Much of a teacher’s work happens after class ends.
Samantha Keppler, NBD Bancorp assistant professor of technology and operations at the University of Michigan, examines if AI can help.
Samantha Keppler is the NBD Bancorp Assistant Professor of Technology and Operations at the University of Michigan Stephen M. Ross School of Business. Her expertise is in education operations, or the mechanics of teaching and learning. Her work focuses on teacher resourcefulness, or how teachers find and create resources in innovative and unexpected ways. Recently, her research focuses on technologies such as teacher crowdfunding platforms and generative AI tools. She earned her PhD in Industrial Engineering and Management Science from the McCormick School of Engineering and Applied Science at Northwestern University. Prior to pursuing her PhD, she was a public high school math teacher for three years in the Bronx, New York.
Teacher Experiences with New AI
Much of teachers’ work is done outside the classroom: preparing lessons, grading student work, and talking one-on-one to specific students or parents. This work is what occupies teachers late into the night and keeps them working throughout the weekends. This is true even for teachers in their 20th year. What worked last year may not work this year. Teaching work involves constant reworking and innovating on what has been done in the past.
With two of my colleagues, one from the University of Michigan and another from the University of California – Berkeley, I followed 24 teachers throughout the 2023-2024 school year to understand how generative AI—mostly ChatGPT—was (or was not) impacting their outside-of-the-classroom work. We conducted interviews, observations, and surveys throughout the year. What we find suggests that teachers may find it more helpful to use generative AI for input rather than for output.
Specifically, we observed that the teachers who felt more productive with generative AI used it for input on their teaching ideas and plans. For example, one teacher asked an AI tool, “How do I help my students understand how to add positive and negative numbers?” In response, she got an array of ideas, some of which she had already tried and thought of, but some of which she had not. She changed her lesson as a result, trying something new.
Teachers who didn’t feel more productive with generative AI were only using it for outputs they needed to create anyhow, like quizzes or worksheets. The AI tools available at that time were not that great at this, and often teachers had to edit whatever the AI produced—sometimes making it just as time consuming to use AI as to not use it.
What we find suggests that generative AI might be more helpful to teachers as a source of advice and ideas than as a quiz or assignment generator.
Read More:
[SSRN] – Backwards Planning with Generative AI: Case Study Evidence from US K12 Teachers
Leave a Reply