Instructor Spotlight: Hwanhee Hong, PhD

Hwanhee Hong, PhD, is an associate professor of  biostatistics and bioinformatics at Duke whose path from Seoul, South Korea, to Durham has been guided by both curiosity and compassion. With a love of numbers and a deep interest in improving human health, she has built a research career focused on developing statistical methods that bring clarity to complex biomedical data.  In the classroom, Hong is known for her warmth, enthusiasm, and genuine care for her students’ growth.

Please share some of your background and what led you to the field of biostatistics/bioinformatics?

I’ve always loved numbers — probably a little too much! As a kid, I found comfort in the logic of math problems, and that passion led me to major in statistics in college in Seoul, South Korea. When I started thinking about how to apply statistics in the real world, I debated between careers in finance or economics versus something in biomedical research. The latter won easily. I found the idea of using numbers to improve human health far more meaningful. That curiosity led me to pursue graduate studies in biostatistics, where I earned both my master’s and PhD in biostatistics, followed by postdoctoral research. Eventually, my journey brought me to Duke University, where I now serve as a faculty member.

What is your current research or professional focus?

My research focuses on developing and applying statistical methods for comparative effectiveness research, data integration, network meta-analysis, causal inference, generalizability, external controls, and Bayesian methods — basically, anything that helps make sense of complex biomedical data. I work primarily with clinical trials, observational studies, and EHR data, and I love tackling real-world clinical questions through data.
Recently, I’ve also been exploring how large language models (like ChatGPT) can support data science and research through advanced prompting methods. It’s an exciting frontier where statistics meets AI.

Which course do you teach in the program?

I teach Generalized Linear Models (BIOS 719) in the fall semester. It’s designed for second-year master’s students and covers the foundations of the exponential family, logistic and ordinal regression, Poisson regression, and generalized linear mixed-effects models. These are essential basic models that you should know for real-world biomedical data.

What do you most enjoy about teaching this course?

I love the interaction — that spark when a student asks a great question that makes the entire class pause and think deeper. I usually start by explaining a concept, but it’s the back-and-forth discussions that really make the class come alive. I also share real research anecdotes — small but meaningful stories about challenges I’ve faced, like debugging an impossible model or negotiating model assumptions with collaborators. These moments make abstract concepts tangible, and I hope they show students how statistical thinking plays out in the workplace.

What are the key skills or concepts you hope students will take away from it?

By the end of the course, I want students to feel confident using generalized linear models — the bread and butter of biomedical research. But beyond technical proficiency, I want them to think statistically : to ask good questions, choose models thoughtfully, and communicate findings clearly. It’s not just about p-values; it’s about connecting scientific questions to appropriate models, interpreting results responsibly, and explaining them in plain English to collaborators who may not be statisticians. In short, I want them to leave the course not just as analysts, but as effective scientific collaborators and leaders.

Are there particular projects, examples, or applications that students tend to find especially meaningful or exciting?

Yes! My class includes a group project using a real dataset — drawn from my own research (shared with collaborators’ approval, of course). Students start with the data, define research aims, analyze it, write a report, and present their findings. This hands-on experience is consistently the students’ favorite part. They get to see how messy real data can be, how decisions affect results, and how to communicate those results to a non-statistical audience. Watching them take ownership of their projects and confidently present their conclusions is one of the most rewarding parts of teaching.

What do you enjoy most about working with Duke MB students?

Duke MS students are some of the most driven, creative, and resilient people I’ve met. They’re not afraid to work hard or to try something new, whether it’s testing a new R package, debugging code at midnight, or proposing a clever workaround I hadn’t thought of. Their motivation and curiosity keep me learning too. Seeing that kind of energy every week makes collaboration genuinely fun.

How do you see students grow or change as they move through your course and the program?

It’s incredibly rewarding to see students evolve. At the start, questions tend to be very basic: “What’s the dimension of a specific matrix?” or “What theorem was applied in this proof?” But as the semester progresses, their questions become deeper: “Is this model appropriate for this question?” or “What assumption might be violated here?”
That’s when I know they’re thinking like statisticians. Similarly, in research projects, students often begin a bit uncertain but gradually take the lead — summarizing papers, running analyses, writing reports, and presenting confidently. Growth like that takes time — months, sometimes a year — but when it happens, it’s worth every minute. My main advice: be patient with yourself. Mastery takes time, but it will come.

What advice do you give students for getting the most out of your class?

First, be critical thinkers . Don’t just accept statistical claims at face value — especially those in news articles that make small effects sound dramatic! Ask questions, dig into assumptions, and check if the story matches the data.
Second, see yourself as a collaborator , not just a technician. You’re not “just cleaning data”—you’re a statistician contributing to scientific discovery. Build trust by being reliable, thoughtful, and communicative.
Biostatistics is a team sport, and good statisticians are not invisible — they’re indispensable. Take ownership of your work, and your collaborators will see you as an essential part of the research team.

What is something students might be surprised to learn about you?

I once performed at Carnegie Hall as part of the Choral Society of Durham! Standing on that legendary stage, singing under the lights, was an unforgettable experience.

What do you enjoy doing outside of teaching and research?

I sing in the Choral Society of Durham and love playing the piano. I also enjoy exploring new restaurants around town, as well as crafting and spending long hours building large LEGO sets.