Investigation of the correlation between researchers' properties and productivity through analyzing co- authorship networks

1 Background We investigate and clarify which properties of a researcher, including the importance in collaboration networks, have strong relationships with his own productivity or that of his collaborators. Our analysis serves following purposes; to identify factors affecting productivity, and to grasp the characteristics of research domains in terms of knowledge production. Such information is also useful in more practical applications; for example, when a researcher searches his partners in research collaboration, and when one judges the substitutability between performance indices in research evaluation. A large number of studies have analyzed the correlation of properties between authors and their co-authors [e. reported that there is a positive correlation between the productivity or citation ratio of a researcher and the importance in the research collaboration network. These studies mainly examined the synchronic correlation between researchers' properties. On the other hand, the diachronic correlation of properties, that is, the correlation between their subsequent and precedent activity, has not yet been sufficiently studied using quantitative methods. One of a few exceptions is Yoshikane et al. [2007], who analyzed the correlation between the researcher's productivity subsequent to a collaboration and the collaborator's precedent activity, targeting only newcomers and their senior collaborators. We extend their methodology and analyze the correlation between researchers' properties and each of the following four aspects of productivity: (1)