Assessment centres can replace interviews and a host of other tools as methods of employee selection. They are highly favoured by many large employers and have become the stuff of TV documentaries, offering viewers insight into the selection to regiments and corps like the Royal Marines.

And yet they are hugely misunderstood by most managers and considered highly controversial by academics.

So, do assessment centres work? And if so, how does a manager make an assessment centre work for them?

Firstly, some definition.

Exercises and ratings

An assessment centre is where a number of candidates participate together, undertaking exercises as selection tests while being observed and rated by multiple assessors. The candidates are effectively in competition. Some managers refer to assessment centres as group interviews, though they must not be confused with situations where candidates attend at the same time but have multiple assessments in personal time-slots across the one or two days. This time-slot approach occurs when selecting in the public sector such as when interviewing for head teachers in education.

In an assessment centre, the exercises are discrete, observable simulations that might allow demonstration of a conglomeration of skills, knowledge, intelligence, personality, reasoning, behaviours, motives, values, beliefs and attitudes. Traditionally, the assessors meet together after each exercise and at the end to determine which candidate they thought did best.

Departed from normality

The first thing to recognise is that assessment centres are fictional environments. No workplace is like an assessment centre. In the assessment centre, everyone is in a heightened emotional state, striving to impress observers. They are striving to demonstrate that they are better than their peers. Mass cage fighting might be a close analogy! Or think Big Brother compressed into one or two days but without the dormitory scenes.

But enough of the cynicism!

This departure from the workplace places a high onus on the designers of an assessment centre. The exercises must illustrate particular individual characteristics in the candidates while recognising this artificial setting. Someone who is naturally extroversive, loving discussion with others may be more subdued in this cooking pot. Someone who naturally considers the feelings of others and strives for good inter-personal relations may be forced to abandon their normal agreeable behaviour when faced with crass behaviour from their peer candidates.

Rating complexity

Observers and raters must understand the complexity of the assessment centre. Interpersonal interactions determine behaviour and anyone trying to judge the performance of a candidate must recognise that this causes inconsistencies. The person who might go on to be the best hire might not stand out so clearly in assessment centre exercises. Academics worry about the value of assessment centres because the construct validity is low. Construct validity is a measure of the degree to which the assessment measures what it’s supposed to measure. It’s difficult to uniquely discern individual traits, reasoning abilities, competencies, habitual behaviours and the like.

That said, criterion validity can be high and this is one of the big attractions of assessment centres. Criterion validity is that all-important metric that links how someone does in tests to how they perform when in the job. Whether criterion validity is achieved or not depends entirely on centre and exercise design.

So even if specific traits can’t be discerned, scoring can be developed that predicts well how each candidate will do in the job they’re being assessed for.

So, what about the detail?

Centre and exercise design

Every exercise will be multi-faceted. Every exercise will be multi-dimensional. Those dimensions – the criteria that the hiring manager seeks – must be identified. And he or she must know that the individual traits or constructs sought are embodied within the criteria. This is particularly important where a construct like personality is to be assessed. Traditionally, personality is best assessed using a personality inventory and analysis and it may simply not be realistic to make any personality decisions in an assessment centre.

Candidates will likely do well in some exercises and poorly in others and observers/raters must be ready for this. It’s unlikely that any one candidate will be good in all. This makes scoring difficult and demands a well-structured scoring system. Raters must not confuse individual candidate scores between exercises, giving the benefit of the doubt, perhaps, from one exercise to another.

Scoring without discussion

Scoring will demand an exercise-to-criteria mapping matrix that enables scores to be entered individually and aggregated. It’s critical that observers/raters do not discuss their opinions with colleagues or scores will lose integrity as rater bias and assumptions grow in significance. Simply, don’t talk, just add up.

This last point goes against received assessment centre wisdom. Managers taking the observer/rater role will naturally want to have a wash-up at the end when a big discussion determines who to hire. Discussion can negate the power of the assessment centre, allowing strong, biased opinions to dominate.

So, in conclusion.

Conditionally valid

Assessment centres can be a valid way of determining who will perform well when hired and in the job. But this will only happen if the exercises are carefully designed. Those exercises should be designed from criteria developed from the job description.

Observers/raters must be trained in observing and on the scoring system. The overhead in running an assessment centre is high. It’s not just a question of asking a group of managers to read a brief and take their seats by the cage.

A couple of final comments though. Many candidates do think that assessment centres are a fair way of determining who’s best. Of course, other candidates, on seeing selection by assessment centre will simply not put themselves forward for ‘cage fighting meets Big Brother’. They recognise that their strengths won’t shine in such a public arena.

Assessment centres are easy to organise for internal selection within a large company. And candidates for graduate and apprenticeship positions will eagerly block out set times to attend. But senior candidates already in employment can’t be so flexible. For them, assessment centres blocking one or two days of prime time are just a no-no, when one considers that they might have to attend several before winning a job offer.

In the end, firms must consider carefully the advantages and disadvantages of assessment centres considering the advantages and disadvantages of alternatives.

For most managers, assessment of individual candidates through GMA tests, personality assessments, situational judgement tests and work sample tests balled up in structured interviews will be their preferred selection method.