Indeed,  64 many of the people who had been engaged in the resource industries would be likely to fall into this category. If they are not engaged with their community, and consequently do not become weaved into the web of community reciprocity, trust and cooperation, the level of community social capital would clearly be limited. The relative inaccessibility of human resources would jar the development of organic solidarity, which Durkheim (1984) coined to describe the ways in which specialized skills and uniqueness can be a foundation of integration by motivating exchange. Norms of reciprocity and cooperation are not such an asset if they only exist among a limited number of people that share comparable types of skills. Similar arguments could be made for those who are younger, and may have particular forms of human capital more common in younger generations such as computer skills. In short, the ways that formal participation seems to be practiced by a limited group of people is not without consequences, and could limit the economic development and resiliency of coastal communities.  The issue with the potential capital deficit of particular socio-demographic groups and resulting deficit in macro forms of social capital assumes that there is a relationship between participation and social cohesion and an aspect of that relationship is causal. It is possible that the relationship between formal participation and attitudes towards leaders is spurious. People that are older, more educated and have a higher income, coincidentally, feel more positively towards leaders and better about their communities but there are no real underlying relationships with formal participation. It is certainly possible that there is a spurious aspect of both relationships. All of the different variables simply coincide at points along the life course, so that as someone along the coast ages, they are likely to have a higher household income, hold a degree perhaps, spend time in a greater variety of group activities, and feel more cooperative and better about their communities. Given that the relationship between formal participation and  65 general social cohesion is significant after controlling for community, regardless of age, household income and education, there is due support to suggest that the relationship is not entirely caused spuriously, although there are likely spurious aspects. However, the relationship between formal participation and attitudes towards leaders is not significant above and beyond these socio-demographic factors once community is controlled. This could indicate that the relationship is spurious, or that it is not influenced by differences between communities. An alternative explanation may lie in the reciprocal relationship between individual and societal social capital and the context of coastal life in British Columbia. Were participation generating social cohesion in the community, that social cohesion would then encourage people to participate, and as a consequence, that social cohesion would become embedded in the community factor and be antecedent to the relationship. However, as many of the communities along the coast have undergone such rapid social changes, people’s sense of security and solidarity may have become unstable so that people feel less trusting of other community members and less positive about their communities as a whole. More anxiety about the state of their communities might encourage members to participate more, particularly if formal participation was seen as a normal way to positively develop social cohesion and stabilize social relations. Perhaps the turbulence along the coast has had less of an impact on the majority of people who participate formally, who are typically at a relative socio-economic advantaged, or on community leaders. This relationship between community machers and leaders therefore may be less impacted by these changes on the coast, whereas the ways that machers feel about the general participation may have it ways that are embedded in particular places. Certainly this could be the case if machers were also a relatively closed group of community leaders: their attitudes towards each other may remain the same even while the local community in changing,  66 while their attitudes towards community members who are more diverse and less familiar would be more susceptible to change.  Who is Developing Social Capital? Socio-Demographic Factors Both informal and formal types of participation generate social capital. However, effects of informal participation are more subtle and may be harder to recognize, masked by the effects of factors that lead people to participate informally more but feel less socially cohesive. The effects of formal participation are more overt, but some demographic groups are much more likely to participate and generate a sense of social cohesion. Both forms of participation are part of a process of social capital development, but in each case, who and where the participant is can predict whether the process unfolds. Age  The younger the respondent, the more likely they are to engage in more informal types of social activities. A scatter graph of age and informal participation with a superimposed Loess curve3 (Figure 12) shows that, on average, the younger the respondent, the more they can be expected to participate. A second scatterplot shows the relationship between age and social cohesion (Figure 13) showing that the youngest people have the lowest levels of general social cohesion. The relationship between age and attitudes towards leaders shows an almost identical shape with a very similar Loess curve.  3 “Loess” is a term that refers to “locally weighted scatterplot smoothing” or “locally weighted polynomial regression.” To provide an oversimplified explanation, it uses regression to predict y values for points along the x- axis by calculating a low degree polynomial to represent the relationship at that local point, which is then fit using weighted least squares. This provides a more accurate visual representation of the relationship for different points along the scatterplot.  67 Figure 12: Age and Informal Activities  Figure 13: Age and Social Cohesion   68 In Figure 13, there is a change in slope at around the age of sixty, after which point additional years show only a meagre increase in the levels of general social cohesion. Similarly, at sixty, the slope of the Loess line between age and attitudes towards leaders almost levels off. There is something about youth, particularly before the age of sixty that encourages people to participate informally and also feel less socially cohesive within their communities. The relationship between age and informal participation may represent a pattern of socializing across a lifecourse. It may be a social norm to engage in these more informal types of socializing in younger age groups. Moreover, educational institutions often enable people to develop a broad range of ties, such that network expansion and the more instrumental social actions of formal participation seem unnecessary. The relationship between age and social cohesion presents a bit more of a mystery. The regression models show that informal participation does generate social cohesion, and yet those younger people who participate a lot do not feel relatively better about their community’s leaders, they are not more trusting, and they do not feel more embedded and positive about their community. This may suggest that younger people are participating more exclusively with their age peers, who are less likely to be leaders, and in ways that do not integrate them with a broader spectrum of community members. The relationship could reflect the assertion made by Putnam that there is a generational decline in social capital. In general, societies all over the world may be decreasing in social cohesion over time, and perhaps this change began after the sixty-year olds reached adulthood so that they have been unaffected. However, while there may be a generational decline, on the coast this discrepancy is probably influenced by the shifting economic landscape. Fewer jobs lead to job opportunities, and less money for recreation and spending. This economic recession would likely hit those between the ages of 20 and 40 the most, as the group with less economic  69 resources and the greatest need for employment opportunities. Their reduced spending power would lead to less supply as the infrastructure to support a financially vibrant young-adulthood dwindles. As movie theatres and bowling alleys close, as they have in Port Hardy, and as restaurants and retail stores begin to cater to an older demographic, perhaps younger people feel less in synch with the social sensibilities of their communities. Moreover, jobs can provide people with a sense of identity and contribution, so that if younger people are unable to become enmeshed in an industry, they may feel less integrated with society. Those that would be the least impacted by the changes in job opportunities would be people that are retired, or those who have flourished through thirty boom years of the resource industry. Gender  The influence of gender on the regression models is somewhat easy to miss. In the exploratory tests, women participate both informally and formally more than men, but gender has weaker relationships with attitudes towards leaders and general social cohesion, suggesting that their significance to the models overall might be marginal. However, once other factors are controlled, women do appear to feel more socially cohesive. All else being equal, women participate more and have more socially cohesive attitudes, suggesting that gender is probably antecedent to the relationship to some degree. These effects, however, are difficult to interpret because whatever antecedent effects they have, they are largely outweighed by the concealing effects of other factors. Nonetheless, being gender is particularly notable as the only factor that predicts all four relationships, and thus warrants due consideration. As discussed in the introductory chapter of this thesis, contemporary coastal British Columbia is a unique social environment as a site of considerable economic change and restructuring. One of the most interesting social changes that has emerged is the unique role of  70 women in coastal communities. While men often pursued work in the resource industries, women pursued higher education, earned an income through the informal economy or in administrative jobs, and became involved in the civic lives of their communities. As a consequence, the collapse of the resource industries disproportionately impacted men. This would explain why the decline in average income, outline in Chapter 1, are much more prominent for men. Women’s work was less impacted by the recession, as was their degree of participation and engagement with their communities. As Enns (2008: 4) summarizes from her conversations with women in these rural communities: [The women] talked about the importance of community involvement, and the strong networks they had developed as a result. They talked about their commitment to their communities, and to the lifestyle they had made for themselves, which made the idea of leaving unthinkable. The men, on the other hand, seemed more resigned to the economic decline and affected by the loss of and continuing lack of steady employment.  Just as the economic turbulence may disproportionately impact those of a younger generation, it may also disproportionately impact men. Employment and labour contributions to society are cornerstone of solidarity, and as work declines, so too declines their sense of engagement and integration with their communities, leaving them at a significant social disadvantage.  Men do not engage in as many casual or formal types of participation, have less socially cohesive attitudes, and thus have less social capital. Household Income and Education   The role of household income and education may parallel those of age as factors in which high levels of informal participation seem to be paired with lower levels of social cohesion. People with a lower socio-economic status would be harder hit by the economic recession along  71 the coast, and their increased vulnerability may lead them to be less trusting and feel less positively about their local environments. People who have a lower income and fewer credentials may feel undervalued as their careers fail to advance in the local economy, while polarization between those who have remained well-off and those who are relatively impoverished would increase. As Knack and Keefer (1997: 1282), Schuller (2001: 103) and Coburn (2000) argue, income inequality combined with excessive power imbalances can reduce the levels of trust and norms of cooperation in a community. In a recent article published by Young and Matthews (2007), they show how recent policy changes can contribute to these inequalities within communities. According to one of their interviewees (Young and Matthews 2007: 182), “there are so many small contractors because Interfor [International Forest Products] started to lay off employees and rehire them as contractors…It can work well for some people, like people with heavy equipment like trucks, but for a lot of people it means sitting on the edge [of financial ruin] and waiting for work that may or may not come”  These economic changes can be seen to exacerbate the distinction between the have and have- nots. As the economies of these communities become increasingly liberalized or market based, as Young and Matthews (2007) argue in their paper, social solidarity would also be increasingly determined by what Durkheim (1984) terms organic solidarity, whereby social cohesion would be built by the exchange of specialized skills and products. If people do not feel they have anything to exchange, they may feel excluded from this form of community integration and revert to types of mechanical solidarity based on similarity and shared interests and experience, signified by informal participation. Hence, this economic exclusion from broader society may concurrently lead members to “withdraw into the circle of family and friends” as de Tocqueville (2000: 78) envisioned, spending time together more privately.  While these types of association do ultimately engender greater feelings of social cohesion, as this research suggests, these effects  72 may be initially masked by the socio-economic conditions on coastal British Columbia that encourage some people to bond together in close networks while feeling less trusting, cooperative and positively integrated with their communities.  Keeping these circumstances in mind, it is likely that the correlation between formal participation and social cohesion is partially spuriously caused by the increased likelihood of people with more education and higher household income participating in these group activities and also having relatively higher feelings of trust, cooperation and a positive sense of community. The relatively high cohesive attitudes among these populations may simply be the result of their unnaturally low levels among people who have a smaller income and do not hold a degree. A second source of explanation reinforces a suspicion that the relationship between formal participation and attitudes toward leaders is spurious. These social groups may feel they have more political efficacy, making them feel better about their leaders, and also leading to more motivation to become actively engaged in the social lives of their communities. These social groups are also likely to have greater influence on leaders, or to be leaders themselves. People with more money are likely to have a greater influence on the business leaders in communities due to their higher spending power and are also more likely to be business leaders themselves. People with high levels of education may also have a greater ability to vocalize their concerns into productive political rhetoric. In other words, people with a higher socio-economic status may participate formally more and have more positive attitudes towards leaders because they embody aspects of leadership or are themselves leaders.  While these forms of participation may enhance positive attitudes towards leaders, it seems highly likely that the majority of the relationship is simply a manifestation of these other factors.  73 The relationship between formal participation and general social cohesion, alternatively, is significant once controlling for community, as well as income, education and the rest of the socio-demographic variables. A prudent but cogent argument would follow that the relationship between formal participation and social cohesion is a combination of independent influences of household income and education, and of a genuine influence of participation to which those socio-demographic factors are antecedent. A prudent, but also cogent line argument would follow that the relationship has aspects that are spurious and genuine. Certainly there is the literature to suggest that the relationship exists (Putnam 1993; 2000; Carpiano 2007/9; Stolle 1998; Veenstra 2002), often with the acknowledgement that participation is more likely to be undertaken by the wealthy and well-educated (Putnam 2000:94). People with a higher level of education may have absorbed a normative culture of group participation through a longer period in the educational system. People with more economic resources may have the time and social stability to feel more capable of committing consistently to an organized group. Certainly, there is no shortage of possible explanations for the increase in formal participation among these members of coastal populations. However, as alluded to in the discussion of formal participation at the beginning of this chapter, the discrepancies in levels of participation and cohesion can be problematic. If only certain segments of the population participate, feel engaged and efficacious with their leaders and trusting and integrated with their communities, it will significantly limit the economic, social and political vitality of these communities. Moreover, any rifts between people with a high and low socio-economic status caused by the neo-liberal restructuring of the coast will be exacerbated. These degrees of inequality are properties of communities, and can partly explain why controlling for community may conceal the relationship between participation and social  74 cohesion. However, this will be dealt with in more depth in the following section on the effects of community.  Ethnicity   Ethnicity plays a unique role in this process of social capital development as the only socio-demographic variable to conceal the effects of both informal and formal participation on the social cohesion variables. People from a North American Indian ethnic origin, whom I will also imperfectly4 refer to as Aboriginal or native people, participate both informally and formally more than others but their attitudes towards leaders and general sense of social cohesion are significantly lower. The explanatory line for this disjuncture may be similar to age, household income and education, that circumstances and shared experiences cause native people to participate more than average but feel less positive about their leaders, less trusting of others, and have a lower sense of community. While the conceptual source of these differences may be similar, substantively they are going to be quite distinct.  Like many countries all over the world, Canada is a colonial product. In British Columbia, as across Canada, people who had lived on the land for time immemorial were subjected to aggressive assimilation policies, which disrupted the social, economic and cultural fabrics of their communities. The enduring legacies of these policies include lower levels of educational attainment, higher unemployment and higher levels of poverty (Statistics Canada 2001). Moreover, the disadvantages have been institutionally reinforced and reproduced by  4 The use of the term Aboriginal or native is imperfect for a number of reasons. First, these terms were not included on the survey so the research participants did not identify themselves using this terminology. Second, it does not include people who identified themselves as Inuit/Eskimo or Métis although these ethnic groups could also be considered aboriginal or native.  75 inadequate or discriminatory health services, education systems, legal systems and state provided infrastructures, such as housing and water (Frideres 2001). Aboriginal people are also much more likely to be the victims and perpetrators of crime. According to Statistics Canada, 35 percent of Aboriginal people report having been the victim of a crime, and are most likely to have been the victim of a violent crime (Statistics Canada 2001). These circumstances could easily undermine attitudes of social cohesion among Aboriginal people in these communities who have more reason to be distrustful and independent.  Social exclusion and disadvantages may also encourage Aboriginal people to participate more, informally and formally. Just as younger people and those with a lower socio-economic status may spend more time with close family and friends and also feel less positive about their communities, Aboriginal people may seek the solace of their peers as a consequence of difficult circumstance and experiences, paired with experiences of social exclusion. On the other hand, informal and formal participation may be a representation of social unrest and activism. They may be heavily engaged with their kin and clan, reflecting patterns of socializing that are culturally specific to First Nations in British Columbia, while distancing themselves from the local community. Hence, the increased participation among Aboriginal people may not develop the same sentiments as non-Aboriginal people when they do participate, because the historical context of that participation is quire different. In fact, this project is grounded upon assumptions about what constitutes social capital that may be inappropriate for native populations. This suspicion is supported by the results of the interaction effects, and will be discussed more thoroughly in that section.   76 The Importance of Community   Theoretically, it would make sense that community would be antecedent to the relationship between participation and social cohesion. As discussed in Chapter 1, this research is based a model of social capital largely on the work of Putnam (2000), who suggests that levels of civic participation and the social networks of individuals are micro forms of social capital that are related to societal norms of trust and reciprocity and can be considered a macro form of social capital. The theoretical work of several scholars was considered in the development of this model, including Bourdieu (1985) and Sacks (2002: 148), who suggest that there is a reinforcing link between the relationships people build with one another, and the sense of solidarity and cooperation among community members. From this perspective, the relationship between participation and social cohesion would become a quality of communities, which would then also lead people to participate more, such that it would become antecedent to the development of social capital. Yet, the question that emerges with this research is not with the unique, small antecedent influence of community on formal participation and attitudes towards leaders, as this role should be expected. Instead, the oddity lies with the concealment of the other three relationships with the addition of community. Why would relatively high levels of participation and low levels of social cohesion, and vice versa, be found in particular communities?  I will touch on three interrelated explanations that probably configure to disrupt the relationship between participation and social cohesion in particular communities. First, the circumstances of coastal British Columbia have lead to social fragmentation within communities, whereby increased participation could reflect a desire to re-secure social resources. Second, increased social cohesion in a community may not actually encourage people to participate, and  77 instead, social dissatisfaction could lead to more civic engagement. These explanations illuminate a possible complexity in the feedback loop between micro and macro social capital. Third, levels of social inequality could prevent individual social actions from developing into more general facets of social solidarity as alluded to in the section on socio-demographic factors.  Just as the economic and social turbulence recently experienced in coastal British Columbia could lead particular community members to feel inordinately low levels of social cohesion, it could also lead particular communities to be less socially cohesive. The loss of businesses, deterioration of local infrastructures, declining job opportunities and increased vulnerability would vary according to the different levels of economic recession experienced in communities. Certainly these factors could lead people to feel more critical of local leaders, less trusting, and less content with their communities. This lack of social cohesion may in fact encourage people to increase their social engagement in their communities. A recent article by Hillman (2008), for example, discusses how elites developed diverse networks purposefully to mobilize in the English Civil War. Coalition building is one way in social discontent can actually motivate social participation. Social contentedness could, under some conditions, make people feel less inclined to build and enhance social networks. If people live in a community with high norms of cooperation and trust, and members are generally well integrated with community life, there may be less need for active socializing. More subtle forms of interaction such as chatting with people over the fence or in the grocery store may be enough. This would particularly be the case in communities that draw people who are looking for a quiet or reclusive lifestyle.  Both scenarios whereby low levels of social cohesion lead to increased participation or high levels of social cohesion lead to less participation add complexity to the idea that that there is a reciprocal relationship between participation and social cohesion. It may not always be the  78 case that the better people feel about their communities, the more likely they are to become engaged in community life. The feedback from social cohesion to participation may be more varied, and depend on a variety of conditions including a community’s history and current socio- political environment. While Putnam (2000) writes of waning political engagement in America, discontent with the current government generated the most engaged primary elections ever seen in the USA in 2008.  Lastly, social inequalities may prevent participation from generating better sentiments about the community overall. Regardless of how much time people spend together, building bonds and generating new relationships, if the community has significant rifts that fragment the population it would be difficult to see the community as a unified place. Participation may thus generate better feelings towards particular people in the community, but not a general sense of community wellbeing overall. Certainly, as the rifts are exacerbated by the restructuring of the resource industries, the social distances that exist between community members poses a significant impediment to the development of social capital.   The more isolated people think their communities are, the more they socially engage in a variety of activities but the less socially cohesive they feel. People that feel socially constrained by the location and limits of transportation in their communities may be more actively trying to overcome those constraints by being more engaged. Yet, it may be difficult for people to feel positively about their communities and socially embedded if they feel they have less mobility to go elsewhere. Their sense of relative deprivation may engender negative sentiments towards their communities, even if they do feel that their communities are otherwise positive places to live.   79  Interactions: The Effects of Participation Vary   As the previous sections discussed, socio-demographic characteristics can influence whether a person participates and generates social cohesion and can also lead people to participate a lot but have relatively low attitudes towards leaders, less trust and a lower sense of community. A similar disjuncture occurs in some communities that seem to host inverse pairings of participation and cohesion. While participation does generate social cohesion, embodying a process of social capital development, its practice is highly dependent on who is participating and where. The interaction tests add an important element to this discussion by showing that the effects of participation vary across socio-demographic and community factors. Some people who participate generate higher levels of social cohesion than others, contributing an additional layer to the story of social capital in coastal British Columbia.  Attitudes towards leaders and general social cohesion are indices of two different types of social capital reflecting institutional and general dimensions. Each are developed through different social processes, and conditioned by different types of factors. As will be discussed in the subsequent sections, the interaction tests show that more institutional factors such as education, marital status and work-status interact with participation in their impact on attitudes towards leaders. Alternatively, the effects of participation vary across ethnicity and household income to impact more general forms of social cohesion.  The effects of participation also vary across community. The relationship between formal participation and attitudes towards leaders is significantly stronger in some communities than in Powell River, in which it is relatively weak. However, the meanings that can be derived from these results are limited. The effects of informal  80 participation on social cohesion are also relatively stronger in some communities, and also particularly strong in two, important cases.  Socio-Demographic Factors Attitudes towards Leaders   The role of participation in generating more positive attitudes towards leaders seems to depend on structural and institutional contexts of their participation. The interaction tests show that for every additional casual social activity in which a degree-holder participates, they should feel significantly better about their leaders. In other words, the role of informal participation is shaped by education: if a community member holds a degree, they are much more likely to feel more positive about community leaders when they engage in more casual social activities. Perhaps people that hold a degree are more likely to actually spend time with leaders on a more casual basis.  As previously discussed, they are also more likely to be leaders who would socialize with other leaders  The effects of formal participation on attitudes towards leaders are shaped by two somewhat surprising factors, and one that is much more predictable: marital status, work-status and community. While neither marital status nor work-status can predict how people feel about their leaders, these attitudes vary significantly along each of these dimensions. People that have never been married or in a marriage-like relationship can not be expected to feel any better about their leaders when they engage in more group activities. Perhaps people that are single participate in these formal activities as a way to meet prospective partners, such that their participation leads to different social ends. However, it is equally likely that marriage expands a person’s social network through kinship ties and by creating a link between a person and his or  81 her partner’s ties. In theory, marriage could double a person’s social network. Margaret Mead (1935: 84) notes in her study of the Arapesh or Papua New Guinea, when asked about whether a man would marry his sister, her informant responded: What, you would like to marry your sister! What is the matter with you anyway? Don't you want a brother-in-law? Don't you realize that if you marry another man's sister and another man marries your sister, you will have at least two brothers-in-law, while if you marry your own sister you will have none? With whom will you hunt, with whom will you garden, whom will you go to visit?  In-laws and the expansion of social networks through marriage are valuable to the Arapesh by enabling them to engage in cooperative activities such as hunting and social niceties such as visiting. Bourdieu (1993: 33) sees the family as the main arena in which social capital is developed and transmitted. For people along the coast or British Columbia, the tests show that the expanse of social ties associated with marriage likely leads group activities to be more productive. The broader set of social relationships may mean that people that have been in a marriage or marriage-like relationship are more likely to meet new people, such as community leaders, through group activities. It is much easier to become familiar with people in group settings when mutual ties are shared. The expansion of a social network through marriage may provide people with an advantage in these formal environments by enabling them to more easily gain familiarity with community leaders.  The effects or work-status on the relationship between formal participation and attitudes towards leaders presents a more difficult puzzle to untangle. Largely, this is because there is no statistical distinction between those that are involuntarily unemployed and those who are retired. However, the mean age for people unemployed is 67, and only 25% are below the age of 57. Certainly, people who are 67 may be involuntarily unemployed, but it is reasonable to assume that a large proportion of people in this group are retired. Regardless of whether people are  82 retired or involuntarily unemployed, work provides an important forum from which to develop networks, engage in professional associations, and offer special skills to be incorporated into local industry.  As Putnam (2000: 80) notes, “work-related organizations, both unions and business and professional organizations, have traditionally been among the most common forms of civic connectedness in America.” Canada would certainly follow suit, particularly in these communities where forestry work and mining have been backed by strong unions. The stronger relationship between formal participation and social cohesion for unemployed people presents a bit of a puzzle.  Thinking about the relationship from another perspective might provide possible answers: people who work have no benefit in the development of social capital, when there should be a connection. What is startling about the results is not simply that people who are unemployed seem to generate more social cohesion when they participate, but that throughout all of the relationships between participation and social cohesion tested in the regression models, working seems to generate no advantage. The culprit may be the financial pressures and insecurity of working in coastal British Columbia. With increased stress of trying to gain a sufficient foothold in tumult of the local economy, these community members may be so time pressured that they feel less inclined to participate, more precarious in the social fabrics of their communities and therefore, without and significant benefit over those who are unemployed. Indeed, this stress may mean that even when participating, they feel anxious and perhaps more critical of community leaders. Considering again that many of the people who participate formally are likely to be leaders themselves, people who are leaders and in the workforce may be more impacted by local leadership than those are retired, and more accountable to their employees and colleagues, and therefore more guarded in their socializing with other leaders.  83  General Social Cohesion  As people increase in household income, so too increases the positive social benefits of their informal participation. Those with more resources may be able to commit more to their casual activities, enhancing their participatory experiences. The increased financial security may also encourage people to take greater social risks: they may be more willing to develop trusting relationships and engage in reciprocal exchange because they feel less vulnerable to opportunism. The greater inclination to engage in tests of trust, as Sacks (2002: 148) might say, may help to build the covenantal bonds in which individual action aggregates into more substantial social norms. The role of ethnicity in this research is unique as the only variable to interact with both informal and formal participation. Native people along the coast of British Columbia can be expected to generate significantly less general social cohesion than their non-native counterparts. In fact, while people generally can be expected to feel more trusting and have a better sense of community when they engage in additional social activities, the relationship is inverse for native people, with the striking implication that the more activities native people engage in, the less socially cohesive they may become. This is particularly important because the level of both types of participation is significantly higher for native people. Given the history of colonization in British Columbia, the low levels of trust and sense of community in native populations is understandable and somewhat expected. What is particularly interesting about these results is that engagement in the social life of the community does not help to remedy those attitudes. The distrust native people in British Columbia feel that potentially emerges from betrayal and bad faith may taint social participation, so that the  84 community building benefits do not accrue to them leading to further exclusion, alienation and negative perceptions. Moreover, native people may genuinely not experience social participation in the same ways as non-native people. Their unique history in coastal British Columbia may lead other community members to interact with them differently, and exacerbate any otherness they already feel. Alternatively, clan based social organization may lead native people to be extremely engaged and socially cohesive with a particular group of people determined by kinship and ancestry. These types of social groupings have been duly noted in early anthropologically studies of British Columbia First Nations (Swanton 1902; Sapir 1915; McIlwraith 1948; Lewis 1970) and can be recognized today in many First Nations communities. Indeed, gift-giving, cooperation and reciprocity, often considered cornerstone practices of social capital, are quite normative in coastal populations within clans and are profoundly observable in potlatch practices in which people will contribute a significant proportion of their wealth, sometimes the majority of their wealth, to their clan with the assumption that it would be reciprocated in the future. As a consequence native people may participate in their communities quite extensively, but find it quite unnatural to be trusting and cooperative with people outside of these kin groups. As Lewis (1970 :16) suggests of the cooperative clans of the Coast Salish, “this kinship structure , embodying a network of valued relationships, is one of the important keystones of the culture, as important as the terrain itself.” Moreover, he goes on to stress the primacy of blood kinship, in suggesting that “there was no other social grouping that superseded it; no other organized social institutions to which an individual could have recourse” (18). While many traditions of native culture have been undermined in colonial processes, the importance of kin are likely to remain to  85 some degree, and may still be more meaningful to Aboriginal people than the politically and geographically determined concept of community that is used in this research. The conceptualization of social cohesion as social capital might also be problematic in this context. Hunter (2000: 35-38) suggests that the concept of social capital may need to be modified to be appropriate for different cultural groups and social environments. In his discussion of social capital in the context of Aboriginal Australians, he suggests that the possible negative outcomes for aboriginal populations may be particularly damaging. With reference to Portes’ (1998: 8) four negative outcomes, Hunter notes that the exclusion of outsiders, excessive claims on group members, restriction on the freedoms of individuals, and downward levelling of norms may have more negative impacts on aboriginal communities. The exclusion of outsiders can limit the provision of public services to aboriginal communities, particularly in terms of education and health. However, as this study focuses on native people living in civic communities, and measures attitudes towards people beyond specific social groups, these issues would not be as applicable to this study. However, the potential excessive claims on members and restrictions on individual freedoms could certainly deter native people in British Columbia from engaging in reciprocal exchange and relationships of interdependency, as they may carry particular consequences. Colonial policies that fragmented First Nations and rendered them dependent on the state have meant that dependency and obligation, even if it is reciprocal, may not be social assets but social burdens and exacerbate vulnerability and mental health issues. Community   Not every community seems to foster the same sort of connection between participation and social cohesion. The results also suggest formal participation produces more or less social cohesion in some communities relative to others, while the influence of informal participation  86 does not show the same variability. When people participate more in Mill Bay, Kitimat, Queen Charlotte, Bowser, Alert Bay, Courtenay, Sechelt, Prince Rupert, Bella Coola, and Port Alberni, they feel significantly better about their community leaders than people who participate in Powell River. However, none of the communities show a significantly stronger relationship between participation and these attitudes above the other communities generally. In other words, someone participating in Mill Bay can be expected to feel better about their leaders than someone in Powell River that participates, but that person in Mill bay cannot be expected to generate more social cohesion than another from a randomly selected community. Hence, the implications of these findings are limited: some communities may foster a stronger connection than others between group activities and how people feel about their leaders, but no community really stands out. Instead, these finding simply suggest that the best connection is significantly better than the worst, rather than suggesting that one is particularly good. Given that this relationship is probably largely spurious to begin with, the differences between the movements from formal participation to positive attitudes towards leaders may simply be a representation of how well leaders work together in each community. In Powell River, the reference category, leaders may disagree on every facet of policy, so that the time they spend together in these formal activities only continues patterns of conflict. In Mill Bay, a relatively prosperous community quickly becoming a retirement destination, leaders and people with leadership qualities may get along relatively better than in Powell River. Alternatively, the changes in the relationship between informal participation and social cohesion are considerably more noteworthy. Where a person lives and participates can predict to which these actions will indeed lead him or her to feel more socially cohesive. Relatively speaking a number of communities produce significantly more social cohesion than Port Alberni,  87 the reference category, when they participate. The relationship is similar to that between formal participation and attitudes toward leaders in this regard, and simply suggests that the best is significantly better than the worst. Unlike formal participation, when the interactions between informal participation and community are retested as binaries, so that particular communities are tested against all the others showing the difference against a more general average, Mayne Island and Kitimat show quite significant and important results. On Mayne Island the influence of informal participation on social cohesion is very high. In fact, it is almost one-to-one, such that for every additional casual activity, someone on Mayne Island can be expected to increase their cohesive attitudes of trust and sense of community on the scale by almost one. In comparison to Mayne Island, the slope of the other communities is very small and insignificant. Similarly, informal participation is a strong predictor of general social cohesion in Kitimat, and comparatively quite a weak and insignificant predicator in others. These results show that the utility of informal types of participation in generating social cohesion is highly contingent on community. If Mayne Island and Kitimat were excluded from this study, there may be no relationship at all. What Does this Tell Us About Social Capital in Coastal British Columbia? How to Build a Cohesive Community  Under the rubric of social capital, Putnam (1993) put forth the idea that engagement is linked to democracy. I say, that depends on who is doing the participating. This research indicates that the link between informal and formal participation and positive attitudes towards leaders is tenuous. Among our respondents, people that participate informally really only have better attitudes towards leaders if they are degree holders. These people are likely to be leaders  88 themselves, leading to the suspicion that casual activities only make leaders, and people with leadership qualities, feel better about leaders. Formal participation leads to a similar type of conclusion. People that are older, more educated, and with higher household incomes are more likely to engage in formal types of activities and feel more positively disposed towards leaders. It seems quite likely that these people are also community leaders, suggesting simply that leaders feel better about leaders when they spend more time with one another. Such a system whereby particular socio-demographic segments of the population are more involved with their communities and more engaged in decision-making is highly problematic, as discussed in the first section.  More general forms of social cohesion are tied to both informal and formal participation. Overall, the role of informal participation in generating more social cohesion is more varied, and seems to depend quite heavily on the circumstances in which the socializing occurs. Schmoozing is a varied social action that reflects the characteristics of particular socio-demographic groups or community environments. Potentially, these practices of private, more intimate, social bonding can represent detachment, exclusion and somewhat anti-social behaviour. Yet, while intimate interaction can by intertwined with low levels of social cohesion, there is a significant underlying relationship illuminating the validity of types of expressive social actions for the development of general forms of social cohesion. Similarly, some socio-demographic populations are much more likely to participate in formal group activities and feel more trusting, more embedded in their communities, and more positive about their communities as places to live.   In some ways, these factors may simply lead to increased participation and higher levels of social cohesion independently, yet the significant underlying relationship that is elucidated once community is  89 controlled, above and beyond these socio-demographic factors, suggests that there is a substantial connection between group activities and social cohesion.  Capital and Return Deficits  Persons with certain social backgrounds participate more and feel more trusting, cooperative and have a better sense of community. People that are older, have a higher household income and hold a university or professional degree are likely to engage in instrumental social actions,  have more social ties across different networks, and feel more socially cohesive than people who are younger who have a low income and education. In some regards, this discrepancy is a capital deficit among those socio-demographic populations who are less likely to develop these social assets. Other socio-demographic groups have high levels of participation and low levels of social cohesion, leading to less social capital overall. While these people may have a lot of connections, a lack of trust and lack of propensity to cooperate would likely limit the ability for these people to capitalize on their networks. The ties developed and enhanced through these participatory practices are what Sandefur and Laumann (1998: 484) call “potential social capital,” whose value is determined by accessibility. Without the more ‘attitudinal’ aspects of social capital, the accessibility of the resources embedded in social networks is curtailed. However, this reduced level of social capital is not the result of a capital deficit (Lin 2000: 790 - 791), or a lack of investment. These social groups commit more time to socializing informally than others. Instead, explanations provided thus far have suggested that external factors may influence levels of participation and social cohesion in opposing directions. The result is that while participation does generate social cohesion for these populations,  90 circumstances and experiences have lead to particularly low levels of social cohesion. A second explanation is grounded in the concept of a return deficit (Lin 2000) through which similar levels of investment in social capital lead to different levels of return. Certainly, there is significant evidence to suggest that a return deficit exists for Aboriginal people in these coastal communities. While they participate more both informally and formally than non-Aboriginal people, they produce significantly less social cohesion. Similarly, people with a lower household income generate less social cohesion when they participate in casual, informal social activities, while they also participate in more informal activities than their wealthier counterparts. The lower level of social cohesion for these groups is not the result of less investment, but on less returns on investment. While these groups may have more social ties in terms of quantity, without a sense of trust and cooperation, these ties are less able to be mobilized. Place and Social Capital  The ways in which social capital is practiced and developed depends on the community in which a person resides. While communities on the coast share similarities and may be considered quite similar, in the arena of social capital their differences are significant and important. In some places, both informal and formal participation may be guided by a desire to build trust and change current circumstances, or inequalities may nullify any attempts to build more cohesive communities. In others, private social activities in more bound social networks may be a way to detach from the local community.  Yet on Mayne Island and Kitimat, these casual activities may be the best way to build stronger feeling of trust and a better sense of community. While much more research is needed to really explore these possible differences and  91 their sources, this research can be interpreted quite confidently to suggest that these differences between communities do exist and provide an important setting in which the story of social capital unfolds.  92 CHAPTER 5: CONCLUSION  Insights, Applications and Directions for Future Research The results of this research open a variety of empirical and theoretic doors about social capital, producing more questions and illuminating more gaps in knowledge. Some of these quandaries and quagmires will be posed at the end of this chapter. Yet meaningful conclusions about the nature of social capital can be drawn, particularly in the context of community development and resilience.  More conceptual insights about the variability of social capital development pave the way for practical applications of this research for development in coastal British Columbia. The theoretical undercurrent running throughout the project echo’s the work of other scholars while contributing an important insight: while several authors aim to contextualize social capital by considering how different race (Green, Tigges and Diaz 1999; Lin 1999), class (Campbell 1986; Lin 1999; Veenstra 2000), geographic (Onyx and Bullen 2000) and organizational (Burt 1992) positions can influence peoples stocks of social capital, or by placing different networks of social capital within broader societal contexts (Rubio 1997), this research considers how the process of social capital development is also dynamic and reflects different social contexts. The process I am referring to is not necessarily a linear sequence of events in time, but rather an interchange between two different dimensions of social capital that combined are valuable social resources. Social capital can be difficult to situate within broader notions of capital. It’s nebulous past, in which it emerged from a rather obscure text on education authored by Bourdieu (1985: 248) as “the aggregate of the actual or potential resources which are linked to possession of a  93 durable network of more or less institutionalized relationships of mutual acquaintance or recognition”, has characteristically endured. As Bourdieu’s definition exemplifies with the frequency of the “or” clause, a precise meaning is difficult to pin down. Yet scholars have attempted to iron out the meaning of social capital with some success. Nan Lin in particular makes significant contributions in this regard by describing the evolution of a network theory of social capital from the more classical works of Marx (1999a), and developing it as a theory of structure and action (2001). No small feat, defining social capital within broader sociological traditions can be difficult. Unlike economic capital, social capital is not tangible and cannot be read on a bank statement or seen it a warehouse. Unlike human and cultural capital, it is not captured within individuals. Skills and education can be housed in bodies, while cultural assets are largely manifest in behaviours. Social capital is unique, as an asset that exists among two or more people. It is the associations and relationships that exist in social space that generates benefits, and for this reason, social capital can be seen as a process, practice or dialogue that occurs between individuals and society. This research has attempted to approach social capital at an individual level, while also including the attitudes that are frequently associated with a more societal version of social capital that can be encapsulated in the concept of social cohesion. Literature on social capital has suggested that these two types of social capital operate reciprocally and generate a variety of positive returns (Putnam 2000). In considering social capital in this light, as a process involving individual actions and societal attitudes, this thesis shows that this processes is dynamic. The link between social participation and social cohesion does not occur invariably, but it is embedded in people’s diverse lives and shaped by who they are, socio-demographically speaking, and where they live.  94 Schmoozers and Machers Typically schmoozers, or those who spend a lot of time engaged in more casual activities, are demographically quite different from machers, those who engage in more formal types of activities. They are probably younger, poorer and less educated. The role of schmoozing within social capital development processes depends on the characteristics and location of the schmoozer. The typical schmoozer has relatively negative perceptions of community leaders and low levels of trust and sense of community, with the exception of women who engage in more formal activities and are more likely to feel more socially cohesive. For other schmoozers, their initial low levels of socially cohesive attitudes suggest that they may engage more casually and privately because they feel alienated from their communities. Schmoozing, in these cases, has particular role to play in processes of social capital development – one that may represent a type of fractured social action that may produce bonds, but may not bring people closer to their communities. In other circumstances, schmoozing seems to be a powerful avenue to fostering community solidarity. Schmoozers with a higher household income and who live in Kitimat or on Mayne Island have much higher levels of trust and a better sense of community than other schmoozers, just as degree-holding schmoozers are more likely to feel positively disposed towards leaders and politically efficacious.  Participating informally also seems to have completely different implications for native people in coastal communities, but will be discussed along with conclusions about broader contexts of social capital development.  Schmoozing can have multiple social meanings that reflect different social contexts in which it is practiced. The implications of participating in more formal group activities are similarly variable in ways that reflect socio-demographic factors. Unlike schmoozers, the typical machers of coastal communities, who are older, have a higher household income and hold a degree, also feel better  95 about their leaders and generally carry more socially cohesive attitudes.  In general, maching, seems to synchronize with social cohesion, but it only benefits those who mach, and these seem to be people who may already occupy a priviledged position in coastal communities. Hence, they are also people who may have a greater political and economic influence in their communities, which, as previously discussed, can potentially be problematic and exacerbate community inequalities.  Even within what seems like a relatively advantaged group of marchers, the participation of some is connected to more positive attitudes towards leaders. Interestingly, machers who are married or unemployed may feel more integrated with community leaders than others. Overall, developing social capital in communities may have more to do with impacting the contexts in which the process occurs rather than simply encouraging people to become schmoozers or machers.  Broader Contexts: Restructuring  The changes that have swept through coastal British Columbia impact the ways that social capital is developed through participatory processes. This unique social environment has provided a wealth of opportunities to gain insights into how social capital development is impacted by the contexts in which people live. While it is difficult to see exactly how these broader factors influence the relationship between participation and social cohesion, the impacts of socio-demographic and community factors on the models open ample room for informed speculation. I will briefly discuss two sites of speculation and interpretation. Firstly, economic disruption and discontent could instigate higher levels of participatory action while people feel less socially cohesive.  Secondly, there may be a link between employment and social capital that has more to do with the social identities gained through labour than the networks and money  96 that employment provides. These results show that the broader social contexts in which social capital is developed influence how these processes unfold in some potentially fascinating ways.    The effects of the restructuring in coastal British Columbia will have different impacts for communities along the coast. Some communities may be less dependent on primary resource industries, such that liberalizing the industries have limited impacts in comparison to other communities, some of which may be suffering the compound effects of lost opportunities in forestry and mining as well as restrictions in commercial fishing. As previously discussed, such changes can affect social capital in communities by exacerbating social inequalities that can create rifts between social classes, consequentially preventing participation from contributing to more generalized attitudes of trust and cooperation within the community. This could explain why high levels of participation seem to be paired with low levels of social cohesion in particular communities. However, it may also be the case that these changes and emerging social rifts have lead to a discontent in previously economically and socially vibrant communities, motivating increased social actions. Declining qualities of life on the coast and increased social fragmentation could actually instigate participation, suggesting that social capital need not always be developed through the calm pleasantries of cooperation. There may be no stronger manifestation of community attachment than criticism, anger and frustration when it starts to decay, and no stronger sign of commitment than the effort to change it amidst such emotions.  The second area that offers important implications lies in the potential relationship between employment and social capital. Underlying all of the social and economic changes in coastal British Columbia is simply the massive decline in resource jobs. Many of the people employed in resource work were born into it, in the sense that they grew up witnessing resource work, likely had a close family member in resource work, quite possibly began participating in  97 resource work before adulthood, and naturally adopted the vocation after high-school. As previously discussed, Durkheim (1984) suggests in the Division of Labour in Society that social integration is increasingly shifting from mechanical solidarity, determined by shared interests, to organic solidarity, determined by difference and specialized skills. Surely the neo-liberal restructuring along the coast has accelerated this switch by decreasing the number of shared types of labour and opening the community up to a more market-based economy. As a consequence, the specialized skills of the resource worker no longer make a valid contribution, and as a representative of those skills, as well as a culture and way of life, the resource worker may not have any avenue into these types of social integration. As a consequence, those whose livelihoods have been undermined by these changing social landscapes may participate in more mechanical forms of solidarity based on shared interests in ways that do not develop into the types of broad social cohesion in the community.  Empirically, this insight can be inferred from the impacts of socio-demographic factors on the regression models. Men have a significant social capital deficit, and they are predominantly impacted by the restructuring of the resource industries. The farther people are from retirement, the more they participate informally, but the less socially cohesive they feel, even after controlling for household income and work-status. This younger generation may not have had the opportunity to become socially integrated with their communities while engaged in a fruitful career and instead have been caught in a more precarious labour market. Lastly, people with more education and a higher household income develop more social cohesion than others when they participate, suggesting that those with a lower household income and education, most likely those involved in the resource industries, are not producing the same sense of trust and togetherness when socially engaged in their communities. While it is only possible to suspect  98 that there is a deeper underlying influence of employment on social capital, there is certainly the opportunity to fuel the fire of suspicion.  Aboriginal People in Coastal British Columbia The second broader context that is much more overt in this research relates to native people and social capital. The results suggest that when native people participate, they cannot be expected to feel any more socially cohesive with their communities, while both formal and informal participatory activities can be expected to coincide with higher levels of social cohesion for other community members. These results can be critically considered within broader social contexts leading to greater implications for coastal communities and the concept of social capital overall. Given the colonial histories in coastal British Columbia that undermined social organization in native communities rendered many native people dependent on the state, the concept of social cohesion may not be an appropriate measure of social capital. Native people may be disinclined to engage in relationships underscored with interdependency and obligation, given this history. Hence, ideals of cooperation, reciprocity and integration reflect particular ideologies that others should not be assumed to hold. Many native communities are also clan based, so that while trust, reciprocity and solidarity may exist within kinship, it may be unusual and counterintuitive to hold these types of relationships with members of the broader society. Hence, while native people along the coast may be disadvantaged in the development of social capital as it has been conceptualized in this thesis, I have made particular assumptions about advantageous social resources that may not in fact be ubiquitous throughout coastal populations. These assumptions are pervasive throughout  99 the literature on social capital, and as the case of native populations in coastal British Columbia shows, research could benefit from more reflexive, skeptical and critical consideration. Inequality in Social Capital  The results of the research show that there are social characteristics that influence whether people develop social capital through informal and formal participation, and also the extent to which that participation generates social cohesion. These differences reflect the circumstances of people’s lives that are tied to different socio-demographic and community factors in the context of coastal British Columbia. In conclusion, unequal distributions in social capital will follow its unequal development as a consequence of varying social circumstances. Hence, some populations may be disadvantaged in their ability to access social resources that would provide them with resilience in the economic wake of resource policy changes. These populations are also those who have other socio-economic disadvantage and are at the greatest risk. Younger people, those without a degree and a lower household income are less likely to be machers and more likely to be schmoozers, but also generally have lower levels of social cohesion despite their schmoozing. Men are less likely to engage socially in either activity, and less likely to feel socially cohesive in their communities. These populations are those who would be more likely to suffer the consequences of lost resource jobs and lack the economic and human capital to adapt and bounce back. This research shows that they are also likely to lack the social capital that can be so productive.  This research also illuminates some areas where social capital development could be improved using two types of approaches. Firstly, social capital could be enhanced by encouraging the circumstances that seem to engender participation and social cohesion. In particular, encouraging education and fostering positive community environments seem to  100 emerge as two possible sites of intervention. Secondly, the ways that particular circumstances influence the development of social capital also presents an opportunity to improve social resources. Altering the institutions that shape formal participation presents one compelling avenue to developing more integrated and inclusive coastal communities. These applications for this research will be discussed in greater depth in the following section. Applications: Building foundations for Social Capital Development  Encouraging the Conditions for Social Capital Development From these more theoretic contributions emerges a variety of substantial implications. Social capital, as a process, works in conjunction with other conditions, so that policies aimed at enhancing social capital would be most productive were they to cultivate the conditions it complements. While participation in community life alone is linked to positive attitudes towards leaders, trust and cooperative attitudes, some people engage in these processes more than others and some who engage seem to produce much different effects. Particular conditions seem to be more conducive for social capital practices than others.  It would follow that encouraging the development of social capital in conjunction with these other conditions could be a powerful way to increase social capital along coastal British Columbia. Education as an Avenue to Social Assets People that are older and occupy a higher socio-economic status, that is, people with a higher household income and education, seem to be in a better position for social capital development overall. They feel relatively positive about their leaders, have higher levels of trust, and a better sense of community. Moreover, people with a higher SES are more likely to have  101 more cohesive attitudes than others when schmoozing and maching, suggesting that higher levels of wealth and education can provide fertile ground for social capital development. Moreover, there is certainly the opportunity to improve public education along the coast and is an important focal point given recent debates about rural education in British Columbia.  North Island College was recently criticized for their decisions to cut programs in Port Hardy due to inadequate funding. While local protests seemed to influence decision-makers, it seems coastal education is constantly on the chopping block. This research implies that there is an additional reason to keep programs open: social capital is one byproduct of education, not only as a factor that encourages people to participate in their communities, but also as a factor that enhances the community benefits of participation. Facilitating secondary education may be a particularly productive avenue to developing social capital, and thus community resilience, along the coast of British Columbia. Community as a Conductor of Schmoozing In Kitimat and on Mayne Island, schmoozing really brings people together, and fosters the type of sentiments conducive to community cohesion and solidarity. In others, the effects of schmoozing may be negligible in these regards. Along this particular dimension of social capital development, community can be seen as a conductor, shaping the dynamic between interaction and attitudes. Particular community environments are good conductors, such that schmoozing is more beneficial, and others may be poor conductors. Elucidating the qualities that place a community in either camp is beyond the scope of this project, and yet the results of this research suggest that such qualities do exist. It would be advantageous to identify those qualities and foster those community environments that conduct schmoozing in ways that positively enhance social cohesion.  102 Targeting Missing Links (Not Everyone is Going to Get a Degree)  There are factors that shape whether people participate and invest in social capital and also the ways in which social capital is generated through participatory processes. Those factors that have positive influences should probably be encouraged, particular through public education and by encouraging community environments that are good conductors of social capital. Yet, there are limits to how far this approach to social capital can be applied: there are no policies that can speed aging, and it would be quite problematic to suggest that social capital would flourish if only community members gained a higher SES, got married and retired. Creating a homogenous community is not a goal or a reasonable policy foundation.  While it is reasonable to encourage high school graduation and an decent standard of living, people who choose not to pursue secondary education or who enjoy occupations that garner less income should have an equal opportunity to engage in community life, become leaders, and feel like trusting, contributing members of a positive social environment. Community social capital and resilience could probably be improved if more disadvantaged people were encouraged to become machers; that is, if younger people with less education and a lower income were encouraged to engage in more formal types of public, group activities and assume leadership roles in their communities.  People with a higher household income are more likely to engage in formal group activities and have higher levels of social cohesion, whereas people with lower household incomes are less likely to get involved with these formal activities and generate cohesive attitudes. This may be due to the potential commitment of resources, both time and money, required for participation in more structured group activities. For people who may be struggling financially or engaged in more precarious forms of employment, making such commitments may be difficult. These difficulties could be alleviated by creating subsidies for group activities or by  103 developing group activities with more flexible time commitments, such as “drop-in” programs. Creating more opportunities for people to become engaged in their communities would provide people who may feel economically excluded to contribute other resources to their communities. The idea of contribution may be particularly potent for community development in a place that is increasingly shaped by organic solidarity, or social cohesion fostered through the exchange of goods and services. As Fuller, Kershaw and Pulkingham (2008: 163) show in their research on voluntarism among welfare recipients, the ability to contribute labour and knowledge to a community can foster a sense of social integration and self-worth among people from more disadvantaged social positions. If people with a lower socio-economic status become involved in various community organizations in ways that enable them to contribute they may also feel more integrated with their community and thus develop their own social capital and enhance social cohesion in their communities. Future Research The reader may disagree with many of the conclusions drawn in this chapter.  Certainly, I could be wrong about native people and social capital, and native people may not feel more socially cohesive when they participate because they experience racism, or because the effects of colonialism have been so devastating. Indeed, many of the arguments presented are yet to be empirically substantiated.  This thesis has empirically supported the claim that processes of social capital development vary according to identifiable factors, and as such, some important conclusions about developing social capital on the coast can be drawn. More research is needed to really understand why these differences exist. There are four questions that jump out of this thesis as areas for future research. Firstly, the low levels of social cohesion amongst the typical schmoozer could provide the foundation for  104 considerable research. Why do younger people, people with less education or a lower household income participate so much informally, but feel less trusting, integrated and positive about their communities? I suggested that it is somehow related to an inclination towards more mechanical forms of solidarity for vulnerable populations who are unable to engage in a social economy increasingly based in organic solidarity. An important, interesting and relevant analysis could qualitatively consider whether this is true.   Secondly, it is entirely counterintuitive and inconsistent with other research that machers who are unemployed would have significantly more positive attitudes towards leaders. I am at odds to imagine a possible explanation. The anomaly of retired machers in coastal British Columbia could certainly generate a significant contribution to literature on social capital as well as labour and the life-course. Thirdly, there is some characteristics of Mayne Island and Kitimat that make these communities particularly effective conductors linking schmoozing and social cohesion. What is it about these communities that is so effective? Lastly, more research is needed to explain why native people, who typically participate informally and formally so much more than others, cannot be expected to feel more socially cohesive. This last question is particularly important, as the results may illuminate significant deficits in social capital along the coast or inadequacies in the concept of social capital, or potentially both. Summary  This research has approached social capital as a multidimensional concept, involving people’s formal and informal types of participation in relationship with their socially cohesive attitudes. The outcomes have illuminated the variable nature of the interchange between participation and social cohesion, suggesting that the contexts in which participation occurs shapes the ways that social capital will develop. Moreover, this research has suggested that a  105 definition of social capital is not culturally ubiquitous, but different cultural practices and forms of social organization influences what can be considered productive and beneficial. For Aboriginal people in coastal British Columbia, the concept of social capital and how it is operationalized may need some adjustment. By implication, social environments are hidden actors that aid or hinder the development of social capital, and fostering social capital in communities has as much to do with changing the qualities of these environments as encouraging participation. This research suggests that such sites for fostering social capital in coastal British Columbia exist in public education, institutions that shape group participation, and in the qualities of particular communities. These arenas foreground the symbiosis between participation and social cohesion, and could enable people to get more out of their participatory activities. Fostering this relationship would likely encourage people to participate more, and generate more social capital as both networks and attitudes, weaving a stronger social web in coastal communities. This web facilitates economic exchange and opportunities, contributes to health and wellbeing, enhances the cogency of policy, encourages innovation among leaders, and provides people with social safety net in times of crisis. In other words, enhancing this relationship is a key to development and resilience along coastal British Columbia.  106 BIBLIOGRAPHY  Adger, W. N. 2003. "Social Capital, Collective Action, and Adaptation to Climate Change." Economic Geography 79(4):387-405. Beggs, J. J., V. A. Haines and J. S. Hurlbert. 1996. "Revisiting the rural-urban contrast: Personal networks in nonmetropolitan and metropolitan settings." Rural Sociology 61(2):306-325. Bourdieu P . 1985 . The forms of capital. In Handbook of Theory and Research for the Sociology of Education, ed. JG Richardson , pp. 241 – 58 . New York : Greenwood Bradbury, A. S. C. 2006. "Transport, mobility and social capital in developing countries." Engineering Sustainability 159(2):79-86. Briggs, X. D. S. 1998. "Brown Kids in White Suburbs: Housing Mobility and the Many Faces of Social Capital." Housing Policy Debate 9(1):177–221. Burt, Ronald S. 1992. Structural holes : the social structure of competition. Cambridge, Mass.: Harvard University Press. Campbell, K. E., P. V. Marsden and J. S. Hurlbert. 1986. "Social Resources and Socioeconomic Status." Social Networks 8(1):97-117. Carpiano, Richard M. 2007/9. "Neighborhood social capital and adult health: An empirical test of a Bourdieu-based model." Health & Place 13(3):639-655. Castells, Manuel. 2000. The rise of the network society. 2nd ed. Malden, MA: Blackwell Publishers. Coburn, D. 2000. “Income inequality, social cohesion and the health status of populations: the role of neo-liberalism.” Social Science and Medicine 51: 135-146. Coleman, James. 1990. Foundations of Social Theory. Cambridge : Belknap Press of Harvard Univ. Press de Tocqueville, Alexis and J. P. Mayer. 2000. Democracy in America. New York: HarperPerennial. Devine, Fiona, Nadia J. Britton, Peter Halfpenny and Rosemary Mellor. 2003. "Family and Community Ties in Space and Time." Pp. 172 - 186 in Social Relations and the Life Course, edited by G. Allan and G. Jones. Hampshire UK: Palgrave Macmillan. Durkheim, Emile. 1984. The Division of Labour in Society. Basingstoke: Macmillan.  107 Enns, S. (2008) There is no other land, there is no other life but this: An investigation intothe impact of gender on social capital and resilience in four rural, island communities of British Columbia. (Doctoral dissertation, University of British Columbia). Available at https://circle.ubc.ca//handle/2429/1009. Enns, S., T. Malinick and R. Matthews. 2008. "It’s Not Only Who You Know, It’s Also Where They Are: Using the Position Generator to Investigate the Structure of Access to Embedded Resources."in Social Capital: Advances in Research, edited by N. Lin and B. Erickson. New York: Aldine De Gruyter. Frideres, J. 2001. Aboriginal Peoples in Canada: Contemporary Conflicts. Toronto : Prentice Hall Fuller, S,, Pulkingham, J., and P. Kershaw. 2008. "Constructing' Active Citizenship':Single Mothers, Welfare, and the Logics of Voluntarism". Citizenship Studies. 12(2): 157-176. Giddens, Anthony. 1990. The consequences of modernity. Stanford, Calif.: Stanford University Press. Granovetter, M. S. 1973. "The Strength of Weak Ties." The American Journal of Sociology 78(6):1360-1380. Green, G. P., L. M. Tigges and D. Diaz. 1999. "Racial and ethnic differences in job-search strategies in Atlanta, Boston, and Los Angeles." Social Science Quarterly 80(2):263-278. Hardin, Russell. 2002. Trust and Trustworthiness. New York: Russel Sage. Hobbes, Thomas and Richard Tuck. 1991. Leviathan. Cambridge England ; New York: Cambridge University Press. Hunter, B.H. 2000. Social exclusion, social capital, and Indigenous Australians: Measuring the social costs of unemployment. Centre for Aboriginal Policy Research: Discussion Paper 204. Available at http://www.anu.edu.au/caepr/Publications/DP/2000_DP204.pdf Jaccard, J. and R. Turrisi. 2003. Interaction Effects in Multiple Regression. London: Sage Jenson, J. 1998. Mapping Social Cohesion: The State of Canadian Research. Canadian Policy Research Networks Inc. Kawachi, I., B. P. Kennedy, K. Lochner and D. Prothrow-Stith. 1997. "Social capital, income inequality, and mortality." American Journal of Public Health 87(9):1491-1498. Kliksberg, B. 1999. "Social Capital and Culture: Master Keys to Development." Cepal Review 69:83-102. Knack, S. and P. Keefer. 1997. "Does Social Capital Have an Economic Payoff? A Cross- Country Investigation." The Quarterly Journal of Economics 112(4):1251-1288.  108 Lavis, J. and G. Stoddart. 2003. "Social Cohesion and Health." Pp. 121-149 in The Economic Implications of Social Cohesion, edited by L. Osberg. Toronto: University of Toronto Press. Lewis, C. 1970. Indian Families of the Northwest Coast: The Impact of Change. Chicago: University of Chicago Press Lin, N.. 2001. "Resources, Motivations and Actions: The Action Foundation." Pp. 41 in Social Capital: A Theory of Social Structure and Action"Resources, Motivations and Actions: The Action Foundation." Cambridge UK: Cambridge University Press. ------.2000. "Inequality in Social Capital." Contemporary Sociology 29(6):785-795. ------.1999a. "Building a Network Theory of Social Capital." Connections 22(1):28-51. ------. 1999b. "Social Networks and Status Attainment." Annual Review of Sociology 25(1):467- 487. Lin, N. and M. Dumin. 1986. "Access to occupations through social ties." Social Networks 8(4):365-385. Matthews, R. and R. Côté. 2005. "Understanding Aboriginal Policing in a Social Capital Context." Social Capital in Action: Thematic Policy Studies:134-152. Maxwell, Judith. 1996. "Social Dimensions of Economic Growth." Eric John Hanson Memorial Lecture Series, Volume 8, University of Alberta. McIlwraith, T.F. 1948. The Bella Coola Indians, Volume 1. Toronto: University of Toronto Press. Mead, M. 1935. Sex and Temperament in three primitive societies. New York : William Morrow Narayan, D. and L. Pritchett. 1999. "Cents and Sociability: Household Income and Social Capital in Rural Tanzania." Economic Development and Cultural Change 47(4):871-897. North, D. C. 1990. Institutions, Institutional Change and Economic Performance. Cambridge: Cambridge University Press. Onyx, J. and P. Bullen. 2000. "Measuring Social Capital in Five Communities." The Journal of Applied Behavioral Science 36(1):23-42. Portes, A. 1998. ‘Social capital: Its origins and applications in modern sociology’, Annual Review of Sociology 24(1):1–24. Portes, A. 2000. “The Two Meanings of Social Capital.” Sociological Forum 15(1) 1-12. Putnam, Robert D. 2000. Bowling alone: the collapse and revival of American community. New York: Simon & Schuster.  109 Putnam, Robert D., Robert Leonardi and Raffaella Nanetti. 1993. Making democracy work : civic traditions in modern Italy. Princeton, N.J.: Princeton University Press. Rubio, Mauricio. 1997. "Perverse Social Capital-Some Evidence from Colombia." Journal of Economic Issues 31(3). Sacks, Jonathan. 2002. The dignity of difference : how to avoid the clash of civilizations. London ; New York: Continuum. Sampson, R. J. 2001. "Crime and public safety: Insights from community-level perspectives on social capital." Pp. 89-114 in Social Capital and Poor Communities, edited by S. Saegert, J.P. Thompson and M.R. Warren. New York: Russel Sage Foundation. Sandefur, R.L. and E.O. Laumann. 1998. “A Paradigm for Social Capital.” Rationality and Society 10(4) 481-501 Sapir, E. 1915. “The Social Organization of the West Coast Tribes” Proceedings and Transactions of the Royal Society of Canada, Section 2, Series 3, Volume 9: 355-374. Schuller, T. 2001. “The Complementary Roles of Human and Social Capital” Isuma: Canadian Journal of Policy Research 2(1): 89-106. Available at dgra/publications/books/oecd/en/6-schuller.pdf Social Capital Initiative. 1998. The Initiative on Defining, Monitoring and Measuring Social Capital: Overview and Program Description. Washington, DC: World Bank. Stanley, D. 2003a. It takes two to bowl: Untangling the concepts of social cohesion and social capital. Ottawa: ON: Strategic Policy and Research Department of Canadian Heritage. ------. 2003b. "What Do We Know about Social Cohesion: The Research Perspective of the Federal Government's Social Cohesion Research Network." Canadian Journal of Sociology 28(1):5-19. Statistics Canada. 2001. Aboriginal Peoples in Canada. Canadian Centre for Justice Statistics Profile Series. Extracted June 26, 2008 from http://www.statcan.ca/english/research/85F0033MIE/85F0033MIE2001001.pdf. Stolle, Dietlind. 1998. "Bowling Together, Bowling Alone: The Development of Generalized Trust in Voluntary Associations." Political Psychology 19(3, Special Issue: Psychological Approaches to Social Capital):497-525. Swanton, J. 1905. “Social Organization of the Haida,” Proceedings of the International Congress of Americanists, 13th Session, NY, 1902. Eschenbach  Printing: Easton PA. Tönnies, Ferdinand. 1957. Community & Society (Gemeinschaft und Gesellschaft). East Lansing: Michigan State University Press.  110 Van der Gaag, Martin and Tom A.B. Snijders. 2005. “The Resource Generator: social capital quantification with concrete items.” Social Networks 27(1): 1-29. Veenstra, Gerry. 2005. "Location, location, location: contextual and compositional health effects of social capital in British Columbia, Canada." Social Science and Medicine 60:2059-2071. ------. 2002. "Explicating Social Capital: Trust and Participation in the Civil Space." Canadian Journal of Sociology 27(4). ------. 2000. "Social capital, SES and health: an individual-level analysis." Social Science Medicine 50(5):619-629. Weber, Max. 2004. “Basic sociological concepts” Pp. 311-358 in The Essential Weber, edited by Sam Whimster. London: Routledge. Wilkinson, Richard G. 1996. Unhealthy societies : the afflictions of inequality. London ; New York: Routledge. Woolcock, M. and D. Narayan. 2002. "Social Capital: Implications for Development Theory, Research, and Policy." The World Bank Research Observer 15(2):225-249. Young, Nathan and Ralph Matthews. 2007. "Resource economies and neoliberal experimentation: the reform of industry and community in rural British Columbia." Area 39(2):176-185.   111 APPENDIX 1: SURVEY QUESTIONS FOR SOCIO- DEMOGRAPHIC VARIABLES  Included below is a list of survey questions that inform the socio-demographic variables for this study, and a note on how these questions are developed into variables.  Age  What is your year of Birth?  *Age is calculated by subtracting the year of birth from 2001, the year when the study was executed.  Gender  What is your gender?  - Female - Male  *Gender (male) is one for respondents who marked “Male” and zero for respondents who marked “Female.”  Religiosity  What is your religious background or affiliation?  - Protestant - Roman Catholic - Muslim - Jewish - None - Other (please specify)  *Religiosity is valued at zero for respondents who marked “None” and one for all others.  Marital Status  - Married - Living with a partner - Separated - Divorced - Widowed - Never Married  112  * Marital Status (Single) is equal to one for all respondents who marked “Separated,” “Divorced,” “Widowed” or “Never Married,” and zero for those “Married” or “Living with a partner.”   Household Income  Which of the following categories best describes your household income for the past year? (please select only one)  - Less than $5,000 - $5,000 to $9,999 - $10,000 to $19,999 - $20,000 to $29,999 - $30,000 to $39,999 - $40,000 to $49,999 - $50,000 to $59,999 - $60,000 to $69,999 - $70,000 to $79,999 - $80,000 to $89,999 - $90,000 to $99,999 - $100,000 and over  *Treated as a continuous variable  Ethnicity  To which ethnic or cultural group (s) did your ancestors belong? (Please select ALL that apply)  - African - Canadian - Chinese - Dutch (Netherlands) - English - French - German - Inuit/Eskimo - Irish - Italian - Jewish - Metis - North American Indian - Polish - Portuguese - Scottish - South Asian - Spanish - Ukrainian - Welsh  *Ethnicity (North American Indian) is valued at one for all respondents who marked “North American Indian” and zero for all other respondents.  Work Status  What is your current work situation? (Please select ALL that apply) - Self-employed full-time - Self-employed part-time - employed by others full-time - employed by others part-time - not working or laid off  * Workstatus(unemployed) is one for all respondents who marked “not working or laid off,” and zero for all other respondents.   Education  Please indicate the highest level of education completed by you, by your mother and by your father. Please select only ONE choice per person.  No Schooling Some elementary School Completed elementary school (Grade 8) Some secondary/high school Secondary/high school graduation certificate or equivalent Some university Bachelor’s degree Master’s degree Professional degree or doctorate Other (please specify)  *Education (Degree) equals one for every respondent who marked “Bachelor’s degree,” “Master’s degree,” or “Professional degree or doctorate,” and zero for all other respondents.     114 APPENDIX 2: ETHICS CERTIFICATE 


Citation Scheme:


Citations by CSL (citeproc-js)

Usage Statistics



Customize your widget with the following options, then copy and paste the code below into the HTML of your page to embed this item in your website.
                            <div id="ubcOpenCollectionsWidgetDisplay">
                            <script id="ubcOpenCollectionsWidget"
                            async >
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:


Related Items