Multivariate predictability is one of those terms that can send even the most seasoned HR or talent management pro scrambling for a dictionary. Multi-what-now?
Luckily, the concept itself is more simple than it sounds. We’re going to unpack it for you here, so that you can toss it around in meetings and sound super smart.
Multivariate predictability consists of three elements:
- “Multi” – meaning more than one.
- “Variate” – meaning variable, or aspect.
- “Predictability” – meaning, how is it likely to play out in the future.
Put together, they are more than the sum of their parts—and that’s sort of the whole point of multivariate predictability. We are pulling together multiple views of a potential employee and determining how they act together to amplify or impact one another.
How does it apply to PI tools?
For example, in The Predictive Index (PI) environment, you can look at an employee’s Cognitive Assessment, and get some great information about how quickly they think and problem solve. Cognitive ability is the single biggest predictor of job success. You can also look at the PI Behavioral Assessment—another great predictor of job success—and get more insights about how they might act on your team or in your workplace.
Join 10,000 companies solving the most complex people problems with PI.
Hire the right people, inspire their best work, design dream teams, and sustain engagement for the long haul.
On their own, both tools provide good power towards predicting job performance. But when using multiple predictors (e.g., assessments, interview data, etc.) together, the magic starts to happen. What this means is that when hiring for a particular role, you are able to envision the future performance picture much more comprehensively.
For example, you may start by looking for sales people who are assertive, aggressive and competitive—which we refer to as high Dominance. That is just one Factor of the PI Behavioral Assessment. You might also need someone who is urgent and likes to move at a past-pace. That is low Patience—another Factor of the PI Behavioral Assessment.
Finally, you know the work environment is fairly unstructured and requires someone who learns very quickly on the job. That means you need a Cognitive Assessment. All of the sudden, you aren’t just hiring around one thing; you have two different PI Factors and the score from a Cognitive Assessment.
But wait, why not use a solid structured interview where every candidate receives a score based on their past experience and skills? Now you have four different data points. From a statistical perspective, there are ways you can throw that all into a big equation and typically what you will find is that each piece contributes its own special part to the prediction of future job performance. Not everything will be equal, but each part will be important.
The sum is more than the whole of its parts
In Schmidt (et al)’s meta-study of 100 years of research in personnel selection, they stress that multivariate predictability is a key element of the hiring and selection of job applicants.
GMA, or general mental (cognitive) ability, is by far the most common variable tested and used by psychologists when trying to build an understanding of an employee’s predicted performance. But, say Schmidt and his colleagues, when used in conjunction with GMA, there are other factors that can significantly increase your ability to spot a stand-out candidate. These variables include things like behavioral profile, integrity tests, and structured interviews, and education.
Interestingly, amount of education is valued highly by many employers but only has 1% predictive capability for job performance. Conversely, if you look at cognitive ability and use of a structured interviewed process, the multivariate predictability is 58%. Unfortunately, Schmidt doesn’t look at three different predictors together, but adding a Behavioral Assessment is likely to make this number even higher.
That’s really all there is to it. Simple, right?
Are you using multivariate predictability in your hiring process? How is it working for you? Share your story with us in the comments.
