The company has published a new guide - called How to conduct a validation study - which explains the pros and cons of different types of verification studies and provides step-by-step recommendations on best practice for HR teams. It offers advice on defining ‘performance’; choosing the right criteria, metrics and sample size; matching assessment data with appropriate performance data and how to interpret the findings.
“Conducting a validation study will help you to recruit the right people, develop the right competencies and make better and more confident talent decisions, as you can investigate the correlations between your assessment results and actual performance data,” said Dr Katharina Lochner, Research Director at cut-e Group. “Armed with these insights, you can build a model of the traits and abilities that will predict the future behaviour of your talent. For example, you can look at someone’s actual on-the-job performance and see how accurately their scores on an assessment test or questionnaire predicted that.”
According to the guide, assessment data can be correlated against performance outcomes, such as sales success or manager ratings, and against other outcomes such as job satisfaction, attrition or absenteeism. Deciding on the criteria to use is often the biggest challenge.
“If your criteria isn’t distinct or objective, you’ll waste your time as the results will be inconclusive and meaningless,” said Dr Lochner. “Manager ratings are often flawed as different managers may have different ideas of what ‘good’ looks like and some only use a small range of possible ratings. To make manager ratings more objective and differentiated, we recommend looking at different aspects of performance such as interpersonal (how individuals interact with others), operational (how they approach a task and what kind of outcomes they produce), and motivational (their level of engagement) instead of using just one overall rating. Moreover, it can be helpful to make managers think of situations in which candidates showed the actual behaviour. This can be done for example by asking how frequently they displayed it. Alternatively, they can be asked to compare each candidate to others in the team so that a full range of possible ratings is used.”
The guide also offers advice on how to present validation insights, in an easy-to-understand format, to the senior team and to recruiters and hiring managers throughout the business.
“A validation study will provide evidence that your assessments, your data and your conclusions are valid and reliable,” said Dr Lochner. “This will help you to demonstrate the impact and value of assessment and it will prove that you’re achieving a return on your investment. When you recruit and develop employees at all levels - against a validated behavioural model that highlights exactly what it takes to succeed in each role - you can transform the performance of your organisation.”
How to conduct a validation study: Guidelines for best practice in assessment validity can be freely downloaded from cut-e’s website www.cut-e.com/advisory/predictive-analytics