In the year following the Third International Chordoma Research Workshop, participants formed 30 new collaborations (thick lines) and 174 new relationships (thin lines), increasing network density by 57%. View full-size network graphs: 2011 | 2012
Measuring the evolution of a research network
When the Chordoma Foundation started in 2007, research on chordoma was at a near-standstill, and only a handful of labs spread across the globe were actively studying this rare disease. Among the most important factors that impeded progress in the field – in addition to scarcity of funding and lack of critical biological materials – was the isolation and lack of coordination among these disparate chordoma researchers.
To connect and expand the global chordoma research community, between 2007 and 2011 the Foundation hosted a series of three International Chordoma Research Workshops (ICRW) in partnership with the National Institutes of Health. In total, these meetings brought together over 150 physicians and scientists from nine countries, many of whom were new to the field of chordoma research.
Over the past five years we have observed numerous examples of new research projects being prompted by unpublished data that was exchanged, or new relationships that were formed, during the workshops. For example, at the last workshop, a researcher from Johns Hopkins who developed a mouse model of chordoma connected with scientists from the NIH who had identified approved drugs that showed promise in chordoma cell lines, leading to a collaborative drug screening project now funded by the Chordoma Foundation. However, while there is anecdotal evidence that the workshops stimulated fruitful new collaborations, we want to actually quantify the impact they have on relationships among participants.
To measure how relationships among participants change after attending a workshop, we conducted a survey immediately prior to and one-year following the 2011 ICRW asking all participants to what extent they knew each other workshop participant. A participant could indicate that he or she (i) does not know, (ii) knows of, (ii) has interacted with, or (iv) has collaborated with each other participant.
Based on survey results, 174 new relationships and 30 new collaborations were formed following the workshop. Social network analysis (courtesy of Eric Giannella, UC Berkeley) revealed that the network density (proportion of potential relationships actually established) increased by 57% and the average path length (the number of steps between any two individuals in the network) decreased by 23%.
|Number of relationships||268||442||+ 65%|
|Number of collaborations||158||188||+ 19%|
|Network density||0.081||0.127||+ 57%|
|Average path length||2.7||2.2||– 23%|
Network density is a measure of how “close knit” a network is; it is the proportion of potential connections that are actually established. Characteristic path length is the average number of steps between any two nodes in a network.
Accelerating research by optimizing the network
Creating relationships and collaborations among researchers is not an important end in itself. It is important because each individual instance of collaboration represents investigators coming together to accomplish something that they could not do alone, and each relationship represents the potential for future collaborations or sharing of information, know-how and ideas that could lead to new research and new discoveries. Similarly, studies of scientific networks have demonstrated that higher network density (proportion of realized connections) [1-3] and lower characteristic path length (average number of steps between nodes)  are associated with increased innovation within the network.
Thus, the marked increase in the number of relationships and collaborations among workshop participants, the increase in network density, and the decrease in characteristic path length of the network could portend a further acceleration of progress in the coming years. Over time, we will test whether in fact the observed increase in cohesiveness among workshop participants results in an increase in discoveries, as measured by the number of chordoma-related publications produced by participants after the workshop.
To our knowledge, the effect of a scientific conference on the interconnectedness of a research community, and the productivity of that community, has never been measured. However, we think that understanding how a conference, or any other intervention, affects the dynamics of the network, could help us be smarter about how we focus efforts in the future (e.g. might we design future meetings to intentionally connect certain isolated regions of the network?). Therefore, we will administer the survey again before and after the next research workshop and intend eventually to survey the entire chordoma research community annually.
Our responsibility as a Foundation seeking to accelerate the development of new treatments for chordoma is to use the resources entrusted in us to catalyze meaningful research as efficiently and effectively as possible. The results of this survey suggest that among the possible ways to stimulate research — awarding grants, offering prizes, distributing biospecimens, etc. — convening the global research community produces a large return on investment. In light of these findings, we embrace the role of the Foundation as chief connecter within the chordoma research community, and we are doubly committed to investing in programs to facilitate communication and collaboration among chordoma researchers. The next research workshop will be held March 21-22 in Boston MA. Click here to learn more »
- Singh J. Collaborative Networks as Determinants of Knowledge Diffusion Patterns. Management Science. 2005;51(5):756–770.
- Abrahamson E, Rosenkopf L. Social Network Effects on the Extent of Innovation Diffusion: A Computer Simulation. Organization Science. 1997;8(3):289–309.
- Ebadi YM, Utterback JM. The Effects of Communication on Technological Innovation. Management Science. 1984;30(5):572–585.
- Fleming L, King C, Juda AI. Small Worlds and Regional Innovation. Organization Science. 2007;18(6):938–954.