Contribution to the handbook
Priority · Assessed 2026-10-28
The question guards against treating improved estimates of a study group’s aggregate preferences as proof of accurate predictions about individuals. The supplied results include held-out individual-choice performance, a specific baseline, and subgroup exceptions, which are consequential limits when evaluating models proposed for representing citizen preferences (q-live-f633b050eab07fbd607e4792; s-hidalgo-augmented-democracy-3f23115dcdcd6332; s-hidalgo-augmented-democracy-a0e788f205918f5c; s-hidalgo-augmented-democracy-d3c61d3ab756ba5d; s-hidalgo-augmented-democracy-ea2515219a5f53ac).
Does this challenge the goal?
In-scope gap
It asks how to interpret specific study results and limits for preference estimation, an evidence and testing gap within the existing goal; it does not challenge that goal.
The supplied material describes a study of 267 participants and particular comparisons, not other populations or real governance settings (q-live-f633b050eab07fbd607e4792; s-hidalgo-augmented-democracy-a0e788f205918f5c; s-hidalgo-augmented-democracy-d3c61d3ab756ba5d).
Why this category?
- Distinguishes aggregate estimation from individual-choice prediction, preventing a material overinterpretation of results.
- Includes held-out accuracy benchmarks and reported subgroup exceptions relevant to testing representational claims.
- Keeps conclusions bounded to the reported study and its comparisons.
The evidence is limited to one study and its participants, models, proposals, and comparison methods. The supplied summaries do not establish that aggregate gains cause or predict individual accuracy, or that either result generalizes to public decisions.
This is an AI assessment of contribution to the goal, separate from whether the question has an answer.
- Within the reported study, how should aggregate sample-estimation gains be interpreted alongside individual-choice accuracy, including the reported subgroup exceptions?
- Individual choice predictions
- Testing whether a model improves a sample
- The data used in the comparison
- Small samples improved in this test
- Individual prediction results
- Models outperformed a simple rule
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 Any answer below is provisional; parts whose evidence changed or was withdrawn are withheld.
There is no current reviewed answer to show yet.
← Back to questions