<?xml version="1.0" encoding="utf-8"?>
<TEI xmlns="http://www.tei-c.org/ns/1.0" xmlns:xsi="http://www.w3.org/2001/XMLSchema-instance" xmlns:hal="http://hal.archives-ouvertes.fr/" xmlns:gml="http://www.opengis.net/gml/3.3/" xmlns:gmlce="http://www.opengis.net/gml/3.3/ce" version="1.1" xsi:schemaLocation="http://www.tei-c.org/ns/1.0 http://api.archives-ouvertes.fr/documents/aofr-sword.xsd">
  <teiHeader>
    <fileDesc>
      <titleStmt>
        <title>HAL TEI export of hal-04688498</title>
      </titleStmt>
      <publicationStmt>
        <distributor>CCSD</distributor>
        <availability status="restricted">
          <licence target="https://creativecommons.org/publicdomain/zero/1.0/">CC0 1.0 - Universal</licence>
        </availability>
        <date when="2026-05-13T04:50:52+02:00"/>
      </publicationStmt>
      <sourceDesc>
        <p part="N">HAL API Platform</p>
      </sourceDesc>
    </fileDesc>
  </teiHeader>
  <text>
    <body>
      <listBibl>
        <biblFull>
          <titleStmt>
            <title xml:lang="en">Can Large Language Models Accurately Predict Public Opinion? A Review</title>
            <author role="crp">
              <persName>
                <forename type="first">Arnault</forename>
                <surname>Pachot</surname>
              </persName>
              <email type="md5">cd254612ec967525025cf2bcde11867c</email>
              <email type="domain">pachot.fr</email>
              <idno type="idhal" notation="string">arnault-pachot</idno>
              <idno type="idhal" notation="numeric">1104041</idno>
              <idno type="halauthorid" notation="string">2242098-1104041</idno>
              <affiliation ref="#struct-1207100"/>
            </author>
            <author role="aut">
              <persName>
                <forename type="first">Thierry</forename>
                <surname>Petit</surname>
              </persName>
              <idno type="halauthorid">11290-0</idno>
              <affiliation ref="#struct-1207100"/>
            </author>
            <editor role="depositor">
              <persName>
                <forename>Arnault</forename>
                <surname>Pachot</surname>
              </persName>
              <email type="md5">cd254612ec967525025cf2bcde11867c</email>
              <email type="domain">pachot.fr</email>
            </editor>
          </titleStmt>
          <editionStmt>
            <edition n="v1" type="current">
              <date type="whenSubmitted">2024-09-10 15:28:55</date>
              <date type="whenWritten">2024-09-05</date>
              <date type="whenModified">2025-02-20 18:03:28</date>
              <date type="whenReleased">2024-09-11 09:33:47</date>
              <date type="whenProduced">2024-09-05</date>
              <date type="whenEndEmbargoed">2024-09-10</date>
              <ref type="file" target="https://hal.science/hal-04688498v1/document">
                <date notBefore="2024-09-10"/>
              </ref>
              <ref type="file" subtype="author" n="1" target="https://hal.science/hal-04688498v1/file/Can_Large_Language_Models_Accurately_Predict_Public_Opinion__A_Review__2024_09_10.pdf" id="file-4688498-4081231">
                <date notBefore="2024-09-10"/>
              </ref>
            </edition>
            <respStmt>
              <resp>contributor</resp>
              <name key="1015481">
                <persName>
                  <forename>Arnault</forename>
                  <surname>Pachot</surname>
                </persName>
                <email type="md5">cd254612ec967525025cf2bcde11867c</email>
                <email type="domain">pachot.fr</email>
              </name>
            </respStmt>
          </editionStmt>
          <publicationStmt>
            <distributor>CCSD</distributor>
            <idno type="halId">hal-04688498</idno>
            <idno type="halUri">https://hal.science/hal-04688498</idno>
            <idno type="halBibtex">pachot:hal-04688498</idno>
            <idno type="halRefHtml">2024</idno>
            <idno type="halRef">2024</idno>
            <availability status="restricted">
              <licence target="https://about.hal.science/hal-authorisation-v1/">HAL Authorization<ref corresp="#file-4688498-4081231"/></licence>
            </availability>
          </publicationStmt>
          <seriesStmt>
            <idno type="stamp" n="SHS">Sciences de l'Homme et de la Société</idno>
            <idno type="stamp" n="AO-SOCIOLOGIE">Archives ouvertes de la Sociologie</idno>
            <idno type="stamp" n="SOCIOLOGIE">SOCIOLOGIE</idno>
            <idno type="stamp" n="AO-SCIENCEPOLITIQUE">Archives ouvertes de la Science politique</idno>
          </seriesStmt>
          <notesStmt/>
          <sourceDesc>
            <biblStruct>
              <analytic>
                <title xml:lang="en">Can Large Language Models Accurately Predict Public Opinion? A Review</title>
                <author role="crp">
                  <persName>
                    <forename type="first">Arnault</forename>
                    <surname>Pachot</surname>
                  </persName>
                  <email type="md5">cd254612ec967525025cf2bcde11867c</email>
                  <email type="domain">pachot.fr</email>
                  <idno type="idhal" notation="string">arnault-pachot</idno>
                  <idno type="idhal" notation="numeric">1104041</idno>
                  <idno type="halauthorid" notation="string">2242098-1104041</idno>
                  <affiliation ref="#struct-1207100"/>
                </author>
                <author role="aut">
                  <persName>
                    <forename type="first">Thierry</forename>
                    <surname>Petit</surname>
                  </persName>
                  <idno type="halauthorid">11290-0</idno>
                  <affiliation ref="#struct-1207100"/>
                </author>
              </analytic>
              <monogr>
                <imprint/>
              </monogr>
            </biblStruct>
          </sourceDesc>
          <profileDesc>
            <langUsage>
              <language ident="en">English</language>
            </langUsage>
            <textClass>
              <keywords scheme="author">
                <term xml:lang="en">Natural language processing</term>
                <term xml:lang="en">Survey Methodologies</term>
                <term xml:lang="en">Ideological Bias</term>
                <term xml:lang="en">Demographic Bias</term>
                <term xml:lang="en">Ethical Considerations</term>
                <term xml:lang="en">Artificial Intelligence (AI)</term>
                <term xml:lang="en">Natural Language Processing (NLP)</term>
                <term xml:lang="en">Bias</term>
                <term xml:lang="en">Meta-Analysis</term>
                <term xml:lang="en">Subpopulation Simulation</term>
                <term xml:lang="en">Public Opinion Prediction</term>
                <term xml:lang="en">Survey Research</term>
              </keywords>
              <classCode scheme="halDomain" n="info.info-ai">Computer Science [cs]/Artificial Intelligence [cs.AI]</classCode>
              <classCode scheme="halDomain" n="shs.scipo">Humanities and Social Sciences/Political science</classCode>
              <classCode scheme="halDomain" n="shs.socio">Humanities and Social Sciences/Sociology</classCode>
              <classCode scheme="halTypology" n="UNDEFINED">Preprints, Working Papers, ...</classCode>
              <classCode scheme="halOldTypology" n="UNDEFINED">Preprints, Working Papers, ...</classCode>
              <classCode scheme="halTreeTypology" n="UNDEFINED">Preprints, Working Papers, ...</classCode>
            </textClass>
            <abstract xml:lang="en">
              <p>This article reviews the capacity of Large Language Models (LLMs) to accurately predict public opinion. Through a meta-analysis of recent research, we evaluate metrics such as accuracy, correlation, and bias to assess the implications for survey-based social science.Our findings indicate that LLMs demonstrate high accuracy in predicting demographic responses and aligning with survey data, suggesting their potential in reflecting public opinion trends. Despite this, challenges persist, particularly with regard to biases in race, gender, and ideology, which can distort model outputs. Several studies propose methods to measure and mitigate these biases, aiming to improve the representativeness of LLM predictions.By addressing these limitations and refining the ethical use of LLMs, these models could significantly enhance public opinion research, offering valuable insights for researchers, policymakers, and decision-makers in forecasting societal sentiments.</p>
            </abstract>
          </profileDesc>
        </biblFull>
      </listBibl>
    </body>
    <back>
      <listOrg type="structures">
        <org type="institution" xml:id="struct-1207100" status="INCOMING">
          <orgName>Emotia</orgName>
          <orgName type="acronym">Emotia</orgName>
          <desc>
            <address>
              <country key="FR"/>
            </address>
          </desc>
        </org>
      </listOrg>
    </back>
  </text>
</TEI>