Contribution to the handbook
Supporting · Assessed 2026-10-28
Reported participant information, ethics review, and public data release are useful context for scrutinizing the study’s evidence and reproducibility. They do not by themselves establish representativeness or a governance safeguard.
Does this challenge the goal?
None identified
The question asks what a paper reports about its participants and procedures; it does not contest the exploration’s goal or beneficiary boundary.
The packet reports the authors’ procedures but does not independently verify them or show that the participants represent a wider public.
Why this category?
- evidence provenance
- research transparency
- limits of sample
The reported ethics review and data release are statements about the study’s procedures; they do not establish that its sample represents citizens outside the study.
This is an AI assessment of contribution to the goal, separate from whether the question has an answer.
Authority, accountability and resilience →
Tests, stop conditions and unresolved work →
Read the exploration goalNew evidence is being assessed. Any answer below is provisional; parts whose evidence changed or was withdrawn are withheld.
Describe what participant information the paper reports collecting.
What we can say so far
The paper reports that participants supplied demographic information, including age, political ideology, urban or rural zone, education, sex, and city and state. In the prompt-context categories described in the paper, age was grouped as young or old (using the top and bottom quartiles); ideology as conservative, centrist, or liberal; zone as urban or rural; education as college or non-college; and sex as male or female. Participants’ pairwise policy choices were also used alongside these characteristics to represent their preferences.
What remains open
- The supplied evidence does not explain how missing responses or participant characteristics outside the listed categories were handled.
Conditions and limits
- This describes the information reported and used in this study; it does not establish that participants represent the broader Brazilian population.
- The Portuguese robustness-check prompt describes some categories with different wording, such as “third age” and “left-wing”; that is a prompt design description, not evidence that the underlying categories were identical across prompts.
Evidence and related ideas
Supporting evidence, qualifications and context play different roles in this answer.
supports
The volunteer policy data The study uses volunteer choices between randomly paired policies from the two leading candidates’ 2022 Brazilian presidential programs. The choices came from an experiment involving collaborative construction of a government program.
Reports participant characteristics including age, location, sex, and political orientation, alongside policy choices.
Large Language Models (LLMs) as Agents for Augmented Democracy
Read full source: Large Language Models (LLMs) as Agents for Augmented Democracy (opens in a new tab) ↗Read source discussion: Large Language Models (LLMs) as Agents for Augmented Democracy →
context
Participants compared policy proposals The study uses paired policy choices from an online participatory experiment held during Brazil’s 2022 presidential election.
Places the paired policy choices in the online participatory experiment and election setting.
Describe the reported research procedures, including ethics review and the data’s public release with earlier work.
What we can say so far
The paper reports using volunteer choices collected through Brazucracia, an online platform used during Brazil’s 2022 presidential election. Participants made pairwise choices between policy proposals; the paper also reports collecting participant characteristics, including age, location, sex, and political orientation, and says participation was anonymous. The proposals were selected and organized for the study, which used paired comparisons. The authors report ethics-review approval and that the data were publicly released with earlier work. These are author-reported procedures, not independent verification. The authors also state that this paper collected no primary human-participant data; read alongside their description of the volunteer dataset, this indicates the study analyzed previously collected data rather than newly recruiting participants for this paper. The researchers fine-tuned six existing language models on the volunteer choices. For individual prediction, they trained on some participants’ choices and evaluated predictions on held-out participants; they also compared ordinary and model-augmented samples against the aggregate preferences of the full participant group. The supplied evidence does not identify the ethics-review body or date, or specify the data-release location.
What remains open
- The supplied evidence does not name the ethics-review body or give the approval date.
- The supplied evidence does not specify where or under what access terms the data were publicly released.
Conditions and limits
- Ethics approval and public release are reported by the authors; the supplied evidence does not independently verify them.
- The paper’s statement that it collected no primary human-participant data should be understood alongside its use of previously collected volunteer choices from Brazucracia.
- The reported volunteer data and evaluation concern the study participants; they do not establish representativeness of the broader public.
Objections to consider
- The statement that this paper collected no primary human-participant data could appear inconsistent with its use of volunteer choices. The answer distinguishes the current paper’s analysis from the earlier collection of those choices.
Evidence and related ideas
Supporting evidence, qualifications and context play different roles in this answer.
supports
The volunteer policy data The study uses volunteer choices between randomly paired policies from the two leading candidates’ 2022 Brazilian presidential programs. The choices came from an experiment involving collaborative construction of a government program.