Of course we know today that stress interviews really only tell us how a candidate would react under a certain kind of stress, not how proficient they are in their day-to-day work performance.

As early as 1942, intelligence testers were advocating using a structured set of interview questions so that each candidate could be given equal consideration. However, the traditional fly by the seat of your pants type of interview kept coming to the fore, even though it had an accuracy rating of anywhere from 14% to about 50%. Over the years, structured interviews have been gaining more and more favor with organizations. Structure improves accuracy and validity and makes the selection process more defensible.

It is important to note that today’s research indicates that structured interviews that are based on job requirements and that combine competency-based questions, behavioral-type questions and critical incidents, give organizations an accuracy rating that is far, far higher than ever before. We know that the structured interview, where every candidate has the opportunity to answer the same questions, is more objective and defensible.

We recognize the validity of behavior-based questions because the best predictor of future performance is past performance.

Well-prepared critical incidents (stories of real events that ask for specific behaviors) identify performance behaviors and have about an 85% accuracy rating.

However, interviewing is not a science. Our job is to eliminate as many opportunities for error as possible. Know what the job entails, follow these recommendations for interviewing, and combine the interview with other testing for the best shot at getting a good candidate.