Teaching Philosophy

Science is about more than a series of facts and so teaching science must be about more than conveying facts. Science is a method for learning about the world, rooted in observation, critical thinking, and intentional inquiry. Teaching science, then, is about supporting this process. My primary goal as an educator is to cultivate students’ curiosity and critical reasoning skills. In achieving this goal, I aim to also cultivate students’ sense of identity and agency as a scientist.

Teaching experience

I have years of teaching experience, including as a TA and instructor, and through running informal workshops and lessons for graduate and undergraduate students.

Analytical Methods in Ecology, Evolution & Natural Resources

Instructor; 2023

An introduction to statistics for upper-level undergraduates in the Rutgers Department of Ecology, Evolution and Natural Resources (course # 11:216:369).

I prepared all the lectures and in-class activities, selected readings, and developed active learning modules in RMarkdown to teach students to use R while reinforcing fundamental concepts.

In this course, I cover the core concepts of frequentist statistics, including a treatment of the basic tools of statistical analysis, as well as key concepts in study design and data collection, management and visualization. Throughout, I keep focus on the logic underlying statistical methods, which is rooted in probability theory, and emphasized how we use the logic of probability to make inferences from data.

Fundamentals of Ecological and Environmental Modeling

Teaching assistant; 2021-2023

An introduction to applied mathematical models for upper-level undergraduates in the Rutgers Department of Ecology, Evolution and Natural Resources (course # 11:216:431).

I graded assignments, and provided tailored feedback to the class as a whole and individual students. For the class, I prepared mini (~ 5-15 min) lectures to follow up on each assignment, reinforcing key concepts and correcting common conceptual or procedural errors. I then met with individual students to provide personalized feedback and answer questions. Outside of class, I provided individual tutoring and ran study sessions.

Plant Diversity and Evolution and Plant Systematics

Teaching assistant; 2020-2022

A primer on the taxonomic and morphological diversity of plants for upper-level undergraduates in the Rutgers Department of Ecology, Evolution and Natural Resources (course # 11:216:412) and graduate students in the Ecology & Evolution program (course # 11:215:508).

I led lab classes, which focused on botanical literacy, plant identification, and placing plant morphological traits within the broader context of evolution.

During the COVID pandemic, I helped redesign the labs to be fully remote, leading the production of entirely new materials for remote, at-home labs. These labs were based around students finding plants in their own homes and neighborhoods—in their yards, their kitchens, even the cracks of their sidewalk or driveway! While running a remote lab was challenging, and students were initially pessimistic, we found that students really rose to the challenge, and ultimately were very successful and got very excited at discovering the botanical diversity right on their doorstep.

Back in the classroom, I used the lessons I learned from our remote classes to revise our in-person activities. In particular, more focus on describing plants and plant morphology, and grounding all further lessons in these observations. Students learn best by doing!

Advanced Ecological Data Analysis

Teaching assistant and guest instructor; 2020-2021

A graduate-level survey of statistical theory and techniques common in ecology, with a special focus on application and interpretation in R, for students in the Rutgers Ecology & Evolution Graduate Program (course # 16:215:599). Run with a “flipped classroom” design, in which students watch lectures and read materials before class, then have discussion and work through materials in class.

As a teaching assistant, I provided students one on one support, especially with R. As a guest instructor, I developed lectures and workshops on maximum likelihood and Bayesian methods for linear modeling. In the likelihood workshop, I covered fundamentals of likelihood, how it can be used to estimate model parameters, and how it can be used to compare the efficacy of competing models (i.e., multi-model inference, sensu Burnham and Anderson1). In the Bayesian workshop, I revisited the basics of probability to derive Bayes’ theorem, developed the concept of models as joint probabilities, then built on what we’d learned in the likelihood workshop to derive the likelihood statements of simple and hierarchical linear models, allowing us to code our own models in JAGS2.

Introductory Biology

Teaching assistant; 2016-2018

Exactly what it sounds like: Intro to Molecules, Cells, and Development (BIOL 203) and Intro to Organisms, Ecology and Evolution (BIOL 204), undergraduate courses at the College of William & Mary. For three semesters, I led labs, which ranged from field ecology to zebra fish ontogeny. During this time, I won the Outstanding Teaching Assistant award. For the fourth semester, I was assistant lab manager, which meant preparing reagents and other materials, setting up labs, and supporting the team of teaching assistants.

Statistics and coding workshops

Outside of formal teaching duties, I have developed and led a number of statistics and R coding workshops for both undergradate and graduate students. At William & Mary, I led introductory workshops on frequentist statistics, coding in R, and occupancy modeling for undergraduate members of the ACER Lab, and a lesson on spatial autocorrelation and spatial regression for a mixed grad-undergrad biostatistics class. At Rutgers, as part of the grad student-organized “Data Club,” I led workshops on hierarchical modeling in both maximum likelihood and Bayesian frameworks.

  1. Burnham and Anderson 2002. Model Selection and Multimodel Inference: A Practical Information-Theoretic Approach. Springer, New York. ↩

  2. If I were to teach this again, I am not sure if I would use JAGS or Stan. There a lot of ways that Stan is simpler—a Stan script is certain cleaner!—but there are also ways that JAGS code maps logically back to the mathematical likelihood statements. In particular, the for() loops in JAGS code are direct analogs to the multiplication ($\Pi$) symbols in the likelihood statement. I think this makes it easier to grasp, especially for folks that are newer to coding. Stan code is simpler and cleaner because Stan is vectorized, which makes things easier if you have the right experience, but is also one more thing to wrap you’re head around, if you’re new to stats and coding. ↩