Ethnographic Methodologies Research

Ethnography Assistant Overview

Ethnographic methods, particularly autoethnography, offer insights into lived experiences but demand rigorous self-reflection and narrative analysis. While conducting autoethnographic research, the investigators conceived of leveraging Large Language Models (LLMs) not just for analysis, but as interactive partners in the inquiry process. This project addresses the challenge of facilitating deep reflection and managing complex narratives within ethnographic work. It aims to develop and evaluate an AI Ethnography Assistant, designed to act as an intelligent interviewer, guiding researchers through structured self-reflection and preliminary analysis, thereby exploring novel methodologies for technology-assisted qualitative research.

Goals

The primary goals of this research are:
  1. Develop and pilot test an LLM-based "Ethnography Assistant" tool capable of conducting adaptive, reflective interviews for (auto)ethnographic research.
  2. Develop AI-driven methodologies to assist researchers in gathering “twice told stories” through guided reflective storytelling and conducting interviews, identifying narrative structures, recurring themes, and instances of evolving perspectives across interview sessions.
  3. Explore and document the ethical considerations of using AI interviewing tools in ethnographic research, culminating in a dedicated ethical framework.

Outcomes

Expected outcomes of the project include:
  • A functional prototype of the Ethnography Assistant.
  • Methodological guidelines for employing LLMs in interactive ethnographic inquiry.
  • An ethical framework addressing responsible AI use in this context,
  • Dissemination of findings through conference presentations and potentially a journal article.

Research Team

  • Dr. Leo C. Ureel II
  • Dr. Michelle Jarvie-Eggart
  • Dr. Lynn Albers (Hofstra University)
  • Laura Albrant

Demo

  • Ethnography Assistant Demo: Explore elder care from a faculty perspective using our ethnography assistant. Launch Demo