Skip to main content Skip to navigation

Data Dialogue with ChatGPT: Using Code Interpreter to Simulate and Analyse Experimental Data

Project Overview

The document examines the role of generative AI, notably ChatGPT and its Code Interpreter plugin, in enhancing educational experiences in higher education, with a focus on physics laboratory settings. It underscores AI's capabilities in aiding data analysis and interpretation during laboratory activities, while also stressing the critical need for precise prompts to optimize its functionality. Through an exploration of three distinct approaches to employing ChatGPT for an introductory physics lab activity, the study identifies both the advantages and challenges associated with the integration of AI into educational frameworks. Furthermore, it provides valuable insights into effective communication strategies to maximize the efficacy of AI tools in education. Overall, the findings suggest that while generative AI holds significant promise for supporting learning and improving educational outcomes, careful consideration of its implementation is essential to address potential hurdles and ensure a productive learning environment.

Key Applications

ChatGPT Code Interpreter for data generation and analysis

Context: Introductory physics laboratory for undergraduate students

Implementation: The lab activity, 'Spring Constant,' involved simulating data and performing analysis using ChatGPT's Code Interpreter through varied prompting strategies.

Outcomes: ChatGPT was able to generate plausible data and analyze it for laboratory tasks, providing insights into the capabilities of AI tools in educational settings.

Challenges: The quality of output was highly dependent on the specificity of prompts; novice users may struggle to identify errors and provide detailed prompts.

Implementation Barriers

Technical Barrier

The output quality of ChatGPT is highly dependent on the detail and specificity of the prompts provided, which may be challenging for novice users. Additionally, students may struggle to recognize misinformation or errors generated by AI, leading to potential misconceptions.

Proposed Solutions: Instructors should provide training sessions on effective prompting strategies and critical thinking skills to help students evaluate AI outputs. Educational institutions should also establish guidelines for the responsible use of AI tools and incorporate AI literacy into the curriculum.

Project Team

Andrew Low

Researcher

Z. Yasemin Kalender

Researcher

Contact Information

For information about the paper, please contact the authors.

Authors: Andrew Low, Z. Yasemin Kalender

Source Publication: View Original PaperLink opens in a new window

Project Contact: Dr. Jianhua Yang

LLM Model Version: gpt-4o-mini-2024-07-18

Analysis Provider: Openai

Let us know you agree to cookies