Nandita Mitra

When students walk into my statistics classroom for the first time—especially those coming from clinical or public health backgrounds—they often carry a mix of apprehension and polite resignation. Many are convinced they’ll never use what they’re about to learn, that statistical formulas are for someone else’s job, and that their own strengths lie elsewhere. I get it. When material feels disconnected from students’ sense of purpose, even the most elegant concepts can fall flat.

I also believe that being open about our own learning journeys can be powerful. I often tell my students that I didn’t always picture a future in statistics. I started undergrad at Brown as a premed, dutifully enrolled in biology and chemistry, thinking I’d head to medical school. But somewhere along the way—probably somewhere between organic chemistry labs—I realized what I really loved was abstract problem-solving. I switched to pure math, completely enchanted by number theory and the beauty of logical proofs. But it wasn’t until I encountered statistics that everything clicked. Here was a way to take the structure and precision I loved and apply it to messy, urgent, real-world problems in population health. Statistics became the bridge between my love of theory and my growing interest in impact. That shift didn’t just shape my research—it transformed the way I teach. When students hear this, they begin to imagine their own, perhaps nonlinear, pathways, too.

This recognition led me to redesign my courses around case-based learning. When students engage with real-world scenarios early and often, even before they’ve learned the statistical tools, they begin to understand why those tools matter. They see themselves not just as recipients of knowledge, but also as emerging analysts, investigators, and advocates. My goal isn’t just to teach methods. It’s to help students answer questions they care about—and in doing so, to see themselves as agents of change.

Now I open the semester with a unit on soda taxes. We look at sales data from cities that implemented a beverage tax and ask whether it actually reduced sugar consumption. Then we dig into potential unintended consequences: Did low-income residents bear the brunt of the tax? Did local businesses suffer? Students wrestle with this messy, real-world policy question before they’ve even touched a regression model. So by the time we introduce the statistical tools, they already know why they matter.

Later in the semester, I draw on my experience as an expert witness in a racial discrimination case involving jury selection in North Carolina. I present students with simulated data and ask: Is there evidence of systematic bias in peremptory strikes? What are the statistical—and ethical—implications of such an analysis? This case gives us a platform to discuss sampling, causal inference, and the role of statistics in illuminating injustice. Even students who were initially skeptical of the field often leave that class realizing that numbers, when used thoughtfully, can be a powerful form of advocacy.

To further connect content with personal purpose, I ask students to design their own final projects. They choose a topic that matters to them—often something drawn from their work, clinical practice, or community—and apply the tools they’ve learned to investigate it. One student analyzed disparities in breast cancer screening rates across neighborhoods. Another evaluated the effectiveness of a local food access initiative. I provide structure and feedback throughout the process (commenting on their hypothesis of interest, the design of their study, and their statistical analysis plan, giving careful attention to the underlying statistical assumptions) but the ideas and questions are theirs. These student-driven projects become the capstone of the course, allowing them to create their own case studies that reflect both statistical rigor and personal meaning.

Of course, I’m far from the only one trying to make this connection between content and purpose. Faculty across Penn’s twelve schools are grappling with the same challenge: How do we help students see the relevance of foundational knowledge to their own goals? Whether it’s a philosophy course introducing ethics to future biomedical engineers or a history course taken by students heading into law, the tension between “what we teach” and “why it matters” is universal.

Teaching with purpose doesn’t mean every student will walk away loving statistics. But it does mean we invite them to care. When students care, they learn more deeply, ask better questions, and surprise us in the best ways. And for those of us teaching required courses that are often viewed as hurdles, this can feel like an uphill battle. But it’s a battle worth fighting. Because when students understand why they’re learning something, they start to see themselves differently. And in helping them make that connection, we sometimes rediscover our own purpose, too.

Nandita Mitra is a professor of biostatistics in the Perelman School of Medicine and co-director of the Penn Center for Causal Inference. She received the Lindback Award for Distinguished Teaching in 2025. 

--

This essay continues the series that began in the fall of 1994 as the joint creation of the College of Arts and Sciences, the Center for Teaching and Learning and the Lindback Society for Distinguished Teaching. 

See https://almanac.upenn.edu/talk-about-teaching-and-learning-archive for previous essays.