HUMAN MEASURES · PROFILE 0.2-H1

Ask four things.
Do not manufacture an answer.

At preregistered randomized intervals, participants answer one tiny instrument: keep, value, burden, and trust. Every due prompt remains in the denominator, nonresponse stays visible, and the result never enters W.

4EXACT QUESTIONS
5RESIDENCE PHASES
72HRESPONSE WINDOW
0RANKING EFFECT
01 / CANONICAL INSTRUMENT

Tiny by design.
Exact by necessity.

Changing the wording, scales, timing, or response rules after seeing results breaks comparability. All four answers travel together as one prompt set.

Q1 / KEEP

If you could remove the robot today at no cost, would you?

KEEPREMOVE
Q2 / VALUE

Has the robot made your life better or worse recently?

-2-1012
Q3 / BURDEN

How much work is the robot creating for you?

01234
Q4 / TRUST

How comfortable are you letting the robot operate without supervision?

01234
02 / LONGITUDINAL VIEW

Random moments.
Fixed analysis.

The schedule is frozen before hour one. Delivery can drift by at most 24 hours, participants have 72 hours to respond, and one neutral reminder is the maximum.

01STRANGER0–10H
02ASSISTANT10–100H
03COMPANION100–1000H
04HOUSEHOLD MEMBER1000–5000H
05INDISPENSABLE5000–10,000H
03 / LOCAL REPRODUCTION

Keep every prompt.
Rebuild every phase.

The checker validates schedule completeness, neutral collection, exact scales, explicit nonresponse, evidence bindings, and all overall and phase-level summaries.

LOCAL HUMAN-MEASURES REPRODUCERPROFILE / 0.2-H1
COLLECTION INTEGRITYINVALIDOne or more hard gates failed.
BEHAVIORAL PRIMARYPermanent voluntary rejection

Determines W and public rank.

ATTITUDINAL DIAGNOSTICKeep · value · burden · trust

Explains the trajectory; never changes W.