LOGOS: LLM-driven End-to-End Grounded Theory Development and Schema Induction for Qualitative Research

2025-09-29

AI Summary

Grounded theory (GT) in qualitative research suffers from poor scalability due to its reliance on expert-intensive manual coding, and existing computational tools fail to achieve true automation. This paper introduces the first end-to-end automated GT workflow, integrating large language model–driven initial coding, semantic clustering, graph-structured modeling, and iterative refinement to generate hierarchical theoretical frameworks autonomously. We propose a novel five-dimensional evaluation metric and a reusable codebook construction paradigm. Evaluated across five heterogeneous corpora, our method significantly outperforms strong baselines; on complex datasets, the generated theories achieve 88.2% alignment with expert-derived coding patterns. The approach substantially enhances analytical efficiency, reproducibility, and cross-domain applicability while preserving methodological rigor.

Technology Category

• Knowledge Representation and Reasoning: Qualitative Reasoning
• Natural Language Processing: Code Generation / Program Synthesis from Natural Language
• Machine Learning: Deep Generative Models & Autoencoders

Application Category

• Web Mining and Content Analysis: Robustness and generalizability of Web mining methods
• Search and Retrieval-Augmented AI: Web evaluation methodologies and metrics
• Economics, Online Markets and Human Computation: Trust and reliance of crowd workers and data experts on GenAI

Abstract

Grounded theory offers deep insights from qualitative data, but its reliance on expert-intensive manual coding presents a major scalability bottleneck. Current computational tools stop short of true automation, keeping researchers firmly in the loop. We introduce LOGOS, a novel, end-to-end framework that fully automates the grounded theory workflow, transforming raw text into a structured, hierarchical theory. LOGOS integrates LLM-driven coding, semantic clustering, graph reasoning, and a novel iterative refinement process to build highly reusable codebooks. To ensure fair comparison, we also introduce a principled 5-dimensional metric and a train-test split protocol for standardized, unbiased evaluation. Across five diverse corpora, LOGOS consistently outperforms strong baselines and achieves a remarkable $88.2%$ alignment with an expert-developed schema on a complex dataset. LOGOS demonstrates a powerful new path to democratize and scale qualitative research without sacrificing theoretical nuance.

Problem

• Automating grounded theory development from qualitative data
• Overcoming scalability bottlenecks in manual coding processes
• Generating structured hierarchical theories through computational methods

Innovation

• LLM-driven coding automates grounded theory workflow
• Semantic clustering and graph reasoning build hierarchical theories
• Iterative refinement process generates reusable codebooks