# Can ChatGPT Play the Role of a Teaching Assistant in an Introductory Programming Course?

**Authors:**  Anishka, Atharva Mehta, Nipun Gupta, Dhruv Kumar, Pankaj Jalote  
**arXiv:** [2312.07343](https://arxiv.org/abs/2312.07343)  
**Date:** 2023-12-12  
**Summary page:** https://www.summarizepaper.com/en/arxiv-id/2312.07343v1/

## Key points

- Large language models (LLMs) have the potential to revolutionize education
- ChatGPT, an LLM, is being explored as a virtual Teaching Assistant (TA) in an Introductory Programming Course
- ChatGPT's capabilities are evaluated by comparing its performance with human TAs in various TA functions
- Functions include solving programming assignments, grading student code submissions, and providing feedback to undergraduate students
- Analysis of ChatGPT's effectiveness in solving programming assignments considers code correctness and code quality
- Proficiency of ChatGPT in grading student code submissions using a given rubric is assessed and compared with human TAs' grades
- Quality and relevance of feedback provided by ChatGPT are analyzed, focusing on addressing mistakes and offering suggestions for improvement
- Implications of integrating ChatGPT into computing education include automated grading, personalized learning experiences, and instructional support.

## Summary

, , , , 

The emergence of Large language models (LLMs) has the potential to revolutionize education, and this study explores the use of ChatGPT, an LLM, as a virtual Teaching Assistant (TA) in an Introductory Programming Course. The researchers evaluate ChatGPT's capabilities by comparing its performance with that of human TAs in various TA functions. These functions include solving programming assignments, grading student code submissions, and providing feedback to undergraduate students. To assess ChatGPT's effectiveness in solving programming assignments, the researchers analyze how closely its solutions align with those submitted by students. This analysis goes beyond code correctness and also considers code quality. Additionally, they assess ChatGPT's proficiency in grading student code submissions using a given grading rubric and compare its performance with the grades assigned by human TAs. Furthermore, the quality and relevance of the feedback provided by ChatGPT are analyzed. The evaluation focuses on how well ChatGPT addresses mistakes and offers suggestions for improvement in student solutions from both code correctness and code quality perspectives. The study concludes with a discussion on the implications of integrating ChatGPT into computing education. It highlights potential benefits such as automated grading, personalized learning experiences, and instructional support. Overall, this research investigates the potential of using ChatGPT as a virtual TA in an Introductory Programming Course. By comparing its performance with that of human TAs in various TA functions, the researchers provide insights into how well ChatGPT can solve programming assignments, grade student code submissions, and provide relevant feedback to undergraduate students. The findings have significant implications for incorporating LLMs like ChatGPT into computing education to enhance teaching and learning experiences.