Teaching Statement

Teaching Experience

Training
As a member Certificate in College Teaching Program at Duke, I have participated in teaching observation and mutual feedback through the Teaching Triangles Program. I also attended a course in the Fundamentals of College Teaching (GS750) during the fall of 2020, with emphasis on learning objectives, modern pedagogy, and navigating the virtual classroom during and beyond the COVID-19 pandemic.

Courses Taught
I have worked as a Teaching Assistant for four courses, focusing on the use of programming and statistics to analyze ecological and environmental data.

Environmental Change in the Big-Data Era (2021, online) is a first-year undergraduate seminar that introduces students to data sourcing, strengths and limitations, and data interpretation through readings and discussions of scientific literature and data exploration. The course also introduces basic concepts in R software applied to climate change, human impacts, and biodiversity loss.
Bayesian Inference and Environmental Models (2021, online) introduces graduate students to Bayesian statistical theory and computational techniques through contemporary environmental examples.
Applied Data Analysis for Environmental and Physical Sciences (2019 in-person, 2020 online) teaches applied statistics and programming in R to Master of Environmental Management (MEM) students, from exploratory data analysis to multilevel and generalized linear modeling, cumulating in a final group project exploring the students’ own data and ecological or environmental questions. My duties included teaching weekly labs and holding four office hours a week to advise students on weekly labs and their final projects. In an effort to adjust the course to the pandemic restrictions of fall 2020, I initiated and authored a series of anonymous surveys to gauge classroom climate and student stress levels during the semester. Several students even reached out to thank me for the surveys and request more. The teaching team was then able to flexibly adjust the course as COVID and election pressures mounted.
Environmental Decision Analysis (2020, in-person) teaches students how to assess decision-making factors on an industrial or governmental scale, including weather events, financial obligations, environmental regulations, and data limitations. My duties included demonstrating decision analysis packages offered in R, grading exams, and holding two office hours a week to help students work through homework case studies.

Online Tutorials
Many of my students, in joining the MEM program at Duke, had expressed that they felt unprepared for the steep curve of learning R programming and statistics at the same time. In preparation for teaching ENV710 again in the fall of 2020, I developed an open-source series of introductory modules, allowing students to learn the basics of R at their own pace. Many students have cited my modules as instrumental to their success in a tough course, and two have asked my permission to share my work with other struggling MEM students not taking ENV710.

Guest Lectures
I have prepared a guest lecture for Duke’s spring 2021 Data Expeditions, a program which introduces undergraduate students to real data exploration and analysis. Entitled “Do hurricanes affect bird biodiversity?”, my lecture walks students through the process of obtaining, cleaning, exploring, and presenting large ecological datasets, using hurricane events and eBird citizen science data to explore the titular question.

teaching philosophy

Maggie was excellent. She was very helpful in complementing what Dr. Poulsen had outlined in his labs and showing us additional methods to certain codes. She was very good at deciphering when to help students versus pushing them just enough to figure things out on their own.

I often meet students who have promising thoughts and ideas, but who see themselves as not being “good at” the computational side of science. I encourage and support my students so that they become more self-confident in their ability to learn programming and statistics. My goals are to (1) give my students a data analysis toolkit which they can apply beyond an introductory course to their own research, and (2) to support my students and encourage their ability to learn the “harder” skills of programming and statistics.

I have had many students tell me that, were it not for my gentle encouragement and insistence in their own ability to learn the material, they would have been too overwhelmed to gain anything from the course. It is not enough to provide students with digestible materials; I must also improve students’ confidence in their own abilities.

Teaching is an iterative process of constant improvement. Each semester and course are an opportunity to build upon my current pedagogical methods. For example, during the first fully virtual semester (fall 2020) at Duke, I leaned on regular, anonymous “course climate” surveys to gauge my students’ stress levels, motivation, and comprehension of the material. These surveys were instrumental in flexibly teaching during a wildly unusual fall semester: I varied office hours; sent out lab notes and scripts a few days in advance; adjusted lab texts to better explain statistical models; and worked to pare down coursework to alleviate stress.

None of this would have been possible without real-time student feedback and a willingness to learn from my students just as they learn from me. I intend to implement regular feedback in all future courses that I teach.

learning objectives

Although each course I have taught or will teach has different expectations, objectives, and materials, three learning objectives remain constant.

First, my students should develop a good anticipatory mindset. At its core, programming requires a flexible and creative thought process, and students should know how to approach programming from data collection through analysis. I don’t expect expertise, but knowing what can be done—and puzzling through or researching how to do it—is a powerful tool.

Second, students should gain a proficiency in troubleshooting and problem solving. Self-reliance is key. They should be able to address errors and problems with readying data for analysis, as well as the basics of code diagnosis; without these, programming becomes incredibly difficult for beginners.
Finally, I want my students to know how to ask good questions. Much of data analysis and exploration involves curiosity and creative “what-if” thinking. Students should be able to hone the questions they’re asking of their data and explain succinctly what they're learning and why it’s important.

Maggie does an incredible job explaining the labs and the concepts we cover in the labs. I learned so much about the details of coding from Maggie as well. She is super patient and goes above and beyond to help students with plenty of office hours and help sessions.