Harnessing Little Networks: One Research Team’s Quest to Analyze the Social Networks of Small Groups

By Samantha Devapiriam — December 07, 2020

Building relationships and trust is key when working in groups. Relationships are essential to the growth of a company, but how can we quantify the efficacy of teams forming social networks? Enter Kayla de la Haye, Ph.D. and George Vega Yon, Ph.D., two researchers from USC Keck’s Department of Preventive Medicine who helped develop a new analytical and statistical approach to understand not only how teams form social networks, but how this impacts their outcomes and performance. 

To further operationalize this and empower other teams to work in groups, they created a software to implement it. The software, which they called ERGMitos, builds off of Exponential Family Random Graph Models (ERGM). ERGMs are often used by scientists who want to observe and understand social networks as well as test hypotheses on them. While ERGMs are typically used in analyzing larger groups, the focus of this research was expanding ERGMs to observing smaller networks.

“Our new method extends ERGMs so that they can be more flexibly and reliably used to analyze small networks. This approach lets us test hypotheses like: Are team collaborations more likely to form between people in the network who are similar in gender or similar in their academic discipline? What type of people become isolated vs. central "hubs" in the network and collaborate with many other teammates? What type of team network structures increase team performance?”

In August of this year, de la Haye and Vega Yon had their research published in an academic journal, Social Networks. Regarding the implications of this research to the medical community, de la Haye said, “In science and medicine, there is growing empirical evidence that team science is incredibly valuable for solving complex problems. The methods we developed provide new rigorous statistical analysis tools to study the structure of small team networks, to better understand how these networks form, the role they play in team success, and to develop better training and support for scientific and medical teams that will help them foster 'effective' team networks.”

The work of de la Haye and Vega Yon is already garnering recognition in the research community. In addition to receiving over 20,000 hits for a tweet (pictured) promoting the publication of their research, the software they developed has over 5,500 downloads since its release earlier this year. To learn more about this innovative approach to ERGM, their publication, titled “Exponential random graph models for little networks,” can be found here.

With the future of working environments becoming increasingly virtual, any team would be smart to take advantage of this software and see how they can improve the social networks within their workplace.

NIH Funding Acknowledgment: Important - All publications resulting from the utilization of SC CTSI resources are required to credit the SC CTSI grant by including the NIH funding acknowledgment and must comply with the NIH Public Access Policy.