Llms, Virtual Users, and Bias: Predicting Any Survey Question Without Human Data

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Sinacola, Enzo, Pachot, Arnault, Petit, Thierry
Format: Preprint
Published: 2025
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866915206705184768
author Sinacola, Enzo
Pachot, Arnault
Petit, Thierry
author_facet Sinacola, Enzo
Pachot, Arnault
Petit, Thierry
contents Large Language Models (LLMs) offer a promising alternative to traditional survey methods, potentially enhancing efficiency and reducing costs. In this study, we use LLMs to create virtual populations that answer survey questions, enabling us to predict outcomes comparable to human responses. We evaluate several LLMs-including GPT-4o, GPT-3.5, Claude 3.5-Sonnet, and versions of the Llama and Mistral models-comparing their performance to that of a traditional Random Forests algorithm using demographic data from the World Values Survey (WVS). LLMs demonstrate competitive performance overall, with the significant advantage of requiring no additional training data. However, they exhibit biases when predicting responses for certain religious and population groups, underperforming in these areas. On the other hand, Random Forests demonstrate stronger performance than LLMs when trained with sufficient data. We observe that removing censorship mechanisms from LLMs significantly improves predictive accuracy, particularly for underrepresented demographic segments where censored models struggle. These findings highlight the importance of addressing biases and reconsidering censorship approaches in LLMs to enhance their reliability and fairness in public opinion research.
format Preprint
id arxiv_https___arxiv_org_abs_2503_16498
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle Llms, Virtual Users, and Bias: Predicting Any Survey Question Without Human Data
Sinacola, Enzo
Pachot, Arnault
Petit, Thierry
Human-Computer Interaction
Artificial Intelligence
Computation and Language
Computers and Society
Machine Learning
Large Language Models (LLMs) offer a promising alternative to traditional survey methods, potentially enhancing efficiency and reducing costs. In this study, we use LLMs to create virtual populations that answer survey questions, enabling us to predict outcomes comparable to human responses. We evaluate several LLMs-including GPT-4o, GPT-3.5, Claude 3.5-Sonnet, and versions of the Llama and Mistral models-comparing their performance to that of a traditional Random Forests algorithm using demographic data from the World Values Survey (WVS). LLMs demonstrate competitive performance overall, with the significant advantage of requiring no additional training data. However, they exhibit biases when predicting responses for certain religious and population groups, underperforming in these areas. On the other hand, Random Forests demonstrate stronger performance than LLMs when trained with sufficient data. We observe that removing censorship mechanisms from LLMs significantly improves predictive accuracy, particularly for underrepresented demographic segments where censored models struggle. These findings highlight the importance of addressing biases and reconsidering censorship approaches in LLMs to enhance their reliability and fairness in public opinion research.
title Llms, Virtual Users, and Bias: Predicting Any Survey Question Without Human Data
topic Human-Computer Interaction
Artificial Intelligence
Computation and Language
Computers and Society
Machine Learning
url https://arxiv.org/abs/2503.16498