Academia has a reputation for being a bit of a closed world, a walled garden of knowledge where secrets are jealously guarded. But the truth is that collaboration is at the very heart of research and scientific discovery, and that for science to advance, researchers need to get together, compare notes, disagree, and have their ideas challenged and built upon by others. Often this happens naturally – like in the cafeteria where PhD students will chat about their projects – but in such a hyper-specialized environment, chances are that people who share your particular research interests cannot be found in the same institution or even the same country. What then?

In the same way that social media has revolutionised personal and professional communication and created dynamic global conversations, platforms like Mendeley now bring academics together in groups formed around those research interests, and the implications of that are tremendous for making science more open and accelerating the pace of discovery.This is why the team here at Mendeley is particularly interested in gaining genuine, real-time insight into research collaboration.

Mendeley is involved in several research projects. Particularly fruitful has been an on-going exchange of researchers and Mendeley staff between our London HQ and the Know-Center at Graz University of Technology in Austria. All projects aim to contribute to the improved use of the wealth of Mendeley data for the benefit of our users and the scientific community in general.

Unsurprisingly perhaps, this recent investigation of research collaboration started as a Hack Day project between Mendeley staff and a visiting researcher from the Know-Center/TU Graz in the context of the TEAM project (http://team-project.tugraz.at) which is coordinated by the Knowledge Management Institute of the TU Graz. Sebastian Pöhlmann (Insights and Analytics Manager) and Piotr Drozd (Community and Business Intelligence Analyst) teamed up with Peter Kraker (PhD student, Know-Center/TU Graz) to visualise cross-country collaboration on the Mendeley platform.

An interactive map has been created that aims to shed some light into the intensity of international research collaboration across different countries. Considering that using Mendeley groups is optional for our users, we are excited to have data on 113 countries. For each of those we show the continent, the rank by user count, the number of connected countries and the proportion of foreign (= international) connections.

By browsing the map or making a selection from the list, you can visualise the connections between researchers for any given country. A connection between two countries is established if at least one of each country’s researchers are members of the same Mendeley group. Of the over 200,000 research groups on Mendeley, we’ve selected private groups with at least two members, as that tends to be the most collaborative group type. Our staff is also very active on the platform so we’ve further excluded groups owned by Mendeley staff  and connections where Mendeley staff are involved. We have further excluded countries with less than 10 total connections.

Browsing the map and the data has produced some interesting insights:

• Among BRICS countries, China, India and Russia have a high proportion of international connections whereas Brazil and South Africa seem somewhat more internally focused
• Generally speaking, North America, Europe, and Australia are very well connected, whereas Asia and South America are somewhat lagging behind.
• There are a few small countries that are very internationalized: Netherlands, Sweden, Switzerland, Belgium, Austria, and Denmark. Interestingly, these countries are also at the top of the KOF Index of Globalization: http://globalization.kof.ethz.ch/static/pdf/rankings_2012.pdf

This is early days, but we hope that by learning more about how our users collaborate with each other, we can continue to develop the best tools to help them work even more efficiently. And by sharing some of the insights on Mendeley Labs we want to contribute our part of the picture to the general knowledge of how research works.

As always, we’d love to hear your thoughts. What does collaboration mean to you and how would you go about measuring and visualizing it?