Open Collections

UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

The social emergence of health : a theoretical interpretation and empirical application of Pierre Bourdieu's… Burnett, Patrick John 2011

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

Notice for Google Chrome users:
If you are having trouble viewing or searching the PDF with Google Chrome, please download it here instead.

Item Metadata


24-ubc_2012_spring_burnett_patrick.pdf [ 816.98kB ]
JSON: 24-1.0072453.json
JSON-LD: 24-1.0072453-ld.json
RDF/XML (Pretty): 24-1.0072453-rdf.xml
RDF/JSON: 24-1.0072453-rdf.json
Turtle: 24-1.0072453-turtle.txt
N-Triples: 24-1.0072453-rdf-ntriples.txt
Original Record: 24-1.0072453-source.json
Full Text

Full Text

THE SOCIAL EMERGENCE OF HEALTH: A THEORETICAL INTERPRETATION AND EMPIRICAL APPLICATION OF PIERRE BOURDIEU'S RELATIONAL THEORY OF SOCIAL ACTION IN A THREE-DIMENSIONAL CANADIAN FIELD  by Patrick John Burnett B.A., Simon Fraser University, 2009  A THESIS SUBMITTED IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF  MASTER OF ARTS in The Faculty of Graduate Studies (Sociology)  THE UNIVERSITY OF BRITISH COLUMBIA (Vancouver)  December 2011  © Patrick John Burnett, 2011  ii Abstract Pierre Bourdieu's theory of social action has been the inspiration for an array of diverse health studies seeking to better understand the nature of social stratification and its relation to health behaviours and outcomes. While several of his well-known theoretical concepts, such as social capital, cultural capital and habitus, have garnered a great deal of attention in the health research community, the nature of their application has for the most part been limited to deterministic schemas examining relationships between social position and social action. There are as yet no health-related studies that offer a comprehensive theoretical account of Bourdieu’s ‘constructivist structuralism,’ incorporating all of his theoretical conceptions of field, habitus, capital, doxa and time. In light of these theoretical and empirical oversights, I offer a health-relevant re-envisioning of Bourdieu's expansive body of work and examine the implications of his relational framework for health research. Drawing upon a relational exploratory analytic method called multiple correspondence analysis and using original Canadian survey data from Vancouver and Toronto, Canada, I translate my interpretation of Bourdieu’s theoretical principles into a thoroughly Bourdieusian empirical depiction of a health-relevant three-dimensional geometric social space. The visual mapping of social space revealed seven different groupings of individuals whose common attributes and dispositions are socially patterned around health-related behaviours and outcomes, illuminating distinct spaces of social differentiation within which healthy and unhealthy individuals are located.  iii Preface The data used in this thesis were collected in 2009 by Gerry Veenstra (University of British Columbia, Department of Sociology) under the auspices of a research project titled 'Social class and health: Innovative theoretical exploration and empirical confirmation.' Funding for this research was awarded to Dr. Veenstra by the Social Sciences and Humanities Research Council of Canada. This research was approved by the Behavioural Research Ethics Board at the University of British Columbia (BREB Number: H08-02412).   iv Table of Contents Abstract .................................................................................................................................... ii  Preface ..................................................................................................................................... iii  Table of Contents ................................................................................................................... iv  List of Tables ........................................................................................................................ viii  List of Figures ......................................................................................................................... ix  Acknowledgements ................................................................................................................. x  Dedication ............................................................................................................................... xi  Chapter  1: Introduction ........................................................................................................ 1  1.1  Theoretical and Empirical Landscape ....................................................................... 1  Chapter  2: Bourdieu, Relational Theory and Health Research ........................................ 8  2.1  Introduction: Pierre Bourdieu's Theoretical Paradigm ............................................. 8  2.2  Theoretical Foundations.......................................................................................... 10  2.3  Relational Philosophy ............................................................................................. 15  2.4  Fields ....................................................................................................................... 17  2.5  Capitals ................................................................................................................... 21  2.6  Habitus .................................................................................................................... 25  2.7  Doxa and Time ........................................................................................................ 28  2.8  Conclusion: Bourdieu’s Relational Theory of Social Action ................................. 33  Chapter  3: Methods for Building and Analyzing a Canadian Social Space ................... 35  3.1  Introduction ............................................................................................................. 35  3.2  Analytical Principles of Methodological Relationalism and Field Theory ............ 35  3.3  Multiple Correspondence Analysis ......................................................................... 37   v 3.4  Strategy for Multiple Correspondence Analysis ..................................................... 38  3.4.1  Step 1: Designating Active and Passive Factors ................................................. 38  3.4.2  Step 2: Coding and Cleaning Active and Passive Variables ............................... 40  3.4.3  Step 3: Determining the Number of Dimensions ................................................ 42  3.4.4  Step 4: Exploring and Interpreting a Social Space ............................................. 44  3.5  Sampling and Data Collection ................................................................................ 46  3.6  Operationalization of Social Factors ....................................................................... 46  3.6.1  Conditions of Existence ...................................................................................... 47  3.6.2  Demographic Composition ................................................................................. 48  3.6.1  Dispositions, Values, Attitudes and Tastes ......................................................... 48  3.6.1  Practices and Behaviours .................................................................................... 51  3.6.2  Health Indicators ................................................................................................. 54  3.7  Conclusion .............................................................................................................. 55  Chapter  4: Analysis of a Canadian Field ........................................................................... 57  4.1  Results ..................................................................................................................... 57  4.2  Dimensions of the Field .......................................................................................... 58  4.2.1  Dimension 1: Dispositions .................................................................................. 60  4.2.2  Dimension 2: Positions and Position-Taking ...................................................... 60  4.2.3  Dimension 3: Time ............................................................................................. 62  4.2.4  Dimension 4: Gender .......................................................................................... 64  4.2.5  Final Remarks on the Field as a Space of Play ................................................... 65  4.3  Presenting the Distinct Groups of Social Factors in the 3D Field .......................... 66  4.3.1  Group 1: Wealthy, Well-Educated Professionals with Refined Tastes .............. 68   vi 4.3.2  Group 2: Young and Hip with Educated Parents ................................................ 68  4.3.1  Group 3: Older, Cultured and Opinionated ......................................................... 69  4.3.2  Space of Freedom: Groups 1, 2 and 3 ................................................................. 71  4.3.1  Group 4: Poor, Uneducated Labourers with Strong Dislikes and Poor Health .. 71  4.3.1  Group 5: Young and Single with Health-Damaging Dispositions...................... 73  4.3.1  Space of Necessity: Groups 4 and 5 ................................................................... 74  4.3.1  Group 6: High-Stress Workaholics with Unhealthy Dispositions ...................... 76  4.3.1  Group 7: Elderly Unstressed Homeowners ......................................................... 79  Chapter  5: The Field of Power ........................................................................................... 81  5.1  Introduction ............................................................................................................. 81  5.2  Healthy and Unhealthy Spaces in the Canadian field ............................................. 81  5.2.1  Space of Freedom and Health ............................................................................. 82  5.2.2  Space of Necessity and Health ............................................................................ 84  5.2.3  The Opposing Spaces in Between Freedom and Necessity ................................ 86  5.2.4  Limitations and Key Points to Consider ............................................................. 88  5.3  Conclusion and Thoughts for Future Research ....................................................... 93  Appendices ........................................................................................................................... 105  Appendix A ....................................................................................................................... 105  A.1  Frequencies of Conditions of Existence Variables ........................................... 105  A.2  Frequencies of Demographic Composition Variables ...................................... 106  A.3  Frequencies of Practice and Behaviour Variables ............................................ 106  A.4  Frequencies of Health Indicator Variables ....................................................... 107  A.5  Frequencies of Values, Attitudes, Tastes and Disposition Variables ............... 108   vii Appendix B ....................................................................................................................... 110  B.1  Broad View of Groups 1, 2 and 3 ..................................................................... 110  B.2  Broad View of Groups 4 and 5 ......................................................................... 111     viii List of Tables Table 1: Conditions of existence variables with mapping labels ............................................ 49 Table 2: Demographic composition variables and mapping labels ........................................ 50 Table 3: Value, attitude, taste and belief variables with mapping labels ................................ 52 Table 4: Practice and behaviour variables and mapping labels .............................................. 55 Table 5: Health indicators and mapping labels ....................................................................... 56 Table 6: List of influential variables by dimension ................................................................ 59 Table 7: List of influential categories by dimension .............................................................. 63    ix List of Figures Figure 1: General view of category point distributions in a 3D field ..................................... 67 Figure 2: Group 1 characteristics ............................................................................................ 70 Figure 3: Groups 2 and 3 characteristics................................................................................. 72 Figure 4: Group 4 characteristics ............................................................................................ 75 Figure 5: Group 5 characteristics ............................................................................................ 77 Figure 6: Group 6 characteristics ............................................................................................ 78 Figure 7: Group 7 characteristics ............................................................................................ 80   x Acknowledgements I offer my enduring gratitude to the faculty, staff and my fellow students in the Department of Sociology at the University of British Columbia and past institutions, particularly Sharon Lebenkoff, François Lachapelle, Janine Althorp and Zoe St. Aubin who have inspired and supported me intellectually and emotionally along the way. I owe particular thanks to Dr. Gerry Veenstra, whose continued support through the tough times and constant encouragement to think big, work hard and have fun, have helped me to produce a thesis that I can be proud of. I also wish to thank Dr. Richard Carpiano for his wonderfully detailed, thoughtful and thought-provoking comments along the way that encouraged me to think differently about theories and concepts I have all too often taken for granted.  I owe much gratitude to my friend and mentor Chris Atchison. His selfless and altruistic view on life coupled with his ability to instil a sense of power in those around him has influenced my personal and academic life in ways I cannot explain.  A very special "I couldn't have done this without you" acknowledgment is owed to my unbelievably supportive wife Jocelyn, whose unconditional love throughout my entire academic career is the only reason I have been able to continue on. Much love to the rest of my family as well for their continued moral support and understanding over the past few years.  I wish to acknowledge that this research was supported by the Canadian Institutes of Health Research (CIHR) grant (FRN: GSM-111975). And finally, I wish to thank Miner3D Inc. ( for kindly providing me with the software required to share the 3D model of the Canadian field online, and SQi Canada ( for providing the web space needed to host the model.  xi Dedication I would like to dedicate this thesis to the memory of my loving and supportive stepfather Jochen Hermann Otto Strebe (September 26, 1958 - November 17, 2011) and to my soon to be born daughter. John, your kindness, sincerity, compassion, and love for all things helped to shape the person I am today. In passing along these warm and empathetic qualities to my daughter, I know you will forever live on through her actions. You will be missed.  1 Chapter  1: Introduction  1.1 Theoretical and Empirical Landscape Public health research has had a longstanding debate regarding the role of individuals in health research – namely, does control of one’s health rest in the hands of the individual (agency) or in society (structure) (Cockerham, 2005, 2007; Frohlich et al., 2001; Link & Phelan, 1995; Williams, 2003)? Research structured around individualistic principles has tended to emphasize the ability of an individual to choose behaviours regardless of structural circumstances, captured by the common adage, “no one can take away your ability to make a better life for yourself” (Mirowski & Ross, 2003, p. 27). On the structural side, researchers have generally emphasized the role of structural mechanisms in "contouring individual dispositions and behavior along socially prescribed lines" (Cockerham, 2005; p.51). Individually- and structurally-focused health researchers have together produced many valuable insights into the strength and nature of relationships between health outcomes and material and/or non-material factors such as: economic capital (see Kaplan et al., 1996; Kennedy et al., 1996; Wilkinson, 2005), social capital (see Kawachi et al., 2004; Putnam, 2000), educational capital (see Mirowski & Ross, 2003) and cultural capital (see Abel et al., 2011; Haines, 2009; Khawaja & Mowafi, 2006; Malat, 2006). Some of this research has explicitly attempted to consider the dynamic between structure and agency, e.g., research that links the social capital of places (neighbourhoods, communities) to the social capital of individuals (and back again) and then to individual-level health and well-being (e.g., Carpiano 2006, 2007; Veenstra 2005). While this body of research has brought social factors to the fore in health research, recent critiques call into question its variable-centred methods  2 and substantialist reasoning which treats social phenomena as existing in society, external to the individual, and possessing essential properties that are discernable, measurable, and universal (Larsen & Morrow, 2009; Emirbayer, 1997). Some scholars argue that researchers should bring new conceptual tools and frameworks to the dynamic interplay between environmental attributes/conditions (structure) and individual practices/life choices (agency) (Abel, 2007; Cockerham, 2005; Frohlich et al., 2001; Lynam et al., 2007; Williams, 2003).  Current theoretical initiatives in health research have called for a shift away from research that treats human action and social phenomena as something that can be reduced into discrete, stable, and generalizable properties, towards a contextualized approach to the study of disease and illness that considers the group characteristics of interdependent actors and the collective nature of social life (Cockerham, 2005; Frohlich et al., 2001; Lynam et al., 2007; Williams, 1995; Williams, 2003). It is argued that health behaviours and outcomes are best understood via the collective patterns and relations between “agency (the ability for people to deploy a range of causal powers), practices (the activities that make and transform the world we live in) and social structure (the rules and resources in society)” (Frohlich et al., 2001; p.781), emphasizing the nature of relationships between actions, interactions and outcomes of people in their social and physical environment.  Largely influenced by French sociologist Pierre Bourdieu's ([1979] 1984) theory of social action, these recent initiatives draw upon his theoretical and analytical framework to social enquiry as a foundation for exploring the recursive and co-dependent relationship between social conditions and social practices (Cockerham, 2005; Frohlich et al., 2001; Williams, 1995; Williams, 2003). Representing a novel way of thinking about the social structuring of human health behaviours and health outcomes, Bourdieu's approach provides a  3 theoretical foundation that perceives human behaviour as more than just an individual lifestyle-choice or the direct outcome of structural determinants; it aims to illuminate the complexities common to social circumstances and social action and the ways in which structural circumstances are embodied and inform the health-related behaviours and outcomes of individuals by within a complex social world. While these lifestyle-focused Bourdieusian studies successfully incorporate his theoretical concept of habitus to explain the dynamic relationship between structural circumstances and health lifestyles, they remain fundamentally grounded in substantialist frameworks1 and as such remain at odds with Bourdieu's critical interpretive foundations and the relational principles at the heart of his theoretical paradigm.  As a means to exploring the persistent health inequalities that exist within developed and developing societies, Pierre Bourdieu's expansive theoretical and empirical work has informed many health studies that have drawn upon his work in a variety of ways, some incorporating his theoretical operationalizations of capitals – economic, social (Carpiano, 2006, 2007; Nakhaie, Smylie, and Arnold, 2006; Stephens, 2007; Veenstra, 2000, 2002a, 2002b) and cultural capitals (Abel, 2007, 2008; Shim, 2010; Veenstra, 2007, 2010) – and others drawing upon his concept of habitus (Lynam et al., 2007; Sieger, Fritz, and Them, 2011; Singh-Manoux and Marmot, 2005); and lifestyles more generally (Cockerham 2005; Frohlich et al., 2001; Williams, 1995; Williams, 1995). While these conceptually-focused Bourdieusian studies have provided a solid understanding of the relevance for health of many of his core concepts, they remain generally focused on examining causal relationships  1 "Considers each practice or pattern of consumption in and for itself, independently of the universe or substitutable practices, and conceives of the correspondence between social positions and tastes of practices as a mechanical and direct relation" (Bourdieu, [1994] 1998, p.3).  4 between the concepts and health behaviours and outcomes, and as such have yet to fully engage with the heart and soul of Bourdieu's theoretical paradigm: methodological relationalism and field theory.  The body of research inspired by Bourdieu`s framework that adopts methodological relationalism is small but growing. It incorporates various aspects of his core theoretical principles, such as capitals, habitus and field, into multivariate analyses (e.g. multiple correspondence analysis) to investigate the patterns of health-related behaviours, values and attitudes of similar types of people in social space (Frie & Janssen, 2009; Gatrell, 1997; Gatrell et al., 2004; Jones et al., 2010; Lengen & Blasius, 2007; Tomlinson, 2003; Veenstra, 2007). Through graphical depictions of social spaces, these studies have primarily explored the nature of relationships between a collection of structural- and individual-level factors (i.e., sex, age, social origin, ethnic origin, income, education level and practices/ behaviours, etc.) within a health-specific social space, and have successfully identified social dimensions in which higher levels of capitals (economic, social and cultural), affluence of neighbourhoods and a variety of measures of social stratification were grouped alongside positive health categories (e.g., self-rated health, health behaviours, mental and emotional health, etc.) and lower levels around negative health categories. While this small body of analytically-focused Bourdieusian health research has successfully incorporated the geometric analytic strategies employed by Pierre Bourdieu throughout his research and have added unique 'spatial' evidence of health-related behaviours and structural factors, they are constrained by cursory (or entirely absent in several instances) assessments of field theory and relational logic.  5  In summary, while these Bourdieu-inspired studies have made great strides in testing and exploring the health-relevance of his theoretical concepts and analytic principles, there are as yet no health-related studies that offer a comprehensive theoretical, conceptual and analytic account of Bourdieu's critical theoretical paradigm. While these implementations of Bourdieu's work have proven fruitful for health inequalities research in many ways, substantialist interpretations and neglect of key relational concepts like field, habitus and especially doxa have limited the ability of health researchers to offer contextually meaningful insights as to how and why health inequalities exist and persist over time. In order to fill in the theoretical gaps and take full advantage of Bourdieu's approach to social research, a detailed assessment of his ontology (what is the nature of social reality?), epistemology (what represents knowledge or evidence of social phenomena? how can it be known?) and methodology (how best to disseminate and characterize social phenomena and social reality?) is required in order to determine how best to translate his complex relational framework into a language that is relevant and useful to health researchers seeking to employ an inherently Bourdieusian framework.  In light of these fundamental theoretical limitations, it follows that there has yet to be an example of a Bourdieusian empirical health study which is grounded in a relational logic and field theory and incorporates his concepts of habitus, capital and doxa. Using original Canadian survey data collected in Vancouver and Toronto, Canada, my theoretical discussion will therefore be supplemented by a relationally-grounded empirical analysis of a three- dimensional (3D) Canadian field. Following the principles of relationality and field theory, my exploratory analysis will consider a series of theoretically and empirically relevant concepts (e.g. indicators of conditions of existence; demographic composition; personal  6 dispositions, values, attitudes and tastes; practices and behaviours; health indicators) in an attempt to reveal groupings within a Canadian social space and the characteristics of individuals within the groups. The goal is to present an intuitively appealing and theoretically generative account of a health-relevant social space that reveals the internal logic and general principles of a health-relevant Canadian field (social space). My descriptive analysis of a Canadian field has the potential to illuminate relational properties of concepts and social factors that have been found to be associated with health inequalities by past researchers and theorists (e.g., income, education, values, actions, behaviours, gender etc.) and reveal underlying assumptions, properties and hidden principles of differentiation that may be related to the production and preservation of the stratified nature of health inequalities in differing contexts. By undertaking an in-depth relational analysis of a particular Canadian field, I hope to reveal some of the common factors that form differences among the individuals in the field (or sites of struggle and differentiation), and also offer some theoretically informed insights regarding the ways in which the forces that draw these factors together are related to health.  The remainder of the thesis is structured into four interdependent chapters. Chapter two draws upon the original theoretical writings of Pierre Bourdieu ([1972] 1977; [1979] 1984; [1994] 1998) and Bourdieu & Wacquant (1992) to provide a comprehensive account of the key theoretical principles at the heart of Bourdieu's relational theory of action. Chapter three offers a detailed account of how the theoretical principles outlined in chapter two can be practically implemented into a relationally-grounded empirical health research program. Chapter four presents the results of the analysis. Chapter five discusses the results and then  7 concludes with practical reflections on the benefits and limitations of a Bourdieusian theoretical framework for Canadian health research.     8 Chapter  2: Bourdieu, Relational Theory and Health Research  2.1 Introduction: Pierre Bourdieu's Theoretical Paradigm To date, health research drawing upon the work of Pierre Bourdieu has been largely focused on the structuralist aspects of his theoretical enterprise. Informed by the lifestyle model originally presented in Distinction (1984: p.171), these scholars have adopted a ‘substantialist’ re-interpretation of Bourdieu's theoretical paradigm in search of a better way to understand how structural circumstances are related to healthy and unhealthy behaviours. Typically exploring relationships between class, health and lifestyles, this body of research has tended to interpret Bourdieu's theory of practice as a one-way structuralist paradigm that gives primacy to structural factors over agency-related behaviours. While the theoretical interpretations of this body of 'contextual' health research have offered important insights into how social circumstances 'constrain' the choices individuals can make, health-related and otherwise, the fact remains that Bourdieu's paradigm has been largely interpreted as being trapped within an objectivist point of view that does not give enough respect to the power of agency and the reflexive character of human action (William, 1995: p.588). I argue that this interpretation fails to engage with his work from the appropriate ontological and epistemological standpoint.  In addition, theoretical re-interpretations of Bourdieu's work in health research have almost entirely focused on capitals broadly, and habitus more specifically, treating habitus as the theoretical linchpin that ties together structural circumstance and actions (agency). This limited focus on habitus is, however, directly related to the ‘structuralist’ way in which his work has been interpreted; that is, habitus only allows these researchers to show how  9 structure is turned into action, how history is turned into the present, essentially perpetuating the notion that Bourdieu's theory of action is only good for illustrating how external social ‘forces' are translated in behaviours.  As will become clear in the following sections of this chapter, Bourdieu's theory is much more than a structuralist theory of action. It is a theory for change, a theory that attempts to better understand the processes that give rise to the social forces that structure society (e.g., social spaces of freedom and struggle), and in so doing can offer insight into how they might one day be changed. To interpret Bourdieu's work as a simple 'lifestyle' model of capitals, habitus and behaviour, as many health researchers have done, overlooks the strengths of his broader relational theory.  The following sections of this chapter will move beyond the abovementioned interpretations of Bourdieusian theory which have, to my mind, limited the ability of health researchers to fully unleash the true potential of Bourdieu's framework. This requires a full understanding of the relational philosophy at the heart of his work, his theories of fields, capitals and habitus, and importantly, his theories of doxa and time, both of which are critical to understanding how human agency is turned back into structure and are entirely absent from all of the Bourdieusian health literature. Acknowledging that there are different ways to interpret and incorporate Pierre Bourdieu's expansive body of work, this chapter is meant to be a detailed assessment of the foundations of his theory of action and an illustration of how his theoretical framework can be interpreted and integrated into social research generally and health research specifically. It is my hope that the contents of this chapter will help to clarify what a Bourdieusian approach -- as opposed to a Bourdieu-inspired approach -- to health  10 research might look like, and how it could be utilized to its full potential to better inform future research endeavours.  2.2 Theoretical Foundations Pierre Bourdieu wrote about his complex theoretical enterprise at great length over many years and using many different mediums, including public talks, news articles, debates, films, lectures, journals and books. Because his body of work is remarkably vast and detailed, not surprisingly there is a wide range of conflicting interpretations of his theoretical framework, particularly in the area of health research. Before describing the many facets of Pierre Bourdieu's theoretical paradigm and the theoretical principles that accompany it, as with any reading of theoretical texts it is important to first take a broad look at the general nature of the paradigm.  Broadly speaking, Bourdieu believed that the role of sociology is to be critical of the social world we have created and to identify the sites of freedom and struggle that affect each and every individual, so that people can have "a small chance of knowing what game[s] we play and of minimizing the ways in which we are manipulated by the forces of the field in which we evolve, as well as by the embodied social forces that operate from within us" (Bourdieu & Wacquant, 1992: p.198). At the heart of his intellectual pursuits, Bourdieu saw social theory as a tool for challenging power structures and providing detailed understandings of the laws, norms and symbolic barriers of the social world, that is, to "uncover the most profoundly buried structures of the various social worlds which constitute the social universe, as well as the 'mechanisms' which tend to ensure their reproduction or their transformation" (Bourdieu & Wacquant, 1992: p.7).  11  These two statements reflect what I believe to be fundamental elements of Bourdieu's philosophy of the social, firstly, that he is a conflict theorist2, and secondly, that he shares in part the social constructionist perspective that place social actors at the centre of social processes. Moreover, he holds that social arrangements, and by extension, social inequalities, emerge from the (inter) actions (i.e., conflict) between social beings in a social space. This perspective contends that social reality is constructed within power-laden social fields, whereby social truths are interpersonally negotiated rather than a function of an objective, free-standing reality. These foundational points about Bourdieu's work are far too often overlooked in Western health research, and are where Bourdieu-inspired health theorists such as Cockerham (2005, 2007), Frohlich et al. (2001), Williams (1995) and Williams (2003), for instance, read his critical interpretive approach in an overly deterministic manner.  Where health theorists employing Bourdieu's work have been mistaken is in their overly substantialist interpretations of select concepts in Bourdieu's work, when they consistently assert that Bourdieu awards epistemological priority to objective conditions over subjectivist understandings. They contend that Bourdieu believes that social factors such as income, education and race have an inherent and unconscious influence on social actions, that his concept of habitus, for instance, is "produced by the objective conditions of existence combined with positions in the social structure" (Frohlich et al., 2001: p.789; emphasis added). The problem with these sorts of interpretations is that they interpret structure as the starting point of his theory, that structural factors such as education, income and race 'objectively exist' in society, external to the individual, as ‘functional’ objects that directly  2 Approach to social enquiry grounded in the view that social structures are created through conflict and struggles between people with differing interests, beliefs, dispositions and resources, leading to the unequal distributions of power that form and enforce social inequalities.  12 produce actions. Where Bourdieu makes it clear throughout his writings that structure and agency are dynamic and can never be thought to presuppose one another, ‘Bourdieu-inspired' health researchers give complete primacy to structure which is a fundamentally false reading of Bourdieu. While grounded in a fundamental belief that the social world is highly structured and temporally and spatially bound by social laws (Bourdieu & Wacquant, 1992: p.52), Bourdieu does not, as many of the current Bourdieu-inspired health theorists would lead you to believe, share the deterministic view common to strict structuralist approaches which contend that the social world is the result of the 'one-way' influence of material reality over the action of individuals and evolves by the hand of immanent and unchangeable laws, norms, and coercive powers external to the individual (Durkheim, [1895] 1982: p.51). Rather, he argues that social laws, norms and social regularities are the result of relational human action within social space, and that it is in the actions of individuals, first and foremost, which give such social factors their power in society. It is from the struggles and conflict that take place between people in social space that social stratification and inequalities emerge.  Along these lines, Bourdieu is much more in sync with the relational foundations of Marxist, Simmelian and Meadian philosophies, among others, which are similarly grounded in the belief that social stratification results from interdependent human action whereby a person actively shapes the world s/he lives in at the same time as it shapes her/him. He works from the fundamental principle that people create and enforce the structuration of the social world through the dynamic interactions that take place between individuals in their day-to- day actions and their environment. Furthermore, Bourdieu argues that what people do in practice creates and reproduces the social factors (i.e., system of discourses) that make up the  13 social world, but also believes that because we create that which divides us, individuals also have the power to re-create discourses that resist dominant forms of power in society.  In contrast to the interpretations of the health theorists mentioned thus far, Bourdieu's theoretical vision rejects the supposition that the reality of the social world is the result of either structurally-determinate relations between individuals (i.e., objectivist) or social actors who create their social world (i.e., subjectivist). His 'philosophy of the social' accepts the critical importance of both modes of thought and aims to move beyond the ever-present duality by incorporating both the subjective experiences people have in the social world and the objective conditions that inform and shape their experiences (Bourdieu, [1980] 1990: p.25).  In order to transcend false antinomies between structure and agency, Bourdieu converges 'structuralist' and 'constructivist' approaches into a science of society which encompasses both the structural factors and objective regularities that impose themselves upon social actors, and more importantly, considers the processes by which individuals relationally construct the social structures that inevitably instil durable and transposable dispositions which contribute to the actions and behaviours that reinforce or remake the social world. He argues for the return to a critical theory of practice, investigating the principles of the production and preservation of the stratified nature of the social order, that he believes is a necessary precondition for establishing an experimental social science which considers the social circumstances in which the activity of individuals occurs and conditions their perception of the world in which they live.  In short, the foundation of Bourdieusian theory requires a break from the deep-rooted impulse to assert ontological priority of either structure or agency (as many health  14 researchers tend to do) when investigating the reality of the social world, that is, to accept that all human beings are not only the product of the accumulated culture of generations past but are also interdependent actors who contribute to the further modification of their social world. This means understanding the nature of the classificatory systems (i.e., objective regularities) that emerge as principles of division from the routine interactions between individuals in their daily lives, namely, the symbolic cultural systems of classification which contribute to the organization and reinforcement of social stratification. In its most fundamental form, Bourdieu's critical interpretive theoretical framework contends that social inequalities are formed through conflict and power struggles between individuals, not through the external and mystical influence of social objects such as income, education, neighbourhoods and/or occupation.  Misinterpretation of fundamental principles of Bourdieu's work can lead to a narrow and overly causal understanding of health inequalities, one that tends to treat structural factors such as income, education, occupational status, etc. as having an inherent power that is directly related in one way or another to health behaviours. Such a perception carries with it the inherent belief that if someone procures such factors (i.e., wins the lottery, gets a university education, lands a prestigious job) then they are likely to have better health behaviours and outcomes, as argued by researchers such as Mirowski and Ross (2003), Putnam (2000) and Wilkinson (2005), for instance. While lifestyle-focused health researchers expand upon this narrow view by focusing on habitus as an objective concept that bridges structure and agency, they remain ontologically distant from Bourdieusian theoretical principles, essentially labelling his work with the deterministic terms with which he is fundamentally in opposition. To further elaborate on the theoretical gaps that exists between  15 current health research approaches and Bourdieu's theoretical paradigm, I now refer to the notion of relationalism, another key principle that is almost entirely absent from Bourdieu- inspired health research.  2.3 Relational Philosophy Relationalism is consistent with the ontological principles described thus far, wherein the 'real' foundation of social reality emerges from the relationships of active individuals. Relationalism is essentially a way of thinking about the social world in terms of relations, which means accepting that all that is social, all that occurs in society, emerges in relation to other social factors located in a variety of social spaces. In a more practical sense, to think relationally means that to truly understand the structure of social life one needs to investigate and explain the often unseen relational context among social elements ‘from the inside’ to show the underlying order and functional links within social space that account for inequalities among individuals (Levi Martin, 2003).  In direct contradiction with substantialist perspectives mentioned earlier, relational approaches see social reality, or 'truths' about social reality, to be the emergent result of relationally-embedded human activity. For instance, Marx's fundamentally relational statement about capital offers a perfect example of relational logic: "capital is not a thing, but a social relation between persons which is mediated through things" (Marx, [1867] 1990: p.932). Similar to Marx, Bourdieu breaks with the common-sense realist representation of social concepts such as 'capital' (i.e., treating it as a 'natural' social object existing independent of individuals), to thinking of it in terms of relations. This means understanding capitals as social phenomena that emerge within a generative matrix and ensemble of social  16 factors and forces (structural and individual). This relational logic can be further applied to the concept of 'class,' which, contrary to what many would believe, does not exist as an observable entity in society for Bourdieu. Rather, he thinks of class in terms of the "social space of differentiation and differences in which 'classes' exist in some sense in a state of virtuality, not as something given but as something to be done” (Bourdieu, [1994] 1998: p.12). This essentially means that classes emerge from human relations and actions.  These relational principles are quite closely related to constructivist philosophies which similarly emphasize that social order is a human product, that it is not part of the nature of things or the laws of nature, and that it only exists as the product of human activity (Berger & Luckmann, 1966: p.51-53). Just as philosophers of science ask social scientists to question the historical process from which 'taken for granted' social concepts have been constructed and come to exist as facts (see Fleck, [1935] 1981; Kuhn, [1962] 1996; Woolgar, 1988), the relational foundation of Bourdieu's theoretical paradigm requires a fundamental break from substantialist reasoning.  In contrast with existent theoretically- and conceptually-focused health-related interpretations of Bourdieu's theoretical paradigm, a relational approach to social enquiry requires a transformation of one's whole vision of the social world which questions taken for granted understandings of social facts, this means that we cannot assume that concepts such as menopause, gender, class or health exist and are experimentally verifiable natural elements of the social world; rather, we must take a broad view of these social concepts and attempt to understand them in relation to the social context in which they were shaped while still immersing oneself in the fullest detail of the social phenomena (Bourdieu & Wacquant, 1992: p.252). Thus, relational thinking requires the researcher to deconstruct taken-for-  17 granted relationships, question the objective nature of categories, and show that they are in fact often multidimensional and entirely dependent on social context. Relational thinking intends to remove the thing-like character (i.e., objective understanding) of scientific concepts by conceiving of them as symbols representing orders and links within reality as opposed to objective facticities. It requires the researcher to discover and capture the unseen forces and internal structure of the social 'objects' and 'forces' that constitute the specific strength and form of social objects within society (Bourdieu, [1979] 1984: p.103).  This notion of relational thinking is the pivotal point at which Bourdieu-inspired health researchers who read his work through an overly substantialist lens depart from his theoretical paradigm. To fully understand the usefulness of his core theoretical concepts, I portray them in the relational fashion originally intended by Bourdieu (as I interpret his intentions, of course). In the following sections I provide a brief overview of these theoretical concepts and clarify their importance to the relational logic at the heart of Bourdieusian sociology, starting with a description of the theoretical concept most important to implementing a relational approach to social research, that of 'field.'  2.4 Fields In order to implement his relational theory of action, Bourdieu draws upon field theory to express the relational logic of competition and struggle between individuals within a field of play, and to examine and explain social phenomena as the result of the interdependent nature of human actions and complex social relations within the constraints of a social environment (Levi Martin, 2003). For Bourdieu, field theory provides a relational platform from which to critique and question the processes of classification and operationalization embedded in  18 commonly used concepts in social research. Rather than treat attributes of individuals and groups (e.g. occupation, age, sex, etc.) in terms of discrete and measurable categories, field theory essentially gives life to the relational approach and offers a way to conceive of the attributes as existing relationally within a field, the latter often referred to as a social space of forces (Bourdieu & Wacquant, 1992: p.243). The belief is that the exploration of a social concept as existing within space of relations allows one to discover the boundaries of the field, to unearth the relational properties of the concept, and to reveal the distinctive features and common factors that form the differences and gaps among individuals and groups within a social space (Bourdieu, [1994] 1998: p.6-7).  In an applied sense, Bourdieu sees the social world as consisting of an ensemble of ‘spheres of play’ or ‘fields’ (e.g., academic field, artistic field, medical field, economic field, etc.), each of which consists of a patterned system of objective forces that are imposed upon the individuals who are at play within them (Bourdieu & Wacquant, 1992: p.17). Each field is believed to represent a social arena in which social elements interact and events take place, revealing the orientation of social life within the social space. These fields are conceived as 'snapshots' of social reality (or as Bourdieu often refers to as ‘a special case of the possible’), differently structured and representing distinct worlds with specific features that influence what is possible for actors located in them (Bourdieu, [1979] 1984: p.226-227). Bourdieu's conception of field focuses on the distribution of active properties in social space, stressing the dynamics of conflict, tensions, struggle and forces related to the differential positions held by individuals and groups in the structure of the field of forces (Bourdieu, [1994] 1998: p.32).   19  A helpful way to think of field is in terms of a 'game,' in the sense that a game is described as having certain rules and regulations that govern how it is played.3 A game (i.e., field) is believed to have 'players' who take part in the game. These players are endowed with differing volumes and composition of material and non-material resources that are related to their ability to understand and participate in the game and even excel in it. The game of hockey, for instance, has a set of socially constructed and agreed upon rules and conventions that regulate how the game is to be played and who is able to play. There are individuals who can and cannot play the game depending on whether or not they are able to understand and abide by the rules (e.g., participation in the game of hockey often requires the individual to be mentally and physically – and financially – able4). Among the individuals who can play hockey, there is likely to be a gradation among players (i.e., differentiation in ability, talent, competence and success); as in all games, there are some players who will be very good at the game and some who will not.  Bourdieu would argue that these abilities are not innate to the individual (that some people are not just simply born to be great hockey players); rather, he believes that there are social forces (forces of power) at play within the game that influence an individual's ability to succeed. Along these lines, he would argue that we should not be interested in describing the universal rules that govern the games. Rather, we must investigate the social factors and forces of power that are most relevant in different social spaces and work to understand how and why certain social forces influence the segregation that takes place within different  3 The use of the game metaphor has been the subject of debate in academic circles employing field theory, but remains widely used in cultural research (see Lamont and Small, 2008; Small, 2004; Swidler, 1995; and especially Levi Martin, 2003 for an in depth assessment of the metaphor’s usefulness in social research). 4 This does not mean that disabled individuals cannot play hockey, it simply means that it is much more difficult to be accepted into the game.  20 games (fields). Namely, what are the social forces that influence who dominates in the different games and who is dominated?  In a practical sense, field theory encourages researchers to seek out sources of conflict and struggle in a given field, identify the underlying assumptions and hidden principles shared by opposing actors, and relate these areas of struggle and differentiation to the broader notion of social forces of power and stratification. It requires the researcher to search for the social mechanisms and properties which distinguish agents in a social space who are as similar and as different from one another as possible, and is a key spatial metaphor used to illustrate the principles of differentiation embedded in a social setting in which resources, institutions and organizations interact (Bourdieu, [1994] 1998).  Given that there are very few instances of Bourdieu-inspired health studies that embrace relational philosophy, it follows that there are very few studies that have incorporated a theory of field. Only Veenstra (2007), who explored health behaviours and outcomes in Canada from a relational perspective, has properly approached the concept of field as a distinct space of struggles and used relational logic to interpret empirical findings. Gatrell et al. (2004) discuss5 relational sociology and field but remain bound to a fundamentally substantialist approach that used Bourdieu-inspired methods. Nearly all of the ‘relational’ quantitative health studies adopting Bourdieusian principles (Frie & Janssen, 2009; Gatrell, 1997; Gatrell et al., 2004; Jones et al., 2010; Lengen & Blasius, 2007; Tomlinson, 2003) use multivariate analytic methods to create a social space of relations but remain ontologically and epistemologically at odds with Bourdieu's theoretical conception of field. When discussing the spaces, the fundamental error inherent in nearly all these studies  5 I put 'discuss' in italics because their assessment of relational logic and field theory appeared to be limited to the first 20 pages of Bourdieu and Wacquant's (1992) book on reflexive sociology.  21 are that they rely on relational methodology and Bourdieusian concepts to map out health within a social space without having a full understanding of the ontological and epistemological principles that inform such an analysis. Such methodologically-driven approaches are quite problematic in the sense that they use inductive 'a priori' methods to tease out relationships within data without having the theoretical grounding necessary to appropriately build models and interpret results.  Field has not had the cleanest incorporation into health research but is nonetheless one of the most important concepts to understand before implementing a Bourdieusian research framework. The concept of field is essentially a theory of social space used to examine the social nature of the inequalities in society. Within society, every field constitutes a potentially open 'space of play' whose boundaries are dynamic borders which are the stake of struggles within the larger field of power. Social researchers incorporating field theory must think in terms of forces and power and how they are available to, and used by, individuals in different social spaces. To better comprehend these abstract ideas of forces and power in society we must understand Bourdieu's theory of capital, that which is believed to represent the social forces that inform the structuration of social order and human action.  2.5 Capitals In order to understand how fields take shape and boundaries are formed, we need to start by considering what social factors are relevant and forceful within particular social fields. Capitals essentially refer to the wide variety of material and non-material resources that are constantly at play and influential to the organization and transformation of a social space. Bourdieu conceives of capitals as relationally-interdependent resources that only operate and  22 exist within the structure of, and in relation to, a given field (Bourdieu & Wacquant, 1992: 101).  His conception of capitals encompasses a wide variety of different species of resources which are all relationally interdependent and operative within fields. He generally speaks of three interdependent types of capital: economic, cultural and social capitals (see Bourdieu, 1986, for detailed 'empirical' descriptions of each). While many health researchers have sought to specify the objective elements of each of these forms of capital, it is important to reiterate that these three forms of capital are entirely theoretical concepts that emerged from Bourdieu's relationally-grounded ethnographic field work (see Bourdieu, [1980] 1990, [1972] 1977) and later empirical analyses (Bourdieu, [1979] 1984); above all else, they are qualitatively derived concepts.6 Furthermore, capitals are 'a priori' theoretical concepts used to explain and reveal the symbolic and objective resources and relations that appear to be the most important in shaping the structure of social life within specific contexts. Thus, capitals are not universally relevant concepts – they are only relevant in the specific fields in which they have power, which means that some resources may be important factors in one social space and not another. While this context-specific understanding of capitals is by no means limited to a relational approach, it is a fundamental assumption of field theory, a point which will be made clear in the results and discussion section of this thesis.  Relating the theory of capitals back to the theory of field, capitals are thought to represent social forces and relative power in a social space. For Bourdieu, the composition of the capitals one has at their disposal informs their "relative force in the game, [their] position in the space of play, and also [their] strategic orientation toward the game" (Bourdieu &  6 For instance, cultural capital was originally a theoretical hypothesis developed to explain the specific profits available to children of different class backgrounds that result in unequal educational achievement (Bourdieu, 1986).  23 Wacquant, 1992: p.99). The general composition of the resources possessed informs the moves (actions) that can be made, be they modest or risky, rebellious or conservative, normal or abnormal, etc. Furthermore, he argues that individuals and/or groups who possess different forms and configurations of capitals that are either beneficial or detrimental within a given field engage in ongoing struggles within a field as bearers of different amounts and combinations, some of which yield greater advantages (or disadvantages). As a representation of the power relations that oppose individuals and groups in social space, capitals are seen as "a weapon and as a stake of struggle, that which allows its possessors to wield a power, an influence, and thus to exist, in the field under consideration" (Bourdieu & Wacquant, 1992: p.98). Accordingly, there is great emphasis put on capitals as influential, interdependent factors that partly inform the nature of social stratification within social spaces (fields).  Returning to the hockey example, having established that the game of hockey should be thought of in terms of a relational field of forces, the next step is to think about which social forces are the most influential in the game, namely, which capitals might be the most influential in structuring the social space and an individual's position within it (i.e., what seems to be important to dominate or be dominated in field). It is the duty of the researcher to determine which resources, in terms of volume and composition, are relevant and influential within the game of hockey. For example, being wealthy, having the best equipment, attending the best training camps, having friends who are good at hockey, having parents who fundraise and playing for coaches with expert knowledge of the game, etc., can all potentially influence the way an individual is able to play the game, how successful they will become, and as such, how they will be located in relation to other players within the game.  24  Some resources will invariably be more important than others in the field, whereby having possession and control of certain resources may result in differences and gaps in the field by which certain individuals may have to struggle more than others to enhance their position in the social order, to be a top player in the game. In this respect, capital takes on the form of ‘power’ within social space. That is, to have access to certain capitals can conceivably influence the way one interacts with others and may garner them greater power within the game. We must, however, acknowledge the relational connection that capitals have with field, whereby capitals that are more influential in one field (e.g., hockey) may or may not be relevant or have as much force in other fields (e.g., the academic field), which means "acknowledging that capital can take a variety of forms is indispensable to explain the structure and dynamics of differentiated societies" (Bourdieu & Wacquant, 1992: p.119).  Essentially, Bourdieu's theory of capitals as power, understood in a relational context via his theory of fields, provides a theoretical framework that asks the researcher to: 1) consider the field in which individuals interact when playing the game, and 2) examine which social factors in the form of capitals have the most force in particular fields, that is to say, how the distribution and composition is related to the position a player has in the field. This relational approach allows the researcher to illustrate how and under what conditions individuals and groups employ strategies of accumulation, investment and conversion of various kinds of capital in order to maintain or enhance their positions in the social order (Bourdieu & Wacquant, 1992).  At this point, Bourdieu's argument may seem overly deterministic and heavily focused on the influential force that material and non-material capitals have over individual actors and their chances to succeed. In response to this indictment, let us refer to his concept  25 of habitus, the dynamic concept that is often drawn upon by health researcher to bridge the gap between structural forces related to opportunity and agent-level behaviours related to action.  2.6 Habitus Arguably the most important aspect of Bourdieu's theory of action, habitus is a theoretical concept that considers how the objective chances and opportunities available to an individual or group of individuals are internalized through socialization into relatively durable dispositions that generally inform the choices they make and the actions they take. Simply put, “habitus is history turned into nature” (Bourdieu, [1972] 1977: p.78). Habitus has garnered heavy attention in ‘Bourdieu-inspired' health research as a theoretically viable way to explain how structural circumstances, both past and present, are embodied as a system of schemes or dispositions that generates practices and health-related behaviours. (See Cockerham & Hinote, 2009 for a summary of health-related applications of habitus.) Habitus is a key theoretical concept that remains largely underdeveloped in a relational sense, that is, it is seldom understood in relation to Bourdieu's theories of fields and capitals in health research circles.  Bourdieu considers habitus to be the product of history, where at an early age individuals begin to learn and internalize durable dispositions, values, tastes and distastes that are formed in relation to the needs and interests that emerge from the social context and conditions of existence within which they are born (Bourdieu, [1979] 1984: p.170). Often described as 'internalized necessity,' the system of dispositions that make up the habitus are reflective of the nature of the conditions and resources that are available to individuals in  26 social space, whereby people learn to live with what is available to them and develop a particular understanding of how the world works, how they fit into it and how they should act, which translates into a fundamental sense for what is possible (and desirable) relative to their social circumstances. Bourdieu argues that the relational constitution of the resources (e.g., economic, social and cultural capitals) available to an individual or group of individuals informs the beliefs, dispositions and values that fundamentally inform the nature of their actions. For instance, individuals who cannot afford a mode of transportation, dislike cold temperatures, value non-contact sports and/or dislike team-based sports would be unlikely to actively participate in the game of hockey, though they may encounter situations where they are forced to 'play the game' (e.g., discussing the sport with a friend who is a hockey fan) in which case they may participate but feel at odds with the situation, like a 'fish out of water.'  It is argued that our habitus is present at all times and in all places, finding expression in the way we speak, how we communicate with others, our tastes in clothing, music and food, the values we have and in the ways we reason; it can be identified in the style of action individuals take in their everyday lives, such as their eating habits, dating strategies, sport engagement and job choices. It is in this system of dispositions that we find the distinct markers of social position that constantly remind individuals, and those around them, of the social distance that they have in relation to other actors in social space, thus leading them to know their place and standing in the social order of different fields (Bourdieu, [1972] 1977: p.82). By virtue of the internalization of social upbringing and life experiences, Bourdieu goes on to argue that not all people have the same opportunity to access certain social fields. In other words, not all people have the same opportunities to 'play the game.' For those who can 'play the game,' some will have distinct dispositions (and resources) that give them a  27 certain 'feel for the game' which is related to their capacity and capabilities to dominate (or be dominated) (Bourdieu & Wacquant, 1992: p.128). In terms of health inequalities, habitus is an important theoretical concept because of its usefulness in explaining why and how structural factors such as income, education, parental education and neighbourhood are related to health behaviours and outcomes, and how they interact to generate actions. Habitus reflects an individual's capacity to act based on the structural limitations and embodied history (i.e., what they have been taught to believe and value) and offers a theoretical way to explain how structure transitions to action and how structural and agent-level factors (both material and non-material) may be related to health inequalities. Within the scope of this definition, it is easy to understand why it has been often incorporated into health research as a bridge from structure to agency, as a way to explain by way of embodied dispositions why some people are healthier than others.  In light of overly deterministic interpretations of habitus as a social construct that informs the actions of unconscious individuals, it is important to link habitus to capitals and fields at all times. For instance, individuals encounter many different situations and participate in many different 'games' throughout their lives, and as such, their systems of dispositions are constantly subject to experiences which can either reinforce or modify the structure of their habitus. Thus, habitus is not to be thought of as a genetic imprint that will forever inform the choices of individuals; rather, it needs to be thought as being "durable but not eternal" (Bourdieu & Wacquant, 1992: p.133), as a dynamic social entity that develops through interactions with individuals and different social fields. Experiences inform how interactions between people within a field take place (i.e., how dispositions inform interactions) and also how a person will interact with others in similar fields in the future.  28 While Bourdieu argues that habitus is by no means the 'fate' that some people think it to be, he acknowledges that "there is a probability inscribed in the social destiny associated with definite social conditions" (Bourdieu & Wacquant, 1992: p.133) which can be interpreted as follows: many a child may dream of becoming an astronaut, but the moon will invariably be closer to some than to others.  At this point we might ask, why then is social life so regular and predictable? Why is it that not all people become astronauts, heart surgeons or professional hockey players? While these questions can largely be answered through the implementation of the theories of field, capitals and habitus described thus far, Bourdieu also draws upon the frequently overlooked theoretical concepts of doxa and time. These concepts are needed to understand why social differentiation remains resilient to change over time.  2.7 Doxa and Time Without an understanding of doxa and time, Bourdieu's theory of action loses much of its relevance for health research in that it becomes difficult to understand how structural forces are reconstructed by individuals, how health inequalities persist, and more importantly, how they can be changed. While habitus allows for researchers to connect structure to agency, doxa is a relational theoretic concept that allows agency to reconnect to structure, essentially removing the 'structuralist' label typically attached to Bourdieu's work.  Where the theoretical concepts of fields, capitals and habitus allow us to conceive of the underlying unity of social strategies and actions that take place within the boundaries of a field(s), Bourdieu describes doxa as the primary perception one has of the social world as a result of their life experiences (Bourdieu, [1972] 1977: p.164-165; [1979] 1984: p.471). Not  29 to be confused with habitus ('history turned into nature’), doxa refers to one's sense of the limits and boundaries that exist in society (‘nature turned into future’). Doxa emerges from the relational intersection of structural constraints, dispositions of the habitus and experiences garnered within different fields: when we act, we think about the future but are constrained by past experiences. It is in day-to-day experiences with the social world that people begin to develop a strong sense of their place within it, a hardened (yet malleable) sense of social order, leading them to avoid fields within which they are unlikely to feel comfortable and to seek out fields that are welcoming, relatable and invoke a sense of comfort. For Bourdieu, people are fundamentally reasonable beings who recognize their capacity and capabilities in certain fields and tend to inhabit fields that they perceive to have the most meaning and interest, namely, social spaces that fit them best and in which they feel the most at home, "like a fish in water" (Bourdieu & Wacquant, 1992: p.127).  The internalized recognition of one's place in the social order leads people to reasonably adjust their expectations for the future to coincide with their chances, whereby they "define themselves as the established order defines them" (Bourdieu, [1979] 1984: p.471) and shape their aspirations according to concrete indices of the accessible and the inaccessible, of what is and is not "for them" (Bourdieu, [1980] 1990: p.64). It is this reasonable conformity to the regularities and tendencies of the social order (i.e., acting in a way that is reasonable, necessary and makes sense) that conserves and perpetuates the divisions of the fields and opposes individuals and groups to one another at all times (Bourdieu & Wacquant, 1992: p.138). From these reasoned actions within perceived limits, social boundaries are re-enforced and naturalized to the point that they appear as self-evident  30 to individual actors (Bourdieu, [1972] 1977: p.164), solidifying their beliefs about what is and is not of interest and/or possible for them.  At this point, the notion of time becomes particularly relevant to Bourdieu's theory of social action. For Bourdieu, time is built into his theoretical conceptualization of social space, not as a metaphysical entity that exists outside of the consciousness of individuals but rather as something that is produced by individuals through their actions (Bourdieu & Wacquant, 1992: p.138). For Bourdieu, practical/reasonable action is in part the product of habitus and in part the product of a reasonable anticipation of the 'forth-coming' (i.e., future), whereby "the experience of time is engendered in the relationship between habitus and the social world, between the dispositions to be and to do and the regularities of a natural and social cosmos (or a field)" (Bourdieu, [1997] 2000: p.208). Individual behaviours are thus believed to emerge through the practical mobilization of the past and the practical anticipation of the future which is inscribed in the present as a state of 'objective potentiality' in such a way that present decisions are at all times a reflection of past experiences and future pursuits (Bourdieu & Wacquant, 1992: p.138).  Essentially, what is being argued is that social action is not just based on past experiences and social circumstances, it is also related to perceptions of the future and what people believe to be reasonably possible given their social standing and position in the social order. One's perception of the future is entrenched in all of the relationally-interconnected social factors mentioned thus far (i.e., fields, habitus and capitals), whereby the potential for future actions are constrained or enabled by the nature of social conditions. For example, a child living in subsidized housing with parents who have been laid off from their jobs, who is taught the importance of conserving and using what is necessary and eating foods that are  31 affordable, might act in a way that is reasonably in tune with his/her social situation; that is, people's aspirations for the future remain within the limits of what they consider reasonably attainable, or as Bourdieu asserts, "habitus is [the] 'can-be' which tends to produce practices objectively adjusted to the possibilities, in particular by orienting the perception and evaluation of the possibilities inscribed in the present situation" (Bourdieu, [1997] 2000: p.217).  This discussion of doxa and time illustrates the ways in which all of the theoretical concepts described thus far come together into a critical theory of action which envisions the social world as a consisting of relationally-constituted spaces of force and power, struggles and conflict. Within different social spaces, some people are likely to have less 'actual' and/or 'potential' power in a field, and as a result might have to struggle more to become 'dominant' in a field. They will have to struggle against the social limitations and relational boundaries that are constantly being reinforced by their actions and the actions of others, and they will have to endure more discomfort confronting unfamiliar social fields. Thus, breaking through the emergent social boundaries that demarcate social differentiation among individuals and groups of individuals will be more difficult for them.  From the theoretical assumption that we are at all times in conflict with others in space and time, Bourdieu goes on to argue that over time our perceptions of what is necessary and reasonable may change when the relational composition of our conditions of existence, system of dispositions and available resources change. While Bourdieu argues that time carries with it a certain 'social rhythm' whereby social differentiation tends to persist and become reinforced, he also asserts that time reintroduces the element of uncertainty into his theory of social action (Bourdieu, [1980] 1990: p.99). Though he clearly argues that agents  32 embody much of their structural circumstances, his critical philosophy of social reality is equally adamant that individuals can learn to play other games and change their position in the social order – some will just have to struggle more than others to do so.  The concepts of doxa and time are necessary to answer the question of why social inequalities in general, and health inequalities in particular, persist over time. Where health researchers have limited their focus on habitus to explain social action, incorporating the concept of doxa will help them to better understand how the 'cycle' of inequalities persist. If researchers only focus on habitus as the explanatory concept which connects structural factors with choices of action, they will fail to consider the reflexive character of the individual whose actions are equally dependent on how they perceive the future. Action is not just based on history and experiences, though these are believed to play a large role; it is also based on perceptions of the future, what people think is possible given their past experiences and present situation. Bourdieu's theory of action is not saved from the "charge of social determinism" through his inclusion of 'experience' to his habitus concept (as stated by Cockerham (2005): p. 62) – such statements imply that people act only in relation to their past and present. This narrow substantialist view does not take into account the relational nature of Bourdieu's theory, which holds that field, capitals, habitus, doxa and time all influence the ways individuals act in relation to one another in space and time. Doxa is reflexivity, it means having a reasonable perception of what the future holds based on the experiences of the past and the available powers of the present. It is the concept that binds together Bourdieu's theory of the social and is critical to any social study looking to explain how and why social actions inform and perpetuate social stratification in societies.   33 2.8 Conclusion: Bourdieu’s Relational Theory of Social Action Bourdieu's theory of practice attempts to bring together structure and agency using a relational way of thinking about social action and society in general. While I have described several different concepts that are fundamental to Bourdieu's way of thinking, relationality requires that each be taken in relation to the other, that is, field, habitus, capitals, doxa and time are all integral to his theoretical framework. To fully understand the structuration of society, we must think about how social conditions and perceptions of the future are interrelated with the opportunities and chances of individuals and groups of individuals in society. Bourdieu's theory of action is not meant to pinpoint precisely which social factors are the most detrimental or beneficial to individuals or groups. Rather, it aims to explore the relative composition and distribution of social factors within social space, how they are all related to one another, and how this composition is related to the force and power one has or does not have in different areas of society (fields). Where certain areas of health research have sought to uncover the causal relationships between social factors and/or individual behaviours and health outcomes, Bourdieu's theory of social action requires that we consider the relational nature of the social constructs within a particular 'field of play' and interpret how and why certain factors are related to others in social space. This means thinking of social inequalities beyond measurable and visible factors and in terms of what powers influence the position of individuals within a given field. Rather than asking which factors are common, we must ask how and why particular social factors are together, and theorize as to why and how they remain together over time.  34  As described earlier, this approach requires an entirely different way of looking at the social world. It requires one to think in terms of relations between things, people, space and other social factors. Bourdieu's relational approach requires that we break from substantialist reasoning to offer theoretical insights regarding particular relational cases of social spaces of power and struggle. Considering the nature of the field, the capitals included within the field, and relationally interpreting the nature of the space using theoretical concepts such as habitus, doxa and time will help to illuminate the social forces and powers at play within particular social fields and how they are bundled with factors related to health.  At this point it would be useful to offer an example of how one might incorporate these theoretical principles into a reliably Bourdieusian study of health inequalities. The following chapter will describe how to implement a thoroughly relational analysis of a social space, including a description of how to construct a theoretically accurate social space (field) and how to theoretically interpret and discuss findings in a way that realizes the full potential of a Bourdieusian approach to health research. In order to investigate the relational theoretical principles and conceptualization outlined in chapter one, I use original Canadian survey data from Toronto and Vancouver to implement a descriptive and exploratory multi- dimensional scaling technique called multiple correspondence analysis (MCA). The results of the analysis will be graphically depicted in a three-dimensional (3D) space using advanced modelling software (XLSTAT by Addinsoft, 2007; Miner3D, 2011).  35 Chapter  3: Methods for Building and Analyzing a Canadian Social Space  3.1 Introduction The theoretical principles described in the previous chapter guide model construction and analysis, allowing for a theoretically sound investigation of the structure of forces present in a Canadian social space. The primary goal of the analysis is to identify the structure of relations and complex patterns between the objective positions and subjective dispositions that occupy social space and affect or shape health-related behaviours and outcomes. This mode of analysis facilitates adoption of a relational approach and will permit me to offer insights regarding how the attributes of people and structural resources in their environments are interrelated with social relations, practices and health within a general Canadian social space. Identifying complex patterns and interrelationships among these social mechanisms will contribute to understanding the nature of social stratification in Canada. Such an analysis requires several key considerations, both theoretical and analytical, in order to properly construct and appropriately interpret a social space. The following sections of this chapter describe these principles.  3.2 Analytical Principles of Methodological Relationalism and Field Theory The application of a relational theory requires the use of relational methodologies in order to construct and analyze a theoretically-relevant social space. To implement his relational theory of action, Bourdieu draws upon field analysis which requires the researcher to search for the social mechanisms and properties that distinguish 'similar' and 'different' agents in a social space. Field analysis offers an analytic foundation for the structural mapping of  36 socially oriented arenas of struggle and freedom, power and privilege, and is a key spatial metaphor used to illustrate the principles of differentiation embedded in a social setting in which resources, institutions and organizations interact and the habitus operates (Bourdieu, [1994] 1998: p.6.)  The ultimate goal when implementing theoretical principles analytically is to construct an 'analogical model' of a field, a theoretical representation of a social space, that allows the researcher to explore how social factors in the form of objective properties are distributed among groups of individuals in social space. In order to unite theory and method in such a way, the researcher must use analytic 'tools' to depict a social space, distinguish zones of necessity and freedom, and expose an overall picture of the forces that influence the orientation of fields. Thus, to implement a field analysis I identify the most pertinent indicators, properties or principles of division within a larger Canadian field, and attempt to distinguish the system(s) of criteria (or social factors) that could account for the set of meaningful and significant differences that objectively separate entities within the field or enable differences among them to arise and persist. Drawing from these theoretical and methodological principles will allow me to illuminate the distribution of powers constitutive of the structure of the field and discuss the actual constellation of the field in terms of properties, dispositions and choices/actions (the patterns of social life), to recognize the underlying unity of social strategies that form the boundaries and limits of the field: "it is the state of the relations of force between players that defines the structure of the field" (Bourdieu & Wacquant, 1992: p.99).  Consistent with the relational theoretical principles outlined in chapter 2, I implement a descriptive and exploratory multivariate scaling technique called multiple correspondence  37 analysis (MCA), also known as homogeneity analysis. Pierre Bourdieu regularly used similar types of analyses in his own research, mainly because correspondence techniques are fundamentally relational and correspond with the theoretical principles of relational thought, "a technique which thinks in terms of relations" (Bourdieu & Wacquant, 1992: p.97).  3.3 Multiple Correspondence Analysis Multiple correspondence analysis (MCA) is an exploratory analytic technique used to examine the associations between multiple categorical variables by transforming the many associations present within the large matrix of cross-tabulations into a graphical representation/map of the variable categories as scattered points in a computer-generated two- or three-dimensional social space. This technique allows researchers to inductively discover the structure inherent in the data (Clausen, 1998) and permits the transformation of a table of numerical information into a graphical display that facilitates the interpretation and exploration of the information (Greenacre & Blasius, 1994). This relational technique places the data firmly at the centre of the research and essentially follows the notion that "the model should follow the data, not the inverse" (Greenacre & Blasius, 2006: p.5). MCA follows a set of theoretical and empirical rules necessary to set up an exploratory social space that is robust and operates as a 'snapshot' of social reality. The researcher must attend to several important issues when constructing and interpreting a social space. The following sections will explain the nature of the important decisions that must be made when transferring relational theory to method.    38 3.4 Strategy for Multiple Correspondence Analysis  3.4.1 Step 1: Designating Active and Passive Factors The construction of a social space, as with the construction of any social model, must start with a general understanding of the logic of the space being studied. This means having a theoretical, empirical and intuitive7 understanding of the social factors that are believed to be the most relevant representations of positions within social space (Bourdieu & Wacquant, 1992: p.233), the aim being "to link the pertinent data in such a manner that they function as a self-propelling program of research capable of generating systematic questions [...] which can be put to the test" (Bourdieu & Wacquant, 1992: p.231). An understanding of the underlying logic of a field helps one to better distinguish what social factors are to be included as either objective positions or subjective position-taking. In terms of MCA analysis, one has to make the important choice of which social factors (i.e., variables) will be included as structuring or supplementary factors, also known as active or passive variables, respectively, which are directly related to the theoretical principles of field analysis described by Bourdieu (Greenacre & Blasius, 2006; Le Roux & Rouanet, 2005; Rouanet, 2006).  In analytic terms, the 'active' factors are those that actively determine the geometric orientation of the social space. Conversely, the 'passive' factors, also known as 'illustrative' or 'supplementary' factors, have no influence on the structuration of the social space; they merely "support and complement the interpretation of the configuration of active variable categories" (Greenacre & Blasius, 2006: p.31). In relation to Bourdieu's relational theoretical principles, this means that the active variables should include social factors that are thought  7 Bourdieu brackets this statement as follows: "ordinary intuition is quite respectable; only, one must be sure to introduce intuitions into the analysis in a conscious and reasoned manner" (Bourdieu & Wacquant, 1992: p.108).  39 to be active at all times in social space. Decisions regarding active or passive variables must be theoretical in nature.  There are many conflicting interpretations of how Bourdieu constructed his social space and which factors he made active and passive. For instance, Rouanet (2006) tells us that Bourdieu "puts age, father’s profession, education level, and income as supplementary variables to demonstrate that differences in lifestyle can be explained by those status variables" (p.144), leading us to believe that Bourdieu wanted to see how values and dispositions (i.e. stances) structure social space, and how objective factors are related to the 'stances.' Lebaron (2009) on the other hand, believes that "questions on tastes and cultural practices were taken as active questions of the analysis; socio-demographic and occupational questions were used as supplementary questions" (p.15). While these interpretations make methodological sense, they do not cohere with the theoretical principles outlined thus far.  Recall that social fields are thought to be structured by the relational actions of individuals in social space which are informed by the amount and constitution of their material and non-material properties (capitals), their dispositions (habitus), perception of the social world around them (doxa) and their behaviours (choices and actions); thus relationally- active individuals are the starting point of Bourdieu’s theory of action. The foundation of Bourdieu's theory of action considers all social factors as being potentially influential in any given space, thus, it is not theoretically (or methodologically) appropriate to determine a priori which social factors structure a social space. The only value judgements a researcher should make before analysis pertain to which theoretically relevant social factors are to be included in the analysis, not which factors are more important than others, the latter being in direct opposition to the break from substantive reasoning and common sense understanding  40 of the social world that is central to Bourdieu's framework (Bourdieu & Wacquant, 1992: p.234-235). Furthermore, because we cannot know for certain which factors will be influential in a particular social space (field), we cannot assume that some factors will carry more power and force than others. It is only at the interpretive stage that the most influential social forces reveal themselves as active and pertinent in the field, those which produce the most relevant differences and confers the most power in the social space (Bourdieu & Wacquant, 1992: p.101). While there are many differing interpretations of the methodological decisions made by Bourdieu, I have opted to stay true to the relational principles detailed thus far and consider all social factors included in the analysis as active variables.  The first step in building a social space is fundamentally grounded in Bourdieusian theory and is thus an analytic process that is fundamentally grounded in a logical understanding of the nature of social reality (ontology) and what represents knowledge or evidence of social phenomena (epistemology) which then inform the selection of a proper methodology that allows for the best dissemination of social phenomena and social reality. Once a researcher has gone through the process of choosing active and supplementary variables, the next important step is to code and clean the data so to minimize inconsistencies in the model.  3.4.2 Step 2: Coding and Cleaning Active and Passive Variables There are two important points to be made with respect to coding variables in MCA. First, when working with survey data, one often encounters what Le Roux and Rouanet (2005) call 'junk modalities,' 'other' categories that are present in the data but are not genuine categories.  41 These 'other' categories are only problematic if their variables belong to the active variables, in which case the 'other' categories would negatively affect the structuration of the space. If this is the case, the best way to discard of them is to treat them as passive or supplementary variables, easily accomplished in all MCA data analysis software packages. The same cannot be said for neutral (such as the 'middle' category on a five-point Likert scale), 'don't know' and 'refused' categories. It is important to keep categories that make a statement, that are reflective of an internal belief. When asking someone about their internal values and dispositions, respondents who say that they are 'neutral', 'don't know' or 'refuse to answer' are saying something meaningful, are taking a stance or position on the question. These stances should not be discounted when choosing 'active' categories.  The second important consideration pertains to the relative frequency of categories included in the analysis (Le Roux & Rouanet, 2005). Categories with small frequencies (low masses) can tend to contribute quite highly to total inertia (explained more later) and thus contribute too much to the solution. It is important to search for 'rare modalities' by identifying cell frequencies less than 5% and by comparing the total mass points contributions with the total inertia points contributions in the output, where low mass values and relatively high total inertia values are indicators of rare modalities. In such cases, a simple solution is to re-group (re-code) the categories in a substantively relevant way (Greenacre & Blasius, 2006; Le Roux & Rouanet, 2005). I have taken great care in ensuring that the categories have a sufficient number of respondents, i.e., with cell proportions greater than 5% wherever possible. It should be mentioned that this rule is more important for 'active' than ‘passive’ categories.  42  Based on these two principles, I have decided to include all 'neutral' categories as active. With the exception of parental education which had a high frequency of 'don't know' responses that is included as an active category, I have treated all other 'don't know', 'refused' and 'missing' categories as passive factors due to their small frequency distributions. Once this step is completed, we can then run the analysis and interpret the social space (field). This requires two more important considerations.  3.4.3 Step 3: Determining the Number of Dimensions After running the MCA, we must examine relations between the categories of variables and how they are grouped together, both numerically in table form and geometrically in a statistically generated space. Once the variables have been chosen and positioned as either active or passive factors, the analysis reduces the data into a number of dimensions that each explain different proportions (percentage) of the variance in the data. The objective of the dimension reduction step is to represent the maximum possible variance in as few dimensions as possible (Greenacre & Blasius, 2006), which means that each dimension will explain a certain amount of variance among the active and passive variables. The variance explained in each dimension is represented by a series of eigenvalues which essentially provide a numerical representation of the percentage of explained variance (or inertia) of each dimension in the model. Generally speaking, the higher the inertia, the more spread out and dispersed (distinct and different) are the categories in space, meaning that there is more variance among respondents and the profiles in the multidimensional space will be more distinct. Higher total inertia means that there is a larger amount of variation or difference in  43 the data (Greenacre, 1994: p.12). Blasius (1994) describes three approaches one can use to determine the number of dimensions to include in an analysis: “(a) Consider all those with eigenvalues that explain more than average inertia. (b) Examine a ‘scree plot’ of the eigenvalues to identify the ‘elbow in the descending sequence – consider those eigenvalues at and above the elbow. (c) Use the application-based method of including all dimensions that have a coherent substantive interpretation” (Greenacre & Blasius, 2006: p.19). Essentially, the choice of number of dimensions to include in an analysis should make empirical (i.e., eigenvalues and scree plot) and substantive sense (i.e., make theoretical sense and be meaningfully different). Typically, the decision on number of dimensions has been largely limited by the unavailability of advanced graphical display software capable of displaying three-dimensional spaces, leaving researchers stuck with two-dimensional maps even when consideration of a third dimension was empirically and theoretically warranted (Greenacre & Blasius, 2006). Given the rapid advancement of data modelling programs such as XLSTAT (Addinsoft, 2007) researchers can now incorporate a third dimension into an analysis of social space. A 3D solution can greatly improve the quality of the explained variance in the data and increase the size of the interpretable space, thus allowing for the inclusion of a larger number of relevant variables (Rovan, 1994). As Greenacre and Blasius (2006: p.28) explain, two dimensions are not always sufficient to mirror the structure of variables adequately, where in higher-dimensional solutions certain variables might be further away from one another than they were when confined to fewer dimensions. As will be demonstrated, the inclusion of a third interpretable dimension offers more physical and substantive room for illuminating differences among variables which facilitates deeper theoretical interpretations as well.  44  3.4.4 Step 4: Exploring and Interpreting a Social Space Once we have properly chosen the variables and determined the number of dimensions to include, the final step is the interpretation of the graphical depiction of the social space. The analysis of a 2D or 3D map comprises two stages: 1) interpreting each of the dimensions (axes) by looking at important questions and important categories (interpreting the eigenvalues and factor contributions, and 2) visually inspecting the positions of categories in the graphical map and where they lie relative to one another (Greenacre & Blasius, 2006; Le Roux & Rouanet, 2005; Rouanet, 2006). The first stage of interpretation requires the researcher to investigate the total contribution each variable (and/or category of the variable) makes to each of the dimensions. As a general rule of thumb, Le Roux and Rouanet (2005: p.218) assert that the items that together account for a substantial amount of the variance in each dimension (normally 75% or more) provide a good overall representation of the influential factors in each axis. It is also important to look at the factor contributions of each category in each dimension to see how much each contributes to the arrangement of the dimensions (axes) (Greenacre & Blasius, 2006). The higher the factor contribution, the further away from the center point of the social space (0,0 point on an axis) they will be.  The second stage is to explore the patterns of categories in the social space, where the job of the analyst is to interpret the relative position of each category to another and how they are patterned into clouds or clusters of properties and individuals (property-taking). This visual interpretation of relative distance between points in geographic space allows one to gauge the relative similarity and dissimilarity of individuals based on their location in the field. When describing the nature of social space, Bourdieu affirms that spatial distances on  45 paper are equivalent to social distances, which essentially means that "all agents are located in this space in such a way that the closer they are to one another in [the] two dimensions, the more they have in common; and the more remote they are from one another, the less they have in common" (Bourdieu, [1994] 1998: p.6).  With respect to presenting and interpreting a 3D space, the literature in the social sciences is almost non-existent (although Rovan, 1994 provides a simple description). I have yet to come across a step-by-step approach to analyzing a three-dimensional social space, and have relied on trial and error to come up with a solution which facilitates the presentation and interpretation of a 3D space in a 2D print format. Generally speaking, the space will consist of a series of categories (in the form of equally sized spheres) distributed in a three- dimensional space. I use some of the advanced functions in XLSTAT to present the clusters of categories in the social space in a manner that facilitates interpretation.  Following these four steps allows for the production of a social space that "makes a part or an aspect of the reality in question materially or hypothetically intuitively accessible" (Levi Martin, 2003: p. 36). In order to implement an analysis of a health relevant social space that will facilitate an inductive assessment of the relational properties and social forces that influence the segregation of individuals in the field, and how these forces are related to health, I use original cross-sectional survey data collected in Toronto and Vancouver in 2009. The next two sections describe the data and the operationalization of variables included in the analysis.     46 3.5 Sampling and Data Collection I utilize data from a research project conducted by Gerry Veenstra at the University of British Columbia and funded by the Social Sciences and Humanities Research Council of Canada. With the assistance of the Survey Research Centre at the University of Victoria, cross- sectional survey data was collected from randomly-selected adults in Toronto and Vancouver in early 2009. A random-digit dialling sampling technique was used to obtain residential telephone numbers, a next-birthday strategy followed to select household members to interview and a computer-aided telephone interviewing system was incorporated to conduct the interviews. In total, telephone interviews were conducted with 732 adults aged 19 and older living in the City of Toronto and 863 adults aged 19 and older living in the Vancouver Census Metropolitan Area (n=1,595). The study garnered a 9.3% cooperation rate, with a response rate (completed interviews / eligible respondents) of 4.1%. The data provides detailed measures of health-related values, dispositions, behaviours and outcomes in addition to a rich variety of cultural indicators useful for analysis of a Canadian health-field.  3.6 Operationalization of Social Factors As outlined above, there are several principles to consider when choosing concepts to include in an analysis of social space. I have included theoretically- and empirically-relevant variables that reflect social factors deemed theoretically relevant by Bourdieu such as conditions of existence, personal dispositions and values, behaviours and practices, demographic characteristics and a series of variables found to be useful indicators of health. The following sections provide a brief description of the variables included in each group of social factors, how they were operationalized and how they will be included in the MCA. For  47 each group of variables, I provide a table containing: 1) the survey item used to generate the variables included in the analysis and 2) a shortened version of each variable category for use in the mapping of the social space. Frequency distributions for these variables are included in Appendix A.  3.6.1 Conditions of Existence The first variables are commonly used as key structural level factors in Western societies and are frequently used in social research to reflect general objective properties related to social standing. These variables are known to be important and relevant indicators of social position, including household income, parental education, personal education, occupation type and home ownership (see Table 1 for the full list of variables and variable categories). Household income is included as a seven category indicator of the total income of all household members before taxes and deductions. For parental education status, respondents were asked about the highest level of education achieved by their father and their mother. The responses to these two questions have been combined into a single four-category variable (two parents with university degree, one parent with university degree, no parent with university degree, don't know). Personal education reflects highest level attained and has been coded into five categories. Respondents were asked what their main occupation was at the time of the survey. Their responses were categorized using the Canadian national occupation classification system (CNOCS) which offers a standardized way to organize occupations into 520 occupational group descriptions that can be re-structured into as few as nine broad categories. For the purposes of this analysis, I have grouped the occupations into  48 15 of the CNOCS's parent categories. Finally, home ownership status is categorized as owning a home without a mortgage, owning a home with a mortgage, and renting.  3.6.2 Demographic Composition The next group of variables reflects demographic composition, e.g., age, gender, marital status and immigration status (see Table 2 for the full list of variables). These variables are included because they reflect fundamental elements of opposition and difference in society and may be influential factors regarding the theoretical principles of habitus and doxa (Bourdieu and Wacquant, 1992: p.134). Age has been coded into six categories of roughly equal ten-year intervals. Gender is grouped into the typical male or female categories. Marital status is grouped into married, living with a partner, widowed, separated/divorced, and never been married. Immigrant status is reflective of the number of years the respondent has been living in Canada. Responses have been coded into four categories: born in Canada, immigrated more than 20 years ago, immigrated 10-19 years ago, and immigrated fewer than nine years ago.  3.6.1 Dispositions, Values, Attitudes and Tastes Personal dispositions are included for their theoretical relevance to Bourdieu's theories of habitus and doxa. In order to capture the internalized dispositions of acting individuals, I have included a series of measures pertaining to personal dispositions, values, attitudes and tastes towards social objects (clothing and food), social behaviours (cooking and traveling), other individuals in social space (how they think they will get along with others), and how others perceive them (style of clothing and fashion).  49 Table 1: Conditions of existence variables with mapping labels  Variables Categories Social space mapping labels Household income Less than $40,000 < $40,000 $40,000 - 59,999 $40,000-59,999 $60,000 - 79,999 $60,000-79,999 $80,000 - 99,999 $80,000-99,999 $100,000 - $149,999 $100,000-149,999 $150,000 or more $150,000+ Parental educational status Both parents with university degrees Both uni degree-PAR-ED One parent with a university degree One uni. degree-PAR-ED Neither parent with a university degree No uni. degree-PAR-ED Other (don't know) Don't know-PAR-ED Personal educational attainment Less than high school < high school-PERS-ED High school graduate High school-PERS-ED Community college or technical school diploma College/tech-PERS-ED Bachelor's degree Post-bachelor's degree Occupation type Senior & Specialist Management Occupations Senior/Specialist/Manager-JOB Other Managers n.e.c Other Manager-JOB Professional Occupations in Business and Finance Professional-Busin/Finance-JOB Clerical Occupations Clerical-JOB Professional Occupations in Natural and Applied Sciences and Health Professional- Nat.App.Sci./Health-JOB Technical and Related Occupations in Health Tech.&Related/Health-JOB Judges, Lawyers, Counsellors, and Policy and Program Officers Professional-Judicial /Council./Policy-JOB Teachers (primary/secondary) Teacher (Prim/Sec)-JOB Professor (college/university) Professor (Coll/Uni)-JOB Paralegals, Protective Services, Social Services Workers and Occupations in Education and Religion, n.e.c. Social-JOB Professional & Technical Occupations in Art and Culture Professional-Art & Cult-JOB General Sales and Service Occupations General-Sales & Service-JOB Specialty Sales and Service Occupations Specialty-Sales & Service-JOB Childcare and Home Support Workers Childcare/Home supp.-JOB Trades, Transport and Equipment Operators and Related Occupations Trades/Labour-JOB Home ownership Fully owned Owned-Home Partly owned Partly-owned-Home Rented Renting-Home     50 Table 2: Demographic composition variables and mapping labels  Variables Categories Social space mapping labels Gender Male  Male Female Female Age Age-19 - 30 Age-19 - 30 Age-31 - 40 Age-31 - 40 Age-41 - 50 Age-41 - 50 Age-51 - 60 Age-51 - 60 Age-61 - 70 Age-61 - 70 Age-71 and older Age-71 + Marital status Married Married Living with a partner Living with partner Widowed Widowed Separated/divorced Separated/divorced Never been married Never married Immigrant status Born in Canada Born in Canada Immigrated to Canada more than 20 years ago >20 yrs.-Immigrate Immigrated to Canada between 10 and 19 years ago 10-19 yrs.-Immigrate Immigrated to Canada 9 or fewer years ago < 9 yrs.-Immigrate  In order to measure personal dispositions, respondents were asked a series of statements pertaining to personal dispositions and beliefs about dress/fashion, food/cooking/entertaining and travel (see Table 3 for the full list of variables). For each statement, participants were asked to rank their agreement or disagreement with the statement on a five-point Likert scale (strongly agree, agree, neither agree nor disagree, disagree, strongly disagree, don't know, refused). I have included 17 of these value statements in my analysis. Considering the frequency distributions and substantive value of each category, I have recoded many of the variables into four categories (category recodes are highlighted in Table 3). In addition to these variables, I included a series of questions related to musical likes and dislikes. Respondents were asked “For each of the following types of music, please tell me whether you like or dislike or perhaps feel neutrally about each type: classical music, hip hop, choral music, folk music, rap, opera, country music, pop, jazz, easy listening, reggae, rock, heavy metal, musical theatre, gospel, blues, new age, big band, golden oldies, world/international,  51 disco.” They were then asked “You've mentioned that you like <read list>. Which one of these is your absolute favourite?” and “You've mentioned that you dislike <read list>. Which one of these do you dislike the most?” which enabled the creation of two variables pertaining to most liked and disliked musical genres. With regards to most liked genres, cell sizes enabled me to distinguish between classical, folk, country, pop, jazz, easy listening, rock, blues and golden oldies in particular. With regards to most disliked genres, cell sizes enabled me to distinguish between hip hop, rap, opera, country and heavy metal.  3.6.1 Practices and Behaviours Practices and behaviours are key aspects to Bourdieu's theory of 'action' and must be included to capture the action elements of human behaviour and reflect individual-level actions that people take, or have taken, in their day-to-day lives. The indicators pertain to smoking frequency, exercise frequency, fruits and vegetables consumption, last restaurant visited, recent travel destinations, wine-tasting event attendance and musical instrument ability (see Table 4 for the full list of variables). These variables have been included for their relevance to health (i.e., smoking, diet and exercise) and for their cultural significance as indicators of social differentiation (i.e., restaurant choices, travel choices, special function attendance and musical instrument abilities). With regards to health-relevant practices, smoking is categorized into those who are daily or occasional smokers, former smokers, and those have never smoked. Exercise frequency has been categorized to distinguish between respondents who exercise less than once a week, one to three times per week, and more than three times per week. Diet is operationalized by fruit and vegetable consumption and is  52 divided into respondents who eat fruits and vegetables several times per day, about once per day, and less than once per day.  Respondents were asked an open-ended question regarding the last restaurant at which they ate out. The variable was coded using attributes of the restaurants such as price, value, speed of service, menu selection, comfort, style/ambiance, service, dining experience, food quality, wine selection and chef’s training to distinguish between five categories: 1) fast-food/convenience, e.g., McDonalds, Subway and Burger King, 2) family-style dining, e.g. Pizza Hut and Applebee’s, 3) casual-dining/midscale, e.g., Red Lobster, Sammy J. Peppers and White Spot, 4) casual-fine dining, e.g. The Keg, The Cactus Club, Spring Rolls and Earls, and 5) fine-dining, e.g. Boulevard Club, Oro, West and Bacchus (Muller & Woods 1994; Noone et al., 2007).  Recent travel destination was calculated by coding open-ended variables produced by two survey questions: “Where was your last vacation trip?” and “How about the trip before that, where did you go?” These questions enabled me to create a variable that determined the most exotic and far-flung destination to which a given respondent had travelled in her/his last two vacation trips. Categories reflected local trips (within province), trips within Canada (outside of own province), trips to the mainland USA (including Alaska), trips to Central and South America (Caribbean and South Pacific included) and overseas trips (Europe, Asian, Africa, Australia/Oceania). For its cultural significance, I have also included two dichotomous (yes/no) questions related to wine-tasting, where respondents were asked "Have you ever attended a wine tasting event?", and musical ability, where respondents were asked “Do you play any musical instruments?”   53 Table 3: Value, attitude, taste and belief variables with mapping labels  Dress-related dispositions Social space  mapping labels I prefer to buy clothes that will last a long time Last-Clothes*** I probably spend more money on clothes than I should Spend too much-Clothes* I like to dress fashionably Fashionable-Clothes* I think that dressing well often leads to success Dress = Success-Clothes*** I want my clothes to reflect who I am Reflect me-Clothes*** I like to wear clothes that make people look at me Draw attention-Clothes** I can often tell how well I will get along with someone by what they're wearing Get along-Clothes** Food-related dispositions I like to cook Like to cook-Food* In my home we like to experiment with new foods, new recipes and new ingredients Experiment-Food*** I think it is important that nearly all of the food I eat is good for me Good for me-Food*** I think that it is important to eat at least three good meals a day 3 meals/day-Food*** I have a sweet tooth Sweet tooth-Food*** I prefer to eat at restaurants that serve larger portions so that I get my money’s worth Big portions-Food** Travel-related dispositions The most important thing for me when I go on vacation is to relax Imp. relax-Travel*** I like to learn about new places and different ways of life when I go on vacation Learn of others-Travel*** My favourite vacations are off the beaten path, places where most tourists don’t go Off beaten path-Travel*** I usually visit as many museums as I can when I’m on vacation Visit museums-Travel*** Music-related dispositions Most liked musical genre Most Liked--Classical Most Liked--Folk Most Liked--Country Most Liked--Pop Most Liked--Jazz Most Liked--Easy listening Most Liked--Rock Most Liked--Blues Most Liked--Golden Oldies Most disliked musical genre Most Disliked--Hip hop Most Disliked--Rap Most Disliked--Opera Most Disliked--Country Most Disliked--Heavy Metal * Variable included in map with all five SA (strongly agree), AG (agree), NE (neutral), DI (disagree), and SD (strongly disagree) categories. ** SA and AG recoded into SA&AG (strongly agree and agree), variable included in map with four categories. *** DI and SD recoded into DI&SD (disagree and strongly disagree), variable included in map with four categories.     54 3.6.2 Health Indicators Important to the health-focus of my research, I have included five commonly-used indicators of health status to assess general, mental and physical aspects of health. These measures include self-perceived overall health, self-perceived mental health, level of stress in daily life, frequency of feeling depressed and body mass index (BMI) at time of interview (see Table 5 for the full list of variables). For self-perceived overall and mental health, respondents were asked: "In general, compared to other people your age, would you say your overall health is" and then asked a similar question regarding mental health. Answers to these questions were originally categorized with a five-point Likert scale, and due to response frequencies were each coded into three response categories (excellent/very good, good, and fair/poor). Respondents were also asked: "How often do you feel depressed?” with response categories every day, most days, sometimes, rarely and never. Responses were re-coded into three categories (rarely/never, sometimes, most days/every day). Regarding stress level, respondents were asked "Thinking about the amount of stress in your life, would you say that most days are:" with response categories being not at all stressful, a bit stressful, quite stressful, and extremely stressful. Response categories were left as is. Finally, body-mass index scores were calculated from respondents’ height and weight and were then coded into four distinct categories: underweight (BMI = 16 to 19.9), normal (BMI = 20 to 24.9), overweight (BMI = 25 to 29.9) and obese (BMI = 30 and over).   55 3.7 Conclusion My construction and interpretation of a robust social space follows the steps outlined in chapters 2 and 3 and is implemented in SPSS version 17. The results of the MCA analysis are then imported into a data modelling software package called XLSTAT by Addinsoft (2007) useful for modelling two- or three-dimensional spaces. The dimensions that explained the most variability overall are depicted visually in a 3D correspondence map, and interpretations or speculations regarding the category groupings are presented.  Table 4: Practice and behaviour variables and mapping labels  Variables Categories Social space mapping labels Smoking Daily or occasional smoker Smoker Former smoker Former-smoke Never smoked Non-smoker Exercise Exercise less than once per week Low-Exercise Exercise one to three times per week Med-Exercise Exercise more than three times per week High-Exercise Diet Eat fruits and vegetables several times per day High-Fruits Eat fruits and vegetables about once per day Med-Fruits Eat fruits and vegetables less than once per day Low-Fruits Last restaurant visited Fast-food/convenience Fast-food/conv. Family-style dining Family-style dining Casual-dining/midscale Casual-dining/midscale Casual-fine dining Casual-fine dining Fine dining Fine dining Recent Travel Local trip (within province) Local-Trip Trips within Canada (outside of own province) In Canada-Trip Trips to mainland USA (including Alaska) USA-Trip Trips to Central and South America (Caribbean and South Pacific) Cent/South Amer.-Trip Trips overseas (Europe, Asian, Africa, Australia/Oceania) Overseas-Trip Wine tasting Have you ever attended a wine tasting event? (Yes/No Wine Tasting (Y/N) Any instrument Do you play a musical instrument Yes no    56 Table 5: Health indicators and mapping labels  Variables Categories Social space  mapping labels Self-perceived overall health Excellent/Very Good EX/VG-Overall Health Good Good- Overall Health Fair/Poor F/P-Overall Health Self-perceive mental health Excellent/Very Good EX/VG-Mental Health Good Good-Mental Health Fair/Poor F/P-Mental Health Stress Not at all stressful No stress A bit stressful A bit stressful Quite stressful Quite stressful Extremely stressful Extremely stressful Depression Rarely/never Rare/Never-Depression Sometimes Sometimes-Depression Most days/every day Most/Every day-Depression BMI underweight (BMI = 16 to 19) Underweight-BMI normal (BMI = 20 to 25) Normal-BMI overweight (BMI = 26 to 30) Overweight-BMI obese (BMI = 31 or over) Obese-BMI   57 Chapter  4: Analysis of a Canadian Field  “My entire scientific enterprise is indeed based on the belief that the deepest logic of the social world can be grasped only if one plunges into the particularity of an empirical reality, historically located and dated, but with the objective of constructing it as a "special case of the possible" [...] as an exemplary case in a finite world of possible configurations [...] the aim is to grasp the invariant, the structure in each variable observed” (Bourdieu, [1994] 1998: 2).  4.1 Results This chapter describes the results of the multiple correspondence analysis (MCA). I start with a description of the most influential variables (eigenvalues (EV)) and factor contributions of the categories for each dimension, and then provide a visual description of the relational cluster patterns within the social space. Given the complexity of presenting a three- dimensional space in a two-dimensional format, I provide a series of static 'snapshots' of important areas in the social space from different angles in order to facilitate the presentation of the results. For an in depth perusal of the social space model, please visit which has the full 3D model available for exploration.  In the first section (section 4.2) I discuss the dimensions of the field, essentially providing a general description of the social factors that are most relevant in structuring the field. The second section (section 4.3) explores the internal structure of the field, which means describing the general constitution of each of the different groups located in the space of the field and the common attributes and properties of the actors who are part of each  58 group. In the final section (section 4.4), I pay especially close attention to two of the most distinctly different groups (different in terms of spatial distance and health) in order to offer a deeper understanding of the overall structure of the field.  4.2 Dimensions of the Field The interpretation of dimensions reveals which social factors emerge as influential in structuring the field in question (in this case, a Canadian social space). To expose these dimensions, a total of 1,595 cases were utilized in the analysis of which there were 260 active cases without any missing values and 1,335 active cases with missing values (retained in the analysis but treated as passive). The MCA produced four empirically and substantively meaningful and distinct dimensions which collectively accounted for 13.10% of the variance in the data. Dimension 1 (D1) explained 3.93%, dimension 2 (D2) 3.44%, dimension 3 (D3) 3.00%, and dimension 4 (D4) 2.72% of the total variance. The variables which together accounted for 75% or more of the variance in each of the four dimensions will be discussed here. The list of influential variables in each dimension is described in Table 6. Among these influential variables, the factor contributions of the most influential categories within each of the four dimensions are also discussed (see Table 7 for a list of most influential categories).  59 Table 6: List of influential variables by dimension Dimension 1 Eigenvalue I think it is important that nearly all of the food I eat is good for me 0.321 In my home we like to experiment with new foods, new recipes and new ingredients 0.275 I think that it is important to eat at least three good meals a day 0.246 I like to cook  0.223 I want my clothes to reflect who I am 0.220 Household income  0.175 I like to dress fashionably 0.168 I like to learn about new places and different ways of life when I go on vacation 0.165 I think that dressing well often leads to success 0.145 I prefer to buy clothes that will last a long time  0.131 I probably spend more money on clothes than I should 0.127 Completed education 0.127 My favourite vacations are off the beaten path, places where most tourists don’t go 0.115 I like to wear clothes that make people look at me 0.113 Dimension 2 Home ownership status 0.239 How often do you feel depressed? 0.232 Self-perceived overall health 0.189 Marital Status 0.181 Smoking status 0.165 Self-perceived mental health 0.164 Household income  0.157 I prefer to eat at restaurants that serve larger portions so that I get my money’s worth 0.142 Age  0.141 Fruits and vegetables consumption 0.138 I like to dress fashionably 0.127 Completed education 0.108 Dimension 3 Age  0.522 Marital Status 0.279 I prefer to eat at restaurants that serve larger portions so that I get my money’s worth 0.157 Most liked musical genre 0.155 Home ownership status 0.139 I like to wear clothes that make people look at me 0.122 I can often tell how well I will get along with someone by what they're wearing 0.103 Dimension 4 I like to dress fashionably 0.384 I like to wear clothes that make people look at me 0.294 I probably spend more money on clothes than I should 0.268 I want my clothes to reflect who I am 0.254 Gender 0.218 Occupation type 0.205 I think that dressing well often leads to success 0.185 Completed education 0.119 Most liked musical genre 0.101   60 4.2.1 Dimension 1: Dispositions The first dimension (D1) is influenced primarily by disposition-related factors, the strongest being food-related dispositions such as eating food that is healthy (EV=.321), experimenting with food (EV=.275), eating three good meals per day (EV=.246) and liking to cook (EV=.223). Household income (EV=.175), completed education (EV=.127) and age (EV=.110) have smaller yet notable influences on the variation in D1. The categories among these influential variables are generally distributed in a linear fashion along D1, whereby agreement with dispositional statements related to food, dress and travel, as well as income, education and age, are distributed along the dimension (see the distribution of categories in Table 7).  This first dimension, which I call the dimension of dispositions, speaks to Bourdieu's theoretical concept of habitus. This dimension is the most influential dimension in the field. The social factors common to this dimension tell us that the internalized elements of distinction, values, beliefs, tastes and dispositions (among other factors) are influential factors in the field. These influential categories reflect differentiated and differentiating principles that also indicate the important principles of classification in the field, that is, the different tastes and dispositions that actors have are important for our understanding of their position in the field.  4.2.2 Dimension 2: Positions and Position-Taking Dimension 2 (D2) is largely influenced and populated by status-related variables that reflect social positioning, such as home ownership status (EV=.239), marital status (EV=.181), household income (EV=.157), age (EV=.141) and education (EV=.108), as well as health-  61 related factors, such as feeling depressed (EV=.232), self-perceived overall health (EV=.189), smoking (EV=.165), self-perceived mental health (EV=.164), fruit and vegetable consumption (EV=.138) and to a lesser extent, exercise (EV=.014). Within these influential variables, the categories with the highest contribution to D2 are renters, high depression, smokers, fair/poor mental and overall health, low fruit consumption and low income (see Table 7).  The composition of the variables and contributing categories generally reflect what Bourdieu called positions (i.e., properties of living conditions (capitals): home, income, education, marital status) and position-taking (i.e., choices made by social actors: smoking habits, fruit consumption, exercise frequency), and will thus be referred to as the dimension of positions and position-taking. Bourdieu generally perceives the space of social positions as being continually retranslated into a space of position-taking (i.e., properties inform practices and vice versa), whereby "the system of differential deviations which defines the different positions in the two major dimensions of social space corresponds to the system of differential deviations in agents' properties (or in the properties of constructed classes of agents), that is, in their practices and in the goods they possess" (Bourdieu, [1994] 1998: p.7). This seems to be reflected in the general composition of D2, where objective positional elements such as income, education, parental education and home ownership tend to be relationally located among similar systems of practices and expressions of agents, such as smoking, exercise, diet and health.  Thinking once again in terms of the field, the general composition of D2 tells us that the conditions of one's existence, one’s actions and one’s health are influential secondary factors in the game being played and therefore are important factors to consider when  62 exploring the composition of each group (i.e., observable properties and resources common to people in each grouping) and the general actions of people in their daily lives. The inclusion of this dimension allows us to better understand the objective properties and actions of the people within each grouping.  4.2.3 Dimension 3: Time Dimension 3 (D3) is heavily influenced by life-course factors, the most influential being age which explains a comparatively large 5.2% (EV=.522) of the variance in D3 (age has the largest overall effect of all the variables in the social space). Other influential factors include marital status (EV=.279), home ownership (EV=.139), preference for musical genre (EV=.155), eating at restaurants that serve large portions to get their money's worth (EV=.157) and several other disposition measures related to life-course stage. Among these variables, the most influential categories in dimension 3 are largely composed of the highest and lowest age groups (over 70 and 19-30) and widowed which account for the highest contributions and reflect the lower and uppermost positions on the dimension. Less influential but worth mentioning are strongly disagreeing about visiting restaurants with big portions, owning a home, and never being married which account for a notable amount of variation on D3. While not immediately apparent in Bourdieu's analyses in Distinction, his depiction of social space was a three-dimensional one (see Bourdieu, [1979] 1984: p.114; Bourdieu & Wacquant, 1992: p.137).   63  Table 7: List of influential categories by dimension* Dimension 1 Factor contribution  Dimension 2 Factor contribution SA-Good for me-Food 0.050 Renting-Home 0.047 SA-Experiment-Food 0.045 Most/Every day-Depression 0.041 SA-3 meals/day-Food 0.043 Smoker 0.040 SA-Reflect me-Clothes 0.042 F/P-Mental Health 0.037 SA-Like to cook-Food 0.031 F/P-Overall Health 0.035 SA-Fashionable-Clothes 0.028 Low-Fruits 0.028 AG-Good for me-Food 0.027 SA & AG-Big portions-Food 0.027 SA-Dress = Success-Clothes 0.026 < $40,000 0.026 SA-Learn of others-Travel 0.025 Age-19-30 0.026 $150,000+ 0.025 Never married 0.022 SA-Off beaten path-Travel 0.024 Married 0.020 SD-Big portions-Food 0.020 EX/VG-Overall Health 0.018 Age-71+ 0.020 Owned-Home 0.018 SA-Last-Clothes 0.019 SA-Reflect me-Clothes 0.017 DI & SD-Experiment-Food 0.019 SA-Imp. relax-Travel 0.017 DI-Draw attention-Clothes 0.016 Rare/Never-Depression 0.017 DI-Like to cook-Food 0.016 SA-Fashionable-Clothes 0.016 Widowed 0.016 DI & SD-Good for me-Food 0.016 < $40,000 0.015 0.015 DI-Fashionable-Clothes 0.015 Low-Exercise 0.014  Dimension 3 Factor contribution Dimension 4 Factor contribution Age-71+ 0.066 SA & AG-Draw attention-Clothes 0.061 Widowed 0.059 SA-Fashionable-Clothes 0.059 Age-19-30 0.053 Male 0.051 SD-Big portions-Food 0.033 SA-Spend too much-Clothes 0.041 Owned-Home 0.030 SD-Fashionable-Clothes 0.038 Never married 0.029 SA-Reflect me-Clothes 0.038 SD-Draw attention-Clothes 0.026 SD-Draw attention-Clothes 0.034 Age-31-40 0.024 SA-Dress = Success-Clothes 0.033 SD-Get along-Clothes 0.023 DI & SD-Reflect me-Clothes 0.030 Age-61-70 0.022 Female 0.029 Both uni degree-PAR-ED 0.021 SD-Spend too much-Clothes 0.028 Not stressful 0.018 0.027 SA-Good for me-Food 0.018 AG-Spend too much-Clothes 0.027 SD-Fashionable-Clothes 0.017 DI-Fashionable-Clothes 0.026 DI & SD-Reflect me- Clothes 0.017 NE-Reflect me-Clothes 0.024 SD-Spend too much-Clothes 0.016 Widowed 0.019 < high school-PERS-ED 0.016 DI & SD-Dress = Success-Clothes 0.019 Most Liked--Golden Oldies 0.015 DI & SD-Off beaten path-Travel 0.016 Partly-owned-Home 0.014 Professional-Nat.App.Sci./Health- JOB 0.016 High-Exercise 0.014 NE-Dress = Success-Clothes 0.016 * These labels are used in the social space mapping to facilitate interpretation of the social space.   64 Bourdieu offers very little explanation of the third dimension in Distinction,8 although some information can be gleaned from his works which describe the third dimension as accounting for the principle of time. Time (and doxa) refers to the 'temporality' of practice and takes into account how the volume and composition of capitals evolve over time, "manifested by past and potential trajectory [habitus and doxa] in social space [field]" (Bourdieu, [1979] 1984: p.114). Theoretical interpretation of social space in terms of habitus and doxa (past experience and future potential) adds context to time and what Bourdieu sees as a third interpretive dimension in a field. In terms of the field in question, the dimension of time allows us to further distinguish how the structure of the field differs for younger and older people.  4.2.4 Dimension 4: Gender Lastly, dimension 4 (D4) is influenced by gender (EV=.218), occupation (EV=.205) (dimension 4 is the only one strongly influenced by gender or occupation) and disposition- related variables such as dressing fashionably, wearing clothes that draw attention, spending more money on clothes than they should, wanting clothes to reflect who they are, and thinking that dressing well leads to success. These fashion-related dispositions therefore appear to be gendered and ‘jobbed’ in nature.  While this gender-specific dimension is less influential than the other three in the social space, it nevertheless brings to light an important aspect of difference and  8 Bourdieu provides limited information regarding the three-dimensional space in Distinction and leaves the reader with the following note: "A fuller presentation of the fundamental principles of [a three-dimensional] construction, i.e., the theory of the different sorts of capital, their specific properties and the laws of conversion between these different forms of social energy [...] is reserved for another book, so as not to overcomplicate the present analysis of the judgment of taste" (Bourdieu, [1979] 1984: p.572)  65 differentiation (male/female oppositions) that is undoubtedly relevant to the structuration of most social spaces. For Bourdieu, gender reflects a fundamental objective and symbolic element of opposition, differentiation and domination in any society, arguing that "gender domination constitutes the paradigm of all domination and is perhaps its most persistent form. It is at once the most arbitrary and the most misrecognized dimension of domination because it operates essentially via the deep, yet immediate, agreement of embodied schemata of vision of the world with the existing structures of that world" (Bourdieu & Wacquant, 1992: p.134). While this dimension represents an important dimension to my overall model and speaks to the under-theorized gender element of Bourdieu's larger body of work, due to technological limitations I am unable to include this dimension into the overall 3D social space (the analytic software does not allow for the inclusion of a fourth dimension).  4.2.5 Final Remarks on the Field as a Space of Play While the above descriptions illustrate the most influential factors in each of the four empirically and substantively distinct dimensions and provide a general understanding of the factors that are influential in the field, they do not speak to the relational nature of each dimension and how they structure the overall space of the field. In order to further investigate the relational properties of the influential categories within each dimension, we need an in- depth and nuanced investigation of the social forces and powers that inform the differences that exist within the larger social space, which means investigating the ‘spaces in between’ the categories. Working from the theoretical and methodological principles described earlier, the following section attempts to identify groupings of categories that may actually represent social groupings of people in social space. In other words, I describe the general constitution  66 of each of seven different groups located in the space of this particular field and the common attributes and properties of the actors who belong to each of the groupings.  4.3 Presenting the Distinct Groups of Social Factors in the 3D Field This section presents a visual depiction of the field to illustrate how the variables and their categories are distributed within a three-dimensional social space. While the MCA identified four dimensions, the modelling software can only generate a space with three axes (i.e., three dimensions). As such, I model a 3D space comprised of dimensions 1, 2 and 3. A Canadian social space also depicting variation by gender will have to wait until technology allows for simultaneous consideration of four dimensions.  In order to make the social space more easily interpretable, I have removed categories that are close to the centre point of the social space (coordinate 0,0) from the visual model. This does not affect the positioning of the remaining visible categories in the social space. The visual presentation of the social space will start with an overall image of the distribution of points in social space without category labels to give a general 'bird's eye view' of the space being interpreted, showing how the categories cluster together overall. Then, to simplify the presentation of the space, I utilize the lattice function in XLSTAT which essentially inserts 'walls' into the 3D space to help visually separate the space into sections that isolate clusters of interest. The analysis reveals seven distinct groupings within the overall 3D social space (see figure 1).  In this context, the field that I be interpreting consists of seven different groups, each of which is made up of a set of individual actors who share common attributes and properties. While I have identified seven distinct groupings of individuals located in the field,  67 given the health-related focus of this thesis I have chosen to discuss the five groups which are located in two distinct spaces in the larger field: the space of freedom (groups one, two and three) and the space of necessity (groups four and five). The groups in each of the two general spaces are located furthest from one another in the field and comparatively reflect the greatest difference and differentiation in dispositions, capitals, habitus and health-related behaviours and health status. While the groups in these two spaces will be the primary focus of the discussion, the characteristics of the two remaining groups (six and seven) are briefly presented and are also discussed in chapter 5. The description of the different groups must be read as a description of the various lifestyles of these groups. This is followed by an analysis of locations in the space of positions of power, the so-called field of power (Bourdieu, 1989, p.16).  Figure 1: General view of category point distributions in a 3D field    68 4.3.1 Group 1: Wealthy, Well-Educated Professionals with Refined Tastes The first distinctive grouping of categories pertains to the collection of variables that are higher on D1 (dispositions), lower on D2 (position and position-taking) and generally near the midway point of D3 (time) (see Figure 2). As a whole, this group appears to be wealthier, highly educated (and have educated parents), have high paying and high ranking jobs, believe they are very healthy, and have distinct cultural tastes. The strongest influences in this grouping pertain to education, income and time, where the actors appear to have spent longer in school obtaining graduate degrees, hold jobs that require extensive education (university or college professor, professional positions in the sciences and health field, senior specialists/managers, professional judicial and counselling occupations, and primary or secondary teachers) and have parents with university educations. These educational factors are relationally bound in this social space with influential factors such as high incomes ($150,000 is higher than $100,000-$149,000 on D3). There also appear to be several cultural factors associated with this group, such as attending wine-tasting events, having recently eaten at a fine dining restaurant, preferring jazz music, and feeling neutral about wearing clothes that make people look at them. In relation to health, this group appears to exercise frequently (though this category is the least influential of the influential categories), have high self-perceived overall health, and do not find it important to relax while on vacation.  4.3.2 Group 2: Young and Hip with Educated Parents The second category grouping pertains to people who are located higher on D1 (dispositions), at the midpoint of D2 (position and position-taking) and generally spread out on D3 (time). This group encompasses the dense space on the positive end of D1 and D2, and  69 is closely located in social space to group 1 (see Figure 3). The circled group of categories located at the bottom of Figure 3 includes people who work in professional art and culture occupations, live with a partner, like rock music and dislike country, have strong feelings towards personal dress but feel neutral about what others wear (i.e., don't think clothing indicates whether they will get along with someone else), and have parents who have both been to university (also an influential category in group 1). This grouping consists of younger/middle aged individuals (31-40 years) and is generally located lower on D3, and while less centralized within this cluster, is closely located to social jobs.  4.3.1 Group 3: Older, Cultured and Opinionated The third grouping is located further up D3 (time), meaning that they are older, and is the highest cluster on D1 (dispositions). This grouping of categories (upper circle in Figure 3) contains very strong dispositions towards food, dress and travel. The respondents in this group strongly believe in eating food that is healthy, eating three meals per day, experimenting with new foods, placing little importance on choosing restaurants for larger portions, and enjoying cooking. They feel strongly about visiting museums when travelling, travelling off the beaten path and learning about others when on vacation and have particular values towards dress, such as buying clothes that last, clothes that reflect who they are, and believing that dressing well often leads to success.    70 Figure 2: Group 1 characteristics   71 4.3.2 Space of Freedom: Groups 1, 2 and 3 Generally speaking, the three groups described thus far share D1 (dispositions) characteristics, that is, similar dispositions towards food, dress, and travel, but are generally differentiated along D2 (positions and position-taking). Group one seems to generally have more money, higher education and prestigious jobs; Group two -- who also have educated parents -- are located further from the high income and high education space and hold cultural/artistic and/or socially-oriented jobs; and Group 3 is higher on D1 and D3, a space consisting of strong cultural dispositions held by older respondents (see Appendix B.1 for a broad view of groups one, two and three). As a whole, the three groups are relationally positioned close to one another in the field. They seem to hold similar capitals, have similar beliefs, similarly come from relatively privileged backgrounds, have similar perceptions of the field around them and share similar lifestyle patterns.  4.3.1  Group 4: Poor, Uneducated Labourers with Strong Dislikes and Poor Health The fourth distinct grouping of categories pertains to people who are located low on D1 (dispositions), high on D2 (position and position-taking) and high on D3 (time) (see Figure 4). Members of this group generally have lower education (and also tend to state that they do not know their parents’ education), lower incomes, work in trades or labour occupations, strongly disagree with dispositions related to food, dress and travel, travel locally, have high levels of depression, have poor overall health, are obese and are likely to be separated or divorced.  72 Figure 3: Groups 2 and 3 characteristics  73 Regarding dispositions, this group disagrees and strongly disagrees with liking to cook, experimenting with new foods, eating three meals per day, wearing clothes that reflect who they are, wearing fashionable clothes, and believing that dressing well leads to success. This group’s preferred genres of music are country and golden oldies (the latter is higher on D3 than the former, potentially reflecting an age gap in musical likes).  4.3.1 Group 5: Young and Single with Health-Damaging Dispositions The fifth distinct group is made up of individuals who are located near the centremost point of D1 (disposition), high on D2 (position and position-taking) and low on D3 (time). Holding dispositions similar to those of group four, these individuals are located lower on D3 and grouped nearest to the age 19-30 category (they seem to be the younger actors in the space of necessity). This grouping is somewhat distributed along the positive side of D2: those furthest to the right side of Figure 5 appear to have fair/poor mental health, are smokers, renters, consume low levels of fruits and vegetables, strongly agree that eating at restaurants with big portions is important, strongly disagree that food should be good for them, and are single. Located further down D2 (left side of Figure 5) are people who have general sales/service and clerical jobs, are sometimes depressed, exercise infrequently, have recently eaten at a fast-food restaurant, have recently travelled within Canada, have good mental and overall health, and generally have neutral dispositions towards food (e.g., neutral about liking to cook, finding it important to eat food that is healthy, and eating at restaurants that serve large portions).   74 4.3.1 Space of Necessity: Groups 4 and 5 Generally speaking, groups 4 and 5 are part of a larger cluster of individuals (see Appendix B.2 for a broad view of groups 4 and 5) distributed along D3 (time), and appear to represent younger and older people who have negative dispositions and generally hold a lower position on D1 (dispositions) and higher position on D2 (position and position-taking). Age seems to be a fairly influential differentiating force in this space where the younger actors are located in a slightly healthier social space with dispositions that are less discerning (i.e. are more frequently neutral).  Generally speaking, groups four and five are strongly opposed to believing it is important to eat food that is good for them. They prefer to eat large portions to get their monies worth, eat few fruits and vegetables, rarely exercise (all of which are more prominent among younger actors in group five), are smokers, have most recently eaten at fast-food restaurants and have fair/poor self-perceived overall and mental health, are frequently depressed and have a high body-mass index (obese). The groups in the space of necessity appear to hold a lower position in the field with fewer symbolic capitals and a habitus that is more reflective of dispositions for what is necessary, possible and practical as opposed to that which is distinguished and aesthetically valuable. The relational composition of their social space paints a picture of struggle, with people with lower education who often don't know their parent's education, have low incomes and work strenuous low-paying jobs.  The groups in the space of necessity share similarly oppositional dispositions towards food, dress and travel, exhibiting a strong dislike for experimenting with new foods, recipes, and ingredients, not finding it important to eat three good meals per day, not liking to cook, and not being concerned with learning about other cultures while on vacation (which is  75 Figure 4: Group 4 characteristics  76 located in close proximity to having only travelled locally). The actors in the space of necessity appear to hold little value for actions that extend beyond what is necessary, such as buying clothes that are fashionable and reflect who they are, and do not think that clothing has a value beyond function (disagree and strongly disagree that dressing well leads to success).  The relational unity of these dispositions, social conditions and behaviours reflect a common and pervasive story in the health literature speaking to the relational nature of cultural differentiation generally (Lamont and Small, 2008) and its effects on health more specifically (Backett-Milburn et al., 2010; Larsen & Morrow, 2009; Veenstra, 2007; Wills et al., 2011). The differences between the groups in the space of freedom and the space of necessity are captured most strikingly by their contrasting health-related dispositions, behaviours and outcomes.  4.3.1 Group 6: High-Stress Workaholics with Unhealthy Dispositions The next distinct grouping of categories pertains to respondents who are located higher on D1 (dispositions), at the midpoint of D2 (position and position-taking) and lower on D3 (time) (see Figure 6). This group is located between groups 3 and 4 in the field, and appears to consist of individuals who work in specialty sales and professional business/finance jobs, have extremely stressful lives, are underweight, have a sweet tooth, exercise frequently, strongly disagree that they like to cook, strongly agree that it is important to relax on vacation, dislike opera and strongly disagree with clothing-related dispositions. More specifically, these individuals strongly disagree with dressing fashionably, wearing clothes  77 Figure 5: Group 5 characteristics  78 Figure 6: Group 6 characteristics   79 that draw attention from others, strongly disagree that they spend more on clothes than they should, and strongly disagree that they can tell if they will get along with someone based on what they are wearing. This group is located near the smoking category (which is most influential to group 4).  4.3.1 Group 7: Elderly Unstressed Homeowners The final distinct grouping of categories pertains to people who are located low on D1 (dispositions), low on D2 (position and position-taking) and high on D3 (time) (see Figure 7). This group consists of older respondents, largely containing widowers and respondents over the age of 71. Generally speaking, this elderly group of individuals have strong dispositions towards dress, where they strongly disagree or disagree that it is important to buy clothes that last, important to wear clothes that draw attention, disagree that they spend more money on clothing than they should, and do not find it important to travel 'off the beaten path.' Members of this group also like classical music, own their homes, and generally find their lives to be free of stress.    80 Figure 7: Group 7 characteristics   81 Chapter  5: The Field of Power  5.1 Introduction To this point, I have discussed the theoretical and analytical principles at the heart of Bourdieu's theory of action, described how to empirically implement his framework, presented a relational model of a Canadian social space (field), described the general structure and composition of the field, and briefly discussed the common objective and subjective characteristics of the groups and individuals located within the spaces of the field. I will now discuss the overall model and specific results in terms of the relational theoretical principles detailed in chapter 2 (i.e., relational logic, fields, habitus, capitals and doxa), providing a theoretically-informed interpretation of the space as a whole and explaining what this model of a Canadian field means for our understanding of the power relations at the heart of social stratification and their relation to health.  5.2 Healthy and Unhealthy Spaces in the Canadian field The Canadian social space described above presents itself in the form of relational actors endowed with different properties and pertinent dispositions that are systematically linked together. The ways in which these attributes are distributed in the space of relations reveals the emergent boundaries, that is, the common factors, underlying assumptions and hidden principles shared by opposing actors that form areas of struggle and differentiation which are pertinent to health. In the field presented above, certain objective and subjective attributes and social factors emerged as being the most important in shaping the structure of the social life within this particular 'case of the possible'. The dimensions revealed that dispositions (values, attitudes, beliefs), positions (education, income, parental income) and position-  82 takings (smoking, diet, exercise behaviours), time (age) and gender were all relationally important factors that exist and function in relation to the field in question. These factors represent social properties that shape the structure of the field. While the field consists of several different and distinct groups, the general principle of symbolic differentiation is most clearly pronounced when comparing the space of necessity and the space of freedom, revealing two very different sets of resources and lifestyles consisting of contrasting objective and symbolic capitals, different social backgrounds and histories, and oppositional dispositions that appear to be relationally bound in space to health-related factors. While there are interesting and important nuances present among each of the seven groups, I have chosen to focus here on the space of freedom and the space of necessity as they most clearly identify the social boundaries that structure the field.  5.2.1 Space of Freedom and Health The individuals in the groups that make up the space of freedom include those with sizeable stores of capitals as well as a healthy habitus. The relational composition of the space tells a story of being born into privilege (both parents university educated), having access to materials and knowledge that assist in learning how to act in the field, and of having a practical sense of the constitution of the field (Bourdieu & Wacquant, 1992: p.20). In addition to having parents who are well-educated, individuals located in this space also tend to be educated themselves, and as such, hold high-prestige occupations which require advanced educations and garner high incomes. The people located in this space possess strong healthful dispositions to eat three meals per day, to eat food that is good for them, to enjoy cooking, to experiment with new foods, recipes and ingredients, and to eschew  83 restaurants that serve large portions. They also appear to exercise fairly frequently and travel for reasons other than to relax. As a whole these relationally bound characteristics illuminate a space consisting of individuals who appear to live in a generally well-educated, affluent and culturally distinct social environment in which healthful values and dispositions are patterned around typically healthful practices and behaviours. Similar groups (or classes) emerged in the fields observed by Veenstra (2007), who found the relational characteristics of the professional class (educated, cultured and wealthy) to have similarly high self-rated health and frequent physical activity (see also Frie & Janssen, 2009; Tomlinson, 2009). Using ethnographic methods, Wills et al. (2011) also observed a particular social space in which an upper-class Scottish family (the Connells) exhibited distinguished eating habits and encouraged and instilled these same tastes for distinguished and/or healthy eating habits into their children, "They reported limiting the consumption of foods high in fat, sugar or salt, like pizza, chips and chocolate and focusing on maximising their consumption of ‘healthy’ foods, like fruit and vegetables" (p.736). Wills et al. (2011) and Backett-Milburn et al. (2010) provide evidence of an upper class nutritional discourse regarding the future health of their children that was generally found to be absent among lower class families, where upper class families "seemed to feed their teenagers in a way which moulded their eating practices for the future their ‘good taste’ was being cultivated for future reward. ‘Learning’ to eat in restaurants for example, or learning to like spicier foods meant they could mould a future adult who would need to call on such dispositions if they were to lead a ‘successful’ life" (Wills et al., 2011: p.736). As a whole, the relational nature and unity of the common characteristics among the actors in the space of freedom depicts lifestyles endowed with an ensemble of dispositions  84 which "imply both the propensity and the ability to get into and play the game" (Bourdieu & Wacquant, 1992: p.19); having an understanding of the rules and expectations when eating at fine dining restaurants, attending wine-tasting events, attending museums, having a taste and preference for the sophisticated styling of jazz music and a strong dislike for country, and pursuing occupations with elements of aesthetic value, such as occupations in art and culture. Their healthful dispositions also speak to their habitus, or in this case, a healthful habitus. As illustrated by the common dispositions shared by those in this particular space of our field, there appear to be similar internalized and embedded values that speak to generally healthful lifestyles which consist of healthful actions and positive health.  5.2.2 Space of Necessity and Health In contrast, the groups in the space of necessity are located within a symbolic system of objective attributes and subjective dispositions that appear to be 'in tune' with the social position that they hold in within the broader field, that is, lower on all three dimensions. Central to this space are social factors related to having less-educated parents (this also includes respondents who replied that they do not know their parents' education), who are themselves less-educated, hold physically taxing manual labour jobs and less prestigious general service, sales and clerical jobs, pay rent for their living space, and have low yearly household incomes. In relation to these structural social factors, the characteristics in the space also speak to a propensity for what is necessary and practical, possibly attributed to the nature of the structural circumstances, such as not caring if clothes are fashionable and reflect 'who they are,' travelling locally or within Canada, not concerning themselves with learning of others while on vacation, not caring for cooking, eating three good meals per day and/or  85 food that is good for them, not eating lots of fruits and vegetables or exercising. Within this space of necessity there are also pervasive indicators of practices that have been found to be related to poor health, such as smoking and eating few fruits and vegetables, in turn closely situated to indicators of poor overall and mental health such as high levels of depression and being obese. As a whole, individuals located within the space of necessity share many common dispositions which speak to the generally unhealthy lifestyle that is well documented in other studies of social stratification and health lifestyles (Backett-Milburn et al. 2010; Frie & Janssen, 2009; Gatrell, 1997; Gatrell et al., 2004; Jones et al., 2010; Lengen & Blasius, 2007; Tomlinson, 2003; Veenstra, 2007).  These relational patterns are not specific to this Canadian field: the dispositions towards food, clothing and travel reflect a preference for utility (function) over aesthetic (form) that is shared by the working-class Watson family from eastern Scotland (Wills et al., 2011). The Watsons tended to cook what was fastest (and exhibited less enjoyment cooking) and looked forward to their Saturday take-out meals wherein "no one had to take responsibility for food preparation, no one had to set the table or get out cutlery and the day could move forward with little interruption" (p.735). In comparison to the upper class Connell family who actively shaped their children's eating habits, Wills et al. (2011) found that, although the lower class Watson family would acknowledge that they would like their children to eat healthier and try different foods, on the whole the young people’s tastes and preferences were treated as their own concern (p.734). Such differences in disposition and action appear to reflect the contrasting values between the space of freedom and the space of necessity identified in my model.   86 5.2.3 The Opposing Spaces in Between Freedom and Necessity Thinking of the field as a relational space of action and conflict, that is, as a space where people act not only for themselves but in relation to others, the social conditions, dispositions and actions of the individuals located in the space of freedom and necessity can be seen as classifying (or symbolic) properties (objective and/or subjective) which inform their actions and the actions of those around them. In a way, the properties of the individuals in the field are classified through their objective and subjective positions, dispositions and actions, thus exposing themselves to classification via their choices, their tastes and their diverse attributes. The model presented above exposes quite distinctly oppositional properties (income, education, parental education, home ownership, marital status, occupation, etc.), dispositions (liking to cook, wearing fashionable clothing, believing that dressing well leads to success, travelling for more than relaxation, etc.) and actions (smoking, exercising, traveling, etc.) among actors in the space of necessity and freedom. These distinct oppositional qualities are important to recognize in order to better understand not only the qualities that are common to certain individuals in the field but also the qualities that are different from others located in the same field.  It is also important to recognize the nature of the symbolic boundaries drawn by members of various groups and to reveal how the relational actions of individuals within various spaces give rise to the (us/them) barriers that keep certain people together and keep others apart (Lamont & Small, 2008). In my model of social space, such barriers appear to exist in the form of musical likes and dislikes, where individuals in group four, located in the space of necessity, share a common love for country music, while those in group two, located in the space of freedom, share a common dislike for this form of music that brings the other  87 group so much pleasure. This is akin to the findings of Bryson (1996) who found that patterns of musical tastes and distastes result in culturally isolating practices for the upper and lower classes and thereby create clear us/them boundaries. As a matter of space and difference, all of the oppositional dispositions and symbolic capitals (likes vs. dislikes, strong agreement vs. strong disagreement, action vs. abstinence) speak to positions and oppositions in space, the basic principles of differentiation that keep similar people together and acting ‘reasonably’ given what they 'know,' avoiding social spaces that are unfamiliar and uncomfortable. Thus, in terms of propensity towards action, one can assume that individuals who have a love for country music would be more likely than those who hate it to attend a Garth Brooks concert, for instance. In this sense, shared patterns of tastes and distastes can be thought to result in culturally isolating practices for the privileged and underprivileged alike that can create pervasive us/them boundaries which work to enforce differences over time (Lamont & Small, 2008).  As illustrated in the model, these shared patterns of tastes and distastes, capitals and actions also appear to be relationally bound to health-related practices and outcomes which differ significantly between groups in the space of freedom and the space of necessity. These sorts of health differences among people located in spaces of freedom and necessity paint a picture of groups of people with common social characteristics, interests, values and dispositions that appear to be meaningfully embedded in and tied to health-related actions and outcomes. To explore the ways in which the social factors are patterned in social space is to reveal interesting similarities among structural factors, personal dispositions, behaviours and health that seem to support Bourdieu’s notion of habitus, whereby the values and dispositions that emerge from experience in a social space appear to also be interrelated with  88 individual-level health. It would seem that within the context of this particular ‘case of the possible’ good or poor health can be related to many different factors within many different contexts. This exercise in relational thinking and model building is meant to offer insight into how structural factors, internal dispositions and behaviours can help us better understand how and why certain factors and perceptions might be related to physical, emotional and self- perceived health. With this being said, there are several important limitations to mention before discussing the usefulness of this approach for future health research.  5.2.4 Limitations and Key Points to Consider It is important to discuss some of the key limitations of relational sociology and field theory that are at the heart of Bourdieu's theoretical approach. First, relational field approaches of this kind risk presenting a social field in a way that implies that the exploratory model(s) present fundamental truths about social reality. The danger in this tendency towards tautology is that it risks blurring the relationship between model and reality whereby that which is revealed in the analogical model of the field is passed off as being representative of the wider population (Levi Martin, 2003; Vandenberg, 1999). Sharing commonalities with Weber's conception of 'ideal types' ([1930] 1992) and Simmel's notion of 'social forms,' (1904) both of which reject the assumption of universality and aim to identify social dimensions and describe the social processes inherent in a particular social situation, Bourdieu's field-specific approach similarly does not aim to make an ontological argument about social reality; rather, it simply aims to explore and synthesize the properties of a particular situation in great detail. As such, the relational field approach used by Bourdieu and employed in this analysis is fundamentally an analytic approach that draws together  89 ideas, unifies concepts, describes the structure of relations specific to the field being analyzed and seeks to explain the nature of social relationships between people in a space (Levi Martin, 2003: p. 43-44). The analytic and exploratory nature of this study becomes apparent upon recognizing that, while I was able to offer a more detailed account of a health-related social space than any other study to date by the inclusion of a 3rd dimension, the fact remains that I was unable to include a substantively meaningful 4th dimension into my analysis, that of gender, which, had it been included, might have produced quite a different social space altogether. Thus, in order to provide increasingly useful and relevant models, there is a need to continually develop new technologies and methodological techniques that allow for increasingly complex models to better understand the relational nuances of a social space. Additionally, like any other form of data analysis, this study is inherently grounded in subjective, albeit theoretically-informed, choices regarding variable selection, coding, model development, interpretation and the like. In the case of variable selection, given the focus on developing a better understanding of habitus and doxa, there is a slightly disproportionate number of variables reflecting disposition, values and beliefs compared to other concepts. The inclusion of a disproportionate number of variables can potentially influence the composition of a dimension, that is, a dimension (e.g., the dimension of dispositions) may appear as a prominent structuring force in the field simply because of a relatively high proportion of disposition-related indicators. While I have illustrated the importance for relational analyses grounded in field theory to include all variables as active when constructing the model, so to limit the amount of bias that goes into the model (i.e., choosing which variables are more influential than others), it remains equally important to refrain from  90 influencing the structure of the field by over-emphasizing certain concepts and social factors when developing the model. In terms of coding the variables that were included in this relational field analysis, I made the decision to include 'neutral', 'don't know' and 'refused' categories as active. I believe that the 'neutral' categories added an important contextual element to the field in that they helped me to better understand the context surrounding individuals who do not have strong dispositions, an element that is often missed in typical causal analyses which tend to treat 'neutral' categories as irrelevant problem categories and indicative of 'fence sitters' (i.e., respondents who refuse to take a stand on a topic). As was illustrated in this model, the 'neutral' categories were distributed meaningfully throughout the field and appeared to have important influence on the structuration of certain groups, adding relevant interpretive context that may have otherwise been lost had neutral categories been combined with definitive (yes/no) categories, treated as passive categories or removed from the analysis all together. Regarding the 'don't know' and 'refused' categories, which were also included as active, it is important to seriously consider whether they do indeed reflect value statements and are substantively relevant. Given that these categories often have small frequency distributions, their inclusion as active variables can have a large influence on the structuration of the field and may decrease the overall validity of the field being presented. While my choice to include the 'I don't know my parents education' was substantively warranted and empirically appropriate, future studies should be mindful of the potential consequences of including these types of categories as active in the field. Finally, regarding interpretation, while I have done my best to detail the steps I have taken and the decisions that were made in the process, my presentation of a Canadian social space is shaped by my subjective understanding of its internal logic and the forces that  91 inform health inequalities. For instance, my personal and academic experiences studying the social construction of health narratives among people with rare diseases has given me a particular understanding about the ways in which relational interactions between patients and professionals can influence how health is understood and internalized by both parties. In this way, my ontological position reflects my experiences and belief that health is fundamentally subjective in that it takes shape in context, that is, at the level of the situation. While we all have inherent biases, I feel that it is important for future researchers seeking to employ this method of relational analysis (or any inductive method) to remain wary of their own ontological and epistemological position and be reflexive of how this informs their research.9 Being mindful of these limitations when undertaking health research grounded in field-theory can aid in developing relational models that are rigorous and useful for informing our understanding of health inequalities in the wider population. This point brings up one final and important consideration regarding the applicability of a study that abandons the search for causal generalities in social life and seeks to better understand the complexities of social forces within specific contexts, as is the case with our model of a Canadian field. The analysis presented in this thesis is grounded in an inductive, relational and field-specific analytic technique, does not employ any statistical tests of significance that would allow for population-level explanations of health inequalities, and as such, does not take advantage of the randomness inherent in the survey sample. While it is reasonable to perceive studying the complexities of social context at the level of the situation as a limitation, particularly on the grounds of the commonly held belief that interpretations of  9 Nearing the end of his academic career, Bourdieu wrote extensively on the importance of reflexivity in the social sciences (see Bourdieu & Wacquant, 1992), a point which, thinking particularly in terms of the amount of interpretation and abstract thinking that is needed in relational sociology and field theory, has become abundantly clear as I conclude this thesis and attempt to relate the findings back to 'reality'.  92 patterns in the field are essentially only relevant to the observations revealed in the field, and as such, are not generalizable to the wider population, there are some important strengths associated with a situation-level approach that can be beneficial for future health research and policy initiatives. While unable to provide statistical certainty, a relational field approach nevertheless offers health researchers a new way to bring together ideas and findings from the health literature and incorporate many different theoretically and empirically relevant social factors into a model that allows us to identify social spaces in which poor or good health seems be present (i.e., has emerged). Furthermore, the patterns revealed in the Canadian field presented here appear to be fairly consistent with findings from similar studies investigating social class and health behaviours, and health lifestyles more generally, which have consistently found income, education and occupations to be related to health across cultures. Given these fundamental similarities with the wider body of class-based health research, there is reason to believe that some of the patterns that emerge in the field may warrant further investigation and may indeed be useful to help contextualize the principles that lead to specific health behaviours within Canadian social spaces. Relational studies exploring the spaces in which health emerges, while fundamentally exploratory in nature, can offer important contextual information for other areas of study, health-specific or otherwise. In the same way that Bourdieu's theoretical paradigm and relational findings have inspired a broad range of innovative approaches in health research, developing further relational models of health-relevant social spaces can have a similar kind of effect. Such studies present particular 'cases of the possible' that can ignite thought for further research, present a new way of thinking about health in context, and offer insights that lead to deeper understandings of health as a relational concept located within webs of  93 symbols, properties, dispositions and actions. With this being said, the implementation of other exploratory methods of investigation are certainly warranted in future research to validate and further explore the principles revealed in relational explorations similar to that presented in this thesis.  5.3 Conclusion and Thoughts for Future Research While I have done my best to present a health-relevant theoretical re-interpretation of Bourdieu's work, there is much that I was unable to do within the limits of a Master’s thesis. With regards to the theoretical portion of the thesis, while I have presented a thorough re- interpretation of Bourdieu’s complex theory of action in a health-relevant context and shed light on the theoretical usefulness of his almost entirely overlooked concept of doxa, there remains a need to offer a theoretical account of relational philosophy and field theory that takes into account the theoretical views of other thinkers, including those who informed the work of Bourdieu and the contemporary thinkers who are using these theories in new and important ways. With regards to doxa, while I was able to discuss its theoretical relevance to health research, I was unable to incorporate it into the empirical component of my study. Further research is needed which utilizes the theoretical concept of doxa (and habitus) to better understand the ways in which personal perceptions of one place in society perpetuates inequalities that may be related to health. While there is a growing body of cultural research investigating concepts similar to doxa, such as symbolic boundary formation (see Lamont and Small, 2008, for a useful review of such studies), these have yet to be incorporated into research on health inequalities.  94  With regards to the 3D model, while I spent much time learning how to develop a theoretically sound and empirically robust model of a field, the scope of its complexity became quite difficult to tackle and interpret. While the 3D model proved to be a useful 'tool' for presenting a detailed social space, I had to significantly limit the scope of my interpretation to surface-level interpretations of patterns that did not address interesting nuances manifested by the time dimension. There remains a need to interpret not just how social factors are located in two dimensions but also how these spaces differ with respects to age, that is, how dispositions and actions change over time and how this relates to health. Also, even though interpreting a space with three dimensions was difficult, there is room for future research which incorporates more dimensions, e.g., the gender dimension in my data which I had to leave out of my analysis. As mentioned earlier, Bourdieu was often criticized for leaving gender largely untouched in his analyses; perhaps he was also unable to come up with an appropriate method to account for and scrutinize a fourth dimension. Nevertheless, we should seek to develop new ways of modeling so to be able to incorporate all relevant information into the models.  While re-interpreting a significant portion of Pierre Bourdieu’s corpus of work and then applying his framework into a theoretically-informed empirical analysis of a 3D health- related social space was an ambitious undertaking, I feel that I have been able to offer some important insights regarding the usefulness of a Bourdieusian approach for health research, i.e., how to build and interpret a theoretically sound relational model of a field and how to interpret the results in a meaningful way. While there is much more work to be done to integrate relational thought, field theory and a Bourdieusian theoretical framework into health research, I feel that my thesis has offered some insights as to where we can start and in  95 what ways these approaches can lead us to develop a novel approach to understanding and addressing health inequalities.  96 Bibliography  Addinsoft (2007). XLSTAT – Statistical Software for MS Excel. URL: Abel, T. (2007). Cultural capital in health promotion. In D. McQueen, I. Kickbusch, L. Potvin, J. M. Pelikan, L. Balbo & T. Abel (Eds.), Health and modernity: The role of theory in health promotion (pp.41–71). New York: Springer. Abel, T. (2008). Cultural capital and social inequality in health. Journal of Epidemiology and Community Health, 62(7). Abel, T., Fuhr, D. C., Bisegger, C., Ackermann Rau, S., & The European, K. G. (2011). Money is not enough: Exploring the impact of social and cultural resources on youth health. Scandinavian Journal of Public Health, 39(6), 57-61. Backett-Milburn, K.,Wills,W.J.,Gregory, S., and Lawton, J. (2006). Making sense of eating, weight and risk in the early teenage years: views and concerns of parents in poorer socio-economic circumstances. Social Science & Medicine, 63(3), 624–635. Backett-Milburn, K. C., Wills, W. J., Roberts, M., & Lawton, J. (2010). Food, eating and taste: Parents’ perspectives on the making of the middle class teenager. Social Science & Medicine, 71(7), 1316-1323. doi: 10.1016/j.socscimed.2010.06.021 Berger, P., & Luckmann, T. (1966). The social construction of reality: A treatise in the sociology of knowledge. New York: First Anchor Books.  97 Blasius, J. (1994). Correspondence analysis in social science research. In M. Greenacre, & J. Blasius (Eds.), Correspondence analysis in the social sciences (pp.23-53). London: Academic Press. Bourdieu, P. ([1972] 1977). Outline of a theory of practice. Cambridge University Press. Bourdieu, P. ([1979] 1984). Distinction. A social critique of the judgement of taste. Cambridge: Harvard University Press. Bourdieu, P. ([1980] 1990). The logic of practice. Stanford University Press. Bourdieu, P. (1986). The forms of capital. In J. G. Richardson (Ed.), Handbook of theory and research for the sociology of education (pp.241-58). New York: Greenwood Press. Bourdieu, P. ([1994] 1998). Practical reason. Stanford University Press. Bourdieu, P. ([1997] 2000). Pascalian meditations. Stanford University Press. Bourdieu, P., & Wacquant, L. (1992). An invitation to reflexive sociology. The University of Chicago Press. Bourdieu, P. (1989). Social space and symbolic power. Sociological Theory, 7(1), 14-25. Bryson, B. (1996). "Anything but heavy metal": Symbolic exclusion and musical dislikes. American Sociological Review, 61(5), 884-899. Carpiano, R. M. (2006). Towards a neighborhood resource-based theory of social capital for health: Can Bourdieu and sociology help? Social Science & Medicine, 62(1), 165-175.  98 Carpiano, R. M. (2007). Neighborhood social capital and adult health: An empirical test of a Bourdieu-based model. Health & Place, 13(3), 639-655. Clausen, S.-E. (1998). Applied correspondence analysis: An introduction. Sage Publication, Inc., Thousand Oaks, California. Cockerham, W. C. (2007). Social causes of health and disease. Cambridge: Polity Press. Cockerham, W. C. (2005). Health lifestyle theory and the convergence of agency and structure. Journal of Health and Social Behavior, 46(1), 51-67. Cockerham, W. C., Hinote, B. (2009). Quantifying habitus: Future directions. In C. Sanders (Ed.), Quantifying theory: Pierre Bourdieu (pp. 201-210) Springer Netherlands. Coleman, J. S. (1990). Foundations of social theory. Cambridge: Harvard University Press. Durkheim, E. ([1895] 1982). The rules of sociological method. The Free Press. Emirbayer, M. (1997). Manifesto for a relational sociology. American Journal of Sociology, 103(2), 281-317. Fleck, L. ([1935] 1981). Genesis and Development of Scientific Fact. The University of Chicago Press. Frie, K., & Janssen, C. (2009). Social inequality, lifestyles and health a non-linear canonical correlation analysis based on the approach of Pierre Bourdieu. International Journal of Public Health, 54(4), 213-221. doi:10.1007/s00038-009-8017-5  99 Frohlich, K. L., Corin, E., & Potvin, L. (2001). A theoretical proposal for the relationship between context and disease. Sociology of Health & Illness, 23(6), 776-797. Gatrell, A. C., Popay, J., & Thomas, C. (2004). Mapping the determinants of health inequalities in social space: Can Bourdieu help us? Health & Place, 10(3), 245-257. doi: 10.1016/j.healthplace.2003.09.005 Greenacre, M. (1994). Correspondence and its interpretation. In M. Greenacre, & J. Blasius (Eds.), Correspondence analysis in the social sciences: recent developments and applications (pp. 4-22). London: Academic Press. Greenacre, M., & Blasius, J. (Eds.). (1994). Correspondence analysis in the social sciences: Recent developments and applications. London: Academic Press. Greenacre, M., & Blasius, J. (2006). Correspondence analysis and related methods in practice. In M. Greenacre, & J. Blasius (Eds.), Multiple correspondence analysis and related methods (pp.3-40). Chapman & Hall/CRC. Haines, R. J., Poland, B. D., & Johnson, J. L. (2009). Becoming a smoker: Cultural capital in young women's accounts of smoking and other substance use. Sociology of Health & Illness, 31(1), 66-80. Hennion, A. (2007). Those things that hold us together: Taste and sociology. Cultural Sociology, 1(1), 97-114.   100 Jones, I. R., Papacosta, O., Whincup, P. H., Goya Wannamethee, S., & Morris, R. W. (2010). Class and lifestyle among middle-aged and older men: A multiple correspondence analysis of the British regional heart study. Sociology of Health & Illness, 33(3), 399- 419. Kaplan, G. A., Pamuk, E. R., Lynch, J. W., Cohen, R. D., & Balfour, J. L. (1996). Inequality in income and mortality in the United States: Analysis of mortality and potential pathways. BMJ, 312(7037), 999-1003. Kawachi, I., Kim, D., Coutts, A., & Subramanian, S. V. (2004). Commentary: Reconciling the three accounts of social capital. International Journal of Epidemiology, 33(4), 682- 690. Kennedy, B. P., Kawachi, I., & Prothrow-Stith, D. (1996). Income distribution and mortality: Cross sectional ecological study of the Robin Hood index in the United States. BMJ, 312(7037), 1004-1007. Khawaja, M., & Mowafi, M. (2006). Cultural capital and self-rated health in low income women: Evidence from the urban health study, Beirut, Lebanon. Journal of Urban Health, 83(3), 444-458. doi:10.1007/s11524-006-9051-8 Konlaan, B., Björby, N., Bygren, L., Weissglas, G., Karlsson, L., & Widmark, M. (2000). Attendance at cultural events and physical exercise and health: A randomized controlled study. Public Health, 114(5), 316-319. doi: 10.1038/  101 Kuhn, T. ([1962] 1996). The structure of scientific revolutions. The University of Chicago Press. Lamont, M., & Small, M. L. (2008). How culture matters for the understanding of poverty. In A. Lin, & D. Harris (Eds.), The colors of poverty: Why racial and ethnic disparities persist (pp.76-102) . New York: Russell Sage. Larsen, K., & Morrow, V. (2009). Social positions and young people's health: A Bourdieu'ian critique of dominant conceptualisations of social capital. Praktiske Grunde, 3, 7-30. Lengen, C., & Blasius, J. (2007). Constructing a Swiss health space model of self-perceived health. Social Science & Medicine, 65(1), 80-94. doi: 10.1016/j.socscimed.2007.02.053 Le Roux, B., & Rouanet, H. (2005). Geometric data analysis: From correspondence analysis to structured data analysis. Springer Science + Business Media, Inc. Levi Martin, J. (2003). What is field theory? American Journal of Sociology, 109(1), 1-49. Link, B. G., & Phelan, J. (1995). Social conditions as fundamental causes of disease. Journal of Health and Social Behavior, 35, 80-94. Lynam, M. J., Browne, A. J., Kirkham, S. R., & Anderson, J. M. (2007). Re-thinking the complexities of 'culture': What might we learn from Bourdieu? Nursing Inquiry, 14(1), 23-34. Malat, J. (2006). Expanding research on the racial disparity in medical treatment with ideas from sociology. Health: 10(3), 303-321.  102 Marx, K. ([1867] 1990). Capital: a critique of political economy, Vol. 1. Penguin Books Ltd. Marx, K., & Engels, F. (1978). The Marx-Engels reader. Edited by R. C. Tucker. New York: W. W. Norton. Mason, J. (2002). Qualitative researching. London: Sage. Miner3D (2011). Data modelling software. URL: Mirowsky, J., & Ross, C. E. (2003). Education, social status, and health. New York: Aldine de Gruyter. Muller, C., & Woods, R. (1994). An expanded restaurant typology. The Cornell Hotel and Restaurant Administration Quarterly, 35, 27-37. Noone, B., Kimes, S., Mattila, A., & Wirtz, J. (2007). The effect of meal pace on customer satisfaction. Cornell Hotel and Restaurant Administration Quarterly. 48, 231-44. Putnam, R.D. (2000). Bowling Alone: the collapse and revival of American community. New York: Simon and Schuster. Rouanet, H. (2006). The geometric analysis of structured individuals × variables tables. In M. Greenacre, & J. Blasius (Eds.), Multiple correspondence analysis and related methods (pp.137-160). Taylor & Francis Group. Rovan, J. (1994). Visualizing solutions in more than two dimensions. In M. Greenacre, & J. Blasius (Eds.), Correspondence analysis in the social sciences (pp. 210-229). London: Academic Press.  103 Sieger, M., Fritz, E., & Them, C. (2011). In discourse: Bourdieu's theory of practice and habitus in the context of a communication-oriented nursing interaction model. Journal of Advanced Nursing. Simmel, G. (1904). The sociology of conflict. I. American Journal of Sociology, 9(4), 490- 525. Singh-Manoux, A., & Marmot, M. (2005). Role of socialization in explaining social inequalities in health. Social Science & Medicine, 60(9), 2129-2133. Small, M. L. (2004). Villa Victoria: The transformation of social capital in a Boston barrio. University of Chicago Press. Swidler, A. (1992). Culture in action: symbols and strategies. American Sociological Review, 51(2), 273-286. Tomlinson, M. (2003). Lifestyle and social class. European Sociological Review, 19(1), 97- 111. Veenstra, G. (2000). Social capital, SES and health: An individual-level analysis. Social Science & Medicine 50(5), 619-629. Veenstra, G. (2002a). Explicating social capital: Trust and participation in the civil space. Canadian Journal of Sociology 27(4), 547-572. Veenstra, G. (2002b). Social capital and health (plus wealth, income inequality and regional health governance). Social Science & Medicine 54(6), 849-868.  104 Veenstra, G. (2007). Social space, social class and Bourdieu: Health inequalities in British Columbia, Canada. Health & Place, 13(1), 14-31. Veenstra, G. (2010). Culture and class in Canada. Canadian Journal of Sociology, 35(1), 83- 111. Weber, M. ([1930] 1992). The protestant ethic and the spirit of capitalism. New York: Routledge. Wilkinson, R. G. (2005). The impact of inequality. New York: W.W. Norton & Company. Williams, G. H. (2003). The determinants of health: Structure, context and agency. Sociology of Health & Illness, 25(3), 131-154. Williams, S. J. (1995). Theorising class, health and lifestyles: Can Bourdieu help us? Sociology of Health & Illness, 17(5), 577-604. doi:10.1111/1467-9566.ep10932093 Wills, W., Backett-Milburn, K., Roberts, M.-L., and Lawton, J. (2011). The framing of social class distinctions through family food and eating practices. The Sociological Review, 59(4), 725-740. Woolgar, S. (1988). Science: The very idea. Chichester, Sussex, England: Ellis Horwood Ltd.   105 Appendices Appendix A A.1 Frequencies of Conditions of Existence Variables Variables Categories Social space mapping labels Frequency (%) Household income less than $40,000 < $40,000 246 18.8 $40,000 - 59,999 $40,000-59,999 189 14.4 $60,000 - 79,999 $60,000-79,999 205 15.6 $80,000 - 99,999 $80,000-99,999 164 12.5 $100,000 - $149,999 $100,000-149,999 260 19.8 $150,000 or more $150,000+ 247 18.8 Parental educational status both parents with university degrees Both uni degree-PAR-ED 190 12.0 one parent with a university degree One uni. degree-PAR-ED 303 19.1 neither parent with a university degree No uni. degree-PAR-ED 993 62.6 other (don't know) Don't know-PAR-ED 100 6.3 Personal educational attainment less than high school < high school-PERS-ED 83 5.2 high school graduate High school-PERS-ED 485 30.5 community college or technical school diploma College/tech-PERS-ED 271 17.1 bachelor's degree 459 28.9 post-bachelor's degree 291 18.3 Occupation type Senior & Specialist Management Occupations Senior/Specialist Manager-JOB 130 10.5 Other Managers n.e.c Other Manager-JOB 126 10.2 Professional Occupations in Business and Finance Professional-Busin/Finance- JOB 85 6.9 Clerical Occupations Clerical-JOB 134 10.8 Professional Occupations in Natural and Applied Sciences and Health Professional- Nat.App.Sci./Health-JOB 105 8.5 Technical and Related Occupations in Health Tech.&Related/Health-JOB 70 5.7 Judges, Lawyers, Counselors, and Policy and Program Officers Professional- Judicial/Councel./Policy-JOB 59 4.8 Teachers (primary/secondary) Teacher (Prim/Sec) -JOB 64 5.2 Professor (college/university) Professor (Coll/Uni) -JOB 37 3.0 Paralegals, Protective Services, Social Services Workers and Occupations in Education and Religion, n.e.c. Social-JOB 68 5.5 Professional & Technical Occupations in Art and Culture Professional-Art & Cult-JOB 103 8.3 General Sales and Service Occupations General-Sales & Service-JOB 58 4.7 Specialty Sales and Service Occupations Specialty-Sales & Service-JOB 44 3.6 Childcare and Home Support Workers Childcare/Home supp.-JOB 79 6.4 Trades, Transport and Equipment Operators and Related Occupations Trades/Labour-JOB 76 6.1 Home ownership Fully owned Owned-Home 507 32.3 Partly owned Partly-owned-Home 589 37.5 Rented Renting-Home 473 30.1     106 A.2  Frequencies of Demographic Composition Variables Variables Categories Social space mapping labels Frequency (%) Gender Male  Male 542 34.0 Female Female 1051 66.0 Age Age-19 - 30 Age-19 - 30 165 10.4 Age-31 - 40 Age-31 - 40 227 14.3 Age-41 - 50 Age-41 - 50 381 24.1 Age-51 - 60 Age-51 - 60 355 22.4 Age-61 - 70 Age-61 - 70 286 18.1 Age-71 and older Age-71 + 170 10.7 Marital status Married Married 818 51.5 Living with a partner Living with partner 96 6.0 Widowed Widowed 113 7.1 Separated/divorced Separated/divorced 248 15.6 Never been married Never married 312 19.7 Immigrant status Born in Canada Born in Canada 1098 69.1 Immigrated to Canada more than 20 years ago >20 yrs.-Immigrate 326 20.5 Immigrated to Canada between 10 and 19 years ago 10-19 yrs.-Immigrate 82 5.2 Immigrated to Canada 9 or fewer years ago < 9 yrs.-Immigrate 83 5.2   A.3 Frequencies of Practice and Behaviour Variables  Variables Categories Social space mapping labels Frequency (%) Smoking Daily or occasional smoker Smoker 274 17.4 Former smoker Former-smoke 432 27.4 Never smoked Non-smoker 869 55.2 Exercise Exercise less than once per week Low-Exercise 194 12.3 Exercise one to three times per week Med-Exercise 716 45.4 Exercise more than three times per week High-Exercise 668 42.3 Diet Eat fruits and vegetables several times per day High-Fruits 1029 65.9 Eat fruits and vegetables about once per day Med-Fruits 402 25.8 Eat fruits and vegetables less than once per day Low-Fruits 130 8.3 Last restaurant visited Fast-food/convenience Fast-food/conv. 109 10.6 Family-style dining Family-style dining 295 28.7 Casual-dining/midscale Casual-dining/midscale 285 27.7 Casual-fine dining Casual-fine dining 284 27.6 Fine dining Fine dining 55 5.4 Recent travel Local trip (within province) Local-Trip 119 7.8 Trips within Canada (outside of own province) In Canada-Trip 162 10.7 Trips to mainland USA (including Alaska) USA-Trip 327 21.5 Trips to Central and South America (Caribbean and South Pacific) Cent/South Amer.-Trip 421 27.7 Trips overseas (Europe, Asian, Africa, Australia/Oceania) Overseas-Trip 489 32.2 Wine tasting Have you ever attended a wine tasting event? (Yes/No Yes 680 62.7 No 404 37.3 Any instrument Do you plan a musical instrument Yes  496 31.9 No 1060 68.1   107 A.4 Frequencies of Health Indicator Variables Variables Categories Social space mapping labels Frequency (%) Self-perceived overall health Excellent/Very Good EX/VG-Overall Health 846 53.7 Good Good- Overall Health 491 31.2 Fair/Poor F/P-Overall Health 238 15.1 Self-perceive mental health Excellent/Very Good EX/VG-Mental Health 1052 66.7 Good Good-Mental Health 397 25.2 Fair/Poor F/P-Mental Health 128 8.1 Stress Not at all stressful No stress 268 17.0 A bit stressful A bit stressful 811 51.5 Quite stressful Quite stressful 428 27.2 Extremely stressful Extremely stressful 69 4.4 Depression Rarely/never Rare/Never-Depression 969 61.7 Sometimes Sometimes-Depression 498 31.7 Most days/every day Most/Every day-Depression 104 6.6 BMI Underweight (BMI = 16 to 19) Underweight-BMI 40 2.6 Normal (BMI = 20 to 25) Normal-BMI 718 47.1 Overweight (BMI = 26 to 30) Overweight-BMI 499 32.7 Obese (BMI = 31 or over) Obese-BMI 267 17.5    108 A.5 Frequencies of Values, Attitudes, Tastes and Disposition Variables Variables Categories Frequency (%) I prefer to buy clothes that will last a long time SA-Last-Clothes 548 34.9 AG-Last-Clothes 802 51.1 NE-Last-Clothes 111 7.1 DI & SD-Last-Clothes 108 6.9 I probably spend more money on clothes than I should SA-Spend too much-Clothes 103 6.6 AG-Spend too much-Clothes 231 14.7 NE-Spend too much-Clothes 107 6.8 DI-Spend too much-Clothes 789 50.3 SD-Spend too much-Clothes 340 21.7 I like to dress fashionably SA-Fashionable-Clothes 155 9.9 AG-Fashionable-Clothes 734 46.8 NE-Fashionable-Clothes 281 17.9 DI-Fashionable-Clothes 328 20.9 SD-Fashionable-Clothes 72 4.6 I think that dressing well often leads to success SA-Dress = Success-Clothes 313 20.0 AG-Dress = Success-Clothes 993 63.3 NE-Dress = Success-Clothes 139 8.9 DI & SD-Dress = Success-Clothes 123 7.8 I want my clothes to reflect who I am SA-Reflect me-Clothes 282 18.0 AG-Reflect me-Clothes 927 59.2 NE-Reflect me-Clothes 192 12.3 DI & SD-Reflect me-Clothes 164 10.5 I like to wear clothes that make people look at me SA & AG-Draw attention-Clothes 369 23.5 NE-Draw attention-Clothes 356 22.7 DI-Draw attention-Clothes 717 45.7 SD-Draw attention-Clothes 126 8.0 I can often tell how well I will get along with someone by what they're wearing SA & AG-Get along-Clothes 215 13.7 NE-Get along-Clothes 184 11.8 DI-Get along-Clothes 893 57.1 SD-Get along-Clothes 272 17.4 I like to cook SA-Like to cook-Food 493 31.4 AG-Like to cook-Food 641 40.9 NE-Like to cook-Food 126 8.0 DI-Like to cook-Food 219 14.0 SD-Like to cook-Food 90 5.7 In my home we like to experiment with new foods, new recipes and new ingredients SA-Experiment-Food 426 27.1 AG-Experiment-Food 688 43.8 NE-Experiment-Food 157 10.0 DI & SD-Experiment-Food 299 19.0 I think it is important that nearly all of the food I eat is good for me SA-Good for me-Food 597 38.0 AG-Good for me-Food 784 50.0 NE-Good for me-Food 96 6.1 DI & SD-Good for me-Food 92 5.9 I think that it is important to eat at least three good meals a day SA-3 meals/day-Food 438 28.0 AG-3 meals/day-Food 825 52.7 NE-3 meals/day-Food 97 6.2 DI & SD-3 meals/day-Food 206 13.2   109 Variables Categories Frequency (%) I have a sweet tooth SA-Sweet tooth-Food 406 25.9 AG-Sweet tooth-Food 541 34.5 NE-Sweet tooth-Food 189 12.1 DI & SD-Sweet tooth-Food 430 27.5 I prefer to eat at restaurants that serve larger portions so that I get my money’s worth SA & AG-Big portions-Food 289 18.5 NE-Big portions-Food 198 12.7 DI-Big portions-Food 819 52.4 SD-Big portions-Food 256 16.4 The most important thing for me when I go on vacation is to relax SA-Imp. relax-Travel 509 32.8 AG-Imp. relax-Travel 689 44.4 NE-Imp. relax-Travel 125 8.1 DI & SD-Imp. relax-Travel 229 14.8 I like to learn about new places and different ways of life when I go on vacation SA-Learn of others-Travel 599 38.6 AG-Learn of others-Travel 773 49.8 NE-Learn of others-Travel 101 6.5 DI & SD-Learn of others-Travel 80 5.2 My favourite vacations are off the beaten path, places where most tourists don’t go SA-Off beaten path-Travel 247 16.0 AG-Off beaten path-Travel 525 34.0 NE-Off beaten path-Travel 275 17.8 DI & SD-Off beaten path-Travel 495 32.1 I usually visit as many museums as I can when I’m on vacation SA-Visit museums-Travel 173 11.2 AG-Visit museums-Travel 474 30.6 NE-Visit museums-Travel 277 17.9 DI & SD-Visit museums-Travel 625 40.3 Most liked genre of music Most Liked-- Classical 222 19.6 Most Liked--Folk 60 5.3 Most Liked--Country 85 7.5 Most Liked--Pop 118 10.4 Most Liked--Jazz 113 10.0 Most Liked--Easy listening 80 7.1 Most Liked--Rock 293 25.9 Most Liked--Blues 69 6.1 Most Liked--Golden Oldies 93 8.2 Most disliked genre of music Most Disliked--Hip hop 83 7.7 Most Disliked--Rap 365 33.9 Most Disliked--Opera 69 6.4 Most Disliked--Country 96 8.9 Most Disliked--Heavy Metal 465 43.1  110 Appendix B B.1 Broad View of Groups 1, 2 and 3   111 B.2 Broad View of Groups 4 and 5 


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