Structure of international collaboration in science: Typology of countries through multivariate techniques using a link indicator

In this article patterns of international collaboration in science are investigated using a specific procedure to analyse data collected from theScience Citation Index. We develop an indicator based on the scientific linkages between countries established through internationally co-authored articles (COPs). The credibility, advantages and uses of this indicator are discussed. We apply the Correspondence Factorial Analysis method and the Minimum Spanning Tree classification to this indicator in order to observe the level of resemblance and the main characteristics of the collaboration structured by 98 countries in eight principal fields of science. The results shown summarize the diverse aspects of countries participating in collaborative works and bring into view the cognitive structure of international research. The use of these methods in the investigation of international collaboration contributes to the analysis of the complex structure of the scientific communities of different countries.