CLUSTERING RESEARCH FIELDS FOR MACRO-STRATEGIC ANALYSIS - A COMPARATIVE SPECIALIZATION APPROACH

The goal of this article is to show that it is possible to construct an index to measure a country's relative specialization in different scientific fields in a way which is both reliable and relevant for macro-strategic analysis. We will call this index a "Revealed Scientific Advantages Index". The technical problem to be discussed is one of aggregation: how can we be sure that an index calculated for a small number of relatively large fields does not mask significant differences that might have shown up had a lower level of data aggregation been used? Science policy needs synthetic measures which am easy to interpret. We will show that the "Revealed Scientific Advantages' approach offers the possibility of building them. The study itself is based on figures obtained through an exploitation of the INIST/CNRS PASCAL database classification of science. 107 sub-fields of this classification were initially used to determine the areas of specialization for 11 countries ("revealed national advantages"). Clustering techniques were then used to aggregate this data and 13 specific fields were identified. The science policy information produced during the study concerned these 13 fields. It proved to be both easily understandable and relevant for macro-strategic analysis.