Faculty Spotlight: Jonathan Stewart

Jonathan Stewart is an assistant professor in the Department of Statistics, part of Florida State University’s College of Arts and Sciences. His research focuses on statistical network analysis, which analyzes data for patterns and connections within networks, with applications ranging from infectious disease models to how people connect and interact. In 2024, Stewart earned a three-year, $300,000 grant from the National Science Foundation to develop new statistical models of how social connections form, challenging current statistical theories.

Tell us a little about your background, where you’re from, and what initially brought you to FSU.

I studied statistics at Rice University in Houston, Texas, earning my bachelor's degree in 2013. I also earned my master's degree and doctorate in statistics from Rice in 2018 and 2020, respectively. While on the academic job market, I was drawn to FSU for its strong leadership and supportive academic environment, as well as the long history of excellence that the department of statistics has enjoyed since its founding in 1959. I officially joined FSU as an assistant professor in Fall 2020.

What inspired you to pursue statistics?

I initially pursued music as an undergraduate student, studying viola performance. I had many interests, especially in the social sciences and statistics. I became particularly fascinated with statistics and data visualization after taking a class with data scientist Hadley Wickham, recipient of one of the highest honors in the field of statistics: the Committee of Presidents of Statistical Societies Presidents’ Award. Wickham became my undergraduate research adviser and an adjunct professor at Rice. I loved quantifying things about the world, turning real-world observations into meaningful information, and communicating complex numerical data using visualization. After this class, I shifted my focus entirely to statistics.

Can you describe your main research focus?

My research focuses on statistical network analysis, which uses statistics to study relationships and connections in data. Network data isn’t just about social media; it helps us understand structured relationships such as how researchers collaborate, friendships form or communication flows. My work involves developing statistical models to describe these networks and identifying the best ways to analyze and learn from them.

What do you want the public to know about the importance of your research?

Our world is deeply interconnected, and understanding networks is crucial across disciplines, from biologists studying gene interactions to economists examining financial systems. Statistical methods help unravel complex relationships, track disease spread and manage financial risk. My work ensures these tools are reliable, transparent and scientifically sound, allowing researchers to make informed, data-driven discoveries.

Tell us more about your NSF-funded research.

NSF’s support has allowed me to explore new statistical methods to better understand how social connections form. I’m collaborating with Joshua Loyal, also an assistant professor in the Department of Statistics, on a project challenging the conditionally independent dyad assumption within latent variable models for network data, a ubiquitous assumption in latent variable models for network data which assumes connections, such as friendships, can be modeled independently given latent variables. In reality, relationships are interconnected and dependent, and new connections often emerge through existing ones. With $300,000 in funding, our research challenges this assumption by developing novel frameworks for models that better reflect how connections, such as friendships, influence one another.

In what ways was earning the 2024 Dean’s Travel Award meaningful to you?

The 2024 Dean’s Travel Award supported my attendance of the 18th International Joint Conference on Computational and Financial Econometrics and Computational and Methodological Statistics in London, U.K. I engaged with top researchers, presented new ideas, and brought new insights back to my students and colleagues at FSU. My research field has a strong presence at this conference, and attending was an essential networking and collaboration opportunity. This award is one of the many ways in which I feel incredibly supported by FSU and the Department of Statistics — they genuinely want faculty to succeed and support our research endeavors.

Can you share a project that has been particularly impactful or exciting for you?

Most of my projects are theoretical or methodological, meaning they focus on developing new statistical concepts or improving existing methods rather than directly analyzing data. One significant project is my co-authored paper, "Concentration and Consistency Results for Canonical and Curved Exponential-Family Models of Random Graphs," published in the Annals of Statistics. This work is joint with Michael Schweinberger, a professor of statistics at The Pennsylvania State University. This work was meaningful because we established the first first statistical theory for consistency of exponential-family random graph models, which had been around for over 40 years at the time of publication. This work was instrumental in my career and has provided many avenues of future research that I've explored at FSU.

What’s your favorite part about your job?

As a researcher, I get to solve exciting and meaningful problems every day. As an educator, I enjoy opening students’ eyes to new ways of thinking. I love showing students that they can make sense of uncertainty and gain meaningful insights from data using statistics. My goal is to help students become the best versions of themselves by encouraging them to push their limits, take risks and embrace growth.

Do you have any exciting projects you’re working toward?

I’m exploring how my research can be extended to other fields, such as using networks for causal inference and predicting unseen connections. For example, if we only see part of a network, can we accurately model the rest? These challenges require deep methodological and theoretical work. Additionally, FSU has increased the number of network analysis faculty over the past five years, and I’m excited to collaborate and exchange ideas with them.

If your students only learned one thing from you (of course, hopefully they learn much more than that), what would you hope it to be?

The biggest lesson I hope my students take away is to challenge themselves. Some of the most important skills you can develop are resilience and creativity. Real-world problems are unpredictable and require perseverance. Pushing yourself beyond your comfort zone builds the problem-solving skills needed to navigate setbacks and think outside the box. The ability to persist through adversity and adapt to challenges is one of the strongest predictors of future success, no matter where your path leads.