IJE Advance Access originally published online on July 28, 2004
International Journal of Epidemiology 2004 33(4):700-704; doi:10.1093/ije/dyh263
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
IJE vol.33 no.4 © International Epidemiological Association 2004; all rights reserved.
Article |
Rejoinder: Crafting rigorous and relevant social theory for public health policy
We wish to express our sincere thanks to the editors of the International Journal of Epidemiology for hosting such an interesting exchange on the idea of social capital and its application to public health. We are flattered that scholars and practitioners of such repute have responded so vigorously to our paper, and thank them also for their efforts.
Obviously we cannot hope to respond to every point raised by the discussants: some comments are more constructive than others and, given the relatively short space available to us, some provide a more obvious entry point for a reply than others. Let us begin, however, with several initial clarifying remarks, before proceeding to a more detailed response. It is emphatically not our view, and nowhere in the paper do we claim: (1) that social capital (however conceived) is the sole (or always primary) variable that explains (or should be used to try to explain) public health outcomes (as claimed or implied by three reviewers); (2) that our definition of social capital includes ‘the state’;2 (3) that empirical indicators of ‘linking social capital’ are limited to positive outcomes alone, like ‘good governance’;3 (4) that by adopting the social capital terminology we are arguing (implicitly or explicitly) that ‘the state’ must necessarily retreat;4 and/or (5) that enhancing ‘competitiveness’ and the ‘imperialist’ reach of ‘capitalism’ is the overriding (wittingly or unwittingly) objective when using the language of ‘capital’.4* In the passages below, we endeavour to restate and clarify our position by putting it in a broader context of what social theory in general seeks to accomplish, what our particular articulation of social capital theory seeks to accomplish, and what the methodological and practical implications are of pursuing our line of (evidence-based) reasoning in the field of public health.
 |
Social theory: what should it seek to accomplish? |
|---|
 Top Social theory: what should... Methods, measurement, and...Applications to public health...References |
|---|
The broad dialectical challenge in social theory is (or should be) addressing the structure-agency problem (also known as the micro-macro problem)6—that is, unpacking the interactions and interconnections between individual choices and larger institutional forces.7 Unfortunately, much of the impetus in social theory in recent decades—with self-contained camps emerging around rational choice theory, post-modernism, and (various manifestations of) institutionalism—has conspired against taking up this challenge,
legacies of which linger in the comments of two of the reviewers. In our view, the challenge facing any adequate social theory is to integrate agency and structure, not simply (or simplistically) to infer the characteristics of one aspect on the basis of one's political or methodological convictions about the other. Indeed, this was both the explicit challenge Coleman9 took up in his theoretical writings on social capital and the unfortunate tendency, noted above, to which he fell prey: namely, in his case, giving methodological primacy to the choices of rational, self-interested individuals and then regarding larger institutional forms and functions as aggregated products of those choices. Though Coleman took up the correct challenge, most social theorists conclude that his resolution of it was at best incomplete. We argue (as do many others) that the dialectic between individual agency and social structures still leaves a large and relatively autonomous conceptual space for that most fundamental of human arrangements, namely groups. It is into this conceptual space that the more ‘mainstream’ conceptualizations of social capital first entered and have since flourished. By arguing and then providing empirical support for notions that membership in different types of social groups and networks has real and measurable consequences for a host of outcomes that both societies and the individuals who comprise them care about (e.g. health, education, crime), innovative social scientists captured the imagination of academics, policymakers, and the public more generally. As such, it should be clear that our preferred unit of analysis for conceptualizing and measuring social capital is both an individual and ecological one. As Putnam3 and Kawachi et al.2 correctly note, the benefits (and costs) of having access to a rich storehouse of networks and organizations accrue to both individuals and groups, and it is this empirical reality that lends social capital much of its policy purchase. Doubtless some of the initial enthusiasm was excessive and/or unwarranted, and there remains considerable room for reasonable people to differ on the extent to which social capital scholarship created or rode a broader wave of political-intellectual change, but it is hard to argue that the documented exponential rise in interest in social capital is merely a passing opportunistic fad, a direct product of the alleged ‘extreme conservatism of US academic institutions’ (Navarro,4 though he is evidently happy to be employed by one), or has come at the expense of an earlier round of ‘extremely creative’ research on the evils of capitalism and the international financial institutions (Navarro4 again, but with no citations or examples provided).
By creating conceptual space for a coherent discussion of how individual action, group affiliations, and structural forces shape each other, social capital theory can be at once rigorous and relevant. In the field of public health, a key motivation for our paper was a sense that the ongoing disagreements between the three camps we identified—the ‘social support’, the ‘inequality’ and the ‘political economy’ approaches—was primarily a product of inadequate underlying social theory. We felt this situation was not only a problem for the individual contributors themselves, but for social capital theory more generally, which, in embracing the ‘bonding’ and ‘bridging’ dichotomy as its core conceptual elements, was constitutionally unable to address, in a satisfactory way, those social relationships that form in the context of institutionalized power differentials—a set of relationships that a host of studies from around the world (e.g. refs 10 and 11) have shown to be crucial for shaping well-being. We recognize that the idea of ‘linking’ social capital, at least as we have articulated it, is not without its problems, but provided, as with other dimensions, (1) the definition is lean (and thus, in principle, observable and, ideally, measurable), (2) the substantive focus is on social relationships (not merely a broad category of institutional connections), and (3) due recognition is given to the full range (positive and negative) of outcomes that such relationships can generate, we believe it is an important conceptual advance.
One of the key ways in which we think the idea of linking social capital represents a conceptual advance is because of its capacity to provoke a more integrated discussion of the state and governance. We repeat that our definition of social capital remains essentially the same as most other users' (namely, the norms and networks that enable people to work together for mutual benefit), and that, as such, it categorically does not include ‘the state’. Nevertheless, because any sufficient social theory must address the structure–agency problem, social capital theory needs to have a place for discussing the ways and means by which broader structural features of society, including the state and major institutions of governance, shape and are shaped by the actions of both individuals and, we would argue, groups. A concept such as linking social capital, which explicitly focuses on social relationships in the context of power differentials, facilitates this connection. Even so, it should be seen as part of, not a substitute for, the task of crafting a comprehensive social theory to explain given empirical and policy problems.
Indeed, the sign of important social theory and of the emergence of a valuable new concept in the social sciences (and humanities) is when it has the capacity and the power to challenge our appreciation of the meaning of the existing stock of concepts and to lead us, heuristically, towards a reformulation of our understanding of those other, more established categories, realizing that they have seemingly new aspects which we had previously been unable to specify (and were therefore unable to research at all rigorously). This is what happened when the concept of ‘human capital’ emerged and again with the concept of ‘gender’ (and much longer ago, of course, with ‘class’). Social capital is another such concept, pregnant with new perspectives. It is our view that its potential to reinvigorate our understanding of other long-established concepts should be particularly true of ‘the state’, which has far too often been lazily deployed simply to mean the converse of civil society (another loose notion), or else ‘national government’ (which, of course, is a shifting political coalition in many cases).
The state might more accurately mean the corpus of laws and their enforcement agencies current at any time in a particular polity. But what are we to make of local laws and county constabularies in Britain, for instance, or of state laws in a federal system? What are we to make of the relationship between federal government politicians and officials, and local state or municipal or county elected representatives and their salaried officers? Furthermore, most important actions or implications of ‘the state’, whether central or local, are, ultimately delivered and experienced by both citizens and officials through ongoing face-to-face relations between people, of varying socio-demographic complexions, across institutionalized power gradients of a bewildering variety of forms, from the classroom to the hospital to the law court.12 To summarily refer to all this as a generic and little-studied phenomenon of ‘state–civil society relations’, is profoundly unhelpful for both our theoretical and our empirical understanding of key problems of power, inequality, and class. The concept of linking social capital problematizes all of this and provides a fresh pathway for researchers to begin to examine how these important sets of relationships, involved in ‘state–civil society’ interactions, might relate to a range of other networks and norms (analytically identified as bonding and bridging social capital) involved in structuring the texture of social relationships throughout a society.
However, those deploying the social capital conceptual framework are not—and should not be taken to be—implicitly or explicitly claiming that a study of bonding, bridging and linking social capital (if it could be satisfactorily achieved; as Kawachi et al.2 correctly note, we remain a long way from any such empirical studies) would provide a comprehensive analysis of state–civil society relations. This would still remain a radically incomplete study without full and proper attention to the historically contingent issues of the clash of ideas, the narrative of political conflicts, and the unintended and (often) unmanageable forces of economic change. Each of these is undoubtedly a more persuasive candidate to be assigned the role of true causal primacy in any attempt at a comprehensive theory of social change. Nevertheless, the compelling point being argued by proponents of the social capital perspective is that, without taking into account the independent effects of the workings of all three forms of social capital, our understanding remains equally incomplete. The crucial point about social capital is not that it provides a complete explanation for anything, but that most explanations are incomplete without it.
|
Methods, measurement, and causation |
|---|
|
TopSocial theory: what should...Methods, measurement, and...Applications to public health...References |
|---|
Closely related to questions of theory are questions of methods for research: how one frames the conceptual foundations shapes how one proceeds to investigate causal relationships. Establishing causality is a notoriously difficult task in the social sciences and humanities, and while we certainly do not pretend to have resolved it with our summary of a case study of how the public health crisis in mid 19th century urban Britain was resolved through a social movement (pioneered mainly in Birmingham), it is nonetheless worthwhile giving a moment's consideration to how various manifestations of social capital theory and its empirical investigation, including measurement, implicitly or explicitly address the question of causality.
It is a critical epistemological and methodological problem for those doing research in the social sciences and humanities, including on social capital, that two quite distinct and somewhat contradictory general models of causation (or, rather, conventions of explanation) are equally valid and useful in their appropriate contexts: the teleological model and the Humean causal model of constant conjunction. The crucial difference between them is that the teleological model is an explanation of human purposive action, in which the outcome, or effect, is deemed to be intimately related to its ‘cause’—it is the intended goal of a prior intention (the interesting empirical questions to be investigated are how intentions come to be formed and to what extent and why their aims are achieved or frustrated, due to imperfections of design or unexpected contingent events). The Humean model, by contrast, is set up in an entirely opposite manner, arguing that for ‘A’ to be accepted as a true cause or explanation of ‘B’ we must be entirely certain that ‘B’ is discrete and independent from ‘A’ and not in some way a part of ’A‘ (the empirical question is then to ascertain to what extent ‘B’ tends to follow from ‘A’, which may, if all other known influences have been accounted for, permit Humean causation to be attributed to ‘A’).13
The co-existence of these two equally valid, yet entirely distinct, forms of explanation, and their associated methodologies, is the source of much mutual misunderstanding, since it crucially depends on the precise explanatory goals of any individual researcher which of the two forms of explanation is relevant and valid; even further complications can arise from the fact that the same piece of research by the same author may legitimately employ both forms of explanation in different phases of its analysis. The whole issue is yet further muddied by the fact that the latter, the Humean model, is the one which it has been deemed most appropriate and valid to deploy in most forms of experimental ’normal science‘ in the empirical natural and biological sciences. The issue of elucidating human purposive action is not part of the problem when, for instance, studying what extent or what kind of change in ’B‘ is, or is not produced by a precise amount of chemical or biochemical or physical factor ’A‘. So the enormous prestige of the natural science disciplines, after several centuries of advancing scientific knowledge using this approach, has resulted in this model, alone, characteristically being considered as the ’scientific‘ one—especially among policy makers and their advisors, who are not, by and large, themselves practising academic researchers. By derivation, it is often mistakenly deemed to be the case that the only form of evidence that is ’scientific‘ (and therefore valuable for policy purposes) is that which has been rendered quantifiable, since much of the most valuable science of the last several centuries has progressed through the application of quantitative methods of observation. In fact it is not quantification that is a requirement to be able to truly demonstrate Humean causation, but only the capacity to observe repeatable events. Since this is rarely possible in the study of any important social problem, it is paradoxical that this model of causation should have achieved such high status among social scientists, since its attainment must remain an aspiration, rather than a methodologically achievable possibility for most researchers. In fact, both kinds of explanation are equally rational and valid for application in the social sciences, depending on the nature of the problem being studied and the kind of ‘causal’ understanding being sought.
Thus, in the historical example briefly summarized in our initial article, both forms of explanation and their implied causal models were deployed. Firstly, it was argued (based on historical and epidemiological research) that the failure to provide both ample domiciliary clean water supplies and arterial sewerage systems in temperate urban agglomerations, such as in 19th century Britain, produced a set of identifiable negative population health outcomes and that, conversely, their provision, ceteris paribus, significantly reduced those negative health outcomes. This is primarily a causal story invoking the elegant and fairly direct Humean model. However, in order to further explain why specific, organized human communities in 19th century Britain at first failed for decades to take this action and then subsequently did so, we have to give an account in terms of the more complex and flexible teleological model. The key ingredients in this aspect of the explanation relate to those historically-researched actions and events which can tell us why, in the earlier period, there was insufficient collective political will to implement ’the sanitary idea‘, despite both a full technological blueprint described in 1842 by Edwin Chadwick and also the passage of a national Public Health Act designed to implement it in 1848, whereas, a quarter of a century later, that collective political will cohered and emerged in certain of Britain's large cities.
There were many crucial factors involved; but one of them, we would argue, was a serious problem with linking social capital among the chief protagonists in the earlier period. This was satisfactorily resolved in the later period through a combination of the social movement of the ‘civic gospel’, the Liberal party's ward-level mobilization of urban working-class voters, and the municipal programme of ‘gas and water socialism’ pioneered in Birmingham (and in its own, slightly different, way, in Glasgow, too). Social capital is not the primary explanatory factor in this account but it is a necessary one, without which our understanding of how a much more sanitary city was eventually achieved (and why it failed to be accomplished earlier) will remain incomplete. Policy responses that ignore social capital—especially getting the linking social capital right—will remain fatally flawed. Neglecting to pay attention to a vital and necessary step in the chain of causation—the dynamics of social capital—will probably elicit only counter-productive results, as was the fate of the 1848 Public Health Act, a high-handed attempt by central government, with insufficient consultation, to force the costs of cleaning up British towns onto reluctant and unconvinced ratepayers.
In the historical account, we do not, therefore, present linking social capital only as ‘good government’, nor do we see it as something that is synonymous with, or determined by, the central state. Even in a partial democracy, voting citizens tend to get the state they vote for. The key point about the relatively generous 18th century ‘old’ Poor Law was that it was both funded and administered locally; the paternalistic linking social capital it fostered was primarily that between local dignitaries and the people of their communities, not the central state. It was these cross-class relationships that were altering as the urbanizing propertied class increasingly embraced the morality of the market and voted for a meaner, ‘new’ Poor Law, enacted in 1834. The character of linking social capital must be dynamic and closely related to changes in the balances of bonding and bridging social capital, as well. With bonding becoming more preponderant in the class-divided and sect-ridden new towns at this time, the prediction would be that in British cities by the 1840s, in addition to many class-specific bonding associations and rather few bridging ones, social relations across power differentials would be strongly characterized by the norms and networks of officiousness (which produces covert, patron–client type relations as the main way to negotiate this distance). Certainly the hated overseers and Gradgrind administration of the Victorian New Poor Law confirms this. By the 1870s, however, in leading cities like Birmingham, there was emerging more of a mutual recognition that town councillors and municipal officials should interact with citizens on the basis of mutual respect. The spirit of these new norms and networks can be seen in both the new ‘incarnational’ religious tone (of which the civic gospel was a manifestation) and the emerging ethos of service and community being proclaimed in the journals and articles of association of the new public service municipal professions and the mission settlements, which both came into existence during the last quarter of the nineteenth century.14–16
We have to learn how to use social capital analysis in this dynamic way, aware of the ways in which its influence can change depending on context. This is very different from the kind of determinist causal role that Navarro,2 and apparently even Muntaner,17 too, wish to give to the concept of class. It is otherwise incomprehensible to fathom why Munataner should think there is any value to his claim that the interpretation of the history of politics, social capital and public health, which we sketched for Birmingham in the 1860s and 1870s, could not have any validity because of the character of imperial policies which Joseph Chamberlain developed 20–30 years later at Westminster, having famously crossed the floor of the House of Commons to become a Conservative in 1886. The second half of Chamberlain's career in national politics as a Conservative Unionist, who eventually in 1903 became the main protagonist of tariff reform—the bête noire of free trade radical liberalism—was a quite different story, though equally fascinating, to that of the first half of his career as an industrial manager and leader of provincial radical liberal social reform, who emerged in the 1860s, fired by the civic gospel moral crusade, as Chair of the Birmingham-based Liberal Education League to become elected Mayor of the city for three successive years and then Liberal MP in 1876.18 One of the great benefits and intellectual gains of the half-century of recent historical scholarship and debate that has gone into investigating politics and class relations in modern British history, stimulated by the pioneering work of Marxist historians such as Eric Hobsbawm and EP Thompson, is that simplistic material determinist interpretations of ’class interests‘, have been shown to be completely inadequate.19–23 All historians agree that class is important, but it is only the beginning of our understanding. To assert, as Muntaner17 appears to, that the crude device of labelling Joseph Chamberlain a bourgeois ’class agent‘ provides a satisfactory explanation of his motives and of the historical significance of all the diverse activities in which he was involved throughout his long career is, by contrast, an end, not a beginning, to the attempt to understand causation in history and human affairs.
|
Applications to public health policy |
|---|
|
TopSocial theory: what should...Methods, measurement, and...Applications to public health...References |
|---|
This foray into the broader purposes of social science theory and method is a vital prerequisite for addressing the practical concerns of how to apply the idea of social capital to policy issues, especially public health. It is vital because good and bad policies alike are inherently (though not exclusively) expressions of an underlying set of ideas, and because the efficacy of efforts to change or improve policies deemed to be ‘bad’—and/or to defend policies deemed to be ‘good’—rests in part on their capacity to articulate a plausible account of both the problem and a corresponding solution. It was not the purpose of our paper to demonstrate that one of the three perspectives on social capital and public health was ‘better’ than the others, but rather to argue that (1) the ensuing (and often acrimonious) debate between them was a function of inadequate underlying social theory, and that therefore (2) a more comprehensive theory could not only offer a potential resolution but show how the core empirical claims made by each camp could be harnessed in the joint pursuit of more effective and inclusive public health policies.
To this end, it is worth repeating that we are not arguing that social capital, however conceived, is or should be the sole or even primary variable used to explain all public health outcomes. The range of such outcomes that occupy epidemiologists is surely most likely to be a product of a host of complex and interdependent structural (institutional, inequality), resource (quality and quantity), and social support variables; a debate cast in terms of which of these is ultimately ‘most important’ will be far less fruitful than one which helps identify the conditions under which particular variables become salient. Getting to this point requires a social theory for public health policy that is not constrained to think in terms only of socio-demographically similar or dissimilar group affiliations (important though they may be in certain cases)—or, for that matter, to see all outcomes as inherently the manifestations of individual choices at one extreme or class contests at the other—but one that can integrate a range of group and network memberships (including ones that span power differentials) into a broader, integrated theory of agency and structure. We like to think that our offering, by virtue of attempting such an integration, helps move these important debates into more constructive territory.
Finally, a reinvigorated social capital theory can, at a minimum, help policy makers improve what they are already doing (‘raising the floor’); more ambitiously, we believe it can also help identify new and better ways to respond to our most pressing social concerns (‘raising the ceiling’). Doing either of these things, however, will entail moving the terms of policy debates beyond familiar discussions of ‘what’ and ‘why’ to ‘how’ and ‘through whom’. If social capital's key insight is that social relationships really matter, then the focus in policy debates needs to unpack the black box of process to appreciate just how crucial are on-going face-to-face relationships to the delivery of key public health services, especially in developing countries.24 Any comprehensive strategy for responding to the HIV/AIDS crisis in Africa (and elsewhere), for example, must surely include not only efforts to lower the costs of producing and disseminating anti-retroviral drugs, but a plan for reducing the enormous stigmas (and misunderstandings) that still surround the disease, for helping rebuild the social fabric into which millions of orphans will be raised, and for helping change deeply entrenched normative expectations about male sexual behaviour. Doing any or all of these things will entail recruiting, training, and supporting tens of thousands of front-line field workers—most of whom will be citizens working in their own countries, of course—to build trusting and enduring relationships within communities—that is, linking social capital—that will, in turn, give them the credibility and leverage to help facilitate a long process of social change. On policy issues and in contexts such as this, a social capital conceptual arsenal restricted to a dichotomous ‘bonding’ and ‘bridging’ distinction is rendered needlessly tepid.
Debate over social capital will no doubt continue and efforts to deploy the concept in empirical research will remain a difficult undertaking. However, the pursuit of effective analysis and of improved practice in public health work is never likely to be an activity free from contention, nor from complex intellectual problems. The fact that the concept of social capital is generating vigorous dialectical critique and interdisciplinary exchange we take to be an auspicious sign.
|
Notes |
|---|
* The Soviet Union had capital (in many forms), Marx wrote a book on capital, and indeed, according to an excellent intellectual history, was the first person to coin the phrase social capital.5
This view is consistent with Granovetter's8 argument that contemporary social theory is caught between poles stressing either an ‘oversocialized’ or an ‘undersocialized’ view of human behaviour.
|
References |
|---|
|
TopSocial theory: what should...Methods, measurement, and...Applications to public health...References |
|---|
1 Szreter S, Woolcock M: Health by association? Social capital, social theory and the political economy of public health. Int J Epidemiol 2004;33:doi:10.1093/ije/dyh013.
2 Kawachi I, KIm D, Coutts A, Subramanian SV. Commentary: Reconciling the three accounts of social capital. Int J Epidemiol 2004;33:doi:10.1093/ije/dyh177.
3 Putnam RD. Commentary: ‘Health by association’: some comments. Int J Epidemiol 2004;33:doi:10.1093/ije/dyh204.
4 Navarro V. Commentary: Is capital the solution or the problem? Int J Epidemiol 2004;33:doi:10.1093/ije/dyh179.
5 Farr J. Social capital: a conceptual history. Political Theory 2004;32:6–33.
6 Alexander J, Giesen B, Munch R Smelser N (eds). The Micro-Macro Link. Berkeley, CA: University of California Press, 1997.
7 Mills CW. The Sociological Imagination. New York: Oxford University Press, 1957.
8 Granovetter M. Economic action and social structure: the problem of embeddedness. Am J Sociol 1985;91:481–510.[CrossRef]
9 Coleman J. Foundations of Social Theory. Cambridge, MA: Harvard University Press, 1990.
10 Narayan D, Patel R, Schafft K, Rademacher A, Koch-Schulte S. Voices of the Poor: Can Anyone Hear Us? New York: Oxford University Press, 2000.
11 Marc A, Dudwick N, Gomart E (eds).When Things Fall Apart: Qualitative Studies of Poverty in the Former Soviet Union Washington, DC: The World Bank, 2003.
12 Pritchett L, Woolcock M. Solutions when the solution is the problem: arraying the disarray in development. World Development 2004;32:192–212.
13 Von Wright GH. Explanation and Understanding. London: Routledge and Kegan Paul, 1971.
14 Meacham S. Toynbee Hall and Social Reform, 1880–1914: The Search for Community. New Haven: Yale University Press, 1987.
15 Hilton B. The Age of Atonement: The Influence of Evangelicalism on Social and Economic Thought, 1785–1865. Oxford: Oxford University Press, 1988.
16 Szreter S. Fertility, Class and Gender in Britain, 1860–1940. Cambridge: Cambridge University Press, 1996, pp. 190–97.
17 Muntaner C. Commentary: Social capital, social class, and the slow progress of psychosocial epidemiology. Int J Epidemiol 2004;33:doi:10.1093/ije/dyh200.
18 Marsh PT. Joseph Chamberlain: Entrepreneur in Politics. New Haven: Yale University Press, 1994.
19 Kaye HJ, McClelland K (eds). E.P. Thompson: Critical Perspectives. Oxford: Polity Press, 1990.
20 Reid AJ. Social Classes and Social Relations in Britain, 1850–1914. Basingstoke: Macmillan, 1992.
21 Savage M, Miles A. The Remaking of the British Working Class, 1840–1940. London: Routledge, 1994.
22 Joyce P (ed.). Class. Oxford: Oxford University Press, 1995.
23 McKibbin R. Classes and Cultures: England 1918–1951. Oxford: Oxford University Press, 1998.
24 World Bank. World Development Report 2004: Making Services Work for Poor People. New York: Oxford University Press, 2003.
CiteULike 
Connotea 
Del.icio.us What's this?
This article has been cited by other articles:
|
|
 |
|
  G. DAVEY SMITH Cultural climate, physical climate, life, and death Int. J. Epidemiol., April 1, 2006; 35(2): 211 - 212. [Full Text] [PDF]
|
|
|
|
|
|
B. Harris Health by association Int. J. Epidemiol., April 1, 2005; 34(2): 488 - 490. [Full Text] [PDF]
|
|
|
|
|
|
S. Szreter Reply to Bernard Harris Int. J. Epidemiol., April 1, 2005; 34(2): 490 - 492. [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||