Do psychometric tests work?

Good question. What people usually mean when they ask if they work is: do tests predict anything useful about future work performance? The short answer is a resounding ‘yes’. As long as a test is used to measure an ability that is actually required of a particular job, then predictive validities are often in the 0.5-0.6 range. What this means is that at the top end of the scale, a test (the predictor) explains 36% ((0.6 x 0.6) x100) of the variance in the criterion – the criterion being something like a measure of productivity. By way of contrast other assessment methods such as the interview are often far less effective. A semi-structured interview would weigh in at 0.38 (14%) or thereabouts. And to get the whole thing in perspective, just in case you’re not impressed, in other fields such as the drug industry, predictive-type validities are often lower. For example, the association between Ibuprofen (the well-known anti-inflammatory) and pain reduction is in the region of 0.14 (2%) – see Robert Hogan’s article, details below.

Want to know more, here are some key references:

  • Bertua, C., Anderson, N., and Salgado, J.F. (2005). The Predictive Validity of Cognitive Ability Tests: A UK Meta-Analysis. Journal of Occupational and Organizational Psychology, 78(3), 387-410.
  • Hogan, R. (2005). In Defense of Personality Measurement: New Wine in Old Whiners. Human Performance, 18, 331-341.
  • Hunter, J.E, & Hunter, R.F. (1984). Validity and Utility of Alternative Predictors of Job Performance. Psychological Bulletin, 96, 72-98.
  • Schmidt, F.L, & Hunter, J.E. (1998). The Validity and Utility of Selection Methods in Personnel Psychology: Practical Implications of 85 Years of Research Findings. Psychological Bulletin, 124, 262-274.
About these ads

One Response to Do psychometric tests work?


Get every new post delivered to your Inbox.

Join 89 other followers