Weekly Blog #43 - Predicting performance...

This weeks article is related to a certain field of my current studies: appraisal of aptitude and related performance measurement tools. I felt the need to process some of the things I've read since I've gained some insights and data on how metrics to predict performance are used and NOT used by corporations which was kind of eye-opening...

IQ-Testing...


How do you feel about IQ-tests? Did you ever get your intelligence tested properly? Or your EQ (emotional intelligence) for that matter? And let me ask another question: how much time does your organization spend on predicting performance of new hires? And how often do you get proper, fact-based feedback regarding your performance accompanied by a thorough development plan?


Most of us have taken different types of intelligence tests during assessment centers or job interviews. Although you won't get a plain number as a results of these tests you get some sort of feedback since you are allowed to the next step of the process or not.


To generic intelligence testing in contrast, there are some specifics and boundaries. For example, you are basically not allowed to test the g-factor (a general construct developed in psychometric investigations of cognitive abilities and human intelligence) in Germany because tests have to be designed specifically in relation to the job the participants apply for. Therefore, in most cases only some dimensions get tested. Usually, these IQ-tests consist of different elements from different fields, e.g. maths, logical thinking, linguistics, verbal comprehension, abstract reasoning, spatial/figurative elements, ....


There are many upsides to intelligence testing, but one that is statistically significant and should change the way many of us think about it: it predicts success at work. Meta-analysis and data from different regions of the world have shown that. There is a significant correlation between intelligence and good performance at work which increases with more complex jobs of r=.51 (0 being no correlation and 1 being 100% correlation) - while the stability of general intelligence as a personality trait is very high at r=.66 (measured over a life span of almost 70 years).


So, intelligence is important to predict success at work. But why don't we use intelligence testing more often? In Germany, IQ tests are used primarily in the selection of apprentices and junior staff, although in less than 20% of the cases. For management levels the amount of testing varies between 3-4%. There are a couple of misconceptions why they are not being used at a higher level:

  • This form of testing is unattractive to potential candidates - they get "scared"

  • Other forms of testing for other competencies are being neglected

  • All of the candidates are intelligent anyways

  • Very intelligent people are displaying behavioral problems ore have autistic tendencies

All of these misconceptions have been debunked. Intelligence testing has been rated by applicants at the same level or higher than interviews and job references by former employers in regards to their validity and acceptance. And it is usually combined with other forms of testing, therefore different competencies can be tested equally in properly designed application processes.


Also, while it's true that applicants for management jobs are more intelligent than the "average" (IQ between 85-115), there are significant variations. Intelligence levels for applicants at low level management positions range between 90 and 120, applicants for higher positions display even higher levels (105-130). These differences matter! Interestingly, it has been shown that the difference between 105 and 130 is more significant in regards to job performance than the difference between 130 and 160. The additional job performance that is correlated to higher IQs seems to cap out at 130-135.


And I think I don't have to mention that there is 0 correlation between intelligence and displaying behavioral problems...


Personality-Testing...


There are additional traits that predict good performance at work. One of these is part of the Big-Five personality traits: Conscientiousness. The Five Factor Model (FFM) is a collection of personality traits that sufficiently and exhaustively characterizes human traits. The FFM (also known as OCEAN Model) consists of the dimensions Openness, Conscientiousness, Extraversion, Agreeableness and Neuroticism.


Conscientiousness reflects the tendency to be responsible, organized, hard-working, goal-directed, and to adhere to norms and rules. And it can be tested with valid personality scales as well! I am not talking about personality types (MBTI, 16PT, ...), even though knowing "types" can be helpful while assembling project teams. I am rather referring to personality scales that are accompanied by inventories of proven items collected and assembled in questionnaires that have been properly and empirically tested with significant samples in various populations.


Testing personality traits is usually very efficient and doesn't take a lot of time so it should be part of on-boarding and recruiting processes. Of course, the amount of items on personality scales can vary from 10 to over 200, and the more items per dimension you test the better and more accurate the results will be, but also hyper efficient tests like the BFI-10 (a 10 item version of the Big Five Inventory) can help due to their proven validity to get a solid understanding of ones personality traits.

In the past I have been relying a lot on my gut feeling while judging someones performance potential, be it in interviews with new hires or during the assembly of my project teams. Learning about proper scales as well as good and standardized tests with a high statistical significance and validity has definitely broadened my horizon.


Adding these methods to standard procedures like interviews, analysis of biographic data (e.g. CVs) and work samples and simulations should be a must in every proper recruiting process. Additionally, tests that predict ones level of emotional intelligence - where empirically validated scales are currently under development - and other factors, e.g. the "dark triad" of personality traits (Psychopathy, Machiavellianism and Narcissism) can deliver a whole new perspective on how capable we really are on judging future performance.

17 views0 comments

Recent Posts

See All