Cognipsy - Blog #scientific assessment

Why do new hires fail?

Wed, 11 Nov 2020 13:21:42 +0530

According to the U.S. Department of Labor, the price of a bad hire is at least 30 percent of the employee's first-year earnings. Another study pegs this cost at USD 15000 per wrong hire. While most hiring managers agree that wrong hires are costly and draining on organizational resources and morale, there is little agreement about why this happens?

We all can agree that no hiring manager or recruiter would hire a new employee only to set them up to fail. Yet, new hires do fail and often in spite of remedial strategies. A common reason for this is that often role requirements are not clear and at other times there is too much focus on skills that can be acquired with time. What many hiring managers fail to recognise is that there are some fundamental tendencies that are hard to fix or acquired due to sheer individuality of people.

SPS Scan identifies 5 clusters of reasons why people fail at jobs and struggle to remedy or learn key behaviours. These behaviours are so fundamental that in the short term they are hard to imbibe often because they come with their well entrenched belief systems which are hard to change. Let's take a quick look at these fie clusters.

1: Antagonism - Pattern of behaviours that antagonise others - Burning bridges and inability to sustain relationships.

Antagonism is a common cause for non-performance in roles requiring social mediation, customer service, account management, teaching/training, helping and care-giving. Antagonism has strong negative relationships with Agreeableness and Conscientiousness.

2: Detachment - Pattern of behaviours that lead to isolation - Inability to socialise with others and create favourable social reputation.

Detachment is a common caruse for non-performance into roles requiring frequent social interaction, information sharing, initiating contacts, speaking publicly and asserting self over others. Detachment has strong negative relationships with Extraversion and Emotional stability.

3: Inhibition - Pattern of behaviours that inhibit appropriate actions - Inability to start tasks and maintain goals.

Inhibition is a common cause of non-performance in roles requiring responsiveness, being busy, doing a large volume of work, stressful timelines and energy/endurance. Inhibition has strong negative relationship with Emotional Stability, moderate positive relationship with conscientiousness and moderate negative relationship with Extraversion.

4: Disinhibition - Pattern of behaviours that exhibit inappropriate actions - Inability to exercise caution and deliberation in taking prudent actions.

Disinhibition is a common cause for non-performance in roles requiring high stakes risk, procedural action, bigger responsibilities and work which may undergo legal/regulatory scrutiny. Disinhibition has a strong negative relationship with Conscientiousness and moderate positive relationship with Extraversion and Openness.

5: Incomprehension - Pattern of behaviours that impact cognitive efficacy - inability to solve problems and think properly.

Incomprehension is a common cause for non-performance in roles requiring sound judgement, analytical thinking, complex problem solving, dealing with novelty, dealing with ambiguity and decision-making with long term consequences. Incomprehension has strong negative relationship with Openness and Emotional stability.

As expected, our research shows that Disinhibition and Inhibition explain maximum variance in performance at workplace as described in Bartram's great eight framework. Statistically this is a great validation for these constructs. But from a practical stand point it's the granular 30 scales that best describe more specific risks. E.g. Managers who are very high on openness are more likely to have risks in areas like "emotional avoidance" and "self-awareness". This is one of the many reasons why understanding and avoiding wrong hires requires must more granular and data based intelligence.

What we know about predicting leadership

Thu, 29 Oct 2020 18:33:10 +0530

Leadership, as you may already in know, is a rather complex phenomenon. But what you should also know is that great amount of money is spent by companies to hire leaders, train/develop them and deal with them should they not perform. Of all costs the latter is the greatest and is seldom accurately reported.

In this summary of research we briefly discuss the science of predicting leadership emergence and effectiveness. An emergent leadership is a very different from effectiveness model of leadership. However, this difference is not the scope of the article.

Intelligence

Better described as general cognitive ability, intelligence is a great predictor of leadership. In fact of all psychological factors g can predict leadership emergence the best (r=0.4). Ability to work with complexity, think about future, build scenarios, learning things quickly and solving problems on the fly are all much needed competence in a leader and g predicts just that. Needless to say g alone is not a mark of a good leader, but from statistical point of view, g predictss leadership the best.

Personality

Personality is a good predictor of job performance. Leadership requires different behaviours at different instances. The ability to exercise one's traits is a mark of good leadership. However, a great deal of research has proven that conscientiousness, neuroticism and openness to idea are all good predictors of leadership emergence. Extraversion and agreeableness have mixed results. Surprisingly, agreeableness has been found to have negative correlations with leadership emergence and positive correlations with effectiveness.

Handling Stress

Leaders are less impacted by stress. And the proof is irrefutable. Biological evidence from cortisol level (stress causing hormone) suggest that the basal levels of cortisol in leaders are higher. Which entails that leaders are used to high stress and their cortisol levels do not increase much with stress. Emotional stability factor from personality also has positive correlations with leadership emergence. Hence, ability to not be perturbed by stress is also a sound predictor of leadership capability.

Emotional Intelligence/Ability

While studies show that EI predicts leadership effectiveness, EI itself is highly correlated with g and big 5 factors. Hence , from psychometric standpoint it doesn't offer any additional insights into leadership capability. Alarmingly, some studies have shown that high EI leads to lower leadership effectiveness.

Creativity

Creativity both in terms of a trait as well as ability is a strong predictor of leadership effectiveness and not emergence. Surprisingly this is true across industries and sectors. This could be a reflection of the fact that strategizing and exploring new avenues of growth do require creativity.

In all, psychology has enough insights into what would make a great leader. Furthermore, definitions of what effective leadership looks like across cultures is similar. This doesn't mean that culture has no impact on leadership. Adequate evidence exists to suggest that socially acceptable leadership behaviours differ from culture to culture. E.g. in US a good leader is expected to be an extrovert while in Japan a good leader is culturally perceived to be introverted. That said, a great amount of leadership can be scientifically predicted with analytical models which are customised to an organization, industry and culture.

What's wrong with hiring?

Thu, 29 Oct 2020 18:14:05 +0530

In a world where human resources is trying hard to justify its existence while business quite literally ridicules the people practices, we are often left to introspect about reality. There are two main reasons for this, one, measures of people practices are very subjective (not that they can't be measured) and two, the impact of people practices are not realised immediately and even when realised, this happens very indirectly.

But not all is baseless. One aspect of HR which receives the least criticism and in my view is the least evolved is the hiring function. Addressed as hiring, recruitment or talent acquisition, practices pertaining to the function are in fact the most baseless and unfounded in management science. The measures of hiring function on the other hand are much more objective and can be measured with much more accuracy in a short frame of time.

Lack of Strategic Principles

Would you hire any different if you were to work for a competitor?

Ask this question and if your answer is no, well you are definitely not hiring right. But this clarity is a part of a bigger problem i.e. HR function not being strategic enough and/or not connecting to the business. The kind of people a company acquires is a single sustained source of advantage and a single source of loss too (once hired law makes it difficult to rid of a resource). While you may struggle to meet numbers for the right cost and time, without the broader principles in place, hiring may just not deliver what business needs.

Use of Resumes

It isn't uncommon to come across recruiters who would say that a glance at a resume would clearly tell them about the fitness of a candidate. Well thought out lab studies with resume have provided poor results. Resume/CV do not predict performance at work (r = 0.05 - 0.08), even with scientific methodology, let alone intuitive gut feel. Structured form based applications on the other hand are moderately better, yet disappointing (r = 0.12). 78% of resume/cvs have been found to be not simply inaccurate but fraudulent (SHRM study). So ultimately impressive resumes predict impression management, verbosity, some intelligence but neither competence nor performance. Unfortunately, this still persists as a screening process for hiring at almost all levels.

Unscientific Hiring Criteria (GPA, Campus, Credit Check...)

Use of any hiring criteria must be based on sound evidence. Credit checks, GPA, campus of graduation are criteria that have no basis in management science. Credit checks have small prediction value of (r=0.11) for entry level roles and negligible (r=0.05) for managerial roles. Similarly, GPA only predicts performance in training and the performance in first year post graduation. Post that GPA has meager correlations with job performance across roles. Campus of graduation and GMAT scores can only predict the income of an individual (r=0.49). However, they have no correlation with performance (r=0.007).

Numerous research papers cite that unstructured interviews are the least effective tools for hiring and many have found them to have negative correlations with long term performance. Similarly work experience in years, beyond three years in a role, doesn't predict performance across levels. In spite of their ineffectiveness in predicting work performance, these criteria are basic requirements for entering most organizations. Professional qualifications, reasoning tests, structured interviewing, personality assessments and/or situational judgment tests are far more predictive of performance and yet used the least.

Unrealised value of Talent

While many say its ok to not get the best talent as long as we get the talent we want, it is a grave mistake. If as a company you are that pays at 70th Percentile in the market, having a talent below 70th Percentile is a bad deal. But how do we know the caliber of talent? No easy answers but a good indicator is the selection ratio. If you have had valid screening and selection mechanisms, you should not settle for hiring one of the three best out of 10 candidates. For mass roles within organization, this should be a golden rule when hiring. Usually hiring teams try to focus on meeting numbers for such roles. Besides, every bit of compromise is costly. Consider the fact that with every ten percentile increase in the caliber of talent, productivity increases exponentially. Employee at 90th percentile is at least seven times as productive as the average employee at 50th percentile.

Taming the Gut Feel

Gut feel in all fairness can only be recommended when you have all candidates who seem to do equally well in all criteria of selection. That is the only possible scenario where hiring decisions based on gut can or should be acceptable. Eventually all gut feel amounts to individual projection biases. Any sustainable talent strategy should steer clear from it, not just when hiring, also while making other people decisions.

Qualified Professionals

One reason why hiring professionals excessively rely on gut feel and unscientific practices is because there are no definitive qualifications for being one. One would assume that all recruiters should at least have basic understanding or personnel or occupational psychology. However this is far from what the market demands out of them. Other skills like behavioural interviewing and observation are also equally scarce. The danger of relying on experience eventually leads to a misplaced sense of what is right and acceptable in hiring.

Misleading Measures

The basic metrics that define hiring should be aggressively measured. While most hiring departments today measure, cost per hire, attrition and average time to hire, more important metrics like base rate (what percentage of screen pool was good enough for successive stages) and hit rate (what percentage of hired candidates were good performers) are rarely tracked.

Hiring right is the pre-requisite for all business both operationally and strategically. While numbers and cost factors are important to support business, the strategic component when ignored can corrode the business slowly. A scientific and proven approach to measurement however is necessary to deliver business impact. Its time to transform both our hiring as well as hiring talent.

References and Suggested Reading:

Credit Check Study

The Blackwell Handbook of Personnel Selection , 1st Edition

Oxford handbook of Personnel Assessment and Selection , Nov, 2012

GMAT study