Assessment Centres are often considered to be the “Rolls-Royce” of selection processes. The opportunity to test participants on a wide range of exercises directly related to the job, combined with in depth observation over a number of days must, surely, produce the best results. But does it? Too many centres are flawed, both in design and implementation, often rendering them completely useless. We show you what the flaws are and how to avoid them.
If you were Chief Executive of a large organisation and you were told that your Assessment Centres, on which you had been spending hundreds of thousands of pounds, were unreliable, inacurate and useless at predicting performance, how would you react? Fury? Despair? But this is precisely what is almost certainly occurring up and down the country at this moment.
We know that Assessment Centres are great ways of predicting performance. The research says so (time and time again) and the fact that their use is widespread in both the private and public sectors indicates the level of confidence people have in their accuracy (or validity as the techies would say). The trouble is that this is only true if the Centre has been carefully designed. There are many pitfalls which can invalidate your carefully thought-through (and expensive!) assessment centre.
All right, you have heard this before and think you do it already…but do you? Tailoring an Assessment Centre to the job does not just mean thinking of a few things that the jobholder does and then finding an exercise which gets somewhere close. Tailoring involves looking at:
- the demands of the job
- the culture of the organisation and team
- how relationships are dealt with
- the level of support typically encountered from the boss
- the organisation’s communication style
- the stresses and pressures encountered day to day
…and ensuring that all these are replicated in the exercises. We know that the closer the exercise is to real-life demands, the more likely it is to test the competencies under the conditions in which they operate. If I need a high level of personal organisation and I am tested on a task where everyone likes me, nobody is under pressure, everyone does what I say, I will probably perform very differently than when I am loathed, everyone is under severe strain and I am completely ignored. If this is what the organisation is like (heaven forbid), then these are the conditions you need to replicate, possibly using actors, dealing with a real-life problem, working out what peoples’ likely reaction is going to be. Only then will the exercise test what you really want it to test….real life!
We all think we are bright….unfortunately, for Assessors, the demands of observing can tax us to the limit. And if we are unable to cope with all the demands on us (writing notes, observing behaviour, trying to find the right place on the response sheet, etc), what do we do? We guess! Can any observer put their hand on their heart and honestly say that there has never been one moment when they might have missed a behaviour by a participant on a centre. I doubt it.
The trouble often lies in the design of the Centre itself, where (especially on a group exercise) an observer might be required to look at two or even three participants, analyse their behaviour, write down their comments and even participate in the exercise itself. This simply means that the demands on the observer become too great and their accuracy diminishes. This means that, if people are not analysed correctly, they probably obtain an incorrect assessment and the wrong person gets chosen for the job.
The answer is to make things easy for the observer and ensure that you only measure the most important dimensions. If your Centre is measuring more than 8 dimensions, it will probably be invalid because your observers will not be able to cope. Ideally, you should only measure between 4 and 6 dimensions on a centre and, ideally, each exercise should only measure one dimension (don’t take my word for it…look at the research!)
On the same lines, another way in which you can make life easy for observers is to ensure that they know what it is they are looking for. We know that, if you take two people, ask them to observe a person’s behaviour and then assess their leadership skills, they are likely to be measuring very different things. One could be measuring empathy with others and vision, whereas the other could be measuring power and assertiveness.
Assessment Centre designers need to itemise precisely the behaviours that observers should be looking for and show the level which each behaviour achieves. The behaviours need to be capable of being observed without ambiguity and it should be clear what level of competency each behaviour is achieving.
Some people think that you should keep the content and process of the centre as secret as possible. This is to ensure that participants can not “bone up” on what the centre is going to test and, thereby, gain an advantage over other participants. Others think that you should tell people in advance what to expect as it lessens nervousness and people can perform better.
It is always better to tell people beforehand what is going to happen. In point of fact, it is very difficult to rehearse the behaviours required on the centre – after all, if you have always lacked empathy with others, getting out the textbooks and studying it the night before is unlikely to help much. If everyone knows what the process will involve, their nervousness will decrease and they will be able to concentrate much more on the task in hand and give of their best. This means you will not only have a more reliable and accurate centre but also a happier one!
We all know that people are different. It, therefore, follows, that groups are different. Which means that individuals operating in groups are likely be impacted differently according to those around them.
Let’s say you are measuring asseriveness. Person A is in a group full of extremely powerful people and, therefore, is unable to get a word in edgeways. This means that the observer, noting a lack of participation, marks them down as “unassertive”. But what if Person B were in a group full of quiet empathetic types. They are able to display their prowess in assertiveness, the others acquiesce and listen and the observer marks them down as “highly assertive”.
In any exercise, people need to have the opportunity to shine and it is the responsibility of the centre designer to make sure that this happens. In group exercises, this can often be achieved by assigning roles (thus ensuring that the candidate has a role that nobody else has) or by providing information selectively around the group.
The best thing to do, believe it or not, is get rid of it altogether! The most accurate results are those which come from the observers themselves – and not from the wash-up session, occurring sometime after the event. It stands to reason – it is the observer themselves who have the best chance of analysing and assessing an individual’s behaviour. But, so often, particularly if a very high or low score has been allocated, people will say in the wash-up session “Are you sure…that seems very high (or low)”….and the observer becomes very unsure as a result of all this peer pressure and alters the mark.
Of course, it may not be appropriate to bin the wash-up session altogether. But centres should try to avoid those long, protracted discussions where everyone analyses each participant’s abilities – whether they have seen them perform or not. Unless there is a problem, simply accept the observer’s analysis – they are more likely to be right than you are – after, they are the ones who saw the candidate perform.
In our business as consultants who run many centres a year, we feel that it is an excellent idea to include people from the organisation as observers. They understand the culture, they can often bring a great deal to the design of the centre and they are able to answer questions about what life in the business is actually like.
The only difficulty is that, where people are not knowledgeable enough about the function and process of an assessment centre, they may not observe and assess as effectively as they should. This means that candidates are incorrectly assessed which may mean that wrong decisions are made.
It is preferable to use highly experienced observers. At the very least, assessment centre leaders need to make sure that training given to inexperienced observers is of the highest order and that only those who are capable of undertaking the job should participate.
Assessment centres are a great way of ensuring that you have the best people in the job. But only if they are done well. It was found fairly recently that the difference between a middle ranking and a high performing manager earning £30,000 amounted around £25,000 difference on the bottom line. That is that, in general, a high performing manager will either bring in or save your company £25,000 each year. That is a saving of around £125,000 over 5 years. Multiply that by the number of employees in your organisation and you will get the picture….choosing the wrong person will cost your organisation big time.
Use assessment centres well and you will have the right people in the right place at the right time. But you need to make sure you use them well.
Tailored to the needs of the client
Well-documented and thoroughly thought-through behaviours
Honesty in communication to candidates
Exercises which match real-life as much as possible
Opportunity for all to excel
Carefully designed and objective scoring system
Slick and efficient administration
Inclusion of feedback process
Appropriately run “wash-up” session which explores the issues
This article presents a series of methods and tools that can be used in the process of assessment. However these guidelines do not guarantee the success of setting up assessment actions, but an assessment professional such as Sinclair can.