Reproducibility is not construct validity: LLM measurement of institutionally situated communication
Abstract
High annotation reproducibility does not necessarily imply that an LLM-inferred measure captures the construct it is intended to measure. We test this distinction using a dataset from the European Commission's AI Act consultation, linking structured survey responses to free-text consultation submissions from the same stakeholders. LLM annotations of consultation submissions are highly reproducible (intraclass correlations > 0.99), yet show limited convergence with survey-reported measures of the nominal construct they were intended to approximate. Divergence between survey-and LLM-inferred text-based measures varies systematically across stakeholder groups: business associations express greater concern about AI risks in text-based consultations than in survey responses (ḡ = +1.0), whereas public authorities and several nonbusiness groups show smaller or negative divergences. Divergences between scores suggest positive spatial autocorrelation across European countries (Moran's I = 0.347, p = 0.036), indicating that stakeholders from neighboring countries tend toward more similar text-based stances towards AI safety concerns. Despite divergence, survey-reported concerns remain strongly associated with support for explainability across all divergence levels. These results demonstrate that LLM annotation reproducibility can coexist with poor construct correspondence and motivate validation procedures that distinguish reproducibility, construct validity, and communication context variation when LLMs are used as measurement instruments.
Related
- Measurement reliability, construct validity, and transparent reporting in original and replication psychological research
- Recovering Expert Critic-Sourced Network Adjacency between Musical Artists from Acoustic Distributions: A Construct-Validity Approach
- On the use of the Belopol'skaya-Daletskii representation of a diffusion on a Riemann manifold to construct path integrals
- Using the slowest observable in one-dimensional Markov processes to construct quasi-exactly-solvable generators with two explicit levels
- How to Construct High Barrycades
BibTeX
@misc{batzdorfer2026reproducibility,
title = {{Reproducibility is not construct validity: LLM measurement of institutionally situated communication}},
author = {Veronika Batzdorfer and Carlo Romano Marcello Alessandro Santagiustina},
howpublished = {HAL},
year = {2026},
month = aug,
url = {https://hal.science/hal-05745153v1},
note = {CC BY-NC-ND 4.0},
}
RIS
TY - UNPB
TI - Reproducibility is not construct validity: LLM measurement of institutionally situated communication
AU - Veronika Batzdorfer
AU - Carlo Romano Marcello Alessandro Santagiustina
PY - 2026
DA - 2026/08/31
PB - HAL
UR - https://hal.science/hal-05745153v1
AB - High annotation reproducibility does not necessarily imply that an LLM-inferred measure captures the construct it is intended to measure. We test this distinction using a dataset from the European Commission's AI Act consultation, linking structured survey responses to free-text consultation submissions from the same stakeholders. LLM annotations of consultation submissions are highly reproducible (intraclass correlations > 0.99), yet show limited convergence with survey-reported measures of the nominal construct they were intended to approximate. Divergence between survey-and LLM-inferred text-based measures varies systematically across stakeholder groups: business associations express greater concern about AI risks in text-based consultations than in survey responses (ḡ = +1.0), whereas public authorities and several nonbusiness groups show smaller or negative divergences. Divergences between scores suggest positive spatial autocorrelation across European countries (Moran's I = 0.347, p = 0.036), indicating that stakeholders from neighboring countries tend toward more similar text-based stances towards AI safety concerns. Despite divergence, survey-reported concerns remain strongly associated with support for explainability across all divergence levels. These results demonstrate that LLM annotation reproducibility can coexist with poor construct correspondence and motivate validation procedures that distinguish reproducibility, construct validity, and communication context variation when LLMs are used as measurement instruments.
ER -