Algorithmic Fidelity of Large Language Models in Generating Synthetic German Public Opinions: A Case Study

Fuente: arXiv
Gespeichert in:
Bibliographische Detailangaben
Hauptverfasser: Ma, Bolei, Yoztyurk, Berk, Haensch, Anna-Carolina, Wang, Xinpeng, Herklotz, Markus, Kreuter, Frauke, Plank, Barbara, Assenmacher, Matthias
Format: Preprint
Veröffentlicht: 2024
Schlagworte:
Online-Zugang:
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
_version_ 1866913869226573824
author Ma, Bolei
Yoztyurk, Berk
Haensch, Anna-Carolina
Wang, Xinpeng
Herklotz, Markus
Kreuter, Frauke
Plank, Barbara
Assenmacher, Matthias
author_facet Ma, Bolei
Yoztyurk, Berk
Haensch, Anna-Carolina
Wang, Xinpeng
Herklotz, Markus
Kreuter, Frauke
Plank, Barbara
Assenmacher, Matthias
contents In recent research, large language models (LLMs) have been increasingly used to investigate public opinions. This study investigates the algorithmic fidelity of LLMs, i.e., the ability to replicate the socio-cultural context and nuanced opinions of human participants. Using open-ended survey data from the German Longitudinal Election Studies (GLES), we prompt different LLMs to generate synthetic public opinions reflective of German subpopulations by incorporating demographic features into the persona prompts. Our results show that Llama performs better than other LLMs at representing subpopulations, particularly when there is lower opinion diversity within those groups. Our findings further reveal that the LLM performs better for supporters of left-leaning parties like The Greens and The Left compared to other parties, and matches the least with the right-party AfD. Additionally, the inclusion or exclusion of specific variables in the prompts can significantly impact the models' predictions. These findings underscore the importance of aligning LLMs to more effectively model diverse public opinions while minimizing political biases and enhancing robustness in representativeness.
format Preprint
id arxiv_https___arxiv_org_abs_2412_13169
institution arXiv
publishDate 2024
record_format arxiv
spellingShingle Algorithmic Fidelity of Large Language Models in Generating Synthetic German Public Opinions: A Case Study
Ma, Bolei
Yoztyurk, Berk
Haensch, Anna-Carolina
Wang, Xinpeng
Herklotz, Markus
Kreuter, Frauke
Plank, Barbara
Assenmacher, Matthias
Computation and Language
In recent research, large language models (LLMs) have been increasingly used to investigate public opinions. This study investigates the algorithmic fidelity of LLMs, i.e., the ability to replicate the socio-cultural context and nuanced opinions of human participants. Using open-ended survey data from the German Longitudinal Election Studies (GLES), we prompt different LLMs to generate synthetic public opinions reflective of German subpopulations by incorporating demographic features into the persona prompts. Our results show that Llama performs better than other LLMs at representing subpopulations, particularly when there is lower opinion diversity within those groups. Our findings further reveal that the LLM performs better for supporters of left-leaning parties like The Greens and The Left compared to other parties, and matches the least with the right-party AfD. Additionally, the inclusion or exclusion of specific variables in the prompts can significantly impact the models' predictions. These findings underscore the importance of aligning LLMs to more effectively model diverse public opinions while minimizing political biases and enhancing robustness in representativeness.
title Algorithmic Fidelity of Large Language Models in Generating Synthetic German Public Opinions: A Case Study
topic Computation and Language
url https://arxiv.org/abs/2412.13169