Situational Judgment Tests in Hiring: A Guide for Employers
What a situational judgment test is, what it predicts and how to develop one for your roles. With a sample scenario, scoring and a step-by-step plan.
Door Ingmar van Maurik · Founder & CEO, Making Moves
Short answer
A situational judgment test (SJT) presents candidates with realistic situations from the role and asks which response is best. It measures practical judgment in the context of the work. An SJT predicts job performance reasonably well (average validity about 0.26), typically shows smaller group differences than a cognitive ability test and is seen by candidates as fair and relevant. You get the most value from an SJT built on scenarios from your own practice.
This article is part of our complete guide to assessments in hiring.
What is a situational judgment test?
An SJT consists of a series of short scenarios, each with several possible responses. An example from customer service:
Scenario: A customer calls for the third time about the same delayed delivery and gets angry. The colleague who manages the case is absent today. What is the best response?
|---|---|---|
Which option is best depends on the organization's way of working and values. That is why experts from within the company set the scoring key.
Formats and response instructions
Text, video or animation
Most SJTs are text-based. Video or animated versions feel more realistic and require less reading, which is why they typically show smaller differences between groups. They are more expensive to produce, though.
"What is best?" or "What would you do?"
This looks like a detail, but it matters:
In selection, the first format is usually the safer choice.
Response formats
You can ask candidates to choose the best response, the best and the worst, or rate each response on a scale. Rating each option gives the most information per scenario.
What does an SJT predict?
According to the meta-analysis by Sackett, Zhang, Berry and Lievens (2022), the average validity of SJTs for job performance is about 0.26. Important to know:
How reliable is an SJT?
Because each scenario taps something different, the internal consistency (Cronbach's alpha) of an SJT is often lower than for other tests, usually between 0.50 and 0.70. That does not automatically mean the test is poor: test-retest reliability says more here. Use enough scenarios (often fifteen to twenty-five) and never let the result count as the only criterion. More on judging quality in what makes an assessment valid and reliable.
Developing an SJT in six steps
1. Collect critical situations
Interview experienced employees and managers about situations in which the difference between good and poor behavior became clearly visible. This is called the critical incident technique.
2. Write realistic scenarios
Turn the situations into short, recognizable scenarios without unnecessary detail or jargon.
3. Collect response options
Have employees with different levels of experience write down how they would respond. That gives you realistic good and less good options, not straw men.
4. Build the scoring key
Have a group of experts rate each option. Keep only scenarios on which the experts agree sufficiently.
5. Run a pilot
Have current employees and candidates take the test. Check difficulty, reliability and group differences, and drop or rewrite weak scenarios.
6. Validate and calibrate
Link scores to performance and turnover, and refine the scenarios and weighting as data comes in. See also how to design company-specific assessments.
SJTs and AI
AI can help write first drafts of scenarios and response options faster, but experts from your organization must always review them and set the scoring key. Also watch the other risk: language models often pick the socially desirable, common-sense answer on generic SJTs. With company-specific scenarios, where the best answer depends on your way of working, and with time limits per scenario, that is harder.
Where in the process?
A short SJT of twenty to thirty minutes is a strong first filter at high volume: job-relevant, pleasant for candidates and with limited group differences. For specialist roles you can use it later, alongside a work sample or structured interview. Which combination works for which role is covered in the guide to assessments.
Off-the-shelf or custom?
Standard SJTs exist for broad job families such as customer service or management, but the real strength of an SJT lies in recognizable scenarios from your own practice: the complaints your customers have, the safety protocols at your sites, the trade-offs your managers make. Those predict better, feel more logical to candidates and are harder to fake. See how we develop custom assessments.
Frequently asked questions
What is a situational judgment test?
A test with realistic work situations in which candidates indicate which response is best. It measures practical judgment in the context of the role.
How is an SJT scored?
With a scoring key set by experts from the organization. The closer the candidate's answer is to the experts' judgment, the higher the score.
How reliable is a situational judgment test?
Internal consistency is often 0.50 to 0.70 because scenarios measure different things; test-retest reliability is more meaningful. With enough scenarios and in combination with other methods, an SJT is very usable.
How long does an SJT take?
Usually twenty to forty minutes, depending on the number of scenarios and the format.
Can candidates prepare for an SJT?
Practicing the format helps candidates feel at ease, so offer a few sample scenarios. With company-specific scenarios, the best answer cannot be learned from generic practice sets.
What is the difference between an SJT and a personality test?
A personality test asks about general traits ("I like to work carefully"). An SJT presents concrete work situations and measures how well someone judges them.
Summary
Want an SJT with scenarios from your organization? See how we develop custom assessments or get in touch.