||In socially distributed models of knowledge production (Mode 2), internationalisation plays a major role in the production of scientific knowledge (Gibbons et al., 1994). Although there exist profound disputes over the historical and conceptual accurateness of this model (Etzkowitz and Leydesdroff, 2000; Pestre, 2003), the increasing intensity and heterogeneity of international collaboration, a key component of internationalisation strategies, is a major point of agreement among scholars. Disagreement, however, still revolves around the methods that can be used to study internationalisation dynamics; whether they should be based on qualitative or quantitative techniques. The recent debate on the responsible use of scientometrics in research evaluation, I argue, could help shed some light into the use of scientometrics in the social study of science, technology and innovation (SSTI). More specifically, by arguing in favour of developing a multi-level analysis and a mixed method approach, I show how the study of international scientific collaboration offers an ideal setting where both qualitative and quantitative traditions can start a dialogue. This requires, nevertheless, understanding first the ontological and epistemological basis of scientometric indicators and the implications of its use in the social study of science.