ABSTRACT
This chapter begins with a consideration of the methodological issues complexity raises. It reviews some of the underlying thinking, particularly the deductive nomological model, and the assumption of linearity and fixed description, that is the methodological basis for much quantitative research. This fails because social categories are multi-dimensional, fluid and may be known under different descriptions, or the same description can apply to different phenomena. This is not an argument against quantitative research, but rather the basis of the claim then developed, that we can only ever know the world under a description. This is achieved through operationalisation, but this is not the same as operationalism. The chapter then goes on to propose a realist theory of operationalisation which depends on three principles: (1) Some measurements measure what they are supposed to measure; (2) Some measurements must be developed within specific contexts – a form of “local realism”; (3) Measurements are interpreted within a social or historical narrative.
