I have five years of interdisciplinary teaching experience spanning international politics, economics, data science and quantitative methods. I have designed and taught courses meeting the unique needs of varying degree levels, including undergraduate, masters, mid-careers and doctoral candidates. 

My teaching philosophy has evolved around my increasingly negative reaction to the quote “I can teach you, but I cannot learn for you.” Admittedly, I liked this idea when I first heard it. It stresses the primary importance of the individual student in learning, and I think it underscores how necessary it is that students buy into the course material and are willing to put in effort. While the core foundations here are sound, I have seen the same ideas lead instructors to wipe their hands of blame and shift responsibility back on to the students when there is a disconnect in learning.

This thinking ignores the powerful and transformative impact an instructor can have on a classroom’s learning environment. The best instructors that I strive to emulate create learning environments where students lean into the course content. I work to replicate this approach in my teaching by ensuring course concepts are deeply connected to real-world issues and by building a space where students feel comfortable participating, making mistakes, and sharing their own experiences. This requires designing lessons with a degree of fluidity that affords students opportune chances to shape the direction of discussions. 

I believe simulation and visualization are key to enhancing learning and intuition about statistical ideas. When learning statistics, we are limited if we rely exclusively on real-world data where the true parameters are unknown. By guiding students through exercises with simulated data, I present them scenarios where we know exactly what the “true” relationship between variables should be. Once we know this relationship in simulated date, it becomes much easier to see and appreciate how our estimates can be biased due to choice of models and control variables.

Teaching students how to simulate problems empowers them to explore new ideas inside and outside of classroom topics. During my teaching, I have created a series of R Shiny Apps to demonstrate key concepts. These apps allow students to change the sample size and other key data parameters and dynamically re-run statistical calculations and update visualizations.

R Shiny Apps for Statistical Instruction

Central Limit Theorem

Power Analysis - Graphical Intuition

Regression - Sampling Variation in Coefficient Estimates

Sample of results from the Power Analysis - Graphical Intuition simulation.