Centres of attention
Assessment centres are widely recognised as a superior selection method. We investigate whether their reputation lives up to scrutiny and consider the most recent trends in their use.
The multiple application of different selection tools, combined with the extended period of time available for observation, contributes to the effectiveness of assessment centres.
The resource-intensive nature of assessment centres makes their use inappropriate for some vacancies. They are commonly used for selection to graduate, professional and management positions.
The design of a centre can be a complex undertaking, and the starting point should be the skills and attributes needed for the role.
Assessors play a key role in the assessment centre process, and should be trained in a number of areas.
The candidate experience is an important consideration. Participants should be fully briefed beforehand and, if possible, be provided with feedback afterwards.
Evaluations should be carried out to check that the assessment centre is performing fairly and effectively.
|
It is not without good reason that assessment centres are seen as the "Rolls-Royce" of selection methods. But unlike the luxury car, which is universally known but bought by relatively few car owners, assessment centres are held in high regard and are in widespread use.
The Chartered Institute of Personnel and Development's (CIPD) 2004 Recruitment, retention and turnover survey found that overall slightly more than four in 10 (43%) employers use assessment centres1.
IRS's recent research into graduate recruitment and selection highlights the strong position that this technique has gained among graduate recruiters. Assessment centres are used by just over half (52.7%) of such employers2. For large organisations employing more than 10,000 people, their usage in graduate recruitment rises to 95.2%. This makes assessment centres by far the most popular selection tool for this size of employer.
Almost three-quarters (74.1%) of graduate recruiters rate assessment centres as their single most effective selection method. In contrast, face-to-face interviews come a poor second, rated effective by less than half (46.4%).
So, if assessment centres are commonly regarded as the ultimate selection method, why do so few employers use them? And why do organisations that use them in their recruitment of new graduates not employ them to fill other types of vacancy?
The answer is simple - they can be costly and complex to design and run, which explains the IRS finding that their usage is greatest among the largest category of employer. Only in a minority of instances can the case be made that assessment centres represent a cost-effective option. The business case is not always straightforward, though. Costs can be reduced if an organisation has access to suitably experienced and trained expertise in-house. And their effectiveness can be improved where the outcome of selection decisions has a strong bearing on the future success of the organisation.
Proving their worth
Stephen Taylor's key resourcing text3 defines an assessment centre as "assembling in one place several candidates who are applying for the same position and putting them through a variety of different tests". It is important to appreciate that a centre is not a place, but a process that candidates must go through. The period of time over which the centre is held - typically between one to two days - provides assessors with a large window ofopportunity through which to observe and rate each candidate as they perform the set exercises.
Well-designed and run assessment centres provide a coherent approach to selection. The role for which the centre is being used to identify recruits is carefully analysed, and the various exercises and other selection methods are chosen for their ability to assess important aspects of the role. Ideally, each aspect of the role will be assessed by at least two different selection methods during the centre. And each candidate's performance in each selection method will be observed and assessed by at least two trained selectors. This is an evidence-based approach to selection, with objectivity and consistency as its watchwords.
As Sue Sherliker, a consultant with Anne Shaw Consulting which specialises in assessment, development and training, points out: "An assessment centre is a systematic collection of activities which identify the critical competencies that effective employees exhibit, and they are undoubtedly the best selection tool we have at the moment. This technique has proven to have the highest levels of predictability for future job performance, as the selection process is not having to rely on the application of one method alone. Rather, it is many methods put together to produce a much more thorough process on which to base appointment decisions."
The objectivity and fairness associated with assessment centres also owes much to the evaluation of candidates' performance by more than one assessor, and the fact that participants are assessed against a set of predetermined criteria. The battery of techniques typically involves a combination of collective and individual selection methods and should be designed around the actual demands of the role.
John Milsom, senior consultant with Wickland Westcott4 - an HR consultancy that specialises in the design and implementation of assessment centres - highlights a number of additional factors that contribute to the effectiveness of assessment centres.
As he comments: "Compared with your average recruitment tool, much more thought and preparation goes into designing and running assessment centres. For that reason alone, they are more tailored to the vacant role. On a practical level, they also promote a certain level of discussion and debate that provides a sounder basis for a selection decision. For example, at the end of a centre, there are typically four or six people with data on candidates' performance challenging each other to justify their views of each candidate's suitability."
A new development
Assessment centres have evolved considerably since their inception during the Second World War for officer selection in the armed forces. Not only has their use in an occupational setting increased considerably over the past two decades, but their application has become both more sophisticated and more varied. For example, some organisations have adapted the technique to design development centres, as illustrated by the Vodafone case study in box 1 . Unlike centres for selection that aim to appoint individuals to vacant or new posts, development centres are designed to identify existing employees who have the potential to be developed for senior positions within the organisation. Although the overall approach may differ slightly, such centres can still be designed around key competencies and include many of the same exercises that are used for selection centres.
"It is critical that employers who use development centres are upfront and honest about whether they are to be used for assessment or development purposes," Milsom points out. "It is perfectly acceptable to use an assessment centre to develop people but you shouldn't use a development centre for selection. Participants are likely to be much more open and take risks in development centres, but knowing that there will be any type of selection decision at the end of it will change the way they behave."
Work focus
Another notable trend in assessment centre usage is to include a greater degree of work relevance in their design.
Dr Charles Woodruffe is managing consultant at Human Assets, a firm of business psychologists, and is also the author of one of the standard works on assessment centres5. In his view, centres are most effective if they are tailor-made and replicate the role as closely as possible. As he explains: "Some centres based on off-the-shelf, generic exercises may look plausible but have nowhere near the predictive qualities of those that closely simulate the role and are designed around the organisation's own competencies."
Woodruffe refers to an article on the Human Assets website (www.humanassets.co.uk) that draws on a recent article in the Journal of Applied Psychology by Charles E Lance. It outlines the latest thinking on assessment centres. Previously, it was assumed that different exercises provided opportunities for candidates to exhibit relatively stable competencies, with the expectation that each competency would be marked relatively similarly across exercises. Repeated studies have shown the opposite to be the case, with the same competency often producing different scores in different exercises.
"This is known as the 'exercise effect' and can be understood as a genuine reflection of both candidates' variation in performance across exercises and the relative sameness of their performance across dimensions in any particular exercise," Woodruffe explains. "For example, a candidate might perform well on all competencies during a presentation but could equally score poorly on some of the same competencies in a group exercise. The practical implication of this is that, if they are to work properly, assessment centres must be comprehensive samples of the role that the candidate will be doing. With so much variation across exercises, it is absolutely vital that the exercises fully reflect the role."
John Milsom believes that competencies are still the best and most accurate solution on which to base assessment. "While it may be possible for evaluation for some narrower roles to rely on task-based exercises alone, this is not really a feasible option for most jobs that have so many different dimensions," he says. "Task-based definitions could not capture the complexities of a management role, for example. Can you imagine how many exercises a recruiter would have to ask candidates to do in order to reflect the diverse range of activities that a manager is required to perform?"
Sue Sherliker also refers to the "exercise effect" and the need for exercises to closely reflect the role. As she explains: "Competencies are the link between the culture, values and attitudes of the organisation and the specific behaviours and skills required. They are the basic building blocks upon which all assessment is measured. However, there has been criticism levelled at the ineffective application of competencies in assessment centres and it is important that employers are aware of the pitfalls. It is essential that the exercises contained within a centre closely reflect both the role and the competencies needed to perform that role."
According to Woodruffe, competencies should be assessed with reference to a pre-agreed rating scale and assessors' guide, but each competency should ideally be measured at least two or three times.
Centre design
It is easy to point out how not to design and run an assessment centre, according to Woodruffe. "There are a number of potential pitfalls, for example, if the competencies are not clear, if there is not enough time built in to the design of the centre to measure them properly and if the overall timetable is rushed. If there is no consideration of equal opportunities, no validation of the centre's results and no feedback to candidates, doubt is also cast on its effectiveness. The detail is very important."
The factsheet on assessment centres for recruitment and selection issued by the CIPD6 states that essential design criteria should include:
the duration of the centre (one day might be sufficient for more senior posts);
location (on-site giving an insight into the reality of the organisation or optimum off-site surroundings, together with ensuring accessibility for candidates with disabilities);
the number of candidates brought together (five may be too few for comfort under close observation and more than eight gives problems in sharing the assessment time);
candidates' background and comparability of past experience; and
the number, mix and experience of the assessors.
"Understanding the job is the starting point for design of the assessment centre and there are two sides to this," Milsom says. "The first is understanding what people need to be good at to perform well in the role - and this is important for setting the scoring criteria - and the second part is understanding what people do in the job. This means picking out representative tasks that will determine the kind of exercises that candidates are asked to carry out at the centre." In his view, it is also vital that there is mix of both group and individual exercises.
The choice of activities and tests is an obvious part of the design process for the centre. And there is an extensive list to choose from - for example, group exercises, written tests, role play, ability and personality tests, and presentations. The final choice of selection methods from the list should be based on the nature of the job and what elements need to be assessed. Therein lies the potential advantage of using assessment centres for selection; they afford the recruiter a generous degree of flexibility and scope to gear the activities to reflect the role. So while many of the same types of tests will be common to all assessment centres, no two centres are likely to be identical.
Sherliker advises that the assessment centre has to be well planned since time is limited. As she explains: "It is critical that detailed attention is paid to the timetable to ensure that it works in practice.
The timing can look fine on paper but there must be sufficient opportunity, for instance, for assessors to make notes, or for briefing candidates on each exercise. Candidates' perceptions of the day are very important and will be influenced by how smoothly the day ran and how comfortable they felt with the arrangements."
The CIPD guidance also points out that an assessment centre should reflect:
the ethos of the organisation;
the actual skills required to carry out the job;
potential sources of recruits;
the extent to which recruitment is devolved to line managers; and
the HR strategy of the recruiting organisation.
The above extract makes clear that there is more to the design of an assessment centre than a series of exercises, even if the activities are based on key competencies and closely simulate the job itself. As well as reflecting the role, whoever is responsible for designing the centre needs to ensure that the values and culture of the organisation are also portrayed as far as possible through participants' experience of the centre. One approach could be to design all the activities around a business context that mirrors the organisation's own. If every opportunity is taken to relay the reality of the job and the environment they will be working in, it is more likely that candidates will be in a position to judge whether or their own values match that of the recruiting organisation.
One of the key issues that employers will need to address early on in the design process is whether to develop assessment centres in-house or turn to an external source. The main considerations here are likely to be cost and the internal sources of expertise available to the organisation.
Measuring up to the role
The assessors who score and evaluate candidates' performance have a crucial role to play in an assessment centre. In Woodruffe's view: "It is vital that assessors are adequately trained and perform consistently. Organisations also need to consider the composition of the assessing panel, that it is not dominated by one gender and that it has no adverse impact on other under-represented groups." Woodruffe advises that a ratio of two candidates to one assessor is ideal, but this quotient can be reduced slightly with "a little bit of ingenuity".
According to Sherliker, training the assessors to a high level of competence is vital, and they must be fully briefed. As she cautions: "There is a lot more involved in the assessment process than is sometimes realised, and so the briefing should include a proper introduction on the organisation's competencies, for example, as well as a thorough examination of the exercises and rating guides."
Good practice
Both the British Psychological Society (BPS)7 and the Association of Graduate Recruiters8 have prepared good-practice guidelines on assessment centres.
The BPS points out that there are a number of key roles that need to be considered in terms of training requirements for assessment/development centres - assessors, observers, facilitators, role players and designers. Its guidelines describe assessors and observers as "those charged with evaluating the behaviour demonstrated in the exercises", and that their training needs to take account of the following:
assessment/development centre principles;
specific materials;
practical work;
skills decay;
feedback;
the organisational context in which the centre is to operate;
equal opportunities issues in assessment; and
confidentiality.
The candidate's perspective
Assessment centres have the potential to be a daunting prospect for those individuals who are invited to participate in them. While it is not feasible to eradicate entirely the feelings of apprehension experienced by candidates, it is possible to follow good practice and make them candidate-friendly. It is worth bearing in mind that attendance at the centre - probably the first in-depth contact that prospective recruits have with the recruiting organisation - affords recruiters an extended opportunity to leave people with a positive impression of the organisation. Their experience on the day may well influence a participant's decision whether or not to accept a job offer.
"Briefing the candidates beforehand is vital," says Milsom. "While assessment centres have become quite standard for some areas of recruitment, this does not mean that candidates will not be completely fazed by them if they do not understand what to expect and what is expected of them."
The main emphasis of that section of the BPS guidance dealing with ethical, professional and legal considerations is on candidate/participant issues, giving a strong indication of how important it is to get this aspect of assessment centres right. The guidance states that the information provided to the candidate beforehand should place them in a position to decide whether or not to attend the centre. If participation in the centre is part of their condition of employment, participants have a right to be fully informed of the purpose of the centre and why they are attending.
All of the four organisations featured in our case studies (see box 1, box 2, box 3, and box 4 ) view their pre-centre communication with candidates as being a crucial activity. The BPS advises that, ideally, communication should take place at least two to three weeks before the centre and specifies a comprehensive list of information that should be covered in the briefing, such as a broad overview of the types of tests or simulations covered, and when - and what kind of - feedback will be provided.
Feedback to candidates forms the other main focus of the BPS guidance relating to candidate/ participant issues. A number of best-practice points are highlighted. These include the legal requirement through the Code of Practice accompanying the Data Protection Act 1998 to provide meaningful feedback to candidates if the results have been stored and feedback is requested.
The BPS recommends that all participants should be given feedback and be informed of any recommendations made, and that this should be done within four weeks of the centre taking place. GKN - one of our featured employer case studies (see box 4 ) - believes that it is very important to provide feedback to both successful and unsuccessful candidates. Typically, it provides this by telephone.
Final review
Evaluation of the effectiveness of their assessment centres is a common undertaking for the majority of employers who design and run centres. The IRS survey of graduate recruitment and selection practices, mentioned above, found that more than eight in 10 (81.4%) graduate recruiters conduct an evaluation.
"Evaluating the performance of the centre is a very important activity," Woodruffe says. "We always try to conduct a thorough evaluation using both qualitative and quantitative approaches. For example, it is possible to undertake a statistical review, such as analysing the ratings to see if some competencies were marked harder than others, or comparing how candidates performed at the centre to how they perform in the role. A qualitative approach could include a review against best-practice guidelines, or elicit views from all the stakeholders."
"It is good practice to carry out a thorough statistical evaluation, particularly if an employer makes extensive use of assessment centres and there are large amounts of data," Milsom says. "This need not be expensive - for example, there are plenty of postgraduate students who are keen to take advantage of real research opportunities and have the skills to carry out a statistical evaluation."
Assessment centre performance can be evaluated in a number of different ways, ranging from the analysis of feedback from candidates and/or assessors, to the longer-term monitoring of retention rates. Equal opportunities monitoring data can also be analysed to verify that the use of assessment centres in recruitment is not adversely affecting any specific group of candidates.
All four case study organisations featured here carry out an evaluation exercise following the running of an assessment centre. For example, GKN takes into account a range of indicators, such as retention rates and the progress and performance of new recruits. The fact that many of the individuals who entered the organisation through the assessment centre route are now reaching very senior levels within the company is also viewed as a real measure of success.
According to Sherliker, an assessment centre is only as good as the results it achieves. Careful follow up of participants' subsequent performance is therefore vital.
Sherliker recalls one large corporate employer that put in place a long-term strategy to evaluate the results of its centres that were run to appoint heads of functions. "The performance of these new managers who had been appointed through the assessment centre process was evaluated at regular intervals over a period of time, and compared with assessment data from the centre," she explains. "The company also mapped the performance of those internal candidates who had not been successful at the centres. This approach to evaluation was very effective and showed a close correlation between assessment at the centre and performance on the job."
Assessment centres undoubtedly represent a potentially powerful selection strategy. While the significant financial outlay typically associated with their use may render them inappropriate for some categories of vacancy, their flexibility and sophistication make them an unrivalled form of selection for key posts. But assessment centres do come with a caveat: employers should not be misled into assuming that they represent an easy selection option. Adherence to good practice, attention to detail and the necessary investment of resources and expertise are just some of the elements that cannot be ignored if assessment centres are to live up to their reputation.
1Recruitment, retention and turnover 2004, a survey of the UK and Ireland, June 2004, CIPD, www.cipd.co.uk.
2Graduate recruitment 2004/05: upturn and optimism.
3Employee resourcing, Stephen Taylor, CIPD, 1999.
4Wickland Westcott, telephone 01625 508100; email: john@wickland-westcott.co.uk.
5Development and assessment centres; identifying and developing competence, Charles Woodruffe, CIPD, 3rd edition, 2000.
6Assessment centres for recruitment and selection, CIPD, June 2004, www.cipd.co.uk.
7Design, implementation and evaluation of assessment and development centres: best-practice guidelines, British Psychological Society, www.psychtesting.org.uk.
8Assessment centres, Association of Graduate Recruiters, July 2001, www.agr.org.uk.
This article was written by Rachel Suff, a freelance employment researcher and writer, rmsuff@dsl.pipex.com.