How Changing and Improving is scored
Behaviour statements are usually scored on a 1–7 scale. This guide explains what each band looks like for Changing and Improving, what lifts a 4 to a 5 or beyond, and how moderation keeps marking consistent.
The 1–7 scale in brief
Most departments score each behaviour from 1 (no or negative evidence) to 7 (excellent evidence), with 4 as the typical pass mark for a single behaviour. The exact thresholds and how scores combine vary by campaign, so check the job advert and any candidate guidance you are sent.
Assessors score against the level descriptors for the grade. They are not comparing you with other candidates; they are matching your evidence to a standard.
Scoring is evidence-based, so a modest example told precisely can outscore an impressive one told vaguely. Assessors are matching what you wrote to the descriptors, line by line.
What each band looks like for this behaviour
| Score | Band | What the evidence looks like |
|---|---|---|
| 1–2 | Poor | Change happened to you; no personal action or a negative tone |
| 3 | Limited | Some personal involvement, but vague actions or no result |
| 4 | Acceptable | You suggested or supported an improvement, with a clear action and outcome |
| 5 | Good | You initiated the improvement, adapted to feedback, and measured the result |
| 6 | Strong | As 5, plus you involved others and embedded the change so it lasted |
| 7 | Excellent | As 6, with wider impact: adoption elsewhere or a clear organisational benefit |
What lifts a 4 to a 5
The jump from acceptable to good is usually about ownership and adaptation. A 4 says what you did. A 5 shows that you drove it: you noticed the problem yourself, asked for views, and changed course when something was not working.
A measured result matters here too. 'Waste fell by 40%' lets an assessor justify a 5; 'it worked well' forces them back towards 4.
What earns a 6 or 7
Higher bands need evidence of scale or lasting effect. A 6 typically shows you brought others with you — training, briefing, handling resistance — so the change survived. A 7 adds reach: your improvement was adopted by other teams, or it changed how the organisation measures or delivers something.
Tip
Do not inflate your result to chase a 7. Assessors see exaggerated claims at interview. A genuine, well-evidenced 5 beats a story that collapses under one follow-up question.
How moderation keeps it fair
Statements are often scored by more than one assessor, and panels compare notes where scores differ. Many campaigns also sample-check marking across assessors, so one marker's generosity or harshness does not decide your outcome.
Your behaviour scores are then combined with the other elements in the advert, such as a personal statement or experience questions, and candidates above the cut-off go forward. Strong marks in one behaviour rarely compensate for a missing one — give every behaviour your best example.
This is also why the instructions in the advert matter so much. If it asks for a specific level or word count, meeting that instruction is the first thing the panel checks.
Frequently asked questions
It varies by campaign. Four is a common benchmark per behaviour, but competitive campaigns can have higher effective cut-offs because scores are ranked. The job advert or candidate pack usually explains how scoring works for that recruitment, so read it carefully before assuming.
No. Assessors score the evidence of behaviour, which lives mainly in your actions. A spectacular result with vague actions tends to score around 3, because there is no proof you caused it. Spend most of your words on what you did, in first person.
Most Civil Service behaviour sifts use a 1–7 scale, but some campaigns use different bands or score behaviours in combination. The approach on this page reflects common practice as of July 2026; always follow the scoring guidance in your specific job advert.
Next step
Get the free STAR template — the same structure used in every example on this site.
Download the template pack