Over the last couple of years you might have heard the phrase “Evidence Based Recruitment” (EBR, for short) here or there. It gained visibility specially around conversations related to candidate tests for logical ability or personality traits, stirring some debate about their relevance, and confusion on how it relates to the well-known competence based recruitment.
The reality, however, is much simpler: EBR is the natural evolution of competence based recruitment, and cognitive & personality tests are just a component of EBR but not the totality of it.
In the words of well-renowned scientific expert and professor of Organizational Psychology, Paul Sackett: “Evidence based recruitment is the practice of making hiring decisions founded on reliable and validated data of actual work performance”.
So how has recruitment evolved thanks to science and technology?
Like most everything else in organizations, digitalization has had a big impact on how we understand and execute recruitment. We’ve moved from an intuitive, inaccurate approach, to a systems-enabled, data-informed approach that allows us to predict work performance for specific roles and contexts with fair reliability. Let us walk through it step by step.
Traditional recruitment
Back in the day recruitment was done without much or any structure. A hiring manager would do it on their own, with little or no support from a recruitment specialist, and do their best to figure out who is the right candidate through an interview with questions they thought (or heard) would be relevant, without clear success requirements. Examples of some typical ones include:
- Why do you want this job?
- Where do you see yourself 5 years from now?
- How many tennis balls would fit in an airplane?
The interviewer would get mixed signals from the answers, and without an assessment rubric, they would use their intuition to interpret them based on how they feel and think about the candidate, thus adding their biases and subjectivity to the mix.
Due to a lack of a robust assessment framework for success requirements, hiring managers would often use proxy signals such as years of experience, or pedigree from certain universities or previous companies, as an attempt to identify who could be a top performer.
The thought process was “if someone has been for long enough in a role, or was a part of XYZ well-known company/school, then they must be good”; but now we know for a fact that tenure isn’t equal to skills proficiency, and that past success in a certain prestigious academic or work context doesn’t guarantee future success in the specific context of a role in your company. Contextual (environmental) variables such as culture, team, manager relationship, etc. have a strong influence in actual work performance.
This approach is what we call traditional recruitment, and it’s characterized by (1) choosing inaccurate requirements (based on the role and context), or not choosing at all; (2) using inaccurate assessment methods; and (3) being highly subjective and biased.
Research shows that this approach has an estimated success rate of around 20%~[1], where “success” means the hired candidate has above average performance after the first year and a half on the role.
Competence based recruitment
Realizing the shortcomings of the traditional approach to recruitment, around the mid 2000s~ many companies started their journeys towards a more structured approach centered around “competences”.
Competences are loose terms used to roughly describe individual characteristics that a person could possess to facilitate their work performance. Typical examples would be: teamwork, leadership, time management, stakeholder management, etc.
The usage of competences was a massive improvement for its time in comparison to the previous traditional approach:
- It provided structure by grouping certain behaviours under a competence (eg: “someone who’s highly proficient in “stakeholder management” behaves ‘this’ and ‘that’ way, and does ‘this’ and ‘that’”).
- It allowed a better alignment of requirements for all roles (eg: “for a candidate to be successful in this role, they need to be highly proficient in competence X, moderate proficiency in competence Y, and beginner in competence Z”).
- It started the conversation around “hard skills” (functional competences) and “soft skills” (interpersonal competences), providing a common language to the behaviors that make someone successful in a role beyond the subject matter expertise in a certain function (eg: coding, accounting, design, etc.)
- Specific questions (eg: using the behavioral STAR method) could be created to attempt to assess competence proficiency, making interviews more focused and objective.
These improvements led to an estimated improvement in hiring success rate to 50%~ [2].
While more than double of the previous estimated 20% of traditional recruitment, 50%~ is not much better than chance, and this is partly due the shortcomings of competence based recruitment:
- The competences are not well established psychometric constructs. This means that their definitions, the behaviors they encompass, and the proficiency degrees, are all open to the relative interpretation of each organization, team, and individual. Eg: someone who’s considered to have a highly proficient “Leadership” competence in an organization, can be considered to have low proficiency somewhere else.
- Even when an organization builds their own internal competence model, full adoption and maintenance is challenging, and cross-org transferability is limited.
- The competences chosen for each role are based on assumptions made from experience, which result in some of them being accurate, but not all.
- The assessments used for the competences during the recruitment, while more focused and structured than before, aren’t always measured against a criteria rubric, nor are they often psychometrically valid and reliable. This means they aren’t highly accurate in measuring the competences they seek to assess.
- The output of the assessments during the recruitment rarely is quantified as structured data points, much less stored and analyzed accordingly for analytics.
The transition towards Competence based recruitment was a fundamental step in the evolution of recruitment. It paved the way of a more structured, objective approach which set the bases for Evidence based recruitment. Competence based recruitment walked, so that Evidence based recruitment can now start to run.
Evidence based recruitment
Now we’ve made it to EBR in the present. During the era of ‘big data’, this is the methodology that many forward-looking organizations have started to implement.
So what exactly is EBR?
In short, EBR is about identifying which traits and skills have made someone a high performer in a specific role and context, and then hiring candidates that demonstrate quantifiable evidence of those traits and skills during the recruitment process.
EBR provides further structure and accuracy to the previous competence based process through applied organizational psychology and data science.
Its defining characteristics are:
- Data-driven: both the key requirements needed for the role, and the candidates' traits and skills are quantified as structured data points.
- Valid & Reliable: the assessment methods used for quantified candidate evaluation are proven to be accurate through rigorous psychometric scientific standards.
- Closed-feedback loop: work-performance data is correlated with candidate recruitment scores data, which provides insights into which traits and skills have higher predictive power for different roles in different contexts.
- Objective & unbiased: the structured, self-correcting approach reduces human bias during both the assessment process and decision making, enabling fairer outcomes.
A systematic, data-enabled approach is what sets EBR apart. The evidence comes from the relationships found between candidates’ traits and skills, and actual work performance measured on the job.
Therefore, EBR can’t exist in a vacuum of the recruitment/talent acquisition function, but as a people analytics insight when the employee lifecycle is connected across talent functions (eg: talent management, performance appraisal, rewards, etc.).
This implies a bigger effort, but the pay off is well worth it: EBR is estimated to increase hiring success to an estimated rate of 70% to 80%[3], significantly above its predecessor.
The core steps of Evidence Based Recruitment
There are 5 steps that must happen for EBR to properly take place:
- Define performance outcome: clear understanding of what “strong performance” looks like for a certain role and context, and the behaviors needed to achieve that.
- Choose critical requirements: define the key skills and traits that enable the desired behaviours.
- Apply reliable selection methods: evaluate candidates with proven assessments of high accuracy and precision.
- Use unbiased decision-making: consider all data points obtained from the candidates' assessments and weight them against the predefined role requirements to facilitate objective decision making.
- Validate performance: Analyze job-performance data against recruitment data to further validate critical requirements as predictors of performance for specific roles and contexts.
In Academic Work, all these steps are facilitated through our own in-house technology, from our own ATS, a mix of our own assessments and from qualified vendors, matching scores, modeling algorithms, etc.
What makes us unique at AW are our own internal tech capabilities that allows us specific insights to predict work performance based on evidence, focusing on assessing the characteristics that truly enable top performance in a candidate, instead of assumptions.
In the next blog posts of this series, we’ll provide more detailed perspectives into each of the five steps of EBR, with actionable suggestions you can do now in your team or organization to start working more evidence-based.
If you have doubts or want to know more, don’t hesitate to reach out!
References
1. Estimating Traditional recruitment success:
- Murphy, M. (2011). Hiring for attitude, McGraw-Hill. Education, Leadership IQ study.
2. Estimating Competence based recruitment success:
- Stroo, M., Asfaw, K., Deeter, C., Freel, S. A., Brouwer, R. J. N., Hames, B., & Snyder, D. C. (2020). Impact of implementing a competency-based job framework for clinical research professionals on employee turnover. Journal of Clinical and Translational Science, 4, 331–335.
- Kolibáčová, D. (2014). The Relationship Between Competency and Performance. Acta Universitatis Agriculturae et Silviculturae Mendelianae Brunensis, 62(6), 1315-1320.
- Dadwal. S, & Arya. P (2024). Impact of Competency Based Recruitment and Selection on Retention of Employees. International Journal of Research Publication and Reviews, Vol 5, no 3, pp 3545-3548
3. Estimating Evidence based recruitment success:
- Sackett, P. R., Zhang, C., Berry, C. M., & Lievens, F. (2022) Revisiting Meta-Analytic Estimates of Validity in Personnel Selection: Addressing Systematic Overcorrection for Restriction of Range. Journal of Applied Psychology
- Sjöberg, S. (2014) Utilizing research in the practice of personnel selection: General mental ability, personality, and job performance. Doctoral thesis. Faculty of Social Sciences, Department of Psychology, Stockholm University, Sweden
- Kuncel, N. R., Connelly, B. S., Klieger, D. M., & Ones, D. S. (2013) Mechanical versus Clinical data combination in Selection and Admissions Decisions: A MetaAnalysis, Journal of Applied Psychology.