Jaya M. Satagopan, Rutgers School of Public Health
Every year, I prepare to teach biostatistics. Every year, I discover that I must also prepare to learn. The classroom is my laboratory, and teaching is my continuous experiment. Each semester provides new opportunities to observe, reflect on what works and what does not, and refine my approach to better support student learning. With each new semester, I return to the same questions: What inspired students? Where did they struggle? What should I do differently this time? These questions continue to shape my growth as an educator.
When I first began teaching, I introduced confidence intervals much the same way I had learned them decades earlier. I explained that a confidence interval for a population mean is the sample mean ± margin of error, discussed the formula for the margin of error, defined what a confidence interval represents, and worked through several examples. My students dutifully copied the formulas onto the cheat sheet they were allowed to bring to the exam and became proficient at the calculations. Yet many still could not explain what a confidence interval actually meant. They had procedural fluency—the “how”—but not the conceptual understanding—the “why.”
The confidence interval example is just one manifestation of the broader lesson I was beginning to learn. I began to notice that, across many topics in statistics, students could often reproduce formulas correctly without fully understanding the concepts behind them. Once I realized this, it reinforced the importance of emphasizing conceptual understanding alongside procedural fluency, and it continues to shape my teaching. My goal has become to help students not only understand the concepts underlying statistical formulas, but also recognize those concepts when interpreting research articles, news stories, and reports.
Moving from Delivering Content to Developing Understanding
Initially, my focus was on covering the syllabus, addressing all the competencies, delivering polished lectures, ensuring students had opportunities to ask questions, and clarifying their doubts. I used in-class exercises to confirm students could perform statistical calculations and write appropriate interpretations of their results. For a while, I thought this approach was sufficient. However, my understanding of what students were learning in the classroom began to evolve as my interactions with them extended beyond the classroom.
Over time, I occasionally ran into former students who told me how they were using concepts from the course in other classes, research projects, and capstone experiences. Some returned to discuss their projects or sought guidance as they applied statistical methods to real-world questions. These conversations were rewarding. But they also prompted me to reflect more deeply: Were my students simply applying the procedures they had learned, or were they developing the ability to think statistically? Was I emphasizing the “how” more than the “why” in my teaching? These questions gradually shifted my focus from ensuring students could correctly perform statistical procedures to helping them understand the reasoning behind those procedures and recognize when and why to apply them. I also began to think intentionally about what students would retain long after the semester ended and apply years later. My goal became not only to teach statistical methods, but also to spark curiosity and help students think statistically.
Learning from Students
Students have become one of my greatest teachers. Through their questions, struggles, and successes, they have helped me recognize the gaps between what I thought I was teaching and what they were actually learning. One example that reinforced this lesson involved teaching the two-sample t-test midway through the semester. I explained that calculating the standard error of the difference between two sample means depends upon whether the population variances can be assumed to be equal. By this point in the course, students use computer output to conduct hypothesis tests and write scientifically appropriate interpretations of the results. Although the course emphasizes R, I sometimes use SPSS output because many students from allied disciplines initially find it more intuitive. Many students quickly learned the decision rule: If Levene’s test for equality of variance was significant, use the second row of the output; otherwise, use the first. They became remarkably efficient at applying this rule. Yet, despite repeated reminders to read the entire output rather than focus on a single p-value, many could not explain why two rows existed or what assumptions distinguish them. This revealed a gap between what I intended students to understand about the two-sample t-test and what they had actually learned.
I changed my approach. Rather than emphasizing the statistical procedure, I decided to return students to the framework introduced in Week 1, when they learned how to formulate research questions and identify variables. I began asking students to start with the research question: What are we trying to learn? What variables are needed to answer the question? Which variable defines the two groups? What is being compared between the two groups? What assumptions are required before applying a statistical method? Only after working through these questions did we turn to the analysis and its output. I began applying this framework across statistical topics throughout the course, helping students see methods as tools for answering research questions rather than as isolated procedures.
Every year, I prepare to teach biostatistics. Every year, I discover that I must also prepare to learn.
Over time, conversations with students about their ongoing projects and capstone presentations became an important opportunity for me to learn whether the change worked. I noticed many students were becoming more comfortable explaining the rationale behind their analytical decisions and interpreting the results in the context of their research question. While this growth reflects the collective efforts of many instructors and mentors, the conversations reinforced my belief that helping students develop statistical thinking must be at the center of my teaching. Paying attention to how students reason helped me recognize whether my teaching was moving them toward statistical thinking.
Adapting to Today’s Students
Improving statistical instruction is not a one-time task. Every cohort of students brings different experiences, strengths, and challenges. Every cohort also teaches me something new. The only constant in teaching is change. The challenges and opportunities facing educators continue to evolve. We are entering an era of artificial intelligence, when students have unprecedented and rapid access to information from multiple sources. The challenge is no longer in helping students find information. It is in helping them evaluate information, interpret evidence, and use statistical thinking to make informed decisions. This requires sustained engagement, deeper learning, and continuous adaptation from both students and educators.
At the same time, data and quantitative information are becoming increasingly ubiquitous, bringing students with a wide range of preparation, experience, and confidence in statistical methods. Each generation of students arrives with its own strengths and challenges. Effective teaching requires adapting to these changes while maintaining high expectations and standards.
Good teaching adapts without lowering standards.
Building confidence is also an important part of good teaching. Many students approach statistics with uncertainty and anxiety, especially when they encounter unfamiliar quantitative concepts. By explaining why we use certain methods, creating opportunities for students to engage with real-world problems, and allowing them to see the relevance of statistics in practical scenarios, I aim to help students recognize that they are capable of thinking statistically.
The Experiment Continues
My own teaching continues to evolve. I have incorporated a flipped classroom approach and developed short instructional videos that students watch before class. In the classroom, I use newspaper articles, public reports, and peer-reviewed publications to help students connect statistical concepts with real-world questions. This approach allows me to spend more time helping students develop statistical thinking while building their confidence in interpreting statistical evidence and applying it to real-world decisions. I also emphasize critical evaluation of evidence, including vigilance about the quality and reliability of published findings. Through real-time activities, I encourage students to see statistics not as abstract methods, but as tools for understanding and addressing public health challenges. Over the years, my students have taught me that effective teaching requires staying curious, listening carefully, and continuously refining my approach. Teaching is never finished. After multiple years in the classroom, I no longer measure success by how well I taught, but by how much my students learned—and by how much I learned alongside them.
Leave a Reply