Best response
A scenario and four or five responses. You pick the single best way to handle it.
Tip: Look for the option that fixes the real problem now and stays inside policy. Polite but passive answers are rarely the best.
Customer service assessment guide · Checked November 6, 2026
Situational Judgment Test: Formats, Scoring and 10 Worked Examples
A situational judgment test (also spelled "judgement" and called an SJT) shows you a realistic work problem, like an upset customer or a coworker conflict, and asks how to handle it. You pick the best response, the most and least effective, or rate or rank every option. Answers are scored against a key set by people who know the job. Call center, retail, bank and federal employers use them, and so does Amazon. OphyAI Customer Service helps on the live test and in practice.
Situational judgment test at a glance
How situational judgment tests work across test makers and employers, from OPM, published research and the employers' own guides, checked November 6, 2026. Your invitation names the exact test and its rules.
| What it is | A short work scenario with several plausible responses. You judge which responses work best. |
|---|---|
| Answer formats | Best response, most and least effective, rate each response, rank the responses. Many employers add work-style statements. |
| How it is built | Scenarios come from real incidents collected from people who know the job, who also judge the options (OPM) |
| Scoring | Against an expert key. Most and least picks are scored right or wrong; ratings earn partial credit for being close. |
| What it measures | Interpersonal skills, conflict management, problem solving, teamwork and judgment (OPM) |
| Length | Set by the employer. AT&T's call center tryout asks you to block 35-45 minutes; Amazon says its work simulations take 20 minutes to an hour. |
| Timing and navigation | Often untimed, but many do not let you skip or go back (AT&T) |
| Who uses it | Call centers and retail (AT&T), banks (Capital One), Amazon, federal agencies (OPM's USA Hire), and airlines such as Delta (reported in 2023) |
| Pass mark | None published. Each employer sets the score it needs for the role. |
Situational judgment test question types
Most SJTs use one or more of these formats. The same scenario can be asked in any of them, so learn all of them.
Most and least effective
Two picks per scenario: the most effective response and the least effective one. Some tests say "most likely" and "least likely" to do.
Tip: Find the worst option first. Responses that blame, argue, ignore the problem or break a rule are usually least effective.
Rate each response
You rate every option on a scale, usually 1-5 or 1-7, from very ineffective to very effective.
Tip: Rate each option on its own and use the ends of the scale when you mean it. Some tests stretch the key so that hedging toward the middle costs points.
Rank the responses
You put all the options in order, from most to least effective.
Tip: Place the best and the worst first, then sort the middle two. The middle is where near misses cost the least.
Work-style statements
Agree or disagree statements, or "most like me / least like me" screens. They describe how you work and have no single right answer.
Tip: Answer as you really work and stay consistent. Some questionnaires run a consistency check on your answers.
"Should do" vs "would do"
Questions ask either what you should do (knowledge) or what you would do (behavior). Research finds both predict job performance about equally.
Tip: In a hiring test, answer both the same way: pick what a strong employee in that job would do.
Video, chat and email scenarios
Some SJTs play the scenario as a video clip or show it as a customer chat or email, with candidate replies to choose from. Some also record a short spoken answer.
Tip: Read or watch the whole message first. The customer's last line usually tells you what they need most.
SJT questions and answers: 10 worked examples
Ten situational judgment test questions and answers written by us in the common SJT formats, across retail, pharmacy, call center, bank, hotel, healthcare and office settings. None are from a real test. Try each one before you read the answer.
Retail store. A customer at your register says the shelf sign showed a jacket at $40, but it rings up at $60. Four people are waiting in line behind her.
What is the best response?
- ATell her the register price is the correct price.
- BAsk a coworker to check the sign while you serve the next customer, then honor the posted price if store policy allows.
- CGive her the $40 price right away without checking, to keep the line moving.
- DSuggest she come back later when a manager is free.
Best response: B.
B checks the facts fast, keeps the line moving and follows the pricing policy. A dismisses her without checking, C gives away money on a guess, and D pushes the problem back onto the customer.
Pharmacy counter. A customer comes to pick up a prescription. Your screen shows it is not ready because his insurance claim was rejected. He says he needs it today and starts raising his voice.
Pick the most effective and the least effective response.
- ATell him insurance problems are not the pharmacy's fault.
- BCalmly explain what the rejection means and offer to call the insurer or check the cash price while he waits.
- CAsk him to call his insurance company himself and come back later.
- DTell him to lower his voice or you will not help him.
Most effective: B. Least effective: D.
B stays calm, explains the problem and owns the next step. D escalates the conflict and refuses service. A and C are not rude, but they leave him to solve it alone, so they sit in the middle.
Call center. A caller says she was charged twice for this month's bill. You can see both charges on her account, and you are allowed to refund billing errors.
Rate each response from 1 (very ineffective) to 5 (very effective).
- AConfirm the double charge, apologize, and start the refund while she is on the line.
- BTell her the second charge will probably drop off on its own.
- CTransfer her to the billing team without explaining why.
- DApologize, take her number and promise someone will call her back.
A: 5. B: 1. C: 2. D: 3.
A fixes a confirmed error on the first call. D is polite but delays a fix you could make now, C hands her off with no context, and B guesses and does nothing. Ratings earn partial credit: a 4 for A is close to the key and costs far less than a 2.
Bank teller. A long-time customer asks you to cash a large check. It looks fine, but the amount is above the limit you can approve on your own.
Rank the responses from most to least effective.
- ACash it, since you know the customer well.
- BExplain that checks this size need a supervisor's approval, and ask your supervisor right away.
- CTell him you cannot cash it and suggest he try another bank.
- DAsk him to wait while you find a supervisor, without saying why.
Ranking: B, D, C, A.
B follows the approval rule and keeps the customer informed. D follows the rule but leaves him confused, and C gives up on a customer you could serve. A breaks a control that protects the bank and the customer, so it ranks last even though it feels friendly.
Hotel front desk. A guest arrives at 11 p.m. and the room she booked is not ready because of a plumbing problem. She has an early meeting tomorrow. Other rooms of the same type are free.
What is the best response?
- ATell her maintenance should finish in about two hours.
- BApologize and give her an available room of the same type or better right away.
- CExplain the plumbing problem in detail so she knows it is not the desk's fault.
- DOffer a discount on a future stay and ask her to wait in the lobby.
Best response: B.
B solves her real need, a room and sleep before her meeting, right now. C explains but solves nothing. A and D both leave her waiting late at night, and D offers something she did not ask for.
Healthcare front desk. A patient arrives 20 minutes late. Clinic policy says patients more than 15 minutes late must reschedule. She says her bus was late and she took time off work to come.
Pick the most effective and the least effective response.
- ATell her the doctor might squeeze her in and ask her to take a seat.
- BExplain the late policy kindly and offer the earliest open appointment, including any opening today.
- CCheck her in anyway, since her reason is a good one.
- DTell her she should have planned for the bus and call the next patient.
Most effective: B. Least effective: D.
B shows care and follows a policy that keeps other patients on time. D blames her and ends the conversation. C breaks the rule for one patient at others' expense, and A makes a promise you may not be able to keep.
Work-style screen. Statement: "I stay calm when a customer is angry with me."
How much do you agree?
- AStrongly disagree
- BDisagree
- CNeither agree nor disagree
- DAgree
- EStrongly agree
No single keyed answer. Agree or Strongly agree fits most service jobs, if it is true for you.
Work-style answers are compared with what the job needs, not marked right or wrong. Stay consistent: if you agree here but later say you snap back at rude customers, the mismatch can count against you.
Work-style screen with four statements about how you like to work.
Pick the statement most like you and the one least like you.
- AI finish one task completely before I start the next one.
- BI enjoy helping people sort out their problems.
- CI prefer to work alone without interruptions.
- DI double-check my work for mistakes.
No single keyed answer. For a customer-facing job, B is a natural "most like me" if true, and C a common "least like me".
These screens make you choose between good traits, so you cannot say yes to everything. Pick what is true, keeping the job in mind: a role built on helping customers suits people who enjoy it.
Office team, no customers. You and two coworkers share a weekly report. The other two keep arguing over who owns the data section, and this week the report went out late.
Pick the most effective and the least effective response.
- AStay out of it. It is their problem, not yours.
- BAsk both of them to meet for 15 minutes, split the sections, and set a deadline for each part.
- CTell your manager the report was late because of them.
- DQuietly take over the data section yourself from now on.
Most effective: B. Least effective: A.
B fixes the cause, unclear ownership, and gets the report out on time. A lets a team problem keep hurting your work. C and D each deal with the late report but skip talking to the people involved, so the conflict is still there.
Online store chat. A customer writes: "My order was supposed to arrive yesterday for my son's birthday. Tracking hasn't moved in three days. This is ridiculous." Tracking shows the package stuck at a local depot. You can reship orders with overnight shipping.
Which reply is best?
- A"Per our policy, deliveries can take up to 10 business days. Please wait for it to arrive."
- B"I'm so sorry, I know how much a birthday gift matters. I can see it's stuck at your local depot. I can send a replacement by overnight shipping today. Would that work for you?"
- C"So sorry! Unfortunately there's nothing I can do once a package has shipped."
- D"Please contact the shipping company with your tracking number for an update."
Best reply: B.
B names the feeling, shows you looked, and offers a real fix with a clear next step. A hides behind policy, C apologizes without helping, and D sends the customer away to solve it alone.
Want more like these, scored? Try 3 free practice questions in OphyAI
How situational judgment tests are scored
SJTs are scored against a key, not a textbook answer. Test makers collect real incidents from people who do the job, write several plausible responses for each, and ask experienced employees and supervisors to judge how effective every option is. OPM says scores are based on these experts' judgments of the best and worst options. Options the experts disagree on are usually dropped, so the responses left in the test have a clear better and worse (Whetzel, Sullivan and McCloy, 2020).
Each format is scored a little differently. Most and least picks are usually scored right or wrong against the keyed options. Rating scales are usually scored by how far your rating is from the experts' average, so a near miss earns partial credit. Because expert averages tend to sit near the middle of the scale, some tests stretch the key or standardize your ratings, so rating everything in the middle does not pay.
Work-style statements have no right answer. Your pattern is compared with what the job needs, and some questionnaires check consistency: Saville says its check flags profiles that read as overly positive or negative. There is no universal pass mark. Research reviews find SJT scores moderately predict job performance, and OPM notes that applicants usually see SJTs as fair.
How to prepare for a situational judgment test
- Step 1
Find out the format and the job
Your invitation or the employer's prep guide names the sections. AT&T, for example, publishes a free guide for each of its tryouts. A pharmacy SJT and a call center SJT reward the same habits, but the details of the job matter.
- Step 2
Learn the five habits keys reward
Across settings, the best answers stay calm, show care for the person, own the problem, follow the policy, and escalate only what truly needs a manager. The worst ones blame, argue, ignore or break a rule.
- Step 3
Practice every format, not just one
Work through the 10 examples above in each format. Rating and ranking feel different from picking one answer, and partial credit means close answers still count.
- Step 4
Answer work-style items honestly and consistently
Go with your first true reaction, as AT&T and Amazon both advise, and keep the same story across similar statements.
- Step 5
Set up for one sitting
Use a quiet room and a reliable connection. Many SJTs do not let you go back, so read the whole scenario and every option before you answer.
Are situational judgment tests timed or proctored?
It depends on the platform, not the SJT format. Many SJTs are untimed or loosely timed and taken online at home, at a time you choose. AT&T's customer service tryout asks you to block 1 hour 30 minutes, does not time its scenario section, and does not let you skip questions or review earlier answers. Its newer call center tryout times only one section, a call-monitoring task.
Some SJTs show scenarios as video clips or animated characters, which candidates in studies rated as more realistic than text. Retakes are set by the employer: AT&T lets you retest after six months. Saville says it compares score trends over time, and it advises employers to tell candidates they may be asked to turn on a webcam and to retest finalists under supervision before an offer.
Situational judgment test answers, AI help and cheating
Are there situational judgment test answers online?
Not for your test. Real SJTs use scenarios written for the job and keys set by people who know it, so there is no answer sheet to find. What you can learn are the habits the keys reward, which the examples above teach. On the live test, OphyAI Customer Service reads each question on your screen and shows the best answer.
Can AI help on a situational judgment test?
Yes. SJT questions are written scenarios with set options, which is what OphyAI Customer Service reads. It shows the best response, the most and least effective, a rating for each option or a full ranking, and keeps work-style answers consistent. It works in Chrome or Edge on a computer and in the Mac and Windows desktop app, not in the phone apps.
Can a situational judgment test tell if you use AI?
The test format does not watch you; the platform around it might. Saville says it compares score trends over time and uses rating formats it says are harder for AI, and it advises employers to warn candidates about webcams and to retest finalists under supervision. In the desktop app, OphyAI's answers show in a popup that macOS and Windows leave out of screen sharing and recording. It does not work inside lockdown or proctored browsers, it does not take typing-speed tests, and hiding a window from screen share does not hide it from a webcam, a person in the room, or proctoring software.
Compare every platform: Can online assessments detect AI?
OphyAI on the web and on Mac and Windows
OphyAI Customer Service on a situational judgment test
OphyAI Customer Service reads each question of an online customer service assessment on your screen and shows the best answer in a small window only you can see. Practice mode gives you realistic mock tests that explain why each option is better or worse.
OphyAI Customer Service works on the live test and in practice. On test day, share the test tab once in Chrome or Edge on a computer, or run it in the OphyAI desktop app for Mac and Windows. It reads each SJT question as it appears and shows the best answer in seconds: the best response, the most and least effective, a rating for every option, or the options in order, best first.
It also answers work-style statements and keeps them consistent across the test, gives you a short script when the test records a video answer, and writes ready-to-paste chat and email replies. To practice first, take a 5, 10 or 20 question practice test that explains why each option is better or worse, or try the free 3-question sample with a free account.
Best response
Pick the single best way to handle the customer.
Most and least effective
Two picks per scenario: the most and the least effective response.
Rate each response
A rating for every option, from very ineffective to very effective.
Rank the responses
The options in order, best first.
Store associate assessmentPart 2 of 4
Scenario 7 of 20
A customer wants to return a jacket bought 34 days ago. The return window is 30 days and she is upset. Pick the most and least effective response.
- ATell her the policy is 30 days and there is nothing you can do.
- BApologize, check whether a manager can approve an exception, and explain her options.
- CProcess the refund right away so she leaves happy.
- DAsk her to come back when the store manager is in tomorrow.
Sharing entire screen
Store associate assessmentPart 2 of 4
Scenario 7 of 20
A customer wants to return a jacket bought 34 days ago. The return window is 30 days and she is upset. Pick the most and least effective response.
- ATell her the policy is 30 days and there is nothing you can do.
- BApologize, check whether a manager can approve an exception, and explain her options.
- CProcess the refund right away so she leaves happy.
- DAsk her to come back when the store manager is in tomorrow.
OphyAI · Customer Service Auto
Most and least effective
Most: B
Least: A
B shows care and owns the problem within the rules. A shuts her down.
Works in Chrome or Edge on a computer and in the OphyAI desktop app for Mac and Windows. It is not in the iPhone or Android apps. A free 3-question sample is open to anyone with a free OphyAI account, on any plan, with no credits. Customer Service is on the Premium plan ($59 a month, 500 credits). A live session uses 15 credits and stays open 2 hours; a practice test uses 5 credits. See everything it does
What it is hidden from, and what it is not
It does not work inside lockdown or proctored browsers, it does not take typing-speed tests, and hiding a window from screen share does not hide it from a webcam, a person in the room, or proctoring software.
Situational judgment test questions
A situational judgment test, or SJT, is a hiring test that describes a realistic work problem and gives you a few ways to handle it. You pick the best response, the most and least effective, or rate or rank them. OPM calls SJTs low-fidelity simulations: you judge the situation instead of doing the task. Outside the US it is often spelled situational judgement test. Employers use it to measure judgment with people, like calming an upset customer or settling a coworker conflict.
Most people find it less stressful than a timed math or logic test, but it is easy to lose points. Every option sounds reasonable, so the difference between the best and second-best answer is small. AT&T says job experience is not essential to do well.
Yes. Employers set the score they need, and if you fall short you do not move on. AT&T, for example, says candidates who do not qualify can retest after six months. Partial credit on rating scales means one weak answer rarely sinks you.
Judgment with people and problems. OPM lists conflict management, interpersonal skills, problem solving, negotiation, teamwork and cultural awareness. Customer service SJTs focus on staying calm, showing care, owning the problem, following policy and knowing when to escalate.
Many US employers. AT&T uses them in its call center and retail tryouts, Capital One in its Virtual Job Tryout and Amazon in its work simulations, and Delta Air Lines has used one in flight attendant hiring. Federal agencies use them through OPM's USA Hire, which includes a Supervisory Situational Judgment Test.
It depends on the employer. Amazon says its work simulations take 20 minutes to an hour. AT&T asks you to block 35-45 minutes for its call center tryout and 1 hour 30 minutes for its longer customer service version.
"Should" questions ask what the right action is; "would" questions ask what you would actually do. Research finds both predict job performance about equally, and in a hiring test people answer both the same way. Pick what a strong employee would do.
Start with the 10 SJT questions and answers on this page. SHL Direct has a free SJT example, AT&T publishes free PDF study guides for its tryouts with sample questions, and OPM's USA Hire has practice for federal applicants. OphyAI Customer Service has practice tests of 5, 10 or 20 original questions with explanations, and a free 3-question sample with a free account.
Sometimes. AT&T's newer tryout guides say a PC, Mac, tablet or phone works, while its older customer service version needed a computer. OphyAI Customer Service runs in Chrome or Edge on a computer and in the Mac and Windows desktop app, not in the phone apps.
Sources
Read on November 6, 2026. OphyAI is not affiliated with any test publisher or employer named on this page.
- OPM: Situational judgment tests
- Whetzel, Sullivan and McCloy (2020): SJT development practices and psychometric characteristics, Personnel Assessment and Decisions
- OPM: USA Hire assessments
- SHL Direct: SJT example question
- AT&T: Customer Service Rep Virtual Job Tryout test guide (PDF)
- Amazon: assessments (work style and work simulations)
- Saville: personality assessment (consistency check)
- Saville: the integrity of assessments
Other customer service assessments
Virtual Job TryoutThe HireVue job simulation: sections, timing, retakes and 6 examples.
Walmart assessmentThe hourly retail associate assessment: formats, results, retakes and examples.
CVS assessmentThe CVS Virtual Job Tryout for store, pharmacy and call center roles.
Target assessmentThe Virtual Job Preview, work style questions and the HireVue video interview.
Home Depot assessmentVideo scenarios and work style questions: about 20 minutes, 96 hours to finish.
USPS assessment testThe USPS postal exam (474-477) explained, with a free practice test.
Kroger assessmentThe Kroger hourly assessment: what to expect and example questions.
Walgreens assessmentThe Walgreens store and pharmacy assessment, step by step.
Teleperformance assessmentCall center assessment: typing, customer scenarios, chat and spoken replies.
Concentrix assessmentThe Harver test Concentrix sends: chat simulation, scenarios, typing, language.
Walmart team lead assessmentThe Teaming Employment Assessment: sections, "Competitive" results, retakes and examples.
Amazon warehouse assessmentThe hourly hiring test: work style pairs, the bin test, results and the 90-day wait.
Alorica assessmentCall center test: typing, scenarios, chat and email writing, mock calls and voice.
Harver assessmentEvery Harver module: chat simulation, SJT, typing, games and cognitive tests.
Publix assessmentThe Publix Retail Employment Test (PRET): formats, results and the 6-month retake.
Ross assessmentThe short Ross Stores application assessment: what it asks and what comes next.
TTEC assessmentTTEC call center assessment: scenarios, typing, computer skills and the system check.
Foundever assessmentThe Maki skills test Foundever (formerly Sitel) sends: parts, timing and retakes.
Video interview guides
HireVue interviewRecorded answers, employer-set prep time and retakes, optional AI scoring.
HireVue interview questionsCommon question types, STAR structures, example answers and a practice plan.
Spark Hire interviewThink time, takes and time limits per question, practice, and AI review.
One-way video interviewsHow recorded interviews work on HireVue, Spark Hire, Willo and VidCruiter.
Walk into your situational judgment test prepared
Create a free OphyAI account and try 3 practice questions at no cost. On test day, OphyAI Customer Service reads each situational judgment test question on your screen and shows the best answer in a window only you can see.