Teaching and Learning Community of Practice (CoP)
Department of Economics
Summer(ish) 2024
Our Teaching and Learning Community of Practice (CoP) is our chance to talk teaching experiences and strategies with our colleagues over lunch. We typically meet over summer, however this year we are starting over Winter 2024.
The sessions are open to Economics department faculty members on either the UTM or downtown campus and sessional instructors on either campus. If you want to get on the mailing list please email Kripa Freitas (k.freitas@utoronto.ca).
Other CoP’s on campus: Arts & Science.
We meet in-person on the dates below in GE 106.
The sessions are 100% in-person, no recordings or virtual attendance options.
More events to be added over time. Check back for the latest schedule. If you have ideas or suggestions or want to colunteer to lead a session, send Kripa an email (k.freitas@utoronto.ca).
Monday April 15, 10:30-12noon (Coffee provided)*unusual time*
ChatGPT and Beyond: Ethical Approaches to AI in Economics Education
Discussion Leader: Marina Adshade (UBC)
Abstract: It is hard to imagine a technological innovation that has that has infiltrated our classrooms with the rapidity and ubiquity of ChatGPT. Within just weeks of its release, universities worldwide were contending with a critical question: “How do we prevent our students from using this tool?”
Eighteen months later, with the upcoming release of ChatGPT-5, it’s clear that this initial question was perhaps misguided, even short-sighted.
The better question might be: “How can we adapt our learning approaches to incorporate this beneficial technology while simultaneously upholding academic integrity?”
The first half of this seminar will explore how AI technology can serve as a multifaceted educational tool, aiding in the development of essential economic competencies, including analytical thinking, software proficiency, quantitative analysis, research acumen, and communication.
An interactive demonstration will provide a view of AI’s utility in constructing arguments, analyzing and visualizing economic data, synthesizing existing economic research, writing code for statistical analysis, and shaping policy recommendations.
The second half of this seminar acknowledges that while AI technology presents a substantial array of benefits for university teaching, these do not come without substantial vigilance costs to educators striving to maintain academic integrity.
Here will discuss insights gleaned from challenge undertaken last summer at the Vancouver School of Economics: “Faculty vs. AI: Is your assessment unbeatable?” Here, we’ll discuss best practices for preventing the unauthorized use of AI in student work and strategies for detecting academic integrity infractions.
I look forward to joining you as we navigate the complexities of AI in education, fostering a dialogue on its responsible use to enhance learning and uphold the standards of academic integrity.
Monday April 15, 3:30-4.30pm *unusual time*
COMET: Learning with Open-Source Empirical Research Tools
Discussion Leader: Marina Adshade (UBC)
Abstract: Traditional classroom environments often fall short in equipping students with the necessary skills to undertake full-scale empirical research projects. Recognizing this critical gap, we have developed an innovative pedagogical toolkit that transforms how students develop the skills necessary to undertake real-world applications of economic theory. This seminar presents an open-source collection of Jupyter Notebooks designed to enable self-directed learning through interactive and hands-on engagement with real-world economic data and models. These Jupyter Notebooks are available for learners of both Stata and R and are suitable for use in research-based courses at the senior undergraduate level, as well as in preparation for graduate school.
These toolkits are part of a larger project that integrates practical, hands-on learning at each level of the economics curriculum to provide students with the opportunity to improve their technical data skills. These modules focus on economic questions, models, and data, using the interactive notebook software, Jupyter Notebooks, to synthesize theoretical learning, practice, and analysis into a single learning experience that is appropriate for laboratory, flipped classroom, or self-directed learning environments.
Central to our mission is the democratization of education, achieved by minimizing the financial and technical barriers to learning. As such, we have made this library of resources available to instructors and students at any institution. You might consider reading our Using COMET for Teaching page.
I welcome the opportunity to discuss the design philosophy behind these toolkits and to high-light strategies for educators who might wish to integrate them into their empirical research courses.
We welcome any feedback on how our project might be more accessible. This can be done by submitting an issue to our GitHub directory.
Tuesday February 20, 12:00-1pm
Supporting Students in Data Courses
Discussion Leader: Michael Stepner
Abstract: Demand for data analytics skills is high and growing among our students and among the employers and academic programs they are pursuing after graduating. Meanwhile, the technological context is changing rapidly:
programming language preferences are shifting, version control and reproducibility are increasingly emphasized in research and industry, generative AI is in the hands of every programmer, and more and more students can’t locate a file within their computer’s directories. Every professor teaching an economics course with data analytics currently wages a solo effort to get their students through the mechanics of programming—how to install software, load data and run code—before we can focus on the main ideas of the course. Yet many students in 300-level and 400-level classes still arrive at the start of the semester struggling with the mechanics of getting their code running.
In this talk, I’ll present the software environment I developed to teach students empirical data analytics skills in ECO481 this fall semester.
Students programmed in Python, R or Stata to complete two assignments and write two empirical papers. They wrote and ran their code in a web browser using Github Codespaces—which eliminated issues with students having software problems on their personal laptops and facilitated assistance from the professor, TA and fellow students in debugging code in a single shared environment. Students submitted their code via Github, and Github ran their code in the cloud after each submission—which allowed students to get instant feedback on their work and iteratively debug their errors, while teaching them how to create reproducible research. At the end of the term, 75% of the students in ECO481 submitted code that ran cleanly from top to bottom and produced the empirical results for their final essay.
I’ll conclude by proposing that we make this software environment available in other economics courses in the Data Analytics sequence. I’d like to work with professors that are interested in using this toolkit in their courses, providing assistance with setting up assignments and troubleshooting issues that arise during the semester. Ultimately, I believe that a common toolkit will make life easier for us and our students—helping professors spend more teaching time on economics, while helping students produce their data analysis code more reliably and spend more time learning economics.