Civil Service Judgement Test Tips From the Behaviours and Code

The scale is not “good, okay, bad, terrible”. Each rating has one official sentence, and the differences between the sentences are where the judgement lies.

Our reading of those four sentences, offered as a way to think and not as the marking scheme: the top two ratings are both for actions that help. The difference is between “some help” and helping “to resolve the situation”. The bottom two are both for actions that do not help. The difference is between an action that leaves things as they were and one that would “make the situation worse”. So for any action you can ask two questions in order: does it help at all, and then how far does it go, or how much harm does it do. The Civil Service does not publish how ratings are marked, so treat this as a reading aid only.

The guidance could not be plainer: “When reviewing an action, look at it independently from the others. Do not assume that only one action is either effective or ineffective - you can rate more than one action the same.” Candidates who have practised ranking tests tend to hand out one of each rating per scenario. Nothing in the CSJT asks for that. If two of the four actions would both help resolve the situation, both are Effective.

Three Civil Service judgement-style tests are described on official pages, and each is answered differently. Preparing with the wrong format in mind is an avoidable error.

Note the third rating in particular. The CSJT says “Fairly Effective”; the Fast Stream page says “Partially effective”. They sit in the same place on the scale, but they are the wording of two different tests, and a practice set written for one is not a copy of the other.

GOV.UK says the CSJT “measures your ability to demonstrate judgement and decision-making regarding specific Civil Service Behaviours, which have been selected as important for success in your chosen job”, and that “You will only be presented with scenarios that are relevant to the job level you have applied for.” The behaviours document says the same from the other side: “Civil Service Behaviours are specific to the grade level of the job role”, and “You will not be asked to demonstrate all Civil Service Behaviours for one role.”

So the single most useful preparation is to open the job advert, note which behaviours it lists, and read the published examples for those behaviours at that grade. These are the nine behaviours and their official definitions.

The same source as tables, for readers on a page where the tool does not load. Each table shows the first two examples GOV.UK lists for each behaviour at that grade; the full lists are in the tool above and on GOV.UK. The document has six grade bands; the four below the Senior Civil Service are shown here, and the tool has all six.

Read one behaviour across the grades and the expectation changes in a way that matters for a scenario. Take Making effective decisions, using the first example GOV.UK gives at each band.

Our reading: at AA and AO the example is about using guidance and asking colleagues for input; at EO it is about taking responsibility for a timely decision; at HEO and SEO it adds empowering others; at Grade 7 and Grade 6 it starts with clarifying stakeholder needs before deciding. An action that is exactly right for one grade (ask a colleague before deciding) can be an incomplete answer at another, where the published examples expect you to own the decision. That is our inference from the descriptors, not a statement about how any scenario is keyed. It is the reason to study your own grade and not a general list.

The Civil Service Code sets out four core values and the standards under each. Its definitions are short enough to learn.

The CSJT guidance does not mention the Code, and we are not claiming the test is marked against it. Our suggestion is narrower. The Code describes what every civil servant must and must not do, so an action in a scenario that would mean hiding an error, misleading someone, ignoring an inconvenient fact or favouring one person unfairly is hard to describe as one that would “help to resolve the situation”. Use the Code to spot actions that cross a line, then use the behaviours to judge the rest.

The Code also says what to do when something is wrong: “If you have a concern, you should start by talking to your line manager or someone else in your line management chain.” That is a description of the proper route for a concern about conduct. It is not a rule that telling your manager is always the most effective action in a scenario; the behaviours at EO and above expect you to take responsibility for decisions that are yours.

The first part asks you to rate “how ambitious, motivated and flexible you are in the workplace”. It “makes up 15% of your overall CSJT score”, and the guidance says: “You should answer the questions honestly, as you may be asked about them at interview.” There is no published guidance on which answers score well, and we will not invent any. The practical point is consistency: describe yourself as you would be prepared to describe yourself, with examples, to an interview panel.

The two official pages disagree on how many times you can relaunch the test, and we print both. The safe plan is the stricter one: assume three launches in total, start only when you have an uninterrupted hour or more, and do not leave the page idle for long.

The practice test says what it contains: “This practice test comprises assessments that measure: Your preferences and behaviours.” It then lists decisions about typical workplace situations “presented as video and text” and decisions “presented as text”. For the first part, the example statement is “I step in whenever someone in my team needs help”, and the options run from Completely Disagree to Completely Agree on a seven-point scale. Its own tips for that part are: “Think about your actual behaviour at work rather than how you would like to behave” and “remember that the test is not timed or trying to trick you”. It adds that you can view the practice test “as many times as you need”, and that no personal information is needed to open it.

On the videos in the test itself: scenarios for three of the behaviours are presented as short videos. Each video has subtitles and British Sign Language translation, written transcripts are available within the test, and “each video can be replayed as needed”. Replaying costs nothing on an untimed test.

The preparation list on the guidance page is short and worth following exactly. It is reproduced in the checklist below, with the technical points from the same page.

If you have a disability, the reasonable-adjustments guide says to tell the recruiter what help you need. Its case studies include a candidate who was given a one-to-one supported session with a trained support worker for the Judgement Test, and another whose test was reset after she became unwell and was timed out. Transcripts and subtitles are available to every candidate without asking.

One more rule belongs here because it protects your result: “Your test answers should be your own; you must not ask for input from anyone.” Hiring managers “may require candidates invited to interview to retake the test under supervised conditions”.