Study guide
Independent Rating Rules
College of Policing National Sift Police Constable Online Assessment
Track your progress as you read
Finished sections get a check mark.
1 Introduction & Exam Overview
About the College of Policing National Sift Eligibility, CVF Level 1, and What Is Assessed How to Use This Guide
2 Format, Scoring & Policies
SJT and BSQ Format Overview Practice Area, Platform, and 7-Day Window Scoring, Percentiles, and Results Resits, Reasonable Adjustments, and Appeals
3 CVF Framework & Values
CVF 2024 Overview and Code of Ethics Value: Courage Value: Respect and Empathy Value: Public Service
4 CVF Competencies I
We Are Emotionally Aware (Level 1) We Take Ownership (Level 1) We Collaborate (Level 1)
5 CVF Competencies II
We Support and Inspire (Level 1) We Analyse Critically (Level 1) We Are Innovative and Open-Minded (Level 1)
6 SJT Mechanics
SJT Structure and Timing The Five-Point Effectiveness Scale Independent Rating Rules
7 SJT Decision Strategy
Priorities, Consequences, and Who Is Affected Professionalism and the CVF Lens Common SJT Traps and Myths
8 Public-Facing Scenario Themes
Community Concerns and Anti-Social Behaviour Vulnerability and Safeguarding Mindset De-escalation, Fairness, and Public Service
9 Team & Ethics Scenario Themes
Colleague Performance and Ownership Scenarios Collaboration and Prioritisation Scenarios Ethics, Integrity, and Challenging Unprofessional Behaviour
10 BSQ & Integrated Prep
BSQ Format, Scale, and Ranking Honesty, Consistency, and Self-Profiling Study Plan and Next Steps After the Sift
Independent Rating Rules
Key Takeaways Explain This Topic AI
Rate each of the four actions independently of the other actions—do not convert the task into a forced rank order by default.
You may allocate the same rating to more than one action when that judgement is appropriate.
You do not have to use the full range of the rating scale for the actions in a scenario.
Base ratings only on scenario information and focus on what is most professional, respectful, and appropriate; no policing experience is required to rate well.
Avoid artificial distribution (spreading 1–5 just to ‘look balanced’); that is rank-ordering disguised as rating.
Last updated: August 2026
The Rule Set You Must Not Freestyle
The five-point scale is only half of SJT mechanics. The other half is how ratings relate to each other . College of Policing candidate guidance states the process rules in plain terms. Learn them as non-negotiable operations—not optional tips.
Official independent rating rules (candidate guidance)
Rate each action independently of the other actions.
You can allocate the same rating to more than one action if appropriate.
You do not have to use the full range of the rating scale for the actions in a scenario.
Base judgements on information in the scenario only.
Focus on what is most professional, respectful, and appropriate .
No policing experience is required to rate well.
Add the structure rule from the first section: even if you would take a different unlisted action, still rate the four provided .
Rule Correct behaviour Common violation
Independent rating Judge action B without forcing it to differ from A Ranking A best, B second, C third, D worst by habit
Same rating allowed Two actions both “effective” if both reasonably help Forcing unique ranks so every action gets a different number
Full range not required Four ratings that only use two bands if that is honest Spreading 1–5 artificially “to look thoughtful”
Scenario-only facts Use the stem as written Importing TV procedure or unstated offences
Professional lens Prefer respectful, appropriate, public-interest first responses Choosing aggressive theatrics as “more police-like”
No experience required Reason from CVF-level professionalism and the stem Assuming only ex-officers can see effectiveness
Independent Rating vs Disguised Ranking
Many aptitude tests train people to rank options. The national sift SJT is an independent rating task. Ranking can be a private mental check after you rate, but it must not replace independent impact judgements.
- What independent rating sounds like internally
- “Action A as a first response: does it help, fail to help, or make things worse?”
- “Action B as a first response: same questions, using the same scale definitions.”
- “If A and B both reasonably help, they can both be Effective.”
- What disguised ranking sounds like
- “I already used Very effective on A, so B cannot also be Very effective.”
- “I need one of each label or the system will flag me.”
- “D has to be Counterproductive because something has to be worst.”
Those thoughts are process errors . Official rules explicitly allow same ratings and explicitly say you need not use the full scale range .
Artificial distribution: the quiet trap
Artificial distribution means forcing variety across the scale so your four ratings “look balanced.” Example of a bad habit:
- Action Forced spread habit Honest independent alternative
- A Very effective Effective
- B Effective Effective
- C Slightly effective Slightly effective
- D Counterproductive Ineffective
In the forced-spread column, the candidate demoted or promoted actions so that labels would differ , not because impact demanded it. That is rank-ordering wearing a rating costume.
Worked Contrast: Same Scenario, Two Processes
Teaching stem (not an official keyed item): A member of the public is upset about a delayed appointment. They are raised-voice frustrated but not threatening. You can resolve the delay with a short explanation and a rebook.
Four teaching actions:
Listen, apologise for the inconvenience, explain the delay factually, and rebook promptly.
Offer a clear rebook and a note of who to contact if problems continue.
Tell them to “calm down or leave” without addressing the appointment.
Walk away mid-sentence to chat with a colleague about weekend plans.
Process 1 — Independent rating (correct approach)
- Action Impact reasoning (framework) Independent label
- 1 Directly resolves need with respect Effective or Very effective
- 2 Also resolves need professionally; may be similarly strong Effective or Very effective (may match action 1)
- 3 Escalates tone, fails to fix the problem, disrespectful Counterproductive
- 4 Ignores the person and the task Ineffective or Counterproductive (worsens experience)
Notice actions 1 and 2 can share a high rating. You are not required to invent a difference.
Process 2 — Forced ranking habit (incorrect)
- Action Forced unique ranks Why this is wrong
- 1 Very effective OK if truly best-in-set
- 2 Effective Demoted only to avoid matching action 1
- 3 Slightly effective Softened despite clear harm
- 4 Ineffective Maybe under-calls harm to force a ladder
The second process optimises for pretty spreads , not truthful effectiveness . Official rules forbid needing that ladder.
Same Rating Allowed: When It Is Appropriate
Use the same rating when two (or more) actions produce similar impact quality on the situation.
- Appropriate same-rating cases
- Two different wordings of a respectful, helpful first response
- Two weak options that both fail to help without clearly worsening things → both Ineffective
- Two harmful options that both escalate unfairly → both Counterproductive
- Inappropriate “same rating” laziness
Do not give all four actions Effective because you are tired. Independence is not indifference. Each action still needs its own impact check. Same ratings are allowed when deserved—not as a bulk default.
Full Scale Range Not Required
You might face a scenario where all four listed actions are poor. You might face one where three are solid and one is weak. Official guidance does not require you to manufacture a Counterproductive or a Very effective just to “complete the set.”
Scenario pattern (teaching) Honest rating shape
One strong, one moderate, two harmful e.g. Very effective / Slightly effective / Counterproductive / Counterproductive
- Two similarly good, two similarly useless e.g. Effective / Effective / Ineffective / Ineffective
- All modest help only e.g. Slightly effective across several actions
- One catastrophic, three merely weak Counterproductive once; Ineffective elsewhere—if that matches impact
The pattern must come from the actions , not from a template you apply to every item.
Professional, Respectful, Appropriate—Without Prior Experience
Guidance directs you to focus on what is most professional, respectful, and appropriate . That is the CVF-aligned compass at practitioner level:
- Professional: calm, fair, ownership-minded, evidence-based on the facts given
- Respectful: dignity for the public and colleagues; no humiliation “lessons”
Appropriate: proportionate first response; neither reckless cowboy action nor total avoidance when help is needed
Why “no experience required” is liberating
If you have never worn a uniform, you can still judge:
- whether mocking a vulnerable person makes things worse
- whether listening and clarifying next steps helps
- whether ignoring a clear problem fails to resolve it
- whether inventing punishments not supported by the stem is inappropriate
If you do have experience, use it only to recognise professionalism—not to override the stem with unstated powers or local customs.
A Repeatable Four-Step Method Per Scenario
Use this sequence every time until it is automatic:
Read the stem for constraints — who is affected, what is known, what is not happening, first-response horizon.
Cover the other actions mentally — rate action 1 alone against the official scale table.
Rate actions 2–4 alone — allow equal ratings; do not force unused bands.
Optional consistency glance — only after independent ratings, check you did not accidentally rank; adjust only if an impact reason was wrong, not to beautify the spread.
Micro-checklist before you leave a scenario
Did I use only stem facts?
Did each action get its own impact decision?
If two ratings match, is that because impact matches?
Did I avoid forcing all five scale points?
Would a respectful professional first response look like my high ratings?
Would my low ratings truly fail to help or actively worsen things?
Integrating Rules With the Scale Table
Keep both tools together:
- If independent impact is… Choose
- Inappropriate / worsens situation Counterproductive
- Poor / will not help Ineffective
- Small positive Slightly effective
- Reasonable help Effective
- Among the best responses Very effective
Independence decides whether two actions can share a row . The scale decides which row each action earns.
Final mechanical myths to delete
“The algorithm requires one Very effective per scenario.” — Not an official rating rule.
“Two Effective ratings cancel each other.” — Same ratings are explicitly allowed.
“I must use Counterproductive somewhere.” — Full range is not required.
“Experienced officers should invent extra facts.” — Scenario-only information still binds everyone.
“Independent rating means I never compare quality.” — You may notice relative quality, but you still assign absolute effectiveness labels, not forced unique ranks.
When structure, scale, and independence work together, the SJT becomes a clean measure of judgement. Later strategy chapters will deepen what professional priorities look like; this section ensures how you enter those judgements into the platform matches College of Policing rules exactly.