Comments

Lei, Z., Zakharov, W., & Lu, Y. (2024). Leveraging ChatGPT for qualitative data analysis: A case study on data management practices among computer vision scholars. Presented at the 2024 Teaching and Learning with AI conference, Orlando, FL.

Abstract

Qualitative data analysis plays a crucial role in deriving meaningful insights from research data. However, conventional software tools like NVivo present challenges such as high costs and complexity (Dalkin, et al., 2021). This study advocates for integrating ChatGPT, an AI technology, into qualitative data analysis workflows to overcome these challenges. Focusing on the data management practices of Computer Vision professors, the study investigates how ChatGPT enhances human analysis by streamlining processes and uncovering hidden patterns within datasets. Structured interviews were conducted with six participants from research institutions (R1). The transcripts underwent manual scrutiny to identify recurring themes and patterns. Subsequently, the results were compared with ChatGPT analysis to evaluate its efficacy in qualitative data analysis. The findings illustrate the effectiveness of ChatGPT in augmenting traditional qualitative data analysis methods. By leveraging AI capabilities, ChatGPT facilitates a more efficient and comprehensive analysis, enabling researchers to uncover nuanced insights that may have been overlooked through manual analysis alone. This case study contributes to the ongoing discourse on AI's role in research, demonstrating how ChatGPT can enhance qualitative data analysis and drive advancements in academic research methodologies. The study also revealed certain limitations of AI as an analysis tool, such as potential inaccuracies, biases, and as well as ethical concerns. Therefore, while AI aids in analysis, manual intervention remains crucial to ensure accuracy and comprehensiveness in research methodologies.

Date of this Version

8-23-2024

Download

Share

COinS