"We must encourage international co-author networks"

(From "Horizons" no. 108 March 2016)

What methods did you apply?

Working with my colleague Fabiana Visentin, we tried to isolate the effect of the mode of recruitment of doctoral students on their scientific productivity between 2000 and 2008. This was measured by the number of publications. We wanted to examine the influence of two factors on the productivity of a doctoral student: having come from another institution​​​ and having a supervisor with an external research network.

What was the result?

Postdocs recruited externally are more productive than those recruited internally. The most prolific come from other institutions with which their supervisor maintains ties, particularly co-authorship ties. They are 10% more productive than students coming from a university completely outside of the professor’s network, and 16% more than local students who stayed put after finishing their PhD.

How do you interpret this?

It shows the role played by social and professional relationships, particularly for recruiting good postdoc students, something that is often a difficult task. A professor’s network can help in selecting good candidates.

What are the consequences for elite universities?

Our results are interesting for institutions, such as EPFL, that deploy global marketing strategies for recruiting talent. Whilst it’s good to have open doors and to be able to double-check the authenticity of unknown certificates, it’s also important to encourage researchers within an institution to develop their external co-author networks, particularly in recruitment countries. Some American universities employ this strategy; they hire Chinese researchers on the basis of the quality not only of their work but also of their networks. These researchers and their connections in China can prove useful when it comes to taking on yet more new postdocs.