ABSTRACT
Our search for parallels between the construction of empirical data in natural science and sociology reaches its conclusion as we now come to consider the actual act of data collection. When comparing the specific measurement operations of physical and social science, one’s first (and quite possibly second) thoughts must be to admit defeat and abandon any pretence that there are worthwhile analogies to be drawn between what are essentially disparate practices. Natural science measurement, as we have seen in Chapter 4, consists of a whole series of physical manipulations, the crucial one being the transformation of the property to be measured into some other form of energy. This transduction process enables investigation of the measurand via its action on further properties under the closed conditions created within the measurement apparatus. Any notion of social measurement in the form of energy transfer, instrument construction, signal transmission, pointer reading and all the rest is patently fanciful. Sociological data will always emerge in an interactional phase between researcher and subject (directly in the case of interviewing, indirectly with the use of official data). Accordingly, measurement in social research will always be primarily an act of translation. There will always be a need to square the conceptual elements established in relatively closed theoretical models, with the reasoning and language of the respondent, which operates routinely in much more open conceptual systems.
