Abstract

We apply methods and concepts of statistical physics to the study of science & technology (S&T) systems. Specifically, our research is motivated by two concepts of fundamental importance in modern statistical physics: scaling and universality. We try to identify robust, universal, characteristics of the evolution of S&T systems that can provide guidance to forecasting the impact of changes in funding. We quantify the production of research in a novel fashion inspired by our previous study of the growth dynamics of business firms. We study the production of research from the point of view both of inputs (R&D funding) and of outputs (publications and patents) and find the existence of scaling laws describing the growth of these quantities. We also analyze R&D systems of different countries to test the "universality" of our results. We hypothesize that the proposed methods may be particularly useful for fields of S&T (or for levels of aggregation) for which either not enough information is available, or for which evolution is so fast that there is not enough time to collect enough data to make an informed decision.