Collaboration is beneficial throughout the scientific research process, from the inception of an idea through to publication. And yet, we don't always understand the factors that structure collaborative networks and the patterns that emerge from them. Here, I've explored patterns of collaboration in the Canadian ecology and evolutionary biology research community.
What you see below is based on co-authorship networks between over 1000 researchers at 25 universities across Canada over the last 30 years. This should provide a decent overview of some general trends in our community, but I apologize in advance if you or your instititution are not represented here. Please don't hesitate to contact me with questions or feedback!
Collaboration across different universities is becoming increasingly common! The networks below show the number of collaborating pairs of researchers across institutions for three different time periods, and clearly the networks are becomnig increasingly dense.
Collaborations are more likely to occur between universities that have similar research programs, are geographically close, and are of similar sizes (by number of PIs).
We can also explore the level of collaboration within an institution and how that changes over time. Turns out there is a lot of variation in the levels of collaboration within different institutions. Generally, universities with more focused research programs seem to have higher levels of collaboration – but there is still a lot of unexplained variation there. Perhaps different universities simply have different cultures surrounding collaboration?
Check out the collaborative network at your own institution below!
We can also explore the factors that influence whether individuals form collaborations. Unsurprisingly, a pair of researchers is more likely to collaborate if they are at the same institution and have similar research programs. More surprisingly, gender seems to play an important role in collaborations. Collaborations are most likely between a pair of women, intermediately likely between a woman and a man, and least likely between two men.
Check out your own collaborative network in the widget below! Note that the networks and stats displayed are just for exploratory purposes – there is additional filtering and processing for the actual analysis.