Dataset to aid educators in using Generative AI in software design
Data files
Dec 16, 2025 version files 392.27 KB
-
Design_Studio_2.pdf
48.58 KB
-
Prompt_Training.pdf
341.86 KB
-
README.md
1.83 KB
Abstract
This dataset is part of a study that reports on the experience of introducing GenAI into an undergraduate software design class. Students were required to use GenAI (in the form of ChatGPT) to help complete a team-based assignment in which they were tasked with designing the core of the solution for how an educational traffic simulator could keep track of the elements in the simulation, as well as how to advance the simulation over time. The researchers collected and analyzed both the ChatGPT conversation logs and the teams' reflections on their use of ChatGPT for the assignment. The report describes how the teams used ChatGPT, reflects on how they felt ChatGPT helped them and where it had shortcomings, and identifies several key lessons for educators on how to deploy GenAI in a software design class.
This dataset supports the study "Using Generative AI in Software Design Education" recently accepted into the 2026 Designing Workshop, held alongside ICSE in Rio de Janiero, April 2026.
https://doi.org/10.5061/dryad.rr4xgxdjg
Description of the data and file structure
This data was prepared as part of the course instruction for the course described in the related paper, "Using Generative AI in Software Design Education". It is provided for other researchers interested in conducting a similar study or for educators interested in taking a similar approach in their software design courses.
Files and variables
File: Prompt_Training.pdf
Description: This contains the presentation given to the students on how to use ChatGPT to support the generation of a UML model and pseudocode. The presentation and a copy of the prompts used in the training are included within the file.
File: Design_Studio_2.pdf
Description: This is the assignment where students were required to use GenAI for support in their design process. The file contains full details of the assignment, including the design problem and the required deliverables.
Code/software
The two files are both PDF files so a PDF reader such as Adobe Acrobat Reader is required to access the files.
Sharing
The dataset is licensed under a CC0 1.0 Universal (CC0 1.0) Public Domain Dedication license.
Access information
Other publicly accessible locations of the data:
- None
Data was derived from the following sources:
- The design prompt for the student assignment is the same one used in N. Mangano, T. D. LaToza, M. Petre and A. van der Hoek, "How Software Designers Interact with Sketches at the Whiteboard," in IEEE Transactions on Software Engineering, vol. 41, no. 2, pp. 135-156, 1 Feb. 2015, doi: 10.1109/TSE.2014.2362924
Data Preparation
The dataset consists of two files:
- The lecture material illustrating how ChatGPT, alongside PlantUML[1], can be used to help produce UML diagrams
- Their course assignment which asked them to use ChatGPT to help with their design.
This data is provided to support educators who may be considering a similar approach in the future.