Stated Preference¶
Core Idea¶
Stated preference infers what an agent values from what the agent reports when asked. The evidence is a solicited ranking, survey answer, willingness-to-pay statement, policy declaration, or other self-report rather than a costly choice made among real alternatives. This gives the method reach where no choice trace exists: hypothetical products, public goods, future contingencies, and alternatives the agent has never been able to choose.
The same reach creates its characteristic weakness. Reporting is often cheap, and the answer is produced through a designed elicitation channel. A response can therefore reflect hypothetical bias, strategic presentation, social desirability, misunderstanding, or the wording and scale of the instrument. Stated preference names the report-based access mode; it does not claim that the report is the underlying preference itself.
Broad Use¶
- Economics: contingent valuation and stated-choice experiments.
- Survey research: declared priorities, vote intention, satisfaction, and willingness to act.
- Psychology and health: self-reported attitudes, intentions, and treatment preferences.
- Product research: surveys about features or purchases not yet available.
- Machine learning: a user or model reports a ranking or policy instead of being observed acting.
Clarity¶
The abstraction separates a valuation from one method of accessing it. It also completes the evidence-channel pair with Revealed Preference: ask-and-infer versus watch-and-infer. Neither channel is automatically the truth; each has characteristic distortions and blind spots.
Manages Complexity¶
Preference evidence can be organized by channel rather than by instrument name. Reports supply hypothetical reach but weak incentives; enacted choices supply stakes but are constrained by menus, defaults, and opportunity. The distinction tells an analyst which errors to test before treating either output as the agent's valuation.
Abstract Reasoning¶
Represent the latent preference as \(P\), the elicitation channel as \(C\), and the recorded report as \(R=f(P,C)\). Inference from \(R\) to \(P\) is warranted only to the extent that the channel response is understood or controlled. Varying wording, incentives, or response format tests how much of the record belongs to the preference and how much to the channel.
Knowledge Transfer¶
The operation transfers anywhere a report substitutes for behavior: from contingent valuation to polling, patient choice, user research, and preference elicitation for intelligent systems. What transfers is the evidence rule and its trade: access to the hypothetical in exchange for reliance on a low-cost, channel-mediated signal.
Example¶
A city asks residents how much they would pay to preserve a wetland that has no market price. The answers expose valuations that purchases cannot reveal, but the estimate depends on the payment scenario, wording, and whether respondents face any real consequence. The result is a stated preference, not a market demand trace.
Relationships to Other Abstractions¶
Current abstraction Stated Preference Prime
Parents (2) — more general patterns this builds on
-
Stated Preference is part of, typical Elicitation Channel Contribution Prime
A stated-preference estimate usually contains a systematic contribution from the question, prompt, scale, or protocol through which the report was elicited.
-
Stated Preference presupposes Preference Prime
Stated Preference presupposes an underlying preference because the solicited report is evidence about an ordering, not the ordering itself.
Children (1) — more specific cases that build on this
-
Stated–Revealed Preference Gap Prime is part of Stated Preference
The gap contains the report-based estimate whose output is compared with behavior under a real choice opportunity.
Hierarchy paths (3) — routes to 3 parentless roots
- Stated Preference → Preference
- Stated Preference → Elicitation Channel Contribution → Observability
- Stated Preference → Elicitation Channel Contribution → Representation → Abstraction
Distinction from Neighbors¶
Revealed Preference infers valuation from costly observed choice and can be distorted by constraints or framing. Elicitation Channel Contribution explains why a report can partly reflect its instrument. Cheap Talk concerns strategic communication without binding cost; a stated preference can be sincere yet inaccurate or channel-shaped.
Notes¶
Promoted from prime_abstractions/v2_density_pilot/stated_preference.md. The
density-pilot alias say_do_gap was deliberately removed and reserved for the
separate Stated–Revealed Preference Gap identity.