The research productivity of universities. A multilevel and multidisciplinary analysis on European institutions

The paper makes use of a novel dataset at European level which includes data on academic staff and students of universities from official National Statistical Authorities, bibliometric indicators on publications, and socio-economic indicators at regional level. The dataset covers all European countries. The unit of analysis is a combination between teaching activities at the level of Field of Education and publications classified by Field of Science, resulting in five major integrated areas of STEM (Science; Engineering; Computer Science; Agriculture; Medicine). Using a multilevel modeling framework and comparing results across disciplinary areas the paper finds strong support for peer effects at institutional level and for the positive effect of international collaborations and attraction of foreign PhD students. It does not find support for economies of scale in research, institutional age, specialization effects and private vs private governance. The external regional environment has an impact on research productivity only in applied disciplines.