The Natural Language Processing Lab at George Mason University is conducting a research study to investigate how non-expert programmers can benefit from interacting with state-of-the-art large language models (LLMs), such as ChatGPT, in programming tasks. As a participant, you will be asked to complete coding tasks in SQL and Python by interacting with an LLM-based system that we have developed. You do not need expertise in any of the two programming languages; instead, you will only “converse” with our system in natural language. The research study will be conducted in person at the Fairfax campus of GMU.
To participate, you should:
- Be at least 18 years old.
- Understand basic concepts in database querying, such as:
- tables, rows, filtering and sorting, etc.
- Understand the basic mathematics and data structure concepts, such as:
- strings, arrays, maximum/minimum, squares, etc.
- No professional programming experience is necessary.
As compensation, all participants will be paid a $50 Amazon gift card. Study sessions will be around 2 hours.
Interested? Please contact hyan5@gmu.edu about your availability.
About
Toward Understandability and Interpretability for Neural Language Models of Source Code is a research study being conducted at George Mason University. Your identity will be kept confidential, and we will only include your answers in the research data if you consent to participate. The study is being conducted by Hao Yan, Prof. Ziyu Yao, and Prof. Thomas LaToza in the Department of Computer Science at George Mason University. Contact hyan5@gmu.edu for more details. The IRBNet number for this study is 2096275-1.