Do profile tests really work?

Everyday there is a new profile assessment company popping up with grand claims of isolating the best talent with a 30 minute test. Psychometric skill testing has been around for decades, and even longer in the education system. The big question is do they work? Can a hiring manager use them to save time and hone in on the right candidates, and even make a better selection?

The short answer is yes and no. No, psychometric tests are not an effective method of actually scoring candidates in any way. The companies that produce them actually print that on every form, because it is proven the resulting scores are not a true indicator of the worthiness of a candidate. Yes, these tests do come up with valid results on natural tendencies. They can be an invaluable resource in guiding the interview process, giving an interviewer a roadmap into the most important questions to ask.

The large problem with the testing is the high standard deviation. Standard deviation is the amount that a test can vary from actual results, up or down. A good example is if you and a friend take a 100 question math test, and the teacher gives you two scores. One score is based on the first 70 answers the next on the entire test.

Test A: First 70 questions = 80%, next 30 questions = 100%, total = 86%

Test B: First 70 questions = 80%, next 30 questions = 0%, total = 56%

As you can see, there is a high possible deviation. The standard deviation lets you know how much higher or lower a score could actually mean. On a highly accurate profile skill index the standard deviation is 13% +/-. So a candidate that scores a 60 could really be a 47 or a 73. Many of these tests are no where near that accurate, a deviation of 20% +/- is considered acceptable.

This high level of inaccuracy makes the scoring useless for quantifying candidates. In addition, profile tests measure natural tendencies not learned skills. The longer a person is in the workforce the further they may have moved from their natural tendencies. An example could be a sales rep that scores as an introvert. Obviously a terrible trait for a person that has to be outgoing to produce revenue. What if the person has 20 years sales experience? They might have started in sales by accident or necessity (happens to a lot of us) and been forced to learn to act as an extrovert to earn a living. Over 20 years the candidate has learned to overcome their natural tendency to be an introvert, but on a profile test they will still score as one.

So, with all the inaccuracy why do I say to use profile assessments?

They can be an invaluable tool in guiding interview questions, and alerting interviewers as to the specific information they need to hear. Take our friend the introvert salesperson. I might ask for concrete examples of getting sales from cold leads, where the customer had to be approached without the benefit of prior introduction. During the interview I will be watching body language, looking for signs of being closed off like folded arms or poor eye contact.

The bottom line is toss the scoring and use the hard data guide you to find out more about your candidate. Then you will get the full benefit of the assessment.

For more questions about finding or assessing talent inbox me. I am always happy to answer a question or give advice.

Greg Gershman – Employer Brand Ambassador –

Leave a Reply

Your email address will not be published. Required fields are marked *