To test or not to test?

Testing is just one tool recruiters have in their arsenal,but how effective is it?The ideal solution in recruitment is to match the ‘right’ candidate to the ‘right’ job in the ‘right’ organisation. It’s as much an art as it is a science. “Science?” I hear you say. Well, yes recruiters use scientific methods and tools as an adjunct to the total recruitment process.The tools that are most widely recognised are psychometric assessments or ‘tests’ – the mention of which is likely to send some people into spirals. The most-commonly used assess ability and personality, although they are just a couple of many options.Firstly, we have to be clear exactly what we are assessing. An initial step in this process is to have an accurate, current and comprehensive job description. (It’s surprising how often we find these do not exist.) From that we select the most appropriate tools – the right tests, for the right role, for the right reasons. For example, if we need to know how well an individual takes on new information, and draws logical conclusions and inferences from that, then a basic test of verbal reasoning is not going to suffice.We can go further. For example, we can extract the key competencies from a job description and generate customised:?    application forms ?    structured interview questions, and?    specific reference checking questions From a client’s perspective, this is highly effective. We know this is the case because candidates selected using these methods stay for a longer period in their new roles.Recruiters can also generate ‘Ideal profiles’ and validate those against the test results of existing position holders. This is effective because it clarifies to the client exactly what they should be looking for and the best approach to achieve their business goals.A fundamental of any assessment process is that standardised measures and guidelines must be observed. Only tests that have high levels of ‘reliability’ (results are consistent for various groups of people over time) and ‘validity’ (the test measures what it claims to measure) should be used. Test scores are then compared to general population samples, or in some cases to more specific groups, such as an engineering or senior management ‘norm’ group. Company norms can be developed from samples of existing staff and this can highlight hidden strengths among the existing talent pool.But can we really call this process scientific? There are examples where this doesn’t work effectively, specifically if the above criteria (using valid, appropriate tests administered and interpreted by qualified people) are not followed. Participants should also receive feedback. This is not just out of politeness; results need to be discussed in context and the individual given the opportunity to elaborate on the findings. Simply putting all candidates through the same class of tests “because those are the ones we always use” is nonsense. Unfortunately, this is a fairly common practice. Test results can be used to highlight individual training and development needs, and have great use in coaching programmes. They objectively identify areas of need and the same type of test can be re-administered post-programme to critically assess ‘growth’ or adjustment. Used in group settings (as comparisons between individuals and as consolidated group data), tests can be used when working on areas of interpersonal conflict and dysfunctional group dynamics. Different types of tests can also be used effectively in career planning, assessing stress levels, organisational culture and much more.