Scaling Technology Acceptance Analysis with Large Language Model (LLM) Annotation Systems

TL;DR

The paper addresses the cost and heterogeneity of traditional surveys in technology acceptance by introducing LLM annotation systems that convert ecologically valid online reviews into structured TAM/UTAUT data. It designs an end-to-end annotation pipeline with a TAM-focused prompt, and validates consistency across 50 runs and accuracy against human experts using WPA as the primary metric. Results show moderate-to-strong run-to-run consistency and high agreement between LLM annotations and human experts for key variables, with some variability in Social Influence and Facilitating Conditions; temperature tuning further improves stability. The study demonstrates that LLM-based annotation can be a scalable, cost-effective alternative for analyzing user attitudes toward technology, with practical implications for design insights and adoption research, while outlining important methodological and ethical considerations for broader deployment.

Abstract

Technology acceptance models effectively predict how users will adopt new technology products. Traditional surveys, often expensive and cumbersome, are commonly used for this assessment. As an alternative to surveys, we explore the use of large language models for annotating online user-generated content, like digital reviews and comments. Our research involved designing an LLM annotation system that transform reviews into structured data based on the Unified Theory of Acceptance and Use of Technology model. We conducted two studies to validate the consistency and accuracy of the annotations. Results showed moderate-to-strong consistency of LLM annotation systems, improving further by lowering the model temperature. LLM annotations achieved close agreement with human expert annotations and outperformed the agreement between experts for UTAUT variables. These results suggest that LLMs can be an effective tool for analyzing user sentiment, offering a practical alternative to traditional survey methods and enabling deeper insights into technology design and adoption.