Practice for employer tests
Situational judgement test practice: scenarios with the reasoning for every option
Practise 12 work scenarios and see why each response is rated more or less effective. Free, and your answers are never sent.
What this page is and is notIs: free practice in a similar format · reasoning for every optionIs not: an employer's test or its answers · a prediction
Is
- Free practice in a similar format to employers' situational judgement tests
- Original scenarios for contact-centre, office and retail jobs
- Reasoning shown for every option
- The key comes from three separate runs of an AI model, each blind to the others, reconciled by an editor; no human expert reviewed it.
- Your answers are never sent. Your sitting, finished or not, is kept in this browser for 48 hours so you can continue or see the result again; Start over or Forget now erases it.
Is not
- An employer's test or its answers
- The Civil Service judgement test
- A prediction of your result
- A certificate
- Affiliated with any test publisher or employer
- Legal or career advice
How the two formats work
Every scenario describes a work problem and four possible actions. Answer using only the information given, as the instruction says.
Most and least effective: choose the most effective and the least effective action, using only the information given. You score up to 2 points for each pick: 2 when it matches our key, 1 when you pick the action our key rates next in line.
Rate each action: give every action one of four ratings, Counterproductive, Ineffective, Fairly effective or Effective. Each action scores 1 point when your rating matches our key and half a point when it is one step away.
No points are taken off, so each scenario is worth 0 to 4 points. There is no time limit: the clock shows the time so far, and you can switch on a pace of 2.5 minutes a scenario and switch it off again. Nothing submits when that time runs out.
The result says how many points agree with our key, describes each skill in words when a sitting has at least three scenarios on it, and lists every scenario with the reasoning for all four actions. Employers key their own tests for their own jobs; this does not predict your score on any employer's test.
Customer service and call centre SJTs
Contact-centre scenarios are calls, chats and emails: a billing mistake, a cancellation, an angry caller, a request that policy does not allow. They test whether you serve the customer, fix the problem, listen before acting, balance policy against the customer, calm a heated contact and know when to pass something on.
You are a phone agent for an energy supplier. A customer is shouting that his bill is "a robbery" and says nobody has helped him in three calls. You have not yet looked at the account. The customer keeps interrupting you when you start to speak. He also says he is on a short break from work and has only ten minutes to talk.
Choose the most effective and the least effective action, using only the information given.
Our key rates option B Effective because the caller first feels heard and then his ten minutes go on the bill itself.
Our key rates option D Counterproductive because it promises to remove a charge before anyone has read the account. A promise you may not be able to keep turns an angry caller into one who also distrusts you.
Options A and C are both Ineffective: one sounds scripted, the other tells an angry caller to calm down. In this set, our key tends to rate highly the actions that hear the customer out and then act on facts.
Practise the contact centre and customer service set · Build your call centre interview introduction
Retail and hospitality SJTs
Retail and hospitality scenarios happen on the shop floor, at the till, in the stockroom and behind a café counter. They test customer focus, when to stop a task to help someone, honesty with the till and the stock, reliability with shifts, safety and teamwork.
You are walking through the shop with a box of stock when you see a drink has spilled across the floor of a busy aisle. Nobody has noticed. Shoppers are walking past, including a woman with a pushchair. The cleaning trolley is at the far end of the shop, about two minutes away. You are in a hurry because the stockroom needs the box in five minutes. The spill is about the size of a dinner plate.
Choose the most effective and the least effective action, using only the information given.
Our key rates option A Effective because someone guards the spill from the first second while a colleague brings a sign and cleaning. The cost is a short delay to the box.
Our key rates option D Counterproductive because a jacket or paper towels hide the liquid without making it safe, and leave nobody responsible for cleaning it.
Fetching the trolley yourself (B) and passing the message on (C) are both Ineffective: the spill is left unguarded while people keep walking through. In safety scenarios, our key rates highest the action that makes the hazard safe now, not the one that finishes your own task first.
Practise the retail and hospitality set · Practise a personality questionnaire for jobs
Graduate and office SJTs
Office and graduate scenarios cover clashing deadlines, a mistake found late, a plan that falls through, a colleague who disagrees, a setback and a request that bends the rules. They test prioritising, quality, problem solving, working with others, resilience and integrity.
You are an analyst in a finance team. At 1 pm your line manager asks you to finish a budget summary by 5 pm for the board papers. At 2 pm a manager from another team asks you to pull a data extract "as soon as possible" for a routine check. You cannot finish both by 5 pm, and your line manager is now in meetings until 4 pm.
Choose the most effective and the least effective action, using only the information given.
Our key rates option A Effective because it protects the fixed 5 pm board deadline and gives the other manager an early, honest answer and a time.
Our key rates option D Counterproductive because it promises a 3 pm delivery you cannot make alongside the summary, which puts the bigger deadline at risk.
Option C, asking your line manager, is Fairly effective: sensible, but the reply may come too late. Option B is Ineffective because the other manager hears nothing for hours. In this set, our key tends to favour early, honest messages over silent work or promises you cannot keep.
SJT questions and examples, explained
An SJT question gives you a short scene and asks you either to choose the most effective and the least effective action or to say how effective each action is. The instruction to use only the information given matters: do not invent a supervisor who is not mentioned or a rule the scene does not state.
Read each option for what it does to the problem, to the person in front of you and to the team. Then check three things: whether you have the authority to do it, whether it is honest, and whether it leaves anyone without information they need.
Here is a rating question, the format where each action gets its own rating:
You work in a café. A customer orders a drink and tells you he has a serious nut allergy. You are not sure whether the syrup he asks for contains nuts. The label is unclear, and the café is busy with a queue. The manager is nearby making coffee. The customer is polite and says he usually avoids syrups. He is willing to wait but is clearly worried about getting it wrong.
How effective is each action? Use only the information given.
Our key uses all four ratings here. Checking with the nearby manager (A) is Effective because the actual question gets answered before anything is served. A plain drink (C) is Fairly effective: honest, but nothing is checked. Turning him away (D) is Ineffective, and guessing (B) is Counterproductive because it risks a serious reaction to save a little time.
In our key, the actions rated highest tend to be honest about what you do not know, use help that is close by, and keep the customer informed. Actions that promise more than you can deliver, or that hide a problem, are rated lowest.
The Civil Service judgement test format, explained
As of November 2026: GOV.UK says the Civil Service Judgement Test measures judgement and decision-making on specific Civil Service behaviours. There are three scenarios for each behaviour, and each scenario asks you to rate four actions on this scale:
| Rating | How our key uses it |
|---|---|
| Counterproductive | The action makes the problem worse. |
| Ineffective | The action does little to fix the problem. |
| Fairly effective | The action helps, with a clear drawback. |
| Effective | The action deals well with the problem. |
Behaviours: the Civil Service describes nine behaviours in its Success Profiles, among them Making effective decisions, Working together, Managing a quality service and Delivering at pace. A job advert says which behaviours its tests look at.
Timing: GOV.UK says the test is not timed, and that most people take between two and four minutes a scenario.
Self-assessment: GOV.UK says a self-assessment section makes up 15% of the overall score. The score is a percentile against a comparison group, and candidates get automatically generated feedback.
Practise the format here: about four in ten of our scenarios use this rating scale. Our scenarios are not Civil Service items, and our key is not the Civil Service key. This page is not the Civil Service test.
GOV.UK's own free practice tests · GOV.UK's guide to preparing for the judgement test
How our key was made
The key comes from three separate runs of an AI model, each blind to the others, reconciled by an editor; no human expert reviewed it.
Each run rated every option without seeing the writer's draft key. The key is the median of their ratings, checked against published SJT guidance.
A scenario was kept only when the runs' ratings for every option were within one point of each other. Keeping only scenarios the runs agreed on may make the bank easier than a typical employer test. Our key is not any employer's key.
What employers do with SJTs
The U.S. Office of Personnel Management describes SJTs as a description of a work problem or critical situation. It says items are typically built from real incidents described by people who know the job, and that scores rest on their judgements of the best and worst responses.
So an employer's key is written for that employer's job. The same action can be rated differently in a different workplace.
OPM says SJTs look at how people handle work with others, such as conflict, problem solving and teamwork, and that they can be given on paper, by computer or by video.
A 2022 re-analysis by Sackett, Zhang, Berry and Lievens estimated the average validity of SJTs at .26 (ρ): how closely scores tend to line up with later job performance across many studies, where 1 would be a perfect match and 0 no link. That describes SJTs in general, not any one test or your chances.
How hiring tests work: each step and what it measures · Adjustments for hiring tests: how and when to ask
Questions people ask
Is this the Civil Service judgement test?
No. About four in ten of our scenarios use the same four-point rating format, but the scenarios and the key are ours. For the real format, use GOV.UK's own free practice tests.
Does an SJT have one best response?
Employers key their own tests for their own jobs, so the best response can differ from one employer to another. Our key shows which responses our editorial process rated more or less effective, and why.
Is it timed?
No. The clock shows the time so far. You can switch on an optional pace of 2.5 minutes a scenario, and nothing submits when it runs out.
Does my result predict my score?
No. It is a practice signal: how many points agree with our key. Employers key their own tests for their own jobs; this does not predict your score on any employer's test.
Who wrote and keyed the scenarios?
Our editorial process. AI writers drafted original scenarios; How our key was made, above, says how the key was set.
Is anything sent or saved?
Your answers are never sent. Your sitting, finished or not, is kept in this browser for 48 hours so you can continue or see the result again; Start over or Forget now erases it.
Sources
- SJTs present applicants with a description of a work problem or critical situation. Items are typically built from critical incidents described by people who know the job, and scores are based on their judgements of the best and worst alternatives. SJTs assess social functioning such as conflict management, interpersonal skills, problem solving and teamwork, and can be given on paper, by computer or by video. U.S. Office of Personnel Management, Situational Judgment Tests. Checked
- The test measures judgement and decision-making on specific Civil Service behaviours. There are three scenarios for each behaviour, and each scenario asks the candidate to rate four actions as Counterproductive, Ineffective, Fairly Effective or Effective. The test is not timed; most people take between two and four minutes a scenario. A self-assessment section makes up 15% of the overall score. The score is a percentile against a comparison group, and the practice test is not scored. GOV.UK, Preparing for the Civil Service Judgement Test (last updated 24 May 2024). Checked
- Lists the Civil Service Judgement Test among the online tests, links a practice test for each, and says the score is compared with a peer group who took the same test to give a percentile; meeting the minimum percentile is no guarantee of an invitation to the next stage. GOV.UK, Civil Service online tests (last updated 27 January 2026). Checked
- Nine behaviours: Seeing the big picture; Changing and improving; Making effective decisions; Leadership; Communicating and influencing; Working together; Developing self and others; Managing a quality service; Delivering at pace. GOV.UK, Success Profiles: Civil Service behaviours (last updated 29 January 2025; Open Government Licence v3.0). Checked
- Table 3 validity estimates (ρ): situational judgement tests .26; integrity tests .31; conscientiousness .19 in general and .25 when work-contextualised. Sackett, Zhang, Berry and Lievens (2022), Journal of Applied Psychology 107(11), 2040-2068. Checked
Written and checked by the MyTestAtlas editorial process. How we check. Spotted an error? Tell us.