The present contribution aims at discussing some issues on the analysis of co-authorship networks providing empirical results on a case study. Two main methodological issues are related to the heterogeneity of the bibliographic archives available to collect collaboration data, and the disambiguation problem to obtain a correct identification of authors per paper. Within this scenario, we are interested in performing community detection algorithm to discover groups, and in analyzing the changes in the groups’ structure over time as a result of the first research assessment exercise attempted in Italy in the period 2004-2010. The results of Italian academic statisticians and their co-authorship relationships provide a fertile ground for reflection.
