Situational Judgement Tests (SJTs) ask candidates to choose the most and least effective response to realistic workplace scenarios. Unlike personality questionnaires, they are harder to fake and directly measure the judgement skills required in the target role.
Why they work
SJTs have a predictive validity of r = 0.38 for customer service performance and r = 0.34 for overall job performance — comparable to structured interviews, at a fraction of the cost. More importantly, they show low adverse impact across demographic groups compared to pure cognitive ability tests.
Designing effective scenarios
The quality of an SJT depends entirely on the quality of the scenarios. Good scenarios are: specific to the role (not generic), drawn from incidents that actually occur in the job (based on critical incident interviews with high performers), and have a clear "best" response that experts agree on.
The response format matters
The "most effective / least effective" forced-choice format is more predictive than a simple ranking format. It forces candidates to think through the worst option — revealing whether they understand what not to do, which often predicts failure more than knowing what to do.
Combining SJTs with other assessments
The highest-performing recruitment batteries combine: Cognitive Ability (predicts whether they can do the job), Big Five (predicts how they'll do it), and SJT (predicts whether they'll handle the messy human situations). Used together, these three instruments can explain up to 60% of the variance in job performance.