UROP Project

Open-Ended Survey Response Analysis Using LLMs

LLM, AI, survey, machine learning, qualitative, data visualization
brian-wilcoxon-4x5-400.jpg
Research Mentor: Dr. Brian Wilcoxon,
Department, College, Affiliation: Office of the Provost, Graduate Student Resource Center, N/A
Contact Email: bwilcoxon@fsu.edu
Research Assistant Supervisor (if different from mentor): Hugh Long
Research Assistant Supervisor Email: hal20a@fsu.edu
Faculty Collaborators:
Faculty Collaborators Email:
Looking for Research Assistants: Yes
Number of Research Assistants: 1
Relevant Majors: Open to all majors
Project Location: On FSU Main Campus
Research Assistant Transportation Required:
Remote or In-person: In-person
Approximate Weekly Hours: 5, During business hours
Roundtable Times and Zoom Link:
  • Day: Tuesday, September 2
    Start Time: 2:00
    End Time: 3:00
    Zoom Link: https://fsu.zoom.us/j/95742015494
  • Day: Wednesday, September 3
    Start Time: 3:00
    End Time: 4:00
    Zoom Link: https://fsu.zoom.us/j/99198105935
  • Day: Thursday, September 4
    Start Time: 1:30
    End Time: 2:30
    Zoom Link: https://fsu.zoom.us/j/98743139935

Project Description

The Graduate Student Resource Center (GSRC)within the Office of the Provost is conducting quantitative and qualitative research studies on FSU student success. We help the broader university develop new data on graduate students and better understand what may impact graduation from the university, time to graduation, and satisfaction with their overall experience, as well as many questions within and beyond those topics. The GSRC team is currently tackling the following project, which you would assist with: the analysis of survey of current students that has open-ended questions and a large number of qualitative responses. Specific projects that you could pursue include:
- Under what conditions could using a large language model to analyze qualitative data be effective? What are advantages and disadvantages of using such a model? How can we best leverage these models in qualitative research?
- How can we improve upon existing work at FSU using LLMs to classify open ended survey data?? How could it be adjusted to be more effective?



Research Tasks: The student research assistant will join our team to help with the following possible tasks:

• Basic literature review and background reading.
• Review and provide suggestions about the current structure for coding qualitative survey responses
• Using the current FSU model and current qualitative codes, assist in checking the accuracy of the current model
• Planning emendations to the model for the Fall 2025 survey data
• Shadow an analyst running the model for the Fall 2025 survey data
• Drafting visualizations for reporting purposes


Skills that research assistant(s) may need: Required skills
- Self-motivated / can work independently with some guidance
- Communicates effectively, especially when there is a problem
- Interested in studying data analytics, survey methodology

Recommended skills
- Proficient in Microsoft Word, some exposure to Excel
- Exposure to academic journal databases
- Ability to effectively summarize information both verbally and in writing

Mentoring Philosophy

No matter your starting point, working on this project can help you develop research skills that are beneficial to success in a multitude of different academic environments. However, there are a few character traits that I consider universally needed for success in any field, and I aim to help any mentee develop these, both by demonstrating them in my own work, and in more active guidance. First, I want to encourage persistent curiosity and imagination in the way you approach and think through problems. Academic work is rarely about answering a single big question, but about learning how to break something into incremental problems and steps that build toward a solution. We will do this together, as a team, and you will be considered a vital partner in this process. As this work takes place, I want to encourage you to remain humble in knowing that there are always more questions to ask, but at the same time, confident enough to know that a dissatisfying answer is not an indictment of you as a person or thinker. Second, I want to encourage you to feel ok with working as part of a research team and asking simple questions. I do not have the answers, and I will not expect anyone else to. What I will expect is that we talk openly about the answers we are looking for, and explore the different ways we might find them together.

Additional Information

You will need to bring their own laptop to our office. You would sit in a small cube in a quiet space with immediate access to your research team.

Link to Publications


Return to Project Database