Evidence dating back over 4000 years suggests that the first aptitude test was developed and used in ancient China in order to select civil servants. Western diplomats and missionaries eventually discovered this use of testing for selection, and advised the British east India Company to follow suit in 1832. This move was also emulated by the British civil service, leading to the spread of occupational testing across the western world.
Francis Galton, Charles Spearman and Karl Pearson where the first psychologists to scientifically research intelligence testing. Using and inventing statistical methods such as correlation, variance and standard deviation, these psychologists were able to measure, and identify relationships with cognitive ability. Since then, many other psychologists and psychometricians have made huge advances in the realm of psychometric testing, including L.L Thurstone, Raymond Cattell, Hans Eysenck and many others.
The psychometric industry as we know it today however began when Peter Saville and Roger Holdsworth founded the global psychometric test publisher SHL. SHL pioneered the use of occupational psychometric testing for selection and assessment purposes. Since then the industry has grown massively, with virtually all large employers using psychometric testing for recruitment and assessment purposes.
How psychometric work
The reason why psychometric tests are used for recruitment, selection and assessment purposes is that the results have been statistically correlated with high job performance. Correlation is a statistical relationship between two different sets of data, and in the case of psychometric test scores and job performance, this is a moderate-strong positive correlation, meaning the higher the test scores, the higher the job performance tends to be. However the correlation is not a perfectly correlation, and some candidates may score highly on the test and perform poorly at work, or score poorly on the test and perform highly at work.
In order to prove that a test is legally worthy of use in selection, academics, psychometric test publishers and occupational psychology consultancies must prove the reliability and validity of their tests. Reliability is the consistency of a measure, whether it produces similar results under consistent conditions. This may be tested by asking test takers to take the test twice, getting multiple people to take the test, and testing the consistency of the scores within the same test. The test needs to be reliable in order to valid. Validity is the degree in which a measure is supported by the evidence and related theory. The extent in which test performance correlated with job performance is a matter of validity. A test which can accurately predict job performance would be a valid test, and a test which cannot accurately predict job performance, would not be a valid test.
Once the organisation providing the testing has the candidates test results, they are compared against a norm group. This norm group is a large collection of previous results from previous test takers of a similar demographic, for example if the test taker is a professional; a professional norm group would be used etc. This norm group serves as a comparison, contrasting the candidate’s performance against the average performance of the norm group. This score is then reflected by a percentile rank, for example a score in the 50th percentile means that the candidate has scored higher than 50% of the norm group, a score of 75% indicates the candidate has scored higher than 75% of the norm group and so on.
Research by academics, psychometric test publishers and occupational psychology consultancies have found psychometric testing to be the most effective predictor of a candidate’s job performance. It is therefore no surprise that organisations are increasingly using psychometrics to meet their recruitment needs. Psychometric testing has outperformed every other major selection procedure, from interviews to assessment centre exercises in predictive ability. Research has also found that the predictive power of psychometric tests does compound with the addition of other selection procedures, boosting predictive power. Candidates applying at large prestigious organisations will therefore be required to undertake various selection procedures for this reason.
Other selection procedures, such as interviews and assessment centre exercises may be vulnerable to biases. For example during an interview situation, the interviewer may make selection decisions based on irrelevant or irrational factors, such as stereotyping. Aptitude testing however provides an objective method of assessment, unaffected by biases and subjective influences.
Another benefit for psychometric testing is that employers can test the abilities most relevant to the competencies of the role. For example for an investment banking role, testing for numerical reasoning ability would be a better predictor of job performance than in less numerical occupations. Similarity, testing for verbal reasoning ability may be more predictive of job performance in journalism than in occupations requiring less writing.
Psychometrics are increasingly utilised by top employers due to their highly effective predictive power. Similarly, their comparatively cheap price and convenient method of delivery further improve the attractiveness of psychometric to employers. Preparing and researching psychometric tests is a brilliant way of ensuring your readiness for your next big psychometric test, helping you to maximise your performance and increase your chances of success.