Empirical generalizations in management science: Theory and modeling

Abstract

Empirical generalizations refer to a repeating pattern or a set regularity that repeats itself over different conditions of a system. In this paper, our goal is to reinforce the growing perspective mentioned above. Specifically, we highlight the concept of empirical generalizations in the field of management (marketing) science, which avers that the phenomena of theory-building and data generation should go hand in hand in order to build higher-order theories. We begin with the discussion of the scientific approach and mathematical model building. The concepts of empirical generalizations are then discussed to elaborate on their role in theory-building and research. We discuss 2 salient approaches to developing empirical generalizations—(i) First theory, then data, or (ii) First data, then theory. We then use the Bass diffusion model as a base to develop the case for the greater use of empirical generalizations in the field of management. The objective of a diffusion model is to generate a life-cycle sales curve of a new product/service based on a small number of parameters. The parameters may be estimated based on the consumer pre-tests, an analogy with the sales pattern of similar products in the past, or by early sales returns as the new product enters the market. Starting from the seminal paper by Bass, a vast amount of literature has been produced in the diffusion modeling. Finally, we provide a product/service diffusion example of the performing arts company as an extension of the Bass model.