How Valid Are Psychometrics?

Friday, 7 July 2013

Psychometric testing is used commonly in business today for everything from recruitment to professional development. But how much weight should we put on psychometric tests when it comes to hiring, firing and promotion?

Critics believe that psychometrics is an inaccurate science and that data from psychometric testing should not be relied on when it comes to making important business decisions. In fact the law is quite strict on using psychometric testing in the workplace and it is important for organisations to adhere to best practices when it comes to administering tests.

Is psychometric testing effective?

The first concern when considering if a personality test is accurate or not would be is it possible to cheat? After all, in a multiple choice test, it is pretty easy just to tick the choice that you think your manager will want to hear, rather than the one that is closest to the truth.

In actuality, well-designed psychometric personality tests are extensive enough and structured in such a way that inconsistent data is easily filtered out when processing the results. However it is important not to rely on such tests as the only method of assessment, but rather use them together with other techniques.

The results of psychometric testing may not be 100% accurate but neither are other assessment methods. A study by Schmidt and Hunter in 1998 found that psychometric tests were actually more accurate than interviews, job knowledge tests or experience when it came to predicting future job performance. Only work sample tests provided a higher accuracy of judging performance.

Psychometric tests are also designed to give reliable and repeatable results. Provided the candidate has experienced no major life changes, either professionally or personally, or they have not received specific training, you can expect an individual to get the same results from a personality or aptitude test several years later. With other techniques such as interviewing, it is highly likely that results will differ depending on arbitrary factors.

Best practices for psychometric testing

For test results to be meaningful and accurate, the questions must be valid and non-biased. This can be quite difficult especially when it comes to culture bias, which has been shown to have an effect on intelligence tests.

It is important when administering a test that every candidate is subject to the same conditions such as physical environment, lack of interruptions, given instructions and time limit.

If not performed carefully, psychometric testing can discriminate against certain candidates and this can have legal implications. Tests should not favour younger or older employees or be biased against certain ethnic groups. The best way to ensure this is to work with a specialist company that has experience in psychometric testing to ensure that the test is well structured and the resulting data is processed effectively.

Like it or not, the popularity of psychometrics in the workplace shows no signs of declining. For advice on how to manage psychometric projects within your business, please contact us for an informal discussion, or read on to the next article.

Leave a comment