@inproceedings{halterman-keith-2026-protest,
title = "What is a protest anyway? Codebook conceptualization is still a first-order concern in {LLM}-era classification",
author = "Halterman, Andrew and
Keith, Katherine A.",
editor = "Liakata, Maria and
Moreira, Viviane P. and
Zhang, Jiajun and
Jurgens, David",
booktitle = "Proceedings of the 64th Annual Meeting of the {A}ssociation for {C}omputational {L}inguistics (Volume 1: Long Papers)",
month = jul,
year = "2026",
address = "San Diego, California, United States",
publisher = "Association for Computational Linguistics",
url = "https://aclanthology.org/2026.acl-long.92/",
pages = "2043--2059",
ISBN = "979-8-89176-390-6",
abstract = "Generative large language models (LLMs) are now used extensively for text classification in computational social science (CSS). In this work, we focus on the steps before and after LLM prompting: \textit{conceptualization} of the categories to classify and using LLM predictions in downstream statistical inference. We argue these steps have been overlooked in much of LLM-era CSS and LLMs can tempt analysts to skip conceptualization altogether. For example, a political scientist classifying ``protest'' with LLMs may never be forced to craft a definition: unlike human annotators who would ask clarifying questions, an LLM can silently accept an underspecified concept to classify and return plausible-looking labels. Using simulations, we show that conceptualization failures induce downstream inferential bias that cannot be corrected solely by a more accurate LLM or post-hoc bias correction methods. We conclude by reminding CSS analysts that conceptualization is still a first-order concern in the LLM-era and provide concrete advice for pursuing low-cost, unbiased, low-variance downstream estimates."
}<?xml version="1.0" encoding="UTF-8"?>
<modsCollection xmlns="http://www.loc.gov/mods/v3">
<mods ID="halterman-keith-2026-protest">
<titleInfo>
<title>What is a protest anyway? Codebook conceptualization is still a first-order concern in LLM-era classification</title>
</titleInfo>
<name type="personal">
<namePart type="given">Andrew</namePart>
<namePart type="family">Halterman</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Katherine</namePart>
<namePart type="given">A</namePart>
<namePart type="family">Keith</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<originInfo>
<dateIssued>2026-07</dateIssued>
</originInfo>
<typeOfResource>text</typeOfResource>
<relatedItem type="host">
<titleInfo>
<title>Proceedings of the 64th Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers)</title>
</titleInfo>
<name type="personal">
<namePart type="given">Maria</namePart>
<namePart type="family">Liakata</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Viviane</namePart>
<namePart type="given">P</namePart>
<namePart type="family">Moreira</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Jiajun</namePart>
<namePart type="family">Zhang</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">David</namePart>
<namePart type="family">Jurgens</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<originInfo>
<publisher>Association for Computational Linguistics</publisher>
<place>
<placeTerm type="text">San Diego, California, United States</placeTerm>
</place>
</originInfo>
<genre authority="marcgt">conference publication</genre>
<identifier type="isbn">979-8-89176-390-6</identifier>
</relatedItem>
<abstract>Generative large language models (LLMs) are now used extensively for text classification in computational social science (CSS). In this work, we focus on the steps before and after LLM prompting: conceptualization of the categories to classify and using LLM predictions in downstream statistical inference. We argue these steps have been overlooked in much of LLM-era CSS and LLMs can tempt analysts to skip conceptualization altogether. For example, a political scientist classifying “protest” with LLMs may never be forced to craft a definition: unlike human annotators who would ask clarifying questions, an LLM can silently accept an underspecified concept to classify and return plausible-looking labels. Using simulations, we show that conceptualization failures induce downstream inferential bias that cannot be corrected solely by a more accurate LLM or post-hoc bias correction methods. We conclude by reminding CSS analysts that conceptualization is still a first-order concern in the LLM-era and provide concrete advice for pursuing low-cost, unbiased, low-variance downstream estimates.</abstract>
<identifier type="citekey">halterman-keith-2026-protest</identifier>
<location>
<url>https://aclanthology.org/2026.acl-long.92/</url>
</location>
<part>
<date>2026-07</date>
<extent unit="page">
<start>2043</start>
<end>2059</end>
</extent>
</part>
</mods>
</modsCollection>
%0 Conference Proceedings
%T What is a protest anyway? Codebook conceptualization is still a first-order concern in LLM-era classification
%A Halterman, Andrew
%A Keith, Katherine A.
%Y Liakata, Maria
%Y Moreira, Viviane P.
%Y Zhang, Jiajun
%Y Jurgens, David
%S Proceedings of the 64th Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers)
%D 2026
%8 July
%I Association for Computational Linguistics
%C San Diego, California, United States
%@ 979-8-89176-390-6
%F halterman-keith-2026-protest
%X Generative large language models (LLMs) are now used extensively for text classification in computational social science (CSS). In this work, we focus on the steps before and after LLM prompting: conceptualization of the categories to classify and using LLM predictions in downstream statistical inference. We argue these steps have been overlooked in much of LLM-era CSS and LLMs can tempt analysts to skip conceptualization altogether. For example, a political scientist classifying “protest” with LLMs may never be forced to craft a definition: unlike human annotators who would ask clarifying questions, an LLM can silently accept an underspecified concept to classify and return plausible-looking labels. Using simulations, we show that conceptualization failures induce downstream inferential bias that cannot be corrected solely by a more accurate LLM or post-hoc bias correction methods. We conclude by reminding CSS analysts that conceptualization is still a first-order concern in the LLM-era and provide concrete advice for pursuing low-cost, unbiased, low-variance downstream estimates.
%U https://aclanthology.org/2026.acl-long.92/
%P 2043-2059
Markdown (Informal)
[What is a protest anyway? Codebook conceptualization is still a first-order concern in LLM-era classification](https://aclanthology.org/2026.acl-long.92/) (Halterman & Keith, ACL 2026)
ACL