Home | Contact

The Relationship between Reasoning Ability and Job Performance

     

In their seminal paper, Schmidt & Hunter (1998) reviewed over 85 years of research into the validity of different selection methods. They concluded that reasoning tests have consistently been found to be the best predictors of job performance, with graphology (not surprisingly) having been consistently found to be the least valid predictor of job performance. They also reported that in addition to predicting job performance, reasoning tests have consistently been found to predict the effectiveness of staff training programmes, with those staff who have higher levels of reasoning ability benefiting more from training than those who have lower levels of reasoning ability. Using statistical procedures (meta-analysis) to aggregate results across different studies, Schmidt & Hunter (1998) concluded that reasoning tests have validity coefficients in the order of 0.51 for predicting job performance and 0.56 for predicting trainability. (These coefficients were corrected for attenuation due to measurement error and restriction of range, and should therefore be considered as providing an upper limit of the validity of reasoning tests.)

 

However, these findings are not as clear cut as they may at first seem. While they clearly demonstrate that reasoning tests have a useful role to play in many selection decisions, the size of these effects is still modest, with reasoning tests typically predicting less than 30% of the variability in job performance.

 

Not surprisingly, Schmidt & Hunter (1998) also found that reasoning tests were much more predictive of a person’s performance in professional/managerial roles (with aggregate validities of 0.58) than they were predictive of a person’s performance in unskilled jobs (with aggregate validities of 0.23). Moreover, they found that the inclusion of a job knowledge test or job sample test in the selection procedure, alongside a reasoning test, further improved the prediction of job performance, as did the inclusion of a personality test or structured job interview alongside a reasoning test. This clearly demonstrates that, as the theory of multiple intelligences would predict, reasoning tests only assess one aspect of the many skills, abilities and aptitudes that determine job performance. As such, other selection methods (such as personality tests, work sample tests, structured interviews, etc.) have to be used alongside reasoning tests in order to assess the other skills and abilities that contribute to successful job performance.
 
 

 

 
 

 
 

 

   

Copyright Psychometrics Ltd 2006 | Privacy policy
Disclaimer | email: info@practicetests.co.uk