Studying and Shaping Networks of Scientific Collaboration at UF

Primary tabs

Publication Date: 
Monday, December 9, 2013

The goal of this project is to map the social networks of professional collaborations among researchers in all disciplines and colleges at UFMost academic research is the result of collaboration among scientists. Collaboration occurs in teams and networks of researchers: Studying these networks helps monitor and drive research productivity. In 2012, BEBR and the UF Clinical and Translational Science Institute (CTSI) started a project aimed at mapping, visualizing, analyzing, and shaping collaboration networks at the University of Florida. The goal of this project is to map the social networks of professional collaborations among researchers in all disciplines and colleges at UF over the years, and use network knowledge to enhance research collaboration and scientific productivity in the University.

Network data can be used to construct collaboration metrics and use them to monitor and shape the pattern and evolution of collaborationsSocial Network Analysis (SNA) is the main methodology that we are using . Social networks are sets of actors and the relations among them: actors at UF can be researchers or organizations (departments, institutes, centers), while relations are professional collaborations—for example, in the case of individual researchers, co-authoring a publication, applying for a grant together, or filing a patent together. Our main data sources are the VIVO database on UF publications, for co-authorship networks; and the database of all awarded grants at UF, from the UF Division of Sponsored Programs, for networks of collaborations on research grants. The first result of mapping collaboration networks is the capability to locate people and organizations in the networks: researchers and research centers can see where they are, where their collaborators are, and where people who are working on similar things are in the UF scientific network over the years. A free look-up application will be offered on the BEBR website to interactively show users the position of people and organizations in the UF network. From the standpoint of research centers, network data can be used to construct collaboration metrics and use them to monitor and shape the pattern and evolution of collaborations inside and outside the centers.

CTSI gathers some of the most central scientists, who tend to be better connected to the University scientific networkPart of the analysis involved measures of network cohesion and network distance among researchers, and showed how the UF scientific network became more cohesive in the last five years, that is, the overall tendency to professional collaboration increased in the University. In the same years, the CTSI began to cover an increasingly larger and more central part of the UF network. Different metrics can be used to assess researchers' centrality in the network. “Degree centrality” is the number of collaborators a researcher has, that is, his degree of connectedness to other scientists in the University. “Betweenness centrality” is a measure of brokerage quantifying the extent to which a researcher falls on the network paths between other researchers, and is a bridge between separate areas of the network. Finally, “closeness centrality” measures the extent to which a scientist is close to every other actor, and can easily reach to the rest of the University network. CTSI researchers showed higher centrality values on all these measures in the last five years, compared to researchers outside the CTSI. In other words, the CTSI gathers some of the most central scientists, who tend to be better connected to the University scientific network, more often located in brokering positions between different research groups, and able to more easily reach to any other UF scientist.

We also defined measures on interdisciplinary research. We started by looking at the "neighborhood" of a researcher in the collaboration network, that is, the set of all his collaborators in a year. Diversity in a researcher's neighborhood, in terms of collaborators' departments, colleges or disciplines, measures how interdisciplinary that researcher's work was in a year. Using this index, CTSI affiliates show an increasingly higher diversity of collaborators compared to non-CTSI researchers, which reveals that the CTSI has functioned as a hub for interdisciplinary research at UF in the last five years. This is confirmed by the sheer number of authors per publication and investigators per grant in the CTSI network, which is on average higher than in the rest of the University. Furthermore, the CTSI network tends to include increasingly more "open triads" than "closed triads" of researchers over the years. Closed triads are groups of three researchers in which everyone collaborates with each other, whereas in open triads two of the three researchers do not collaborate, and the third one is a bridge between them. Because open triads tend to be bridges that span separate and distant areas of the network, this structural feature of the CTSI network suggests a growing diversity of backgrounds, methods, and substantive topics in CTSI-supported research activities.

So far, the BEBR/CTSI social network project has provided insights on several aspects of the UF scientific community and its workings, such as:

  1. The evolution of scientific collaborations at UF in the last years;
  2. The drivers of this evolution;
  3. The centrality and connectedness of specific UF centers to the rest of the University;
  4. The development of interdisciplinary research efforts across UF colleges.

However, SNA is not only used to describe and explain networks, but also to intervene on them to achieve desirable goals in organizations. One component of our project is to design ways of changing the network, that is, adding or rewiring collaboration links, with the goal of enhancing the scientific productivity and overall quality of the University research network. Using criteria based on network structure, we identified missing collaborations among specific researchers, whose development could be particularly beneficial to both the collaborators themselves, and the whole UF network, for example by connecting entire diverse scientific communities in the University. BEBR carried out a survey among these potential collaborators to explore the possibilities of connecting them, and actually creating the missing links, in an innovative strategy to support scientific research at UF through network interventions.

Results and updates from this ongoing project can be found in a dedicated page in the BEBR website.

 

The first use of Social Network Analysis in the Social Sciences can be traced back in the 1930s, but SNA did not really develop into a recognized academic fields until the 1970s. Ever since, this set of analytical methods, and the scientific theories it underpinned, became increasingly popular in such diverse disciplines as sociology, physics, anthropology, political science, psychology, marketing, biology, and public health. A good introduction to the history and development of SNA is this Science article: Borgatti, Stephen P, Ajay Mehra, Daniel J Brass, and Giuseppe Labianca. “Network Analysis in the Social Sciences.” Science 323, no. 5916 (2009): 892–5.

VIVO is an open source software system that provides both open-access data on people, organizations, research and events at UF, and a search engine to browse professional profiles in the University. It was originally conceived as an information resource for scholarship, that researchers could use to find potential collaborators and mentors.

 

Publication Types: 
BEBR Division: