In a typical Situational Judgment Test (SJ), an applicant is presented with a series of business situations.  Depending on the job for which the applicant is applying, these may include an irate customer, sales resistance for a product you are attempting to sell, or contention with a co-worker.  The presentation modality may be video, computer, or paper-and-pencil.  The applicant is then presented with several options and asked to choose what he or she would do next.  The applicant would receive a higher score on the item if the option selected was the one favored by job "experts."  In some studies experts are those who have performed successfully on the job.  In others, experts are those with more job experience.

Part of the appeal of Situational Judgment Tests is the possibility that they may measure common sense, or "street smarts."  Job experts often express the opinion that common sense is at least as important for successful job performance as academic intelligence.  In addition, majority-minority differences may be lower.

In a series of studies, various Situational Judgment Tests have produced positive relationships with job performance.  However, the reported validity coefficients usually range between .20 and .30.  This is in line with Biodata and Personality Test results but falls short of results commonly found for GMA and Work Sample Simulations.  Majority-Minority differences in SJ scores are usually less than for GMA scores, but greater than Personality or Biodata.

Are SJ tests fakeable?  There is little published research on this issue.  But it seems plausible that applicants can distort their responses.  Knowing what an employer would expect you to do in a given situation is not exactly the same as doing it.  You may know that your employer expects you to behave professionally with an irate customer.  In practice, you may or may not do so.

What do Situational Judgment Tests measure?  Is it common sense?  Academic intelligence?  Life experiences?  Or something else?  Research to date has not provided a clear answer to this question.  Further research is clearly necessary.

To summarize, Situational Judgment Tests can be developed which predict job performance.  However, the prediction falls short of that achieved by GMA and Work Sample Simulations.  Majority-Minority differences are less than for GMA but greater than Personality or Biodata.  Responses may be fakeable.