A technological construction of society: Comparing GPT‐4 and human respondents for occupational evaluation in the UK

Despite initial research about the biases and perceptions of large language models (LLMs), we lack evidence on how LLMs evaluate occupations, especially in comparison to human evaluators. In this paper, we present a systematic comparison of occupational evaluations by GPT‐4 with those from an in‐dep...

Full description

Saved in:
Bibliographic Details
Published in:British journal of industrial relations Vol. 63; no. 1; pp. 180 - 208
Main Authors: Gmyrek, Paweł, Lutz, Christoph, Newlands, Gemma
Format: Journal Article
Language:English
Published: London Blackwell Publishing Ltd 01.03.2025
Subjects:
Chatbots
Humans
Language modeling
Large language models
Occupational prestige
Occupations
Prestige
Program evaluation
Respondents
Sociological research
Stigma
United Kingdom > UK
ISSN:0007-1080, 1467-8543
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:Despite initial research about the biases and perceptions of large language models (LLMs), we lack evidence on how LLMs evaluate occupations, especially in comparison to human evaluators. In this paper, we present a systematic comparison of occupational evaluations by GPT‐4 with those from an in‐depth, high‐quality and recent human respondents survey in the UK. Covering the full ISCO‐08 occupational landscape, with 580 occupations and two distinct metrics (prestige and social value), our findings indicate that GPT‐4 and human scores are highly correlated across all ISCO‐08 major groups. At the same time, GPT‐4 substantially under‐ or overestimates the occupational prestige and social value of many occupations, particularly for emerging digital and stigmatized or illicit occupations. Our analyses show both the potential and risk of using LLM‐generated data for sociological and occupational research. We also discuss the policy implications of our findings for the integration of LLM tools into the world of work.
Bibliography:ObjectType-Article-1
SourceType-Scholarly Journals-1
ObjectType-Feature-2
content type line 14
ISSN:0007-1080
1467-8543
DOI:10.1111/bjir.12840