Data-driven Assessment: What it is and how to use it

81% of companies use psychometric assessment to make more reliable hiring decisions. So, when it seems that everyone else is on the same bandwagon, how do you hop on the trend too?

The world of hiring has evolved drastically since its inception. Hiring decisions were and still are made on bias and insights rather than data, making ad-hoc processes time-consuming to gather, review, and agree on. This meant that rather than relying on a solid strategy, hiring teams had to assume that their hiring methods were effective. Nowadays, however, with the abundance of tools available on the market, it's as easy as ever to make data-driven hiring processes a pillar of your talent team. 

Above others, assessment methods such as ability and work-related tests are some of the most popular tools used in the hiring process. They allow the hiring process to be taken to the next level by providing measurable, objective data that gives those hiring the ability to more accurately analyze a candidate's potential success by providing insights into their capabilities to perform specific tasks. As psychometric assessments continue to increase, hiring managers are either already using them or considering incorporating some testing in their hiring process.

So, when it seems that everyone else is on the same bandwagon, should you hop on the trend too?

What is data-driven recruitment?

Data-driven recruitment means utilizing facts and statistics to make objective hiring decisions, from choosing a candidate to making hiring plans. However, there is not only one "right" way to incorporate data into your hiring process. For some organizations, being data-driven means collecting and acting on various hiring metrics, such as time-to-hire, cost-per-hire and offer acceptance rates to track the success of the hiring process and make the process more effective. For example, organizations may track the offer acceptance rate and see that it is significantly lower this quarter than last. This may indicate that something recently changed at this point in the process (i.e., longer delays between the offer call and contract send-out, a change in your employment contracts, etc.).

For other organizations, being data-driven can include standardized interview guides, cases, and evaluation frameworks to ensure that all candidates are treated fairly in the evaluation process.

At Amby, we use process data (i.e., dropout rates, time-to-hire, etc.) to flag blindspots in our hiring process and make minor optimizations that can make a big difference. We also standardize evaluation material to create a fair and unbiased process for each candidate.

Lastly, we also use data specific to candidate potential. This is known as psychometric assessment and is the icing on the cake for incorporating data into our hiring processes. Based on ability and work-related tests, these methods are the most common way to assess a candidate's potential with data rather than intuition. They are sure to weave robustness and data-driven decision-making into your hiring process.

Why is psychometric assessment important?

So, how widespread are these tests? Globally, psychometric assessments are currently being used by over 80% of the Fortune 500 companies in the US and over 75% of the Times Top 100 companies in the UK. Furthermore, according to Thornton (2021), 81% of companies use psychometrics to make more reliable hiring decisions, as opposed to only 67% of companies in 2010. This increase shows that psychometric assessments are increasing in popularity among recruiters and HR teams, but does it work scientifically?

Ability tests

The first category of psychometric testing is ability tests, which allow you to assess a candidate's general mental ability (GMA). GMA refers to the level at which an individual learns, understands instructions and solves problems. It was and still is considered the best predictor of job performance, as people with higher GMA acquire more job knowledge faster (Grobelny, 2018).

To dive even more into the science aspect of psychometric assessments, we suggest looking at Schmidt and Hunter's (1998) meta-analysis based on 85 years of research on how well assessments predict performance. Not only does this research show that GMA is the best predictor of job performance, but it also indicates that when paired with other forms of assessment, such as a work sample test (.63), an integrity test (.65) or a structured interview (.63), GMA's predictive validity of the candidate's job performance further increases. However, for these tests to yield accurate assessments of ability, the test itself must be reliable.

There are many, for lack of better words, "amateur" psychometric tests out there that have no research to back them up. If you are looking for ability test recommendations, we suggest AoN, our assessment partner. Their GMA tests are built on many years of psychological research, and there is a wide variety to choose from, all of which are reliable and relevant to work performance.

Personality tests

The second category of psychometric testing is personality tests, which allow you to examine the innate personal competencies needed to perform well in a job. It enables recruiters to assess the candidate's behavior in certain circumstances - helping us assess how committed they may be to the role and their compatibility with the company culture. Prior research shows that if an employee is placed in a role that does not fit their character traits, it could lead to lower engagement. This lower engagement, in turn, also leads to lower productivity and a higher chance of turnover.

Like GMAs, when choosing which personality tests to use in Amby's recruitment process, we considered how the test was developed, the reliability information and the technical features of the test. In addition, we partnered once again with AoN as their personality assessment tool has been psychometrically validated and is based on 50 years of research (AoN, 2020).

However, not everything is rainbows and butterflies, and personality tests' potential drawbacks make them less reliable. Some of these drawbacks are:

• Screening out qualified candidates. There is no mainstream, or one size fits all personality type for the role in question. Personality tests may also exclude talented candidates who think outside the box.
• Flawed results. Candidates may respond or choose specific answers based on what they think the employer is looking for based on research into which qualities and skills your company looks for in its people. This knowledge affects the authenticity of candidates' responses in personality tests.
• Test not suited to your hiring process. Many popular personality tests were not designed to be used for hiring purposes. An example is the Myers-Briggs Type Indicator which is more suited for development and training. Therefore, using a test unsuited to your hiring process could result in skewed results.
• Candidates' emotional state. Your candidate's emotional state could vary from what it would be in the work environment. This may lead to inaccurate results due to making certain decisions in the moment of writing the test due to emotional state. This could lead to a result that is not sufficient, leading to the organization filtering good candidates from the hiring process and vice versa.
• Legal risks. Candidates in the past have sued potential employers because the test they participated in discriminated against individuals with specific difficulties. For example, asking questions about mood changes could be against candidates who suffer from bipolar disorder or depression.

Other than avoiding basing the hiring decision solely on the results of a personality test, different actions can be taken to avoid encountering the above drawbacks.

These include:

• Not using personality tests after the interview stage.
• Selecting unreliable tests due to being cost-conscious.
• Forgetting to communicate the test's purpose.
• Hiring a particular personality type time and time again based on the current internal team.

It is essential to note that personality tests are a tool, not a solution. Therefore, they are not helpful when it comes to finding a candidate from a cultural standpoint, and hiring decisions should NOT be based solely on the results of the tests. Instead, these tests provide insights into the candidates' minds, behaviors, and preferences, and they can help enrich learning and development and improve workplace relationships.

To sum up - the rise of data-driven assessment comes down to time efficiencies and bias reduction. From a process standpoint, the data-driven assessment saves us time by eliminating candidates who are not a fit early in the process. From a bias standpoint, data-driven assessment equips the interviews with unbiased and structured test results to evaluate skillsets and traits the same - giving every candidate an equal chance of success. In that sense, it levels the playing field for candidates and removes the possibility of unconscious bias from interviewers.