How do we know our assessments are reliable? A comparison with CANTAB

As part of its Global Cognitive Empathy Study (GCES), the Cognisess Research team, led by Chief Scientific Officer Dr Boris Altemeyer, has been conducting research with a number of UK Universities and Colleges to further confirm the reliability of Cognisess’ assessments by comparing them with the clinically approved CANTAB assessments.

We all want to find the perfect job – a job that captures our full attention and gives us an environment where our values align with our actions. But the perfect job is hard to find. Choice, rather than being an advantage, is one of the biggest obstacles between the ever-elusive perfect job and us.

We can use self-reflection to point us in the right direction, turning inward to think about our passions, what we are good at and what we would like to be good at. A huge number of different factors could be the reason why we’re currently in a particular career. Whether this career is right for us often depends on no more than luck. Clearly then, the right sort of help is needed to allow us to find right job.

Cognisess provides that help and with our assessments you can see just how well you fit certain roles. Our tools highlight what makes you a unique individual, ranging from your personality – how you react in certain situations and to other people – to your general mental ability. However we are only as accurate as our tools, so to further improve these, at the beginning of this year we began an exciting new research project.

What are CANTAB assessments?

In neuropsychological testing, CANTAB assessments (from the Cambridge Neuropsychological Test Automated Battery) are one of the most extensively validated tools, with more than 30 years of research investigating their reliability and validity. We are now in the process of comparing how participants perform in the gold standard CANTAB assessments, versus how they perform in our own.

How can we prove the reliability of what we measure?

CANTAB and Cognisess assessments both evaluate cognitive and emotion recognition performance. If a participant performs well in CANTAB, they should perform well in Cognisess. In statistical terms, this is known as convergent validity, or construct validity. Researchers typically determine construct validity by answering a simple question, does the assessment measure what it claims to measure?

Anyone can take part in our research project and to ensure that we collect data that is representative, we are conducting the project with an eclectic group of participants. Those taking part range from college and MBA students to CEOs and other individuals in full time employment, and we are working with Bath College, Bath Spa University, the University of Bath and the University of Exeter to achieve this.

Focussing on our users

The project also allows us to gather valuable user feedback on the Cognisess assessments. In the realm of research, fun and assessments are not typically synonyms. When assessments are designed for recruitment, this further reduces the likelihood that individuals going through the process will enjoy what they are doing. Yet it has become apparent that users really enjoy using the Cognisess system and playing our assessments.

Peter Wills, 18, from Bath College said, “My favourite games were ColourSwitch and MatriX, although I swear that sometimes there was no X to find in MatriX. I really liked comparing my results with my friends and that got me to play the games again, so that I could get a better score.”

Our assessments are short and to the point – which is very popular with our users. As Bath Spa Art Student, Sanket Shrestha said, “I was surprised about how much fun I ended up having and it didn’t feel like a drag.”

By comparing our assessments against the CANTAB battery, we can ensure that neither the brevity nor entertainment of our own assessments detracts from their reliability. We’re seeking to make our assessments as good as they can be without impacting on any of the important aspects of our system that appeal to our users.

If you would like to learn more about the study then get in touch with us at support@cognisess.com.