Checked 14 October 2026. Format facts are from the GOV.UK guidance for the Civil Service Judgement Test (last updated 24 May 2024), the official CSJT Practice Test, the official video guide and Success Profiles: Civil Service behaviours (updated 29 January 2025), read on 14 October 2026. The worked scenario is from the ProSyllabus practice set; it is independent practice material and its ratings are ours, not official answers.
Anyone searching for Civil Service Judgement Test questions and answers should know one thing first: the real ones are not public. The Civil Service does not publish the test’s scenarios, and it does not release answers to them. What can be shown honestly is the format of each question type from the official guidance, the handful of official example questions that anyone can see, and a practice scenario worked all the way through on the test’s own four-point scale. This guide does those three things. The worked scenario is ours, from our practice set, and is labelled as ours throughout.
The short answer
No real Civil Service Judgement Test question has a published answer. The official practice test is not scored, and after the real test you are not told which answers were right.
GOV.UK says of the practice test: “The practice test is not scored.” Of the real test it says: “For security reasons, we cannot provide feedback on answers to specific questions.” So every set of “answers” you find online, including the 4 ratings worked on this page, is its author’s reading and not the Civil Service’s key.
Format: GOV.UK, Preparing for the Civil Service Judgement Test (last updated 24 May 2024), read 14 October 2026. The worked scenario is from the ProSyllabus practice set and is not official.
Are the real questions and answers published?
No. Four official statements, read together, settle it.
| Where | What it says |
|---|---|
| GOV.UK guidance, on feedback after the real test | “After taking the test, you will receive your score and automatically-generated feedback.” and “For security reasons, we cannot provide feedback on answers to specific questions.” |
| GOV.UK guidance, on the practice test | “The practice test is not scored. It is designed to show you the look and feel of the real test while keeping the questions easy to respond to.” |
| The official practice test, first section | “You will not receive any performance feedback on these example questions.” |
| GOV.UK guidance, on your own answers | “Your test answers should be your own; you must not ask for input from anyone.” |
So there are two kinds of official example question: those in the practice test, and those the guidance says the real test “also includes” at “the beginning of each section”. Neither comes with a published answer key. Anything sold or shared as “the real answers” is either made up or taken from a test that candidates are told to complete alone, and the guidance adds that hiring managers “may require candidates invited to interview to retake the test under supervised conditions”.
What a Civil Service Judgement Test question looks like
GOV.UK says: “There are two parts to the test.” The two parts ask different kinds of question and are answered in different ways.
| Part | What you are shown | How you answer | Weight |
|---|---|---|---|
| First part: typical behaviour and preferences | Statements about how you behave and what you prefer at work | You say how far you agree or disagree with each statement. The official practice test shows seven options, from Completely Disagree to Completely Agree | 15% of the overall CSJT score |
| Second part: workplace scenarios | A scenario, in text or as a short video, “along with a range of possible actions” | You give each of four actions a rating: Counterproductive, Ineffective, Fairly Effective or Effective | Not stated as a figure. The page says only that the first part is 15%; the remaining 85% is our subtraction |
For the second part the guidance gives the building blocks: “There are three scenarios for each behaviour, and each scenario asks you to rate four actions.” You see only scenarios “relevant to the job level you have applied for”. The test is not timed, and “most people take between two and four minutes to answer one scenario”.
The four ratings, as GOV.UK words them
| Rating | Official description (GOV.UK) |
|---|---|
| Counterproductive | An unacceptable action that would make the situation worse. |
| Ineffective | A poor action that would not help the situation. |
| Fairly Effective | A useful action that would be of some help to the situation. |
| Effective | A good action that would help to resolve the situation. |
These four names are the ones on the guidance page, and they are the ones used for the worked example below. The official video guide names the options differently, as “effective, slightly effective, ineffective, or counterproductive”. We print both because both are official, and we found no explanation for the difference. This guide uses the four names on the guidance page. Our tips guide goes through how to tell the four ratings apart.
The official example questions that anyone can see
Two official example statements for the first part are public. The practice test’s first example is: “I step in whenever someone in my team needs help.” The answer options shown under it are Completely Disagree, Strongly Disagree, Somewhat Disagree, Neither Agree or Disagree, Somewhat Agree, Strongly Agree and Completely Agree. The official video gives another: “I adapt to change quicker than other people”.
There is no right answer to learn for these. The practice test’s own advice is: “Be as honest as you can in your responses.” and “Think about your actual behaviour at work rather than how you would like to behave.” The guidance page gives a practical reason: “You should answer the questions honestly, as you may be asked about them at interview.” We do not offer model answers to self-assessment statements, and we would distrust anyone who does.
For the second part, the official practice test says it will show “an example of each potential question type”, including workplace situations “presented as video and text” and situations “presented as text”. We read its introduction and its first example question on 14 October 2026 and did not answer or submit anything, so this guide does not describe its scenarios. Open it yourself before the real test: that is what it is for. The GOV.UK guidance page itself prints no example scenario.
Worked example: A mailbox nobody owns
This is our practice scenario, not a real test item. It is scenario 1 of the 12 in the ProSyllabus Civil Service Judgement Test practice set, the part of the set that is free to try, and it is reproduced here word for word from the set: the scenario, the four actions, our rating of each on the test’s own scale and the reason for each rating. Behaviour: Managing a Quality Service.
The scenario
You are an administrative officer in a pensions enquiry team. Four of you answer a shared mailbox, and nobody has been given the job of sorting it. This morning it holds 140 unanswered emails. The oldest is 12 working days old, and the team's published standard is a reply within 10 working days. Your team leader, Asha, is on a training course until tomorrow but is reading her messages.
The four actions
| Action | What you could do |
|---|---|
| A | Move the emails that are more than 10 working days old into a separate folder, so that the main mailbox shows the team inside its standard. |
| B | Answer the oldest emails first, and message Asha today with the number waiting, the age of the oldest and a suggestion that the team agrees a daily rota for sorting the mailbox. |
| C | Carry on answering emails in the order they appear on your screen, and bring up the backlog at next week's team meeting. |
| D | Spend today answering the oldest emails yourself, without telling Asha or your colleagues about the backlog. |
Before you read on, decide a rating for each action: Counterproductive, Ineffective, Fairly Effective or Effective. On the full page the tool below does this for you; on a page without it, note your four ratings and compare them with the table that follows.
Rate it yourself: A mailbox nobody owns
Rate each of the four actions for the scenario above on the test’s own scale, then check against our ratings. Look at each action on its own: more than one can share a rating.
A. Move the emails that are more than 10 working days old into a separate folder, so that the main mailbox shows the team inside its standard.
B. Answer the oldest emails first, and message Asha today with the number waiting, the age of the oldest and a suggestion that the team agrees a daily rota for sorting the mailbox.
C. Carry on answering emails in the order they appear on your screen, and bring up the backlog at next week's team meeting.
D. Spend today answering the oldest emails yourself, without telling Asha or your colleagues about the backlog.
Scenario 1 of our practice set. The ratings and reasons are ours, not official Civil Service answers. Nothing you choose here is saved or sent.
Our ratings and the reason for each
| Action | Our rating | Reason |
|---|---|---|
| A. Move the emails that are more than 10 working days old into a separate folder, so that the main mailbox shows the team inside its standard. | Counterproductive | Our reading: the customers who have waited longest are now out of sight and will wait longer still, and the mailbox gives a false picture of the service. That makes the situation worse, not just no better. |
| B. Answer the oldest emails first, and message Asha today with the number waiting, the age of the oldest and a suggestion that the team agrees a daily rota for sorting the mailbox. | Effective | Our reading: the people outside the standard are helped first, the person who can change how the mailbox is run gets the facts today, and a workable fix is on the table. It deals with the breach and with its cause. |
| C. Carry on answering emails in the order they appear on your screen, and bring up the backlog at next week's team meeting. | Ineffective | Our reading: nothing changes for the customers already outside the standard, and the one person who could reorganise the work hears about it a week late. The situation is left as it was. |
| D. Spend today answering the oldest emails yourself, without telling Asha or your colleagues about the backlog. | Fairly Effective | Our reading: the customers who have waited longest get a reply, which is of real help. It stops short of Effective because the size of the backlog and the lack of an owner stay hidden, so it will build up again. |
The same four answers, set against the official definitions
| Rating | Official description (GOV.UK) | Action in this scenario |
|---|---|---|
| Counterproductive | An unacceptable action that would make the situation worse. | A |
| Ineffective | A poor action that would not help the situation. | C |
| Fairly Effective | A useful action that would be of some help to the situation. | D |
| Effective | A good action that would help to resolve the situation. | B |
In this scenario the four actions take four different ratings. That is a feature of this scenario and not a rule of the test. GOV.UK says: “Do not assume that only one action is either effective or ineffective - you can rate more than one action the same.” Do not hand out one of each rating by habit.
Reading the four answers as a ladder
Start at the bottom. Action A is Counterproductive because it changes the situation for the worse. In the words of our reason, “the customers who have waited longest are now out of sight and will wait longer still”, and the mailbox “gives a false picture of the service”. That matches the official wording of the lowest rating: an action “that would make the situation worse”. Action C is Ineffective because it changes nothing: “The situation is left as it was.” The official wording of Ineffective is an action “that would not help the situation”. Neither action helps. Only one of them does harm, and that is the whole difference between the two ratings.
Now the top. Actions D and B begin the same way: both answer the oldest emails first. They finish one rating apart because of what happens next. Action D stops there, and the reason given is that “the size of the backlog and the lack of an owner stay hidden, so it will build up again”. Action B adds a message to the team leader with the numbers and a suggested fix, so that it “deals with the breach and with its cause”. That is how our reading separates the two official descriptions at the top of the scale: an action “of some help to the situation” and an action “that would help to resolve the situation”.
Notice what the scenario gives you and what the higher-rated actions do with it. The oldest email is “12 working days old” against a standard of “a reply within 10 working days”, so the breach has already happened and the oldest emails are the urgent ones. Asha is away “but is reading her messages”, which is why telling her today is possible and why raising it at “next week's team meeting” means, in the words of our reason, that “the one person who could reorganise the work hears about it a week late”. And “nobody has been given the job of sorting it” is the cause, which only one of the four actions touches.
All of this is our reading of our own scenario against the published scale. It shows how the four official sentences can be applied to a concrete case. It is not how the Civil Service marks its own scenarios, which is not published.
The behaviour behind it: Managing a quality service
The scenario puts you in the role of an administrative officer and is written around Managing a quality service. GOV.UK defines that behaviour as: “Deliver service objectives with professional excellence, expertise and efficiency, taking account of diverse customer needs.” These are the five examples GOV.UK gives for it at Administrative Assistant and Administrative Officer level, word for word, each with our note on where the worked scenario touches it.
| GOV.UK example at AA and AO | Where the worked scenario touches it (our reading) |
|---|---|
| Gain a clear understanding of customers’ needs and expectations | The team’s published standard, a reply within 10 working days, is the expectation these customers have been given. The scenario starts with it already missed. |
| Plan, organise and manage your own time to deliver a high quality service which gives taxpayers a good return for their money | Choosing which emails to answer first is a decision about your own time. Three of the four actions differ on exactly that. |
| Act to prevent problems by identifying issues, reporting them and providing solutions | The action we rate Effective does all three: it counts the backlog, tells the team leader today and suggests a rota. |
| Keep customers and all colleagues up to date with progress | The action we rate Fairly Effective helps the customers and leaves the team leader and colleagues uninformed. |
| Show customers where to access relevant information and support that will help them to use services more effectively | Not touched by this scenario. |
The right-hand column is our reading, written after the scenario. It is not a claim that the real test is marked against these lines, and GOV.UK says of the examples: “There is no expectation that you need to demonstrate every part of each example to be successful.” The point for preparation is narrower: the published examples at your grade describe the behaviour the scenarios are written around, so it costs nothing to read them before you sit the test.
How to use a worked answer
A worked answer is useful for the reasoning and close to useless as something to memorise, because the scenario in front of you on the day will be a different one. Three habits carry over.
| Habit | Where it comes from | What it looks like in the worked example |
|---|---|---|
| Rate each action alone | GOV.UK: “When reviewing an action, look at it independently from the others.” | Each action gets its own rating and its own reason, judged against the scenario and not against the other three. |
| Ask whether the action helps at all, then how far | The four official descriptions: “make the situation worse”, “would not help”, “of some help”, “help to resolve” | Two of the four actions help and two do not. After that the rating turns on how far the help goes, or on whether harm is done. |
| Use every detail the scenario gives | Our reading; no official page says this | The facts in the scenario decide which actions are possible, which deal with the cause and which come too late. |
If your ratings differed from ours, look at the direction. This is our suggestion, not official advice: rating actions higher than our ratings again and again suggests you are giving credit for good intentions; rating them lower suggests you are marking down actions that do help, only not completely. One scenario cannot show a pattern, which is the case for doing more than one. Our guide to free practice lists what is available without paying, official and ours.
Why a practice key is not the real key
What a worked practice answer can and cannot tell you
- It can show how the four official descriptions apply to a concrete situation, which is the skill the scenario part asks for.
- It cannot tell you the Civil Service’s rating for any real scenario. We found no official key published, for the practice test or the real one.
- It cannot tell you how a near miss is scored. None of the official pages we read says whether a rating one step from the intended one earns partial credit.
- It is written for one job level. The real test shows “scenarios that are relevant to the job level you have applied for”, and what counts as a complete response at one grade may not be at another.
- It is text. The real test presents scenarios for three of the behaviours as short videos, and a job may show only video, only text or a mix.
- It leaves out the first part of the test. The self-assessment is 15% of the score and has no right answers to practise.
Quotations are from GOV.UK, Preparing for the Civil Service Judgement Test, read 14 October 2026.
For what the score is and why no pass mark is published, see CSJT pass mark and scoring. For reading the percentile and the feedback report, see what a CSJT score means. An older guide of ours, eight original practice scenarios, has further teaching scenarios written before the practice set; they are separate from it.
What we could not find
What we could not find
- Any real Civil Service Judgement Test scenario published with its intended ratings.
- An answer key or scoring for the official practice test. The guidance says it is not scored.
- How many example questions the official practice test or the real test contains.
- Which answers to the self-assessment statements score well. The only official advice is to answer honestly.
- How the scenario ratings are marked, including any credit for a rating one step away.
- A reason for the difference between “Fairly Effective” on the guidance page and “slightly effective” in the video.
- Which behaviours are shown as video and which as text.
Read on 14 October 2026: the GOV.UK guidance page, the introduction and first example of the official practice test, and the official video transcript. All are listed under Sources.
Frequently asked questions
Frequently asked questions
Where can I find Civil Service Judgement Test questions and answers?
The real questions and answers are not published. GOV.UK links an official CSJT Practice Test with example questions, which is not scored, and says the real test includes example questions at the beginning of each section. Worked answers on any other site, including this one, are that site’s own practice material.
Does the Civil Service tell you which answers you got right?
No. GOV.UK says you receive your score and automatically-generated feedback, and that for security reasons it cannot provide feedback on answers to specific questions.
What types of question are in the Civil Service Judgement Test?
Two. In the first part you say how far you agree or disagree with statements about your typical behaviour and preferences at work; this is 15% of the overall score. In the second part you read or watch workplace scenarios and rate each of four possible actions as Counterproductive, Ineffective, Fairly Effective or Effective.
What is an example of a Civil Service Judgement Test question?
The official practice test’s first example statement is “I step in whenever someone in my team needs help”, answered on a seven-point scale from Completely Disagree to Completely Agree. For the scenario part, this page works through one of our own practice scenarios; it is not a real test item.
Is there one correct action in each CSJT scenario?
Not in the sense of picking one. You rate all four actions, and GOV.UK says not to assume that only one action is either effective or ineffective: you can rate more than one action the same.
Are the answers on this page official?
No. The 1 scenario worked here comes from the ProSyllabus practice set. The ratings and reasons are our reading of the scale definitions on GOV.UK. We found no answer key published by the Civil Service.
Can I memorise answers for the Civil Service Judgement Test?
There is nothing official to memorise, because real scenarios and their ratings are not published and you see only scenarios for your job level. What carries over from practice is the method: rate each action on its own against the four official descriptions.
More in this series
- CSJT pass mark and scoring
- What a CSJT score means
- CSJT tips from the behaviours
- Seeing the Big Picture: examples by grade
- Changing and Improving: examples by grade
- Making Effective Decisions: examples by grade
- Leadership: examples by grade
- Communicating and Influencing: examples by grade
- Working Together: examples by grade
- Developing Self and Others: examples by grade
- Managing a Quality Service: examples by grade
- Delivering at Pace: examples by grade
- Free CSJT practice: what exists
- Civil Service Judgement Test: 8 original practice scenarios
- Fast Stream online tests: a different test with a different scale
- Fast Stream vs direct entry and the other routes
Sources
- official — GOV.UK: Preparing for the Civil Service Judgement Test (last updated 24 May 2024)
- official — Civil Service: official CSJT Practice Test (unscored)
- official — Civil Service: “Guide to the Civil Service Judgement Test” (the short video linked from the GOV.UK guidance, about four minutes, with English captions)
- official — GOV.UK: Success Profiles, Civil Service behaviours (updated 29 January 2025)
official = a document published by the conducting body. reported = a news or coaching site we could not check against an original. Where sources disagree this page says so rather than picking one.
Ready for more? 12 further scenarios, rated the way the real test asks
Original workplace scenarios at the level of the roles that use this test. Rate each action Counterproductive, Ineffective, Fairly Effective or Effective, then see the reason behind every rating. Start with the free one below.
- 12scenarios
- 48actions to rate
- 4behaviours
- 48ratings explained
Every rating comes with its reason, written against the scale definitions on GOV.UK and the Civil Service Code.
FreeTests: Managing a Quality Service
A mailbox nobody owns
Read the situation, then rate each of the 4 actions from Counterproductive to Effective. You then see the right rating and the reason for each. About 3 minutes · free · no sign-in.
A mailbox nobody owns
You are an administrative officer in a pensions enquiry team. Four of you answer a shared mailbox, and nobody has been given the job of sorting it. This morning it holds 140 unanswered emails. The oldest is 12 working days old, and the team's published standard is a reply within 10 working days. Your team leader, Asha, is on a training course until tomorrow but is reading her messages.
Opens when you unlock all
- The third transfer · Managing a Quality Service
- Two forms, one set of details · Managing a Quality Service
- Sixty applications by Friday · Delivering at Pace
- Two jobs, one afternoon · Delivering at Pace
- The system is down · Delivering at Pace
- Handed over at the last minute · Working Together
- One case in five arrives incomplete · Working Together
- The new starter who works from home · Working Together
- Explaining a refusal · Communicating and Influencing
- The guidance nobody absorbed · Communicating and Influencing
- An out-of-date figure in the room · Communicating and Influencing
- 11 more scenarios with 44 actions to rate on the test's own four-point scale
- 4 behaviours, three scenarios each, the way the real test is built
- The reason behind every rating, so you learn the scale and not the answers
- Your rating pattern: whether you mark too generously or too harshly, and what to fix
- Untimed, like the real test. Work through it in one sitting or several
Independent practice material. Not written or endorsed by the Civil Service. The real test shows only scenarios for your job level, may use video, and opens with a self-assessment section worth 15% of the score that has no right answers to practise. Results are a percentile against other applicants, and no answer key is published.
