Whose Story Gets Told? Positionality and Bias in LLM Summaries of Life Narratives

TL;DR

This paper investigates how Large Language Models interpret human life stories, revealing biases related to race and gender, and proposes a pipeline to identify and analyze these biases for ethical research use.

Contribution

It introduces a summarization pipeline to detect biases in LLM interpretations of life narratives, emphasizing ethical considerations and bias mitigation.

Findings

The pipeline can identify race and gender biases in LLM summaries.

Biases in perspective-taking can lead to representational harm.

The approach encourages bias analysis in LLM-based qualitative research.

Abstract

Increasingly, studies are exploring using Large Language Models (LLMs) for accelerated or scaled qualitative analysis of text data. While we can compare LLM accuracy against human labels directly for deductive coding, or labeling text, it is more challenging to judge the ethics and effectiveness of using LLMs in abstractive methods such as inductive thematic analysis. We collaborate with psychologists to study the abstractive claims LLMs make about human life stories, asking, how does using an LLM as an interpreter of meaning affect the conclusions and perspectives of a study? We propose a summarization-based pipeline for surfacing biases in perspective-taking an LLM might employ in interpreting these life stories. We demonstrate that our pipeline can identify both race and gender bias with the potential for representational harm. Finally, we encourage the use of this analysis in future studies involving LLM-based interpretation of study participants' written text or transcribed speech to characterize a positionality portrait for the study.