Written by: UnconstrainED
The year is 2011. I am sifting through 90 hand-written exit tickets from my lesson on wavelengths, sorting them into piles based on exceeds, meets, and below mastery. I'm also looking for any major misconceptions that indicate the need for a potential re-teach. It's after 7 pm, and I'm just leaving school. My first-year teacher-self could never have dreamed that this entire task could be done in less than 5 minutes with the use of AI.
As teachers, we know that AI is here to stay and will play a large role in our students' future. Developing our AI literacy will support both our work in the classroom and our students' preparation for the future. AI literacy for educators means understanding the capabilities and limitations of AI, using AI tools effectively, and being able to evaluate AI-generated content. We also know that data is paramount. It's how we chart our courses forward, set goals, and celebrate achievements. If you've started school, I imagine you've already begun to collect data on your students–whether it be qualitative or quantitative. While supporting teachers, I've noticed that many are becoming well-versed in prompting Large Language Models (LLMs) like ChatGPT, Claude, or Gemini to aid in resource creation–which is fantastic! However– it's time for teachers to also add data analysis to their AI literacy. AI, particularly LLM's, can seamlessly analyze vast amounts of data, saving teachers time.
In this post, I'll share a few data analysis use cases and prompts that will empower you to use LLMs for your classroom data analysis. Before we dive in, I want to emphasize that data protection is key when using LLMs. It's crucial to ensure that you are sanitizing your data of any student information or identifiers and complying with all relevant regulations (e.g., GDPR, FERPA, COPPA) before sharing it with any LLM. An important part of AI literacy is understanding the ethical implications of using AI in education, including issues of data privacy.
Use Case 1: Large-Scale Data Analysis
LLMs can process large data sets–like your first summative. You can share your test data with an LLM like ChatGPT, which can then identify trends in your multiple-choice, short-answer, and open-ended question data quickly. As you are the expert, you can then use these trends to inform your instructional practices moving forward.
Prompt: "I want you to act as a data analyst and 5th grade ELA teacher. Attached is our summative test, answer key, and assessment data from our Unit on Wonderstruck, the novel. Analyze the data, identifying trends, both positive and negative, and create a list of key insights. Generate some instructional ideas based on the trends that I could implement to support my learners in the next unit."
Use Case 2: Qualitative Data Analysis
LLMs can easily process survey short-answer and open-ended survey data, drawing out major themes. Let's say you wrapped up your first unit and wanted student feedback to determine what worked well to help guide your next unit.
Example Prompt: "I want you to act as a data analyst and 10th-grade Biology teacher. Please analyze the attached student reflections on the engineering design unit and identify the aspects of the project they liked the most."
Use Case 3: Data-Focused Lesson Planning
We've all been there - you're still having the same misconception pop up even though you've reconfigured your lessons on this concept many times. You can share data on the misconception with an LLM so that it can suggest activities, resources, or even pacing based on the information. At a minimum, it might help you think of addressing the misconception in a new way that could lead to stronger results with your students.
Example Prompt: "I want you to act as a data analyst and 6th-grade math teacher. Please analyze the attached student's recent test scores and give me suggestions on how I should structure the next week of lessons on fractions."
Remember that as the teacher, you are the expert on data. AI is just a tool with limitations–in this case, a potentially very time-saving tool that's there to support and enhance your expertise, not replace it. By engaging with AI for data analysis, you not only save time but also enhance your AI literacy. As you use AI for data analysis, you will be able to determine its capabilities and limitations, and further develop your ability to create strong prompts.
Comentarios