Statistics Students Report on Cross-Disciplinary Summer Research

September 17, 2026

Statistics Students Report on Cross-Disciplinary Summer Research

Three Statistics students at The Ohio State University spent the summer applying statistical theory and methodology to research questions spanning network science, educational measurement, and health care.

Arushi Vishwakarma, working with Dr. Dena Asta through the Lubrizol Summer Fellowship, studied spectral dynamics and PageRank centrality in latent space network models. Her work examined the Wigner Semicircular Law and how eigenvalue distributions change when its assumptions are relaxed. She also studied graph Laplacians for different network models and began exploring how spectral distributions could be used to better understand reciprocity in real-world networks.


Alex Nguyen created another Lubrizol-supported project focused on Bayesian model selection for Item Response Theory (IRT) models. Building on previous master's thesis research, the project examined computational methods for comparing complex IRT models using Bayes factors, with the goal of developing the work into a manuscript for publication. The summer also included refining research code, conducting additional analyses, and reviewing related literature.


Jaehoon Kim focused his summer research on high-dimensional regression methods for drug–drug interaction screening. His project examined statistical approaches for identifying medication combinations that may increase the risk of serious adverse events. He investigated Bayesian factor regression models and explored extensions for binary medication and adverse-event data, along with preliminary simulation studies.

Together, these projects highlight the range of research being pursued by Statistics students and the role statistical methodology plays in addressing questions across disciplines. The students also used the summer to strengthen their research skills, refine their methods, and move their work toward future publications and continued study.