Creation and Implementation of Two GenAI Activities

Abstract: In this case study, I reflect on the integration of GenAI activities into two fall 2023 ENG 101 classes and one ENG 102 class. My case study responds to the need for student-centered research related to students’ perceptions and use of GenAI at CSU. This type of teacher-research can enhance the student experiences and increase student success, and due to the proliferation and popularity of GenAI tools, this type of work can also be applied across colleges and disciplines. I discuss the creation and implementation of one formal assignment, an AI summary critique essay, and a fifty-minute “AI Exploration Day” activity. I also include students’ perceptions of the assignments and experience(s) with GenAI, gathered via an informal in-class survey, and reflect on ways to revise and refine each activity for future classes.

Introduction

Generative AI (GenAI), or “deep-learning models that can generate high-quality text, images, and other content based on the data they were trained on”(Martineau, 2023) has been the topic of much conversation in higher education over the last two years. For example, Mills, Bali, and Eaton (2023) write, “With the release of ChatGPT in November 2022, the field of higher education rapidly became aware that generative AI can complete or assist in many of the kinds of tasks traditionally used for assessment. This has come as a shock, on the heels of the shock of the pandemic. How should assessment practices change? Should we teach about generative AI or use it pedagogically?” As a First-Year Writing (FYW) instructor at Cleveland State University, I knew that I needed to address generative AI quickly. Additionally, I understood that GenAI would not be disappearing and in fact, I knew that current and future students will most likely be asked to use and engage with continuously changing forms of GenAI in their future college classes and careers. As such, it became even more imperative for me to consider how to integrate this tool in a way that would help prepare my students for these situations.

In spring of 2023, Dr. Shelley Rose, Interim Director of the Center for Faculty Excellence at Cleveland State University, created a Faculty Learning Community (FLC), “Artificial Intelligence in Teaching and Learning.” The FLC consisted of 30 interdisciplinary faculty members and four instructors as facilitators. I had the opportunity to serve as a facilitator, and my participation in this group inspired me to create and implement a last-minute GenAI reflection requirement for my ENG 100 students in spring 2023. The assignment asked students who used GenAI tools such as ChatGPT to include the prompts they used and write a discussion of how they cited and changed the generated content at the end of their final essay. Arguably, this last-minute reflection assignment demonstrates an attempt to help students understand how and when to use GenAI tools but adding it to the curriculum during the last assignment made it less successful. However, this experience encouraged me to explore ways to proactively integrate discussions of and assignments related to GenAI into my FYW classes, rather than simply reacting to AI use as situations arose.

The assignment I developed drew from common ideas in a variety of higher education blogs and articles, my experience as a facilitator in the FLC, and, later, my attendance at the Teaching and Learning with AI Conference at the University of Central Florida in September 2023. In this case study, I discuss the creation and implementation of one formal ENG 101 assignment (an AI-Generated Summary Critique Essay) and a fifty-minute “AI Exploration Day” activity presented to all of my FYW classes. I include students’ feelings towards the exploration activity, gathered via an informal in-class survey, and finally reflect on ways to revise and refine each activity for future classes.

The ENG 101 AI-Generated Summary Critique Essay

In fall 2023, I taught two sections of ENG 101: College Writing 1 and one section of ENG 102: College Writing II. ENG 101 focuses on preparing students to write for various college situations and asks them to practice various genres while ENG 102 focuses more on research and information literacy skills (“First-Year Writing”). I determined that it made sense for me to develop my first standalone AI assignment for my ENG 101 classes. 

The assignment was fairly simple. Students chose one of two articles provided (e.g., Donald Murray’s “Teach Writing as a Process Not Product” or Anne Lamott’s essay, “Shitty First Drafts”), asked ChatGPT to write a “brief summary” of the article, and then critiqued that summary’s accuracy, neutrality, and tone. Prior to completing the assignment students read about summary writing in our writing program’s free textbook, A Guide to Rhetoric, Genre, and Success in First-Year Writing and about the pros and cons of ChatGPT in the Wall Street Journal Opinion section’s “How do Students Feel about OpenAI’s ChatGPT?” and Naomi S. Baron’s “How ChatGPT Robs Students of Motivation to Think and Write for Themselves”. 

My students noticed that ChatGPT generally produced accurate and neutral summaries, although we discussed that this is not always the case, which supports the importance of fact-checking all generated content. The accuracy of the summaries was likely due to the fact that the articles I asked students to choose from are often assigned in many college writing classrooms across the country and “LLMs [like ChatGPT] are trained on vast bodies of preexisting text (such as content from the Internet), which, to some extent, predetermine their output” (MLA-CCCC Joint Task Force on Writing and AI, 2023, p.6).

We also discussed the style of ChatGPT’s summaries; many students found its language uncreative, overly academic, and boring.

This provided an important entryway into the next assignment which asked students to read and respond to, Sara P. Alvarez, Amy J. Wan, and Eunjeong Lee’s Writing Spaces article, “Workin’ Languages: Who We Are Matters in Our Writing ” which discusses the importance of language diversity in writing. Future iterations of this assignment could also ask students to connect these two assignments by reflecting on how generative AI tools like ChatGPT continue to perpetuate “[…]linguistic injustice because LLMs [like ChatGPT] promote an uncritical normative reproduction of standardized English usage that aligns with dominant racial and economic power structures” (MLA-CCCC Joint Task Force on Writing and AI, 2023, p.6).

The “AI Exploration Day” Lesson

While I created the AI Summary Critique assignment as part of my ENG 101 curriculum, I was inspired to let my students “play” with Generative AI in a safe space after returning from the Teaching and Learning with AI Conference. As such, I developed a 50-minute AI Exploration Day drawing from Minerva University’s panel “Exploring the Role of Generative AI in Active Learning” and Kevin Yee’s presentation “With Great Power Comes Great Responsibility: Examples of ChatGPT Infused Assignments in a Course on Superheroes.”

My lesson included a short presentation, an interactive group activity, and a reflective survey to measure students’ understanding of what they learned. The presentation (see Figure 1) focused on basic concepts such as AI literacy, when and when not to use AI, and the importanceof using it ethically and with integrity.

Figure 1. A screenshot of the AI Exploration Day Presentation

Following the brief lecture, students completed an interactive group activity (see Figure 2). They broke themselves into five groups and chose an AI tool that they were interested in (these had been added by me) or added their own choice of AI tool. They were asked to “play around” with the tool and answer a series of questions before sharing what they learned about the tool with the class (e.g., jigsaw style).

Figure 2. A screenshot of students completing the interactive group activity (names have been removed).

Lastly, students took a short survey asking them to reflect on their usage (or non-usage) of GenAI, understanding of ethical issues surrounding GenAI, and feelings towards the AI Exploration Day activities.

Overall, this lesson was successful. As part of the reflective survey, I asked, “How did you feel about today’s activities? Also, do you have any suggestions for how I could make it better? Or how I could integrate AI in a more effective way into my writing classes?”. Sixty students completed the survey and 49 students reported very positive feelings towards it while 7 indicated less enthusiasm (labeled “neutral” in Table 1), and four either did not respond or did not take the survey. Many students stated that it was “fun,” they “learned a lot,” and the lesson was “interesting.” One factor I noticed was that many students reported not knowing much (if anything) about GenAI prior to the activities. For example, one student wrote, “I think it was fun to be able to learn about different AI’s. Personally[sic] I’d never used/knew much of AI until this class!”

What I Have Learned

Over the course of the semester, I learned a lot about integrating GenAI into my writing classrooms. While I believe the AI Generative Summary Critique assignment was useful for students in that it allowed them to learn about ChatGPT, how it works, and what kind of texts it can produce well, I need to revise it for next semester. I want students to experience how ChatGPT (and other GenAI tools) often produce inaccurate and/or misleading content. I also plan to place this assignment later in the semester, perhaps as the second assignment, so that students gain more experience with summary writing and reading comprehension before critiquing content produced by AI.

I also plan to continue using the AI Exploration Day activity, though I will assign this during the first two weeks of the semester while my students are still learning about important course policies such as plagiarism. The discussions should also include conversations that help students understand exactly how GenAI tools work and help them understand that they are not “’doing’ or ‘intending’ or even ‘thinking,’ what we are witnessing is the production of word sequences that look like intentional human text through a process of statistical correlation” (MLA-CCCC Joint Task Force on Writing and AI, 2023, p.6). I also want to reach out to other colleagues on campus with more expertise in AI to perhaps visit my classes and help students learn about AI literacy.

Generative AI is not going anywhere, and some students will continue to use it in their classes, with or without instructor approval. Considering this changing education landscape, I chose (and continue to choose) to embrace the affordances of GenAI tools such as ChatGPT. Teaching students how to critically engage with these tools while also reflecting on when and if it is appropriate in various contexts will help prepare them for future courses. Further, this type of reflective work will help students understand why it is not appropriate to use GenAI in different learning contexts and classes; this understanding will help students succeed in and beyond their college classrooms.

References

Martineau, K.  (2023). What is generative AI?, https://research.ibm.com/blog/what-is-generative-AI

Mills, A, Bali, M., & Eaton, L. (2023). How do we respond to generative AI in education? Open educational practices give us a framework for an ongoing process. Journal of Applied Learning and Teaching, 6(1), https://doi.org/10.37074/jalt.2023.6.1.34

MLA-CCCC Joint Task Force on Writing and AI. (2023). MLA-CCCC joint task force on writing and AI working paper: Overview of the issues, statement of principles, and recommendations. https://hcommons.org/app/uploads/sites/1003160/2023/07/MLA-CCCC-Joint-Task-Force-on-Writing-and-AI-Working-Paper-1.pdf

Tambasco, L., Levitt, R., & Davis, D. (2023, September 24-25). Exploring the role of generative AI in active learning [Conference presentation]. Teaching and Learning with AI Conference, Orlando, Fl.

Yee, K. (2023, September 24-25). With great power comes great responsibility: Examples of ChatGPT infused assignments in a course on superheroes [Conference presentation]. Teaching and Learning with AI Conference, Orlando, Fl. https://fctl.ucf.edu/wp-content/uploads/sites/15/2023/09/Handout_ChatGPT_Assignments.pdf