A situational judgement test (SJT) describes a moment at work, lists several things a person might do, and asks you to judge them. An employer sets one to see how you weigh customers, rules, colleagues and time before anyone has watched you work.
This guide is for the candidate: the four formats, how each is scored, what research on coaching, retesting and faking found, and how to practise honestly. It promises no pass and no score.
EdProcto publishes this guide, and its page for candidates opens the practice kits described below. For other kinds of employer test, and for test day, read how to prepare for a pre-employment test and aptitude test practice.
What is a situational judgement test, and which formats will you meet?
A situational judgement test is a set of short work situations, each followed by four or five possible responses, in one of four answer formats. The invitation or first screen says which.
| Format | What you do | Usual scoring | Example |
|---|---|---|---|
| Pick the best | Choose one response | A mark if it matches the key | EdProcto's "Situational judgement" kit |
| Best and worst | Choose the most and the least appropriate response | Often a mark for each end | Judicial Appointments Commission (n.d.) |
| Rank | Rank the responses from most to least appropriate | Marks by how near each sits to its keyed place | NHS England (2024) |
| Rate each | Give every response a rating | Marks by how far each rating is from the panel's | Government People Group et al. (2024) |
The last three are official pages. NHS England's practice paper for doctors says near misses in a ranking earn marks; the UK Civil Service test rates every action on four steps, from counterproductive to effective.
How is a situational judgement test scored?
A situational judgement test is scored against a key that people who know the job agreed beforehand. No textbook holds the answer: the key is a panel's view of what works in that role. How an employer builds one is in how to write a situational judgement test.
- Scoring methods vary, and you rarely see them. De Leng et al. (2017) applied 28 scoring methods to one rating test taken by 931 medical school applicants. They differed in whose ratings formed the key, how distance from it was measured and, most of all, how systematic error in a candidate's ratings was controlled.
- The result is usually relative. The Civil Service compares each score with a representative group of applicants and reports a percentile: at 44, you did better than 44% of that group (Government People Group et al., 2024). It gives no feedback on single answers.
What each format means for how you answer
- Pick the best. One decision, all or nothing. Read every response first: two can be sensible, and only the panel's first choice scores.
- Best and worst. Two decisions, made one at a time. For the worst, ask what harm each response could do: a broken rule, something hidden, a person left at risk. The least helpful response is not always the worst.
- Rank. The middle counts as well as the ends. Where nearness earns marks, two neighbours swapped cost least and an end placed at the opposite end costs most.
- Rate each. Judge each response by itself against the scale's labels; the Civil Service guidance says more than one action can have the same rating. Rate what you judge: scoring can adjust for how high and how widely a person rates overall (McDaniel et al., 2011).
A situational judgement test example: one item in four formats
The item is our own example, keyed by us for the arithmetic; it comes from no employer's or publisher's test.
You work on the support desk of a software company. A customer writes that she was charged twice this month, and you can see the second charge of £180. A refund above £100 needs your team lead's approval. Your team lead is in a meeting for another hour. A. Reply now: you can see the duplicate, the refund needs one approval, and she will hear by 3 pm. Send the request to your team lead. B. Refund the £180 yourself so she is not kept waiting, and tell your team lead afterwards. C. Reply that her message has been passed to the right team. D. Wait until your team lead is free, then reply once with the outcome.
Suppose the panel's order is A, C, D, B, and its ratings, from 1 (makes things worse) to 4 (resolves it), are A 4, C 3, D 2, B 1. A candidate is sure A is best and C next, but thinks an hour of silence (D) is worse than an unapproved refund (B).
| Format | Answer | Key | Score |
|---|---|---|---|
| Pick the best | A | A | 1 of 1 |
| Best and worst | Best A, worst D | A, B | 1 of 2 with a mark for each end; 0 where both must match |
| Rank | A, C, B, D | A, C, D, B | 0.67 of the marks with part credit; 0 without |
| Rate each | A 4, C 3, D 1, B 2 | A 4, C 3, D 2, B 1 | 2 points away of a possible 10: 80% |
EdProcto's own marking code marked the best and worst row and the ranking row, with "Partial credit" on and off. It counts a ranking pair by pair: five of the six pairs stand the right way round and one does not, so (5 − 1) ÷ 6 = 0.67. The rating row is our arithmetic, one distance method among many: how far the four ratings sit from the panel's, against the farthest they could.
One judgement, honestly held, earned 100%, 50%, 67% and 80%. Pick the best never found the weak spot; the other three did, at different prices. This panel treats a broken control on money as worse than a slow reply, and an employer's published values usually say which it prizes.
Does coaching, practice or retesting raise an SJT score?
On average yes, by an amount that depends on the test; no average is a promise to one person.
SJT coaching. Lievens et al. (2012) studied coaching in a real high-stakes selection and, after allowing for who chooses coaching, put its effect at about half a standard deviation. Cullen et al. (2006) trained undergraduates in strategies for two tests built for college admission. One was open to coaching, and its scoring could also be exploited, by more than a standard deviation, through a tactic about the answer scale and not about judgement. McDaniel et al. (2011) then showed a scoring adjustment that often removes that gain.
Whose coaching. In medical school admissions, Stemig et al. (2015) concluded that commercial coaching techniques may work less well than had been thought, and the organisation's own preparation better. Begin with what the employer supplies.
Retesting. People who sat a Belgian medical admission exam a second time scored a third of a standard deviation higher on its situational judgement test (Lievens et al., 2005). Among repeat applicants for a firefighter post, two timed ability tests rose by about two fifths of a standard deviation, the untimed situational judgement test by less, and a third sitting added little (Dunlop et al., 2011).
What these studies cannot tell you
They report averages, and people who retake a test or buy coaching differ from those who do not. We read the journal studies cited here in abstract only, and the official pages in full.
Should you answer what you would do or what you should do?
Answer the question on the screen. If it asks which response is most appropriate or effective, judge the responses. If it asks what you would most likely do, say what you would do. NHS England's paper asks outright what a candidate should do in the post.
The wording changes what is measured. McDaniel et al. (2007) pooled the studies: tests that ask for the best response follow cognitive ability more closely, tests that ask what you would do follow personality, and the two predict performance about equally. Ployhart and Ehrhart (2003) set the same items under six instructions: answers under the two kinds of wording agreed only loosely, and there the "would do" versions predicted better. Under real stakes the gap closed. When 2,184 applicants to medical college were handed one wording or the other at random, mean scores did not differ in any way that mattered (Lievens et al., 2009).
Why faking is a poor bet
Faking can move a score, and it empties the score of meaning. Peeters and Lievens (2005) had 293 students answer a situational judgement test either honestly or faking. The faking group scored 0.89 of a standard deviation higher, and the test's correlation with college success fell from 0.33 to 0.09.
An answer that is not yours has to be lived up to. The Civil Service says candidates may be asked at interview about their self-assessment answers, and may have to retake the test under supervision (Government People Group et al., 2024). Learn what the job needs, then answer truthfully.
How do you practise for a situational judgement test honestly?
Practise in this order: the employer, then the format, then the reasons.
- Read what the employer says it values. The job description, the published values, any competency framework. The Judicial Appointments Commission (n.d.) puts each role's framework on the vacancy page.
- Take the employer's own practice test. The Civil Service tells candidates to try its unscored practice test first.
- Learn the format and the clock. The Civil Service test is untimed, at two to four minutes a scenario for most people. The judicial test allows 50 minutes. Work out your minutes for each situation.
- Answer cold, then read the reason. In EdProcto's "Situational judgement" kit, cover the responses and say what you would do. Then write the best, the worst, the full order and a rating for each before you press "Check".
- Write the principle, not the answer. "Tell the person waiting what you know and when they will hear." "An approval rule is not mine to waive." A principle carries to a new situation, and belongs in your error log.
What does not work
- Memorising answers. A key is one panel's view of one role, and another panel may order the same responses differently.
- Scale tactics. They are a bet on scoring you cannot see.
- A guaranteed pass. The pass mark is the employer's. The Civil Service says a recruiter may raise it once every applicant's score is in.
How do EdProcto's practice kits fit?
EdProcto is the best place to build the habit this guide rests on, judging first and then meeting the reason at once: its study screen gives the keyed response and its reason after every answer, as NHS England's practice paper does under each key.
The "Situational judgement" kit. Among the kits, it asks eight questions with a 12-minute limit, drawn from a bank of twelve workplace situations (safety, honesty, priorities, customers, policy), with no situation asked twice in a set. Every question offers four responses and asks "What is the best response?": multiple choice, best response only. Press "Check" and the screen says "Right." or "Not this time.", with the answer and the reason. At the end, under "What you don’t know yet", a button retries only what you missed. It is practice on that kit's own situations, with no norm and no link to any employer's test.
Re-tests. A finished set comes back on the dashboard under "Due for a re-test", two days on after a miss and a week on after a clean run. "Take it again" runs the whole set: see re-tests that come due.
Your own role. Twelve of the 22 "Hiring: …" kits, among them "Hiring: customer support" and "Hiring: any role, from its job description", have Procto AI write ten situational multiple-choice questions, each with its reason, from the job description and the notes you paste. It is asked to write only what your material answers, so read each draft against it.
Both ends, and a full order. The Best and worst and Ranking shapes are in the builder. Write an item of your own in one, as in the worked example, and switch "Partial credit" on so that one matching end, or a nearly right order, earns part of the marks.
What it costs. A free account opens every kit. The "Situational judgement" kit is made by rule, so it is free without limit on every plan and does not count towards the Free plan's ten quizzes a month. A kit written from your own material is a quiz like any other and counts towards the ten. Free has Procto AI write 20 questions a month and Learner 100, with no limit on self-tests: see pricing.
Quick questions
Can you fail a situational judgement test?
Yes, where the employer sets a pass mark or a minimum percentile. What an employer may fairly do with a score is in aptitude tests for hiring.
Should I say what I would do or what I should do?
Whichever the instruction asks. Under real stakes the two wordings produced much the same mean scores (Lievens et al., 2009).
Is paid SJT coaching worth it?
The evidence is mixed. One study put coaching at about half a standard deviation (Lievens et al., 2012); another found commercial techniques weaker than had been thought (Stemig et al., 2015).
Where can I find free situational judgement test practice?
EdProcto's page for candidates opens the "Situational judgement" kit on a free account, with the reason after each answer. Take the employer's own practice test as well.
References
- Cullen, M. J., Sackett, P. R., & Lievens, F. (2006). Threats to the operational use of situational judgment tests in the college admission process. International Journal of Selection and Assessment, 14(2), 142–155. https://doi.org/10.1111/j.1468-2389.2006.00340.x
- De Leng, W. E., Stegers-Jager, K. M., Husbands, A., Dowell, J. S., Born, M. P., & Themmen, A. P. N. (2017). Scoring method of a situational judgment test: Influence on internal consistency reliability, adverse impact and correlation with personality? Advances in Health Sciences Education, 22(2), 243–265. https://doi.org/10.1007/s10459-016-9720-7
- Dunlop, P. D., Morrison, D. L., & Cordery, J. L. (2011). Investigating retesting effects in a personnel selection context. International Journal of Selection and Assessment, 19(2), 217–221. https://doi.org/10.1111/j.1468-2389.2011.00549.x
- Government People Group, Civil Service Resourcing, & Government Recruitment Service. (2024, May 24). Preparing for the Civil Service Judgement Test. GOV.UK. https://www.gov.uk/guidance/preparing-for-the-new-civil-service-judgement-test
- Judicial Appointments Commission. (n.d.). Qualifying test. Retrieved October 10, 2026, from https://judicialappointments.gov.uk/the-application-process/about-online-tests/qualifying-test/
- Lievens, F., Buyse, T., & Sackett, P. R. (2005). Retest effects in operational selection settings: Development and test of a framework. Personnel Psychology, 58(4), 981–1007. https://doi.org/10.1111/j.1744-6570.2005.00713.x
- Lievens, F., Buyse, T., Sackett, P. R., & Connelly, B. S. (2012). The effects of coaching on situational judgment tests in high-stakes selection. International Journal of Selection and Assessment, 20(3), 272–282. https://doi.org/10.1111/j.1468-2389.2012.00599.x
- Lievens, F., Sackett, P. R., & Buyse, T. (2009). The effects of response instructions on situational judgment test performance and validity in a high-stakes context. Journal of Applied Psychology, 94(4), 1095–1101. https://doi.org/10.1037/a0014628
- McDaniel, M. A., Hartman, N. S., Whetzel, D. L., & Grubb, W. L. (2007). Situational judgment tests, response instructions, and validity: A meta-analysis. Personnel Psychology, 60(1), 63–91. https://doi.org/10.1111/j.1744-6570.2007.00065.x
- McDaniel, M. A., Psotka, J., Legree, P. J., Yost, A. P., & Weekley, J. A. (2011). Toward an understanding of situational judgment item validity and group differences. Journal of Applied Psychology, 96(2), 327–336. https://doi.org/10.1037/a0021983
- NHS England. (2024, June 5). Multi-Specialty Recruitment Assessment (MSRA) practice paper: Instructions. Medical Hub. https://medical.hee.nhs.uk/medical-training-recruitment/medical-specialty-training/publications/multi-specialty-recruitment-assessment-%28msra%29-practice-paper/instructions
- Peeters, H., & Lievens, F. (2005). Situational judgment tests and their predictiveness of college students' success: The influence of faking. Educational and Psychological Measurement, 65(1), 70–89. https://doi.org/10.1177/0013164404268672
- Ployhart, R. E., & Ehrhart, M. G. (2003). Be careful what you ask for: Effects of response instructions on the construct validity and reliability of situational judgment tests. International Journal of Selection and Assessment, 11(1), 1–16. https://doi.org/10.1111/1468-2389.00222
- Stemig, M. S., Sackett, P. R., & Lievens, F. (2015). Effects of organizationally endorsed coaching on performance and validity of situational judgment tests. International Journal of Selection and Assessment, 23(2), 174–181. https://doi.org/10.1111/ijsa.12105