Boosting Your Prediction from Your Hiring Assessment Tools

September 5, 2019 Alina Siemsen

First published on LinkedIn

Getting stronger prediction about in-post job performance is the holy grail for recruiters. In this article, Alina Siemsen explains how this can be achieved with talent assessments.

It’s clear that the overriding reason for including pre-hire assessment in the hiring process is to spot those who will do well in the job if hired. I am often asked how organizations can strengthen that prediction.
The answer is by looking not just at the usual criterion-related validity of the assessment used, but also at the incremental validity which could be achieved by introducing additional assessment tools such as another test, questionnaire or assessment center exercise.

What Is Incremental Validity – and Why Is it Important?
Incremental validity is a very specific psychological term. It refers to the additional or incremental gain that can be made in the predictive capability of the hiring process by including extra assessment. It means we can significantly increase our ability to understand which candidates are more likely to perform well if hired. Quite simply, it helps give us a more accurate indicator of a candidate’s future success.

Let Me Share an Example of How Incremental Validity Works.
If you are using a specific test as part of your hiring process, you will already understand the link or correlation between the test results and the performance metric you are using. This could be a sales revenue figure, or customer service score, or something else. You will have carried out a validity study using, say, an ability test and have generated a correlation coefficient.
In practice, the ability test predicts well on its own. But looking at the variety of demands in the position you are wanting to fill, you work out that these other aspects could be measured using an additional assessment (such as another cognitive test, a personality questionnaire or a structured interview). The additional assessment result would complement your previous score. If you now repeat your previous calculation and correlate the combined result with your performance metric, you will see a gain in the correlation coefficient compared that when using only the single test. This gain is called “incremental validity”. You can see this illustrated below.


You can see that there is some overlap measured between the ability test and the personality questionnaire (at point A). But there is an additional area not measured by the ability test (point B), and this is what would be referred to as the “incremental validity of the personality questionnaire”.
The only way to obtain this stronger prediction is to use an assessment tool that measures something different to that you are already measuring but still correlates with high performance in the job. The key is variety of assessment in job-relevant areas.

How Does This Apply to Assessment Early in the Hiring Process, and in the Design of the Assessment Center?
Including extra assessment methods before and/or during an assessment center can improve the accuracy of the performance prediction.  
As an example, Credit Suisse is able to highlight and hire 15% more high performers by using both a verbal and numerical reasoning test during its pre-assessment centre process rather than a single test. But greater prediction can also be gained during an assessment center. Think about adding in a different type of exercise than those already included and work out if it measures something relevant but not yet assessed in the process.

What Are the Possible Pitfalls to Avoid?
It could be easy to get carried away by adding more and more assessments into the test battery or assessment center. You think that adding an extra measure will give you greater insight; that is not always the case.
Any job success is a combination of different skills, abilities and behaviors – and these are what need to be measured. If you introduce a test which looks at, say, an ability not needed for the job, that is not just an invalid use of the assessment, but also consumes time and may distract you from the results of the important skills. Remember, just adding in more and more assessment does not mean you will get a pay back in terms of stronger prediction. Think carefully and carry out a validity study to test which assessments give you the return. Get in touch with us if you would like to know more.

Clients are embracing what is sometimes seen as the ‘next step’ in validity. You can hear in this short video how Beiersdorf is recruiting for a VUCA world and read in this case study the project carried out at Credit Suisse.

About the Author

Alina Siemsen

Alina Siemsen is a product development consultant in the research team at Aon's Assessment Solutions. Alina completed her Master's degree in Business Psychology at the Nordakademie in Hamburg. She is interested in how gaming elements can be applied to assessments. Aon's Assessment Solutions undertake 30 million assessments each year in 90 countries and 40 languages.

Follow on Linkedin Visit Website More Content by Alina Siemsen
Previous Article
What Is the Recruiters’ Role in the Age of AI?
What Is the Recruiters’ Role in the Age of AI?

Scare stories abound concerning the impact AI (Artificial Intelligence) will have by taking jobs and replac...

How Cutting-Edge AI Can Unearth Ignored or ‘Invisible’ Talent
How Cutting-Edge AI Can Unearth Ignored or ‘Invisible’ Talent

Truly diverse talent exists in unexpected places, and AI makes finding that talent possible.

Subscribe to our talentNews