Complete underlying reasoning (57 claims and questions)
The main steps above are a reading path. These are all assessed contributions to this argument, including challenges and questions.
Answered · What data and study setting do the authors use?
Participant group, choices, and setting matter to the scope of conclusions about preference estimates and representation.
What data and study setting do the authors use?
AI browsing question drawn from this source idea; not a quotation or a claim that the authors asked this question.
Inspect the contribution and sourcesAnswered · What did the study report about the stability of participant preferences?
This asks about participant-preference stability and its context, relevant to the narrative’s distinction between within-sample estimates and population representation.
What did the study report about the stability of participant preferences?
AI browsing question drawn from this source idea; not a quotation or a claim that the authors asked this question.
Inspect the contribution and sourcesAnswered · How consistent were the group’s preferences, and what context do the authors give for their estimate?
The question concerns stability within the participant group, which must not be conflated with representativeness or generalization.
How consistent were the group’s preferences, and what context do the authors give for their estimate?
AI browsing question drawn from this source idea; not a quotation or a claim that the authors asked this question.
Inspect the contribution and sourcesAnswered · How did the researchers check whether they had enough participant data?
The sample-size check relates to stability of measured preferences while leaving the broader population scope unresolved.
How did the researchers check whether they had enough participant data?
AI browsing question drawn from this source idea; not a quotation or a claim that the authors asked this question.
Inspect the contribution and sourcesAnswered · What forms of group decision-making did the study leave untested?
The distinction between tested preference aggregation and untested consensus methods qualifies what the study’s results can establish about decision-making beyond its measured preference task.
What forms of group decision-making did the study leave untested?
AI browsing question drawn from this source idea; not a quotation or a claim that the authors asked this question.
Inspect the contribution and sourcesAnswered · Can AI help estimate a group’s preferences from a small sample?
The evidence matches the narrative’s distinction between improved estimates for the studied participant group and population representativeness.
Can AI help estimate a group’s preferences from a small sample?
AI browsing question drawn from this source idea; not a quotation or a claim that the authors asked this question.
Inspect the contribution and sourcesAnswered · Did adjusting the model or adding reference documents better match the full participant data?
The comparison concerns matching preferences in the study sample; it does not establish accuracy for a wider population or beyond the particular model and test setting.
Did adjusting the model or adding reference documents better match the full participant data?
AI browsing question drawn from this source idea; not a quotation or a claim that the authors asked this question.
Inspect the contribution and sourcesAnswered · Does adding a model to a sample improve predictions of the full study group’s preferences?
The question and evidence concern aggregate preferences within the study group, matching the narrative’s distinction between within-sample estimates and population representation.
Does adding a model to a sample improve predictions of the full study group’s preferences?
AI browsing question drawn from this source idea; not a quotation or a claim that the authors asked this question.
Inspect the contribution and sourcesAnswered · How many choices did the study record, and were they complete?
Dataset size and completeness bear on the scope of preference estimates.
How many choices did the study record, and were they complete?
AI browsing question drawn from this source idea; not a quotation or a claim that the authors asked this question.
Inspect the contribution and sourcesAnswered · How did the authors measure whether a sample matched the data for all participants?
The measurement is bounded to reproducing preferences in the study sample.
How did the authors measure whether a sample matched the data for all participants?
AI browsing question drawn from this source idea; not a quotation or a claim that the authors asked this question.
Inspect the contribution and sourcesAnswered · How did the authors judge whether a sample matched the full study’s preferences?
The comparison is bounded to patterns in the study sample, not wider-population representation.
How did the authors judge whether a sample matched the full study’s preferences?
AI browsing question drawn from this source idea; not a quotation or a claim that the authors asked this question.
Inspect the contribution and sourcesAnswered · What data and models did the authors study?
The data and models are tied to collected preferences in a specific election setting.
What data and models did the authors study?
AI browsing question drawn from this source idea; not a quotation or a claim that the authors asked this question.
Inspect the contribution and sourcesAnswered · What data did the researchers use to test the models?
The data source delimits the provenance and setting of the preference evidence.
What data did the researchers use to test the models?
AI browsing question drawn from this source idea; not a quotation or a claim that the authors asked this question.
Inspect the contribution and sourcesAnswered · What does the study aim to contribute?
The stated contribution concerns benchmarks bounded by study data and population.
What does the study aim to contribute?
AI browsing question drawn from this source idea; not a quotation or a claim that the authors asked this question.
Inspect the contribution and sourcesAnswered · Did adding model predictions for more people improve how closely the sample matched the study data?
The question concerns sample augmentation accuracy; the reported improvement is bounded to this dataset and aggregate study-group preference measure.
Did adding model predictions for more people improve how closely the sample matched the study data?
AI browsing question drawn from this source idea; not a quotation or a claim that the authors asked this question.
Inspect the contribution and sourcesPartly answered · What did the authors report for Type III A, which combines predictions for new people and new proposals?
The result concerns augmentation tests on study data; it does not establish population representation.
What did the authors report for Type III A, which combines predictions for new people and new proposals?
AI browsing question drawn from this source idea; not a quotation or a claim that the authors asked this question.
Inspect the contribution and sourcesAnswered · How did the study present policy choices to participants?
The elicitation method helps delimit the collected choices underlying the study’s findings.
How did the study present policy choices to participants?
AI browsing question drawn from this source idea; not a quotation or a claim that the authors asked this question.
Inspect the contribution and sourcesAnswered · What data did the study use?
The source data and setting bound the population to which results can be applied.
What data did the study use?
AI browsing question drawn from this source idea; not a quotation or a claim that the authors asked this question.
Inspect the contribution and sourcesAnswered · Did model predictions improve the group-preference estimate in the reported comparison?
The comparison is against collected study preferences, not an established estimate of the wider population.
Did model predictions improve the group-preference estimate in the reported comparison?
AI browsing question drawn from this source idea; not a quotation or a claim that the authors asked this question.
Inspect the contribution and sourcesAnswered · How do the researchers test whether smaller samples match the group’s preferences?
The benchmark is the study-participant aggregate, not a test of population representativeness.
How do the researchers test whether smaller samples match the group’s preferences?
AI browsing question drawn from this source idea; not a quotation or a claim that the authors asked this question.
Inspect the contribution and sourcesAnswered · What data did the authors sample and augment in their validation procedure?
The validation reference is the full study sample, so results concern reproducing that sample’s aggregate pattern.
What data did the authors sample and augment in their validation procedure?
AI browsing question drawn from this source idea; not a quotation or a claim that the authors asked this question.
Inspect the contribution and sourcesAnswered · How did the authors measure and compare group-wide policy preferences?
The question concerns a preference-data measure whose reported scope is limited to the study sample.
How did the authors measure and compare group-wide policy preferences?
AI browsing question drawn from this source idea; not a quotation or a claim that the authors asked this question.
Inspect the contribution and sourcesAnswered · What kinds of preference data can a language model add?
The augmentation methods concern the scope of preference information, while leaving population representativeness unresolved.
What kinds of preference data can a language model add?
AI browsing question drawn from this source idea; not a quotation or a claim that the authors asked this question.
Inspect the contribution and sourcesAnswered · Which people do the study’s results describe?
This directly concerns the distinction between estimates for the study group and population representation.
Which people do the study’s results describe?
AI browsing question drawn from this source idea; not a quotation or a claim that the authors asked this question.
Inspect the contribution and sourcesStarting point · Ways a language model can add predicted preferences
The types distinguish additions to the study’s preference data, making clear that estimating preferences for additional people is a different task from representing the studied group.
The authors describe several ways to add predictions about preferences. Type I predicts preferences for additional people using demographic information, such as age. Type II A predicts more preferences for people already in the sample about proposals already included. Type II B predicts those people’s preferences about new proposals. Type II C combines those two Type II approaches. Type III adds predictions for both new people and new proposals.
Source coverage: The full readable text of the 37-page research draft, including references and appendices. Figures are not independently interpreted. The passage distinguishes methods by whether they add people, preferences, or proposals. Type II C combines the two Type II methods; Type III combines predictions for new people and new proposals. Addresses: The methods for adding predicted preferences described in the study
Inspect the contribution and sourcesImportant limit · People were represented with limited data
The representation uses a specific, limited set of collected data; it does not establish accuracy for missing information or broader populations.
The study represented each person using a short set of personal background details and preferences between pairs of options. The authors say this is a limited form of a digital twin, meaning a digital representation of a person.
Source coverage: Full readable text of the 37-page research draft, including references and appendices. Figures are not independently interpreted. The authors contrast their limited representation with a more detailed example, but do not test whether a richer representation would improve this study’s predictions. Addresses: Scope of the personal representation
Inspect the contribution and sourcesImportant limit · A check on preference stability
The check assesses stability of preference estimates in participant data, not representativeness of the population or accuracy beyond that setting.
To check whether their sample was large enough to give stable estimates of preferences, the researchers compared equal-sized, non-overlapping random samples. They measured how closely the proposal win rates in the two samples moved together, a relationship called correlation.
Source coverage: Full readable text of the retrieved 37-page research draft, including references and appendices. Figures are not independently interpreted. They used Pearson correlation, a measure of how closely two sets of values move together, and its coefficient of determination (R²), which shows how closely the proposal win rates in the two samples track each other. They plotted the result against sample size. Addresses: Appendix B data adequacy checks
Inspect the contribution and sourcesImportant limit · Group preferences were consistent on retest
The reported reliability concerns the study group’s preferences; it does not establish population representativeness or performance outside the studied setting.
The authors estimate that the group’s preferences were 95.38% consistent when tested again in the full participant sample. They say this is consistent with earlier research using choices between pairs, where rankings tended to settle after about 30 to 40 choices per participant.
Source coverage: Full readable text of the retrieved 37-page research draft, including references and appendices. Figures are not independently interpreted. The comparison with earlier work provides context for the estimate; it does not independently check the estimate or show that the participants represent Brazil. Addresses: Consistency of the full sample’s group preferences
Inspect the contribution and sourcesImportant limit · Explore other data sources
Other collection sources could change the data and setting to which estimates apply, but the authors do not claim they would improve representativeness.
The authors suggest exploring ways to expand the data used by a model, including by using data collected in other ways, such as through voting advice applications.
Source coverage: Full readable text of the retrieved 37-page research draft, including references and appendices. Figures are not independently interpreted. The authors identify these data sources as an avenue for further study, without stating that they should improve the approach. Addresses: A proposed direction for future research
Inspect the contribution and sourcesSupporting point · Recorded preferences may help
The result concerns estimates based on the study’s participant data, not population representativeness.
In tests with GPT-3.5 Turbo, the authors report that Type II A and B augmentation performed better than Type I B augmentation. Here, augmentation means using a model to predict preferences. The Type II tests used all demographic information and predicted additional preferences for people already in the sample. The authors suggest that these tests may benefit from using participants' recorded preferences as well as demographic details.
Source coverage: Full readable text of the 37-page research draft, including references and appendices. Figures are not independently interpreted. The reported comparison is for tests using all demographic information. The authors' explanation about the extra preference information is a possibility they suggest, not a cause tested separately. Addresses: Reported results and proposed explanation for Type II A and B augmentation
Inspect the contribution and sourcesImportant limit · How the proposals were collected
The source and curation define the proposal set on which the study’s preference data are based; they do not establish that it covers the full policy agenda or public opinion.
The study used 67 policy proposals collected from Brazucracia.org and linked to three candidates in Brazil’s 2022 presidential election. A lawyer led the work of selecting and organizing the proposals.
Source coverage: Full readable text of the retrieved 37-page research draft, including references and appendices. Figures are not independently interpreted. This describes where the proposals came from and how they were selected and organized, not a check that they were complete or fair representations of the candidates’ platforms. Addresses: Study materials
Inspect the contribution and sourcesStarting point · Testing whether a model improves a sample
The claim is explicitly scoped to the study group’s preferences.
The authors ask whether a sample selected using chance, combined with a language model trained on that same sample, can predict the preferences of all 267 study participants better than the sample alone.
Source coverage: The full readable text of the 37-page research draft, including references and appendices. Figures are not independently interpreted. This is the study’s stated comparison for estimating group preferences. It concerns reproducing the preferences measured among these 267 participants, not necessarily those of a wider public.
Inspect the contribution and sourcesImportant limit · Demographic details and ethics review
The reported procedures give context for the participant data while leaving population representativeness unestablished.
Participants reported demographic details, and the authors report ethics review approval and public release of the data with earlier work.
Source coverage: Full readable text of the retrieved 37-page research draft, including references and appendices. Figures are not independently interpreted. These details describe what the authors report; they do not show that the participants represent the public. Addresses: Study data and procedures
Inspect the contribution and sourcesStarting point · Testing combined group preferences
The target is the study group’s combined preferences.
The researchers also compare whether ordinary random samples and samples expanded with model-generated choices can reproduce the combined preferences of all 267 participants.
Source coverage: Full readable text of the retrieved 37-page research draft, including references and appendices. Figures are not independently interpreted. The paper describes comparing ordinary random samples and samples expanded with model-generated choices against the combined preferences of all 267 participants.
Inspect the contribution and sourcesImportant limit · How the authors compared the recorded preferences
Matching the full study data does not establish population representation.
The authors organize preference data in a matrix, a table of rows and columns. Each row represents a participant, and each column records a preference between a pair of policy proposals. They take samples and add model predictions to this data, then compare how well those samples reproduce the overall preferences in the full sample.
Source coverage: Full readable text of the 37-page research draft, including references and appendices. Figures are not independently interpreted. This describes how the data are organized and how the authors checked whether a smaller sample, or one with added model predictions, matched the full participant data. Addresses: The structure of the preference data and how the authors checked the results
Inspect the contribution and sourcesImportant limit · The study’s scope
Directly supports the distinction between study-group findings and national population preferences.
The study asks whether models can learn the preferences of a specific group of 267 participants. It does not aim to estimate what people across Brazil prefer.
Source coverage: Full readable text of the retrieved 37-page research draft, including references and appendices. Figures are not independently interpreted. The authors define the 267 participants as the group whose preferences they aim to model. They state that the data cannot support estimates for people across Brazil.
Inspect the contribution and sourcesImportant limit · A limited set of comparisons
The limited data basis cannot establish population representativeness.
The dataset contains 8,719 choices between pairs of proposals from 267 participants. The choices cover 67 proposals, and the authors say the data are sparse, meaning they do not cover all possible comparisons.
Source coverage: Full readable text of the retrieved 37-page research draft, including references and appendices. Figures are not independently interpreted. The authors identify the limited set of recorded comparisons as a reason to study augmentation. Addresses: The study’s data coverage
Inspect the contribution and sourcesImportant limit · Testing individual predictions
Cross-validation within the study does not establish that the data represent Brazil.
The researchers train models on some participants’ choices. They test the predictions against choices from participants whose data were not used to train the models.
Source coverage: Full readable text of the retrieved 37-page research draft, including references and appendices. Figures are not independently interpreted. The researchers split participants into groups for training and testing. They compared the model’s predictions with choices from the test group.
Inspect the contribution and sourcesImportant limit · Six models adapted using volunteer data
The models were adapted using volunteer data, so results from them cannot by themselves establish population representation beyond the collected participants.
The authors fine-tune six existing language models on the volunteer data: Llama-2 7B, LLaMA-3 8B, ChatGPT 3.5 Turbo, Mistral 7B, Falcon 7B, and Gemma 7B.
Source coverage: Full readable text of the retrieved 37-page research draft, including references and appendices. Figures are not independently interpreted. The authors adapt the models using collected policy-preference data, then use them to predict choices and group preferences.
Inspect the contribution and sourcesImportant limit · How the study measured a match
Specifies what the reported sample-match measure captures and limits it to reproducing the study’s participant data, not population representation.
The study compares each proposal’s win rate in a sample with its win rate among all study participants. It then uses R², a score for how closely two patterns match, to measure how closely the sample reproduces the pattern for all participants.
Source coverage: The full readable text of the 37-page research draft, including references and appendices. Figures are not independently interpreted. This measures similarity to the full study sample; it does not directly test whether the preferences are correct for people outside the study. Addresses: The study’s measure of how closely a sample matches the full study
Inspect the contribution and sourcesImportant limit · No new human data collected
It cautions against treating the paper’s proposals or model outputs as new evidence of voter preferences.
The authors say the study involved no original research with human participants and collected no primary data, meaning no new data gathered firsthand for the study.
Source coverage: Full readable text of the retrieved 37-page research draft, including references and appendices. Figures are not independently interpreted. This is a statement about the study’s methods, not proof that the proposals or any model outputs reflect what voters think. Addresses: Study limits
Inspect the contribution and sourcesImportant limit · Type III A combined two kinds of preference prediction
The result concerns the study’s augmentation tests and does not establish population representativeness or performance elsewhere.
The authors also report results for Type III A augmentation, which combines predictions for additional people and additional proposals. Here, augmentation means using a model to predict preferences. They describe its results as similar to the other augmentation tests; the figure caption specifies that this test used all demographic information.
Source coverage: Full readable text of the 37-page research draft, including references and appendices. Figures are not independently interpreted. The passage reports a general similarity to other tests, but gives no separate numerical result here. The Type III category combines predictions for new people and new proposals. Addresses: The reported Type III A preference-prediction result
Inspect the contribution and sourcesImportant limit · A method adapted from urban planning
The method describes how preference data are structured, but does not establish that the participants or preferences represent a wider population.
The study adapts a paired-comparison approach from urban planning, where people compare options and computer systems use those choices to create a picture of which options people prefer.
Source coverage: Full readable text of the retrieved 37-page research draft, including references and appendices. Figures are not independently interpreted. The authors describe this as an analogy for their approach, not proof that the two applications have identical results. Addresses: The study’s methodological model
Inspect the contribution and sourcesSupporting point · Predictions improved some sample matches
The reported estimates concern the study sample’s aggregate preferences, not a broader population.
In the Type I B test, the authors adjusted six models using the same preference data: GPT-3.5 Turbo, LLaMA-2 7B, Falcon 7B, Mistral 7B, LLaMA-3 8B, and Gemma 7B. They tested initial random samples of 5%, 25%, and 50%, adding predictions for 20% of the people who remained outside each sample. For the 5% sample, R², a score for how closely two patterns match, rose from about 30% to about 75%; for the 25% sample, the gain was about 7 to 10 percentage points. The supplied text gives no separate numerical result for the 50% sample.
Source coverage: The full readable text of the 37-page research draft, including references and appendices. Figures are not independently interpreted. These figures describe tests on the study’s data, with models adjusted using the same data and predictions added for a stated share of people outside the initial sample. The text identifies 50% as a tested sample size but gives no separate numerical result for it. These tests do not show how well the method would work in other settings. Addresses: The reported result for adding Type I B predictions
Inspect the contribution and sourcesImportant limit · Participant details used as context
It specifies a limited set of participant attributes used in the study, rather than establishing representation of the wider population.
The participant characteristics used as prompt context included age, political ideology, urban or rural zone, education, sex, and city and state.
Source coverage: Full readable text of the retrieved 37-page research draft, including references and appendices. Figures are not independently interpreted. The table gives age as young or old using the top and bottom quarters; ideology as conservative, centrist, or liberal; zone as urban or rural; education as college or non-college; sex as male or female; and location as city and state. Addresses: Table 2 demographic characteristics
Inspect the contribution and sourcesSupporting point · Small samples gave better estimates in this test
The claim is expressly about aggregate preferences within the participant group, not population representation or individual-response accuracy.
For samples below 30% of participants, the authors found that using models to improve estimates from the small sample gave a more accurate estimate of all participants’ preferences than estimates from random samples without model help.
Source coverage: Full readable text of the 37-page research draft, including references and appendices. Figures are not independently interpreted. The authors compare estimates of all participants’ preferences. They do not describe filling in individual participants’ missing responses. Addresses: Using models to improve estimates from a sample
Inspect the contribution and sourcesImportant limit · Anonymous comparisons of two proposals
This describes how the choices were collected, but does not establish that the resulting data represent the wider population.
Participants compared two policy proposals at a time on an online platform, and their participation was anonymous.
Source coverage: Full readable text of the retrieved 37-page research draft, including references and appendices. Figures are not independently interpreted. The passage describes a screen for comparing two proposals at a time and says participation was anonymous; it does not explain recruitment or voting rules here. Addresses: Brazucracia comparison platform
Inspect the contribution and sourcesStarting point · Where the data came from
The data source is part of the study setting whose participant scope limits generalization beyond the collected group.
The study uses pairwise choices and information about participants’ backgrounds collected through Brazucracia, an online government-program builder used during Brazil’s 2022 presidential election.
Source coverage: Full readable text of the retrieved 37-page research draft, including references and appendices. Figures are not independently interpreted. The passage identifies the online project and describes its role in collecting participant choices and personal details.
Inspect the contribution and sourcesStarting point · Six models tested on collected preferences
The collected data and study setting delimit what the prediction results can establish about other populations or settings.
The authors further trained six existing large language models, which are AI tools trained to work with language. They tested how well the models predicted preferences using detailed political preference data collected during Brazil’s 2022 presidential election.
Source coverage: Full readable text of the 37-page research draft, including references and appendices. Figures are not independently interpreted. The paper describes using detailed preference data collected during the election to study model predictions. This was not a test of a complete voting system. Addresses: Study method and data
Inspect the contribution and sourcesImportant limit · Fine-tuning did better than adding reference documents
The outcome is measured against the preferences of the study participants, not a general population.
In a comparison using GPT-3.5 Turbo, predictions from a model adjusted through fine-tuning with LoRA matched the full participants' preferences better than predictions made using RAG, especially with 25% and 50% samples. RAG, or retrieval-augmented generation, gives a model reference documents to use when answering. The reference document summarized proposals from the candidate favored by each participant: Bolsonaro for conservatives and Lula for liberals.
Source coverage: Full readable text of the 37-page research draft, including references and appendices. Figures are not independently interpreted. This comparison used one model and reference documents selected according to participants' favored candidates. It does not establish that adjusting a model through fine-tuning is always better than giving it reference documents. Addresses: The study's comparison of fine-tuning with retrieval-augmented generation
Inspect the contribution and sourcesImportant limit · Summarizing group preferences with proposal win rates
The metric compares a sample with the full study sample and therefore does not establish preferences of a wider public.
To summarize policy preferences in a group, the authors define each proposal’s win rate as the share of recorded choices between two proposals in which it was selected. They then compare win rates from a sample with those from all participants in the study.
Source coverage: The full readable text of the 37-page research draft, including references and appendices. Figures are not independently interpreted. This approach summarizes choices even when participants did not compare every possible pair of proposals. Its conclusions concern how well a sample reproduces the results for all participants in the study, not necessarily the preferences of a wider public.
Inspect the contribution and sourcesStarting point · Participants compared policy proposals
The study’s data origin and setting are directly relevant to the stated limits on population-level inference.
The study uses choices between pairs of policy proposals from an online participatory experiment held during Brazil’s 2022 presidential election.
Source coverage: Full readable text of the retrieved 37-page research draft, including references and appendices. Figures are not independently interpreted. The data come from the named experiment and election period; they should not be treated as a survey of all people in Brazil. Addresses: The study’s participant data
Inspect the contribution and sourcesStarting point · The volunteer policy data
The volunteer sample, policy set, and election-specific setting define the evidence’s scope and limit population inference.
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.
Source coverage: Full readable text of the retrieved 37-page research draft, including references and appendices. Figures are not independently interpreted. The experiment covered 67 policies. Volunteers chose between pairs and reported characteristics such as age, location, sex, and political orientation.
Inspect the contribution and sourcesSupporting point · Estimates of group preferences
Reports an improvement for the study’s group-preference estimate; its participant and data scope limits broader population conclusions.
Using cross-validation, a way to test a method on separate parts of the data, the authors report that a sample of people selected by chance, with each person's chance of selection known, augmented with model predictions estimated group policy preferences more accurately than the sample alone.
Source coverage: Full readable text of the retrieved 37-page research draft, including references and appendices. Figures are not independently interpreted. The paper estimates accuracy for individual choices and group preferences. The passages state the direction of the group-level result, but not its size or uncertainty.
Inspect the contribution and sourcesImportant limit · Reported stability in the sample
Stability within the sampled participants does not show that their preferences represent Brazil’s population.
The researchers estimated that two random samples of 53 participants each, about 20% of the study sample, would produce an R² of about 0.72 for proposal win rates. R² is a measure of how closely the rates in the two samples track each other. They reported an R² of 0.9538 when using the full sample.
Source coverage: Full readable text of the retrieved 37-page research draft, including references and appendices. Figures are not independently interpreted. The appendix presents the 53-person estimate and the full-sample repeatability result as findings from its check of the data, not as proof that participants represent the wider public. Addresses: Appendix B data adequacy checks
Inspect the contribution and sourcesImportant limit · Comparing proposal win rates in a sample and the full study
The measure concerns aggregate choices within the study sample, not population representativeness or missing-response accuracy.
The authors compare proposal win rates in a sample with the rates for all participants in the study. They use the squared Pearson correlation, a statistical score for how closely two sets of values follow the same pattern. The authors treat a value of 100% as identical win-rate patterns.
Source coverage: The full readable text of the 37-page research draft, including references and appendices. Figures are not independently interpreted. This describes the study’s chosen measure of similarity. A high score on this measure concerns the pattern of proposal win rates; it does not by itself establish that every individual’s choices or every other feature of the group is represented.
Inspect the contribution and sourcesImportant limit · Models beat a simple party rule
The finding concerns group preferences in the study data and a specific baseline, not population representativeness.
The models also predicted group preferences more accurately than a rule that assumes people always choose the policies of their preferred party or politician.
Source coverage: Full readable text of the 37-page research draft, including references and appendices. Figures are not independently interpreted. The authors interpret the comparison as evidence that their models captured preference details beyond a simple left-right political divide. It does not show what specific information the models used. Addresses: Comparison with party-based prediction
Inspect the contribution and sources