Abstract
Classification is integral to comparison. The aim of this paper is to reflect on the nature, purpose and limits of classification in comparative health policy. We begin by describing the role of classification in comparative research design, discussing Weber's concept of the “ideal type” and drawing on the sociology of scientific knowledge to reflect on classification as an essentially social and uncertain process. In the sections which follow, we present an outline history of the classification of health systems, identifying a “normal science” of comparative studies of health policy and exploring a number of theoretical, conceptual and methodological issues which arise from it.
Notes
1. As early as 1973, Mark Field proposed the following “formal definition”: “The health system is that societal mechanism that transforms generalized resources into specialized outputs in the form of health services” (Field Citation1973: abstract and 772).
2. “[T]heories that lie between the minor but necessary working hypotheses that evolve in abundance during day-to-day research and the all-inclusive systematic efforts to develop a unified theory that will explain all the observed uniformities of social behaviour, social organisation and social change” (Merton Citation1968: 39); “[T]estable propositions, derived from fundamental theory, addressing observable phenomena”, http://www.sociology.columbia.edu/about/main/dept_history/index.html (accessed October 24, 2006).
3. Of course, even case studies are cases of something (some category or class).
4. Watkins detects a shift in Weber's conception of the ideal type between the essay on “Objectivity in Social Science and Social Policy” (1949[1904]) and his posthumous Wirtschaft und Gesellschaft, distinguishing between holistic and individualistic versions. The latter text is even clearer that an ideal type is constructed “not by withdrawing from the detail of social life, but by formalising the results of a close analysis of some of its significant details considered in isolation” (Watkins Citation1952: 24).
5. It is for this reason that classification became the object of social scientific attention in anthropology in the work of Durkheim and Mauss, and Lévi-Strauss (Boyne Citation2006).
6. Cited by Barnes et al. (Citation1996: 56). This is a form of the problem identified by Sartori (Citation1970) as conceptual “stretching” or “traveling” and debated by Collier and Mahon (Citation1993); for discussion of its relevance in comparative health policy, see Burau and Blank (Citation2006). Where Sartori is concerned with accuracy, the sociological argument is more radical in claiming that standards of accuracy are themselves socially informed.
7. For a more extended (and more normative) treatment, see Anderson (Citation1972); for an appreciation of the original article, see Freeman and Marmor (Citation2003).
8. In essence, this means that health care is financed from public sources, either through taxation or through compulsory insurance contributions; that health finance is administered by public agencies who are not also health care providers, and that providers, if not independent or even private bodies, are invested with the organizational and managerial autonomy to compete for contracts with purchasers.
9. For discussion of different comparative projects in health policy, see Marmor et al. (Citation2005).
10. For a historical review of this literature, see Abrahamson (Citation1999) for additional critiques, see Carrier and Kendall (Citation1986) and Veit Wilson (Citation2000), and for empirical reassessment of Esping-Andersen's typology, Arts and Gelissen (Citation2002).
11. Fuzzy set theory was conceived to address this problem, though we know of no instance of its being used in comparative health research.
12. There are exceptions, of course, notably Hollingsworth (Citation1986) and more recently France (2006). Gusmano et al. (Citation2006, 2007) take a different line, focusing on the distinctive health politics of cities.
13. This is why, following recent reform of health care in the Netherlands, the difficulty of establishing an agreed description of the new system (public, private or hybrid) has become such a significant public issue (Okma, this volume).