A New Horizon in Mixed-Method Studies: Chatgpt-Assisted Phenomenological Analysis

Background: Phenomenological approaches emphasize a detailed exploration of lived experiences from a first-person perspective. Despite the inherent complexities in traditional phenomenological techniques like epoché and reduction, direct descriptive analyses can address the 'how' of experience. Objective: This study aimed to leverage ChatGPT, renowned for its descriptive generation proficiency, to perform a phenomenological analysis. Methodology: Our research followed four key stages: 1) Preparation of Phenomenological Data, where transcriptions were refined to align with the research question; 2) Individual Analysis, where ChatGPT highlighted experiential nuances from each transcript; 3) Global Analysis, synthesizing insights from individual narratives temporally and transversally; and 4) Structure of the Experience, which synthesized the elemental components of shared experiences. Custom prompts, tailored for each stage, ensured alignment and precision in capturing the experience dimensions. Results: ChatGPT showcased a sophisticated understanding of human experiences, demonstrating an ability to organize themes reflecting the intensity of sensations and variations in empathetic encounters. The tool's proficiency in thematic organization provided a phenomenologically-grounded understanding of data, highlighting how individuals engage with and are affected by experiential stimuli. Discussion: Our findings spotlight ChatGPT's potential in qualitative research, transforming raw input into rich phenomenological accounts. ChatGPT offers a blend of precision and scalability, making it a compelling tool for researchers delving into the intricacies of human experiences. Further research is needed to explore the full potential of ChatGPT in conducting phenomenological analyses and to solidify the methodological framework, ensuring it robustly addresses the nuances and depth of phenomenological inquiry.