LLMRpanel: Benchmarked Silicon Samples for Survey and Experiment Design

Administers survey and experimental instruments to panels of language-model personas, with respondent-level randomization, benchmark comparison against human data, and conjoint estimation from recorded respondent-level profile assignments. Samples of language-model personas follow Argyle et al. (2023) <doi:10.1017/pan.2023.2>; the case for benchmarking them against human data is set out in Bisbee et al. (2024) <doi:10.1017/pan.2024.5>; the conjoint estimand is the average marginal component effect of Hainmueller et al. (2014) <doi:10.1093/pan/mpt024>.

Version:	0.6.1
Depends:	R (≥ 4.2)
Imports:	LLMR (≥ 0.8.9), tibble, rlang, cli, stats, utils
Suggests:	testthat (≥ 3.0.0), ggplot2, knitr, rmarkdown, shiny, bslib, DT, LLMR.shiny (≥ 0.1.2)
Published:	2026-09-12
DOI:	10.32614/CRAN.package.LLMRpanel (may not be active yet)
Author:	Ali Sanaei [aut, cre, cph]
Maintainer:	Ali Sanaei <sanaei at uchicago.edu>
BugReports:	https://github.com/asanaei/LLMRpanel/issues
License:	MIT + file LICENSE
URL:	https://github.com/asanaei/LLMRpanel, https://asanaei.github.io/LLMRpanel/
NeedsCompilation:	no
Materials:	README, NEWS
CRAN checks:	LLMRpanel results

Documentation:

Reference manual:	LLMRpanel.html , LLMRpanel.pdf
Vignettes:	Survey and experiment pilots with LLMRpanel (source, R code)

Downloads:

Package source:	LLMRpanel_0.6.1.tar.gz
Windows binaries:	r-devel: not available, r-release: not available, r-oldrel: not available
macOS binaries:	r-release (arm64): not available, r-oldrel (arm64): LLMRpanel_0.6.1.tgz, r-release (x86_64): LLMRpanel_0.6.1.tgz, r-oldrel (x86_64): LLMRpanel_0.6.1.tgz

Linking:

Please use the canonical form https://CRAN.R-project.org/package=LLMRpanel to link to this page.