We love to incorporate testing into the Selection Processes that we build. The addition of tests can be very easy – not really time consuming – and the benefits are significant. Tests can help you gain valuable insights into aptitudes and personal preferences, but most of all they offer a chance to check your biases. Just when I thought I had the perfect candidate, about the time the halo effect took-over, test results would come back revealing new insights that helped me revisit my thinking. I argue that this critical thought process triggered better choices.
If you are considering the use of tests in the selection process here are four suggestions:
1. Match the test (instrument) to your job specifications or job description. This is a hugely important suggestion. Tests used in selection must be shown to produce information that is directly related to someone’s ability to do the job that you are asking of them. Equal Opportunity Employment legislation in North America requires this general rule. If you incorporate a math test into a selection process, job success must require math.
2. Tests measure different things so you will need to choose either Aptitude or Personality Traits. I like to frame things this way: two types of tests are useful in selection. Aptitude refers to talent or ability to learn and personality traits refer to preferred modes of behaviour. For example, a math or writing test is a simple test of talent. A more sophisticated example would be The Wonderlic Cognitive Ability Test which is used to assess the aptitude of prospective employees for learning and problem-solving. Usually the scores from these tests serve best as a way of ruling candidates out. Test like Hogan or the PI can be used to identify candidates’ work style, understand their core drivers, and recognize patterns of behaviors. Both of these instruments profess “predictive” analysis.
3. Make sure that your instrument has been validated professionally. There are a ton of instruments out there, and more and more every day are popping up online. Anyone can produce a test and the business can be very lucrative so Caveat Emptor. We suggest that the only instruments that you should use, should be validated professionally. The Buros Center for Testing is an independent organization within the University of Nebraska. They are the world’s most widely used test review center. For $15 or $20 you can see the test reviews of all kinds of instruments. If you use a test that has not been submitted to Buros, I would be skeptical. Serious test offerings are well studies and can prove their own validity. I have nothing against Astrology but FYI – it has not been studied by Buros.
4. Integrate the test into a second interview by developing structured questions that dig into key results. Maximize the information that you get from the test scores from personality assessments by using them to guide the questioning in a second in-depth behavioral interview. Two spots are often great places to dig in: they help to find supporting behavioural evidence where the results of the test are strongly in line with what you are looking for and in places where the scores seem different than your personal observations from your earlier interview(s).
Descriptive versus Predictive Validity
When you choose your psychometric instrument keep in mind that just because the tool might offer an accurate or valid description of a person’s current preferences or behaviour it does not mean that that same person will behave in this way in the future under any circumstance. The Myers Briggs is a great example of this issue. In fact the instrument’s authors recommend against using it in selection for this very reason. Proving predictive validity requires a specialized validation study. If your sales organization is large enough (500+ for example) these studies can be done quite easily provided that you have decent records about compliance to the selection process by your sales managers. Just check back over a period of time to see if the instrument predicted performance accurately by sorting out those that the instrument indicated would be most successful. How did they fare?
In our experience HR makes a great partner in the implementation of testing in selection. When you approach them, expect reticence. They have likely seen testing abused in the selection process and they understand the legal implications of systemic bias in selection. If you can demonstrate that you will integrate one or more valid instruments into your process systematically and train your managers in the proper interpretation, they could be your biggest ally.
A final note:
In our experience, these four simple steps make good use of what can be very helpful instruments in the selection process. I for one, would not want to hire without them. And, in the event that you are ever sued for any bias in selection, you will be glad that you can explain how and why you used the instruments in front of the judge.