Supplementary Materials: Experimental Evaluation of ChatGPT for Test Case Generation and Quality Analysis through Test Smells in Software Testing Education

This dataset contains all supplementary materials from the experimental study "Evaluation of LLM Use in Test Case Generation from User Stories: An Experimental Study in Educational Context with Test Smell Analysis". The study investigated the educational application of Large Language Models (LLMs), specifically ChatGPT, for test case generation from user stories and subsequent quality analysis through test smells.The repository is organized into two main folders:1. Experiment ArtifactsPhase 1: Scanned responses from participants, distributed scripts, and activity guides for manual test case generation and ChatGPT-assisted generation.Phase 2: Scanned responses, identification and correction worksheets for test smells, and instructional materials used during the test smell analysis session.2. Results AnalysisCollected Data: Raw data files containing participant evaluations, test smell identifications, and questionnaire responses.Quantitative Analysis: Complete statistical analysis performed using Google Colab and Python, including scripts for data processing, visualization, and statistical tests.Qualitative Analysis: Thematic analysis of open-ended responses, participant feedback, and observational notes from the experimental sessions.Experimental Context:The study involved 25 undergraduate technology students across three consecutive sessions. Participants first generated test cases manually, then with ChatGPT assistance, and finally analyzed test smells using a structured catalog. The materials support transparency, reproducibility, and further research in LLM applications for software testing education.File Formats:Scanned PDF documents (participant responses)Python scripts (.py) and Jupyter notebooks (.ipynb)CSV data filesInstructional materials (PDF)Analysis reportsThese materials enable researchers and educators to replicate the study, examine the raw data, or adapt the experimental design for similar investigations in software engineering education.