Collapsed Versions of the Detailed Class Schema

Abstract

There are both theoretical and practical reasons speaking in favour of a tighter version of our class schema. From a theoretical standpoint, it may be argued that not all the divisions made in the 17-class schema have a class character. In fact, there is most probably no class difference between the position in labour markets and production units held by skilled clerks on the one hand and routine clerks on the other. Likewise, farmers with a couple of employees do not work in a very different setting than farmers without employees (both belonging to what we call the petite bourgeoisie). In our view, the distinctions made in the 17-class schema based on four work logics and four levels of marketable skills are interesting in their own right as they permit an in-depth analysis of the workings of the employment structure. Yet depending on the object of the study, a class schema with fewer categories is more accurate and has furthermore the merit of greater parsimony. In empirical research, however, it is normally not so much for theoretical considerations but for practical constraints that collapsed versions are preferred to the original — detailed — class schema. Thus, in samples that are not quite as big as the German Socio-Economic Panel or the British Household Panel, the reduction of the number of classes responds to the statistical necessity of having sufficiently large cell counts.