How the nature of networks determines the outcome of publicly funded university research projects

We study the influence of network characteristics-breadth, composition, and depth-on outcomes of publicly funded university research projects. These outcomes are classified in Stokes' research quadrant. The article is based on a combined quantitative-qualitative evaluation of a competitive publicly funded research program, known as 'Mobilizing Programs', in Belgium in the period 2002-11. Projects funded by the Programs aim to direct university research towards potential business applications in the medium term. The unit of analysis is the project beneficiary, and research cooperation is an explicit prerequisite for obtaining public funding. The novelty of the article lies in the combination of a refined setting of breadth, composition, and depth of research networks at project level. We find that a high number of partners in a network (breadth) stimulates pure basic research and that importance of partners (depth) is supportive of use-inspired basic research. We highlight the role of public research centres to render university research projects more use-inspired and application-driven. When different types of partners are involved (network composition), joint university-business research teams are not necessarily a recipe for application performance.