Operationalization,a superfluous concept

This paper discusses a number of issues in the methodology of social and behavioral sciences which concern how concepts are defined and related to observations, how theories are constructed, and how we proceed in actual research. Examples are mainly taken from experimental psychology, and from social psychology, but the problems and answers apply to the whole of behavioral and social sciences. These issues are related to philosophy of science. They are, however, presented from the point of view of a researcher who asks himself how to proceed in actual research, and who is critical about the present status of social and behavioral science. The main point is a pledge for a renewed empiricism. This is close to what can be called logical empiricism but should not be identified with any particular-ism. Some issues raised by critics of logical empiricism are addressed in the final section. The first problem deals with how to clarify concepts in behavioral and social science. This will show an unresolvable circularity in operationalization of theoretical constructs. The position and role of measurement is discussed in relation to the objectives of behavioral and social research. Subsequently, it is pointed out that a clear distinction between empirical concepts and theoretical constructs solves many, if not all of the problems of defining and clarifying concepts, thereby obviating the need for a so-called operationalization. It should, however, be very clear from the beginning that we cannot do without operational definitions of theoretical concepts, and that to some extent we must return to the original spirit of operationism, although in a manner which is different from current practice. An important element of the advocated approach is the definition of the empirical domain of a theory in terms of observational categories. Guttman's facet design can be a useful tool to define domains, especially for non-experimental research.