Why Does Our Assessment Test Have Math Questions?

Jim Brown


By Jim Brown

This is a question I get a lot.  “Why does your assessment tool ask so many math and vocabulary questions?  Doesn’t that just intimidate or demean the job candidate?” Well here’s the answer. We love to intimidate and demean your potential employees.

Okay, that’s a joke, of course we don’t want to to do that, there really is a great reason why we ask all those tricky questions and it has to with determining your candidates cognitive ability.  Ask yourself these questions: How do you know if someone can keep up with the pace of change in a role?  Whether or not they’re an effective decision maker?  How good are their problem solving skills?  Well these questions are answered by knowing cognitive ability, and they are really tough to determine from an interview.

Avon CEO Andrea Jung said, “Clear thinking in senior leadership is a primary attribute we look for. I’ve seen little correlation between those who have a formal business education and those who possess clear thinking.…Some people have a knack for this, some don’t.”  For some great insight into the importance of hiring for executive intelligence you should really read “Hiring for Smarts” by Justin Menkes in the Harvard Business Review.

So what are cognitive ability tests and why should they be a key part of your hiring process?  A cognitive ability test is a tool that measures aspects of general intelligence, such as analytical thinking, verbal reasoning skills, the ability to learn quickly, and general mental agility.  These particular attributes can tell you how well someone will be able to use mental processes to solve work-related problems and develop on the job.  They are also, according to psychological research, one of the most powerful predictors of job performance.

Predictive science measures future job performance with a correlation coefficient, essentially a measure of how accurately they predict success.  If you were able to predict future job success perfectly with 100% accuracy you would have a correlation coefficient of 1.0.  Unfortunately, I just don’t think we’ll ever get to 1.0 but the goal is to get as close as we possibly can because it will save us a lot of time, money, and headaches.  Cognitive ability tests have a correlation coefficient of 0.51 to predicting job success.

Now let’s add a little bit of context to this score.  Unstructured interviews have a correlation coefficient of 0.18, the same as years of experience, and reference checks have a coefficient of 0.36.  What are unstructured interviews you ask?  They are the kind of interview that is typical to about 67% of the hiring decisions that are being made out there.  You bring someone in and ask them a few questions and decide to hire them in the first five minutes because you’ve got a good “gut” feeling.

Now let’s take a look at some other research results.  Structured interviews also have a correlation coefficient of 0.51 and years of education have a coefficient of 0.10.  A cognitive ability test with a valid behavioral assessment has a coefficient of 0.67.  But the standard parameters that we’ve measured prospective job candidates for decades on has been: an unstructured interview plus years of job experience plus education. How many billions of dollars has this cost public and private industry over the last few decades.  Enough!  Lets look at a hiring and selection process that works.

Here’s our hiring and selection process:

  1. Use a reliable and validated job-matching assessment tool that measures cognitive ability, behavioral characteristics, and career interests and administer to your short list of candidates.  Just employing this tool will bring up your reliability scores, the tool should also give you the ability to create a focused benchmark for individual roles by testing existing high performers.  This benchmark then becomes the target to compare candidates to and improves your chances for reliability.  For an example of this type of tool, read more about our ProfileXT® job-matching assessment.
  2. Review the reports from your assessment tool, our ProfileXT® supplies interview reports that suggest appropriate behavioral based questions, to create a highly structured interview.  No hiring process should ever rely solely on a psychometric assessment to make the decision, but these tools are invaluable to getting the information you need to conduct an interview that helps you determine job suitability.  You may need to conduct more than one interview with other people in the organization so that you continue to avoid biases unique to you.
  3. Perform background and reference checks to further verify the person is who they claim to be.  This should be a pillar of your hiring process, It’s often a pretty frustrating or boring part of the process and so it gets skipped on a regular basis.  If you don’t like doing it then outsource it, but make sure that it gets done.

These three key pillars support a very effective process.  I wish I could tell you that together they delivered a correlation coefficient of 1.38 but that’s just not realistic. What they do however is create a process that should be around 0.85 and that is pretty good, especially considering the vast majority of hires are made with information that predicts success on a very weak level.  And I haven’t even gone into all of the other benefits these assessments provide when it comes to onboarding, training, coaching, and promoting your new employee.

Employing this process creates an immediate and substantial return on investment because of the increasingly high costs of hiring mistakes.  There are studies that suggest hiring the wrong manager could cost you up to 24 times their annual salary.  I don’t know what you pay your managers but even the cost of a $40,000/year front-line supervisor is well worth adding a $200-$300 dollar assessment and the extra time it takes to do better interviews and reference checks.

If you’d like more information on the ProfileXT® assessment tool, or to try it out as a demo, be sure to contact me to set that up.

Predictors of Performance Information via Robinson and M. Smith Personal Selection (2001) British Psychological Society

Join the Discussion