Section 422 of 440
Complete canonical tutorial. This reader section contains the same teaching body as PWR-240 · Policy adaptation. Open the Power dossier.
PWR-240 · SUPERVISED full tutorial
Adapt an authorised policy through evidence, consultation, implementation checks and a reversible version change
This lesson teaches the full supervised policy-adaptation method: define the problem and current version, confirm mandate, state the theory and evidence, compare options including no change, assess distribution and rights, consult affected people, decide with reasons, implement as a versioned change, monitor outcomes and harms, and review, reverse or sunset. Practice uses a fictional policy.
1 · Permission and limits
Know exactly what you may do
2 · Get ready
Gather what you need and check the starting conditions
What you need
- Qualified facilitator and fictional reading-room policy v1 limiting bookings to one hour.
- Supplied baseline data for four weeks: 40 one-hour bookable slots, 30 used and 10 unused late slots; six people requested two hours for an access reason, only two completed within one hour and four did not. Staff can support at most two two-hour bookings per week. Submissions say: access group A needs a longer-session route; staff group B warns against an open-ended queue; short-session group C wants one-hour capacity preserved.
- Problem/options table, impact and uncertainty screen, decision record, implementation measures, review trigger and v2 template. The supplied trial result is seven completions among eight longer-session requests, while median wait rises from two to four working days. The declared queue trigger is more than three working days, so the result requires adaptation rather than an unqualified success label.
Before you start
- Confirm the policy, institution, people and data are entirely fictional.
- Identify the fictional competent authority and which decisions remain outside it.
- Predeclare evaluation questions, distribution groups, adverse effects and reversal threshold before choosing an option.
3 · The method
Follow these steps in order
- Lock policy and problem
Quote the current version, purpose, rule, date and the observed problem without jumping to a solution.
Why: Version and problem clarity make change testable.
Check: A reader can distinguish policy text from complaint or hypothesis.
- Confirm mandate and rights limits
State who may change the rule, required consultation or review and non-negotiable access, consent or legal constraints.
Why: Good evidence cannot authorise the wrong body.
Check: The fictional authority and stop lines are explicit.
- Build the evidence chain
Separate observations, source quality, causal uncertainty, affected-person experience and missing data.
Why: A policy response can outrun the evidence.
Check: Every claim has a source, uncertainty or unknown label.
- State how change might work
Write the mechanism from rule to behaviour to outcome and identify implementation assumptions.
Why: Without a theory, failure cannot be interpreted.
Check: The chain includes possible unintended effects.
- Compare options
Include no change, non-policy support, limited trial, broader change and repeal; score expected benefit, burden, feasibility, rights and reversibility.
Why: One preferred option is not an evaluation.
Check: Each rejected option has a reason linked to evidence or impact.
- Assess distribution and risk
Ask who gains, pays, waits, loses access or bears enforcement; identify cumulative and accessibility effects.
Why: Average benefit can hide unequal harm.
Check: The impact table names affected and absent groups.
- Consult before decision
Give accessible information, a real open question and safe input routes; record disagreement and what remains outside influence.
Why: Participation must occur while options can still change.
Check: Every submission is traceable and no group is claimed represented by a token voice.
- Decide with reasons
The authorised role selects an option or no change, cites evidence, values, uncertainty and consultation, and states the appeal/review route.
Why: Reason-giving connects evidence to accountable power.
Check: The record shows what changed because of input.
- Implement as a version
Publish effective date, owner, training/communication, accessible formats, transition, complaints and rollback; retire conflicting v1 instructions.
Why: A decision is not adaptation until implementation is controlled.
Check: Only one current version directs the fictional service.
- Monitor and review
Measure uptake, access, burden, outcome and adverse effects; at the trigger choose keep, adapt, reverse or sunset and preserve the history.
Why: Adaptation includes stopping a harmful or useless policy.
Check: The review decision follows predeclared evidence rather than sunk cost.
4 · Worked example
See the whole method used once
Scenario
A fictional reading room limits reservations to one hour. Data show unused late slots, while accessibility submissions say some users need longer sessions.
Walkthrough
- The facilitator confirms the service director’s fictional mandate and separates booking data from the hypothesis that two-hour slots will improve access.
- Participants compare no change, supported exceptions and a four-week two-hour trial, then map staff burden, queue length and accessibility distribution.
- Accessible submissions favour a supported-exception plus trial; the director chooses it, gives reasons and publishes v2 with complaint and rollback routes.
- At review, access improves but queue delay exceeds the predeclared threshold, so the policy is adapted to reserved longer slots rather than simply labelled successful.
Result
The exercise shows evidence changing a version through monitored distribution and reversal logic. It does not validate a real policy or public consultation.
5 · Right and wrong
Compare correct or safer execution with the common wrong version
| Moment | Right / safer | Wrong / riskier | Why it matters |
|---|---|---|---|
| Defining the problem | Separate observed issue from proposed solution. | Write “policy must change” as the problem. | Starting with one preferred fix hides viable booking alternatives. |
| Consulting decision point | Ask while options and mechanism can change. | Seek comments after approval to demonstrate engagement. | Late consultation cannot influence the decision. |
| Evaluating decision point | Measure implementation, distribution, benefit and harm. | Count an evaluation report as adaptation. | Evidence has effect only if a governed decision responds. |
| Versioning decision point | Retire conflicting v1 instructions and keep history. | Publish v2 while old pages remain active. | Competing booking-rule versions create unequal access and staff practice. |
6 · Common mistakes
Spot the error and apply the correction
| Mistake | Fix |
|---|---|
| The active booking rule lacks a version and owner. | Quote the exact active rule, date and owner before analysis. |
| Unused late slots are treated as proof of mechanism. | Write alternative explanations and the implementation chain. |
| An average completion rate hides access and queue effects. | Break out access, burden, exclusion and adverse effects across affected groups. |
| Stakeholder submissions receive no reasoned feedback. | Show what input changed, did not change and why. |
| The trial has no queue trigger or rollback owner. | Set rollback/sunset authority and threshold before implementation. |
7 · Practice
Turn the steps into a usable skill
First session
- Define fictional v1, problem, authority and evaluation questions.
- Sort evidence, uncertainty and affected-person submissions.
- Compare options and distribution, including no change.
- Write the authorised fictional decision and versioned implementation plan.
- Apply supplied monitoring results and choose keep, adapt, reverse or sunset.
Repeat plan
The qualified policy/evaluation facilitator controls cases and complexity. Repeat with a new fictional rule after at least 48 hours; real policy work requires the legitimate authority and public/affected-person process.
Progress when
- Problem, theory, evidence and option comparison remain distinct.
- Affected-person input can visibly change the option or implementation.
- Monitoring covers distribution and adverse effects and triggers a real keep/adapt/reverse/sunset choice.
Do not progress when
- The case becomes a real institutional or public policy question.
- Jurisdiction, consultation rights, equality/access duties or decision authority is unclear.
- Participation would be coercive, tokenistic or unsafe.
8 · Check the result
Measure what changed
Traceable movement from policy evidence to a versioned review decision
How: Score mandate, problem, evidence, theory, options, distribution, consultation, reasons, implementation, monitoring, complaints and reversal across the fictional case.
Good result: A good result produces one current version and a review decision that follows the predeclared outcome/harm rules with reasons visible.
This does not prove: It does not prove causal benefit, legitimacy, fair implementation or authority to change a real policy.
Self-check
- What exact v1 rule and observed problem are being examined?
- Which evidence supports the mechanism and what alternative remains?
- Who gains or bears burden under each option?
- Which predeclared result causes reversal or sunset?
9 · Stop, adapt or get help
Keep the safety boundary practical
Stop and get help
- Stop if the exercise is used for a live employer, public body, service or affected community.
- Stop when competent authority, rights review, consultation or evaluation safeguards are missing.
- Route real legal, public, clinical, labour, safeguarding and policy matters to the accountable authority and independent advice.
Accessibility and adaptations
- Provide consultation information, submissions, reasons, complaints and review in accessible formats and needed languages.
- Offer asynchronous, private, supported and AAC input; public speaking is never the only route.
- Measure process access and distributional burden, not only average policy uptake.
10 · Evidence and limits
Why these instructions are here
- official guidance
The Magenta Book provides official principles for designing and interpreting policy evaluation; an evaluation report does not itself establish adaptation or benefit.
Magenta Book: Central Government guidance on evaluation - official guidance
The Orange Book supports risk, control, monitoring and review in governance, including explicit response to uncertainty and adverse effects.
The Orange Book: Management of Risk — Principles and Concepts
Limits
- All policy, data and participants are fictional.
- Evidence, authority, legitimacy and implementation are separate tests.
- A version change is not improvement until outcomes and harms are reviewed.
- An evaluation report cannot prove adaptation or benefit.
- Official evaluation and review frameworks do not establish that findings were valid, acted upon or beneficial; adaptation may be blocked by incentives and power.
Open the complete canonical research register
- Official normative system supportLimiting / contraryThe Orange Book: Management of Risk — Principles and Concepts
HM Treasury · 2026 · Official standard
- Official normative system supportLimiting / contraryMagenta Book: Central Government guidance on evaluation
HM Treasury and Evaluation Task Force · 2026 · Official standard
- Official normative system supportLimiting / contraryGuidance for after action review (AAR)
World Health Organization · 2019 · Official guidance
Read the complete evidence interpretation on the Power dossier.
Tutorial delivery controls
Learn, adapt, troubleshoot and resume
Progress is saved only in this browser on this device.
Step-by-step learner mode
Each activity includes its success check, a nearby accessible alternative and an “I’m stuck” correction path. Alternatives preserve the target where possible; when they change the task, Titan labels them as related rather than equivalent.
Lock policy and problem
Quote the current version, purpose, rule, date and the observed problem without jumping to a solution.
Version and problem clarity make change testable.
A reader can distinguish policy text from complaint or hypothesis.
I’m stuck on this step
Reset: Re-read this authored instruction — “Quote the current version, purpose, rule, date and the observed problem without jumping to a solution.” — and its success check, then attempt only this step.
Possible snag: The active booking rule lacks a version and owner.
Correction: Quote the exact active rule, date and owner before analysis.
Stop / get help: Stop if the exercise is used for a live employer, public body, service or affected community.
Confirm mandate and rights limits
State who may change the rule, required consultation or review and non-negotiable access, consent or legal constraints.
Good evidence cannot authorise the wrong body.
The fictional authority and stop lines are explicit.
I’m stuck on this step
Reset: Re-read this authored instruction — “State who may change the rule, required consultation or review and non-negotiable access, consent or legal constraints.” — and its success check, then attempt only this step.
Possible snag: The result from “State who may change the rule, required consultation or review and non-negotiable access, consent or legal constraints.” does not yet meet this declared check: The fictional authority and stop lines are explicit.
Correction: Return to the start of “Confirm mandate and rights limits”, reduce complexity or pace, and repeat only the part needed to satisfy: “The fictional authority and stop lines are explicit.”
Stop / get help: Stop if the exercise is used for a live employer, public body, service or affected community.
Build the evidence chain
Separate observations, source quality, causal uncertainty, affected-person experience and missing data.
A policy response can outrun the evidence.
Every claim has a source, uncertainty or unknown label.
I’m stuck on this step
Reset: Re-read this authored instruction — “Separate observations, source quality, causal uncertainty, affected-person experience and missing data.” — and its success check, then attempt only this step.
Possible snag: The result from “Separate observations, source quality, causal uncertainty, affected-person experience and missing data.” does not yet meet this declared check: Every claim has a source, uncertainty or unknown label.
Correction: Return to the start of “Build the evidence chain”, reduce complexity or pace, and repeat only the part needed to satisfy: “Every claim has a source, uncertainty or unknown label.”
Stop / get help: Stop if the exercise is used for a live employer, public body, service or affected community.
State how change might work
Write the mechanism from rule to behaviour to outcome and identify implementation assumptions.
Without a theory, failure cannot be interpreted.
The chain includes possible unintended effects.
I’m stuck on this step
Reset: Re-read this authored instruction — “Write the mechanism from rule to behaviour to outcome and identify implementation assumptions.” — and its success check, then attempt only this step.
Possible snag: Unused late slots are treated as proof of mechanism.
Correction: Write alternative explanations and the implementation chain.
Stop / get help: Stop if the exercise is used for a live employer, public body, service or affected community.
Compare options
Include no change, non-policy support, limited trial, broader change and repeal; score expected benefit, burden, feasibility, rights and reversibility.
One preferred option is not an evaluation.
Each rejected option has a reason linked to evidence or impact.
I’m stuck on this step
Reset: Re-read this authored instruction — “Include no change, non-policy support, limited trial, broader change and repeal; score expected benefit, burden, feasibility, rights and reversibility.” — and its success check, then attempt only this step.
Possible snag: The result from “Include no change, non-policy support, limited trial, broader change and repeal; score expected benefit, burden, feasibility, rights and reversibility.” does not yet meet this declared check: Each rejected option has a reason linked to evidence or impact.
Correction: Return to the start of “Compare options”, reduce complexity or pace, and repeat only the part needed to satisfy: “Each rejected option has a reason linked to evidence or impact.”
Stop / get help: Stop if the exercise is used for a live employer, public body, service or affected community.
Assess distribution and risk
Ask who gains, pays, waits, loses access or bears enforcement; identify cumulative and accessibility effects.
Average benefit can hide unequal harm.
The impact table names affected and absent groups.
I’m stuck on this step
Reset: Re-read this authored instruction — “Ask who gains, pays, waits, loses access or bears enforcement; identify cumulative and accessibility effects.” — and its success check, then attempt only this step.
Possible snag: The result from “Ask who gains, pays, waits, loses access or bears enforcement; identify cumulative and accessibility effects.” does not yet meet this declared check: The impact table names affected and absent groups.
Correction: Return to the start of “Assess distribution and risk”, reduce complexity or pace, and repeat only the part needed to satisfy: “The impact table names affected and absent groups.”
Stop / get help: Stop if the exercise is used for a live employer, public body, service or affected community.
Consult before decision
Give accessible information, a real open question and safe input routes; record disagreement and what remains outside influence.
Participation must occur while options can still change.
Every submission is traceable and no group is claimed represented by a token voice.
I’m stuck on this step
Reset: Re-read this authored instruction — “Give accessible information, a real open question and safe input routes; record disagreement and what remains outside influence.” — and its success check, then attempt only this step.
Possible snag: Stakeholder submissions receive no reasoned feedback.
Correction: Show what input changed, did not change and why.
Stop / get help: Stop if the exercise is used for a live employer, public body, service or affected community.
Decide with reasons
The authorised role selects an option or no change, cites evidence, values, uncertainty and consultation, and states the appeal/review route.
Reason-giving connects evidence to accountable power.
The record shows what changed because of input.
I’m stuck on this step
Reset: Re-read this authored instruction — “The authorised role selects an option or no change, cites evidence, values, uncertainty and consultation, and states the appeal/review route.” — and its success check, then attempt only this step.
Possible snag: The result from “The authorised role selects an option or no change, cites evidence, values, uncertainty and consultation, and states the appeal/review route.” does not yet meet this declared check: The record shows what changed because of input.
Correction: Return to the start of “Decide with reasons”, reduce complexity or pace, and repeat only the part needed to satisfy: “The record shows what changed because of input.”
Stop / get help: Stop if the exercise is used for a live employer, public body, service or affected community.
Implement as a version
Publish effective date, owner, training/communication, accessible formats, transition, complaints and rollback; retire conflicting v1 instructions.
A decision is not adaptation until implementation is controlled.
Only one current version directs the fictional service.
I’m stuck on this step
Reset: Re-read this authored instruction — “Publish effective date, owner, training/communication, accessible formats, transition, complaints and rollback; retire conflicting v1 instructions.” — and its success check, then attempt only this step.
Possible snag: The trial has no queue trigger or rollback owner.
Correction: Set rollback/sunset authority and threshold before implementation.
Stop / get help: Stop if the exercise is used for a live employer, public body, service or affected community.
Monitor and review
Measure uptake, access, burden, outcome and adverse effects; at the trigger choose keep, adapt, reverse or sunset and preserve the history.
Adaptation includes stopping a harmful or useless policy.
The review decision follows predeclared evidence rather than sunk cost.
I’m stuck on this step
Reset: Re-read this authored instruction — “Measure uptake, access, burden, outcome and adverse effects; at the trigger choose keep, adapt, reverse or sunset and preserve the history.” — and its success check, then attempt only this step.
Possible snag: An average completion rate hides access and queue effects.
Correction: Break out access, burden, exclusion and adverse effects across affected groups.
Stop / get help: Stop if the exercise is used for a live employer, public body, service or affected community.
Correct versus incorrect execution
These accessible process diagrams are built from the tutorial’s own right/wrong teaching. They are not anatomical illustrations and do not add technique beyond the canonical tutorial.
Separate observed issue from proposed solution.
Write “policy must change” as the problem.
Ask while options and mechanism can change.
Seek comments after approval to demonstrate engagement.
Measure implementation, distribution, benefit and harm.
Count an evaluation report as adaptation.
Retire conflicting v1 instructions and keep history.
Publish v2 while old pages remain active.
Method-structure checklist
10 of 10 structural checks present
- Ordered, Power-specific instructions — present
- Every activity has a success check — present
- Materials or supplied records are declared — present
- Measurement or assessment rule is present — present
- Tutorial-specific troubleshooting is present — present
- Stopping or escalation boundary is present — present
- Every activity has an adjacent alternative — present
- Correct-versus-incorrect comparison is present — present
- Evidence context is bound to the Power record — present
- Planning metadata is present — present
The method-readiness band and presence checklist assess tutorial presentation and are separate from evidence quality for the underlying Power. They are automated editorial aids, not human approval.
Manual editorial sign-off: Pending. This tutorial must not display a human-approved state until an identified editor signs the exact content hash.