The theory/method relationship

You are here

The theory/method relationship

From the point of view of A-level, and probably for many undergraduate courses, it is still commonly suggested that there is a direct correspondence between theoretical position and types of method. This is the 'classical approach'. But this position is being increasingly attacked as obsolete, a non-debate.

What are the differing viewpoints?

Research is carried out to discover something about the world. The methods used in research are linked in various ways to the researchers commitment to a particular version of reality (ontology) and ways of knowing that reality (epistemology). Research methods are linked to theory. Research methods are seen as the product of different sets of underlying assumptions.

The idea that choice of method involves important theoretical issues and is not simply a matter of technical advantage or disadvantage is suggested by Guba (1985), who argues the impossibility of combining different approaches.

There have been a number of qualifications made to the rather sterile positivist/interpretivist debate:

The difference between normative and empirical generalisations. The difference between 'ought' and 'is' statements. Thus Ackroyd and Hughes, Guba and many feminist sociologists suggest that there 'ought' to be a link between theoretical position and methods of data collection. But, Platt, Bulmer, Pawson and Bryman seem more concerned with how sociologists actually go about their research.

Ackroyd and Hughes distinguish between 'generic' and 'theory specific' methods. Generic methods treat data collection techniques as simple tools to be used or disregarded according to convenience. Any problems are primarily technical and of little theoretical importance. Theory specific methods are methods, which have developed within a particular theoretical tradition, used outside of this context they are seen as a liability rather than an asset.

In the late 1980s, the 'Sociological Review' was the site of a number of articles on the relationship between theory and method. Some of these suggest that empirical evidence does not support the assumed close connection between theory and method. Platt (August 1986) disputed the assumed connection between functionalist approaches and survey data collection techniques. She suggests that historical research reveals not only that not all functionalists used surveys but that surveys were used by non functionalist theorists. Platt argues that it is not that theory and method are never causally related, but that theory and method have no necessary causal relationship. In practice research methods have greater autonomy, they are not tied to a particular theory.

'The tendency to see theory and method as intimately related has in it, both in this case and more generally, more of an ideology about what the relation ought to be than it has of close historical observation of what actually happens.'

Bulmer (Sociological Review, August 1986) makes a similar point - that there is no necessary connection between the use of a particular method and a particular general methodological standpoint. For example, observational methods have been used in both positivist and interpretive fashion, even though their methodological assumptions are very different. Most sociological research, Bulmer argues, involves multi-method approaches, the kind of 'triangulation' suggested by Denzin. Finally, Bulmer suggests that the whole debate is pointless, a kind of 'sociological babble'.

Pawson (Developments in Sociology, Volume five), also sees the theory/method debate as a myth, a non-debate. He argues that the positivist/interpretivist debate is a product of institutional inertia and that what was learned from 1980s sociology was that the options confronting the sociological researcher have nothing to do with positivism versus interpretivism. Pawson argues that the bulk of sociologists today are basically essayists, neither quantitative nor qualitative.

'Which ever way you look at it... it is clear that positivism versus phenomenology is a dead duck... stop pretending that this opposition sets the agenda for sociological method.'(Pawson).

The case for combining quantitative and qualitative approaches is that they complement each other - they give a better general picture. Consequently, the positivism/interpretive debate has now been superceded by a new opposition between purism and pragmatism.