Research shows: cognitive tests strongly predict job performance

 A meta-analysis conducted in the United Kingdom (Bertua, Anderson & Salgado, 2005) has found that intelligence tests and tests of  specific cognitive abilities are strong, reliable and valid predictors of both job performance  and response to training.  Operational validities are in the range of .5-.6, meaning that these  cognitive tests can account for approximately 30% of the variance in job performance between  candidates.  This makes cognitive ability tests the single strongest predictor of job performance, over and above other popular measures such as personality assessments and job interviews. The predictive validity of cognitive ability tests generalises across job types and  settings, meaning that these tests are useful predictors of most jobs in most industries (if not all!).  Their ability to predict job performance is strongest for more complex roles, such as professional and managerial roles.

The meta-analysis involved 56 papers and books reporting 283 samples, including 60 samples with overall job performance as the indicator and 223 samples with response to training/training success as the indicator.  Ability tests used in these studies included general mental ability (overall IQ), verbal skills, numerical skills, perceptual abilities, spatial abilities.

Specifically, the study found that:

For job performance

  • Operational validities for job performance (how good the tests are at predicting job performance) were strong for all types of tests. The best tests were found to be tests of perceptual skills (.5) and general mental ability (.48).
  • When analyses were done by job type it was found that operational validities of tests for job performance were highest for professional jobs (e.g., solicitor, doctor) at a very strong .74
  • The operational validity of job performance was also very high for managerial roles, at .69

For training success

  • Operational validities for training success were also strong and were strongest for tests of numerical skills (.54) and perceptual skills and general mental ability (both .5).
  • The operational valdities were highest for engineers (.64), followed by professionals (.59).

The study concluded that cognitive ability tests are robust predicts of job performance and training success across a wide range of occupations. The findings were in line with earlier meta-analyses conducted in the USA.

Cognitive ability measures are the strongest predictor of job performance, and as shown by this study, employee’s response to training. If you want to access these strong predictors of job performance, including a wide range of cognitive ability measures assessing perceptual skills, numerical skills, verbal abilities and spatial skills, please contact us to find out more.

To find out more about our assessment tools, please visit our website.


Bertua, C., Anderson, N., & Salgado, J.F. (2005). The predictive validity of cognitive ability tests: A UK meta-analysis. Journal of Occupational and Organizational Psychology, 78, 387–409.