How we change what others think, feel, believe and do
‘Work sample’ is a method of testing ability by giving the candidate a sample of typical work to do and evaluating their performance.
Work samples may appear as short questions along the lines of 'What would you do in this situation' or more complex scenarios to analyze. At its most naturalistic, the candidate is put into a the actual job where they may spend some time actually doing real work. The normal situation, however is for the person to be given a role-play or real-life situations where the candidate acts out a realistic situation. This creates a repeatable pattern whereby multiple candidates can be given the same test and hence more easily compared.
Job-knowledge tests focus on specific dimension or content to determine current knowledge, such as a test of knowledge about the highway code.
Knowledge tests such as this may be computerized, enabling them to be taken at any time and even in any place. This also reduces the cost of administration and can reduce security issues (such as loss of exam papers).
Proctoring is a method often used, whereby questions and sequences are regularly changed to reduce copying cheating.
Job knowledge tests are increasing use in professional areas such as medicine and architecture.
Hands-on performance tests
Hands-on performance tests are used to test people for physical capabilities. For example, a psychomotor test, which is characterized by manual dexterity exercises.
Situational judgments tests
In situational judgments tests, people are asked how they would act in a given situation. This may be done with a multiple choice to enable automated marking. It can be used in many different jobs, for example leadership and teaching.
These tests assess job knowledge and the ability of the candidate to apply this knowledge in specific situations, (rather like in situational interviewing). They can be used to assess for aptitude and trainability as well as for current knowledge, and can be helpful in recruiting people with no previous experience.
Work samples, as with other selection methods often start with a job analysis of good performers.
The job is typically broken down into key behavioral components, which are then used to create a checklist of desirable behaviors.
From this, scenarios and case studies may be developed.
Work samples is normally used to test current skill. It can also be used to test for the ability to learn new skills. It is based on the premise of behavioral consistency, where the way a person acts in a simulated situation is assumed to be the same as they might act on the job.
It is useful for reducing bias by assessors and is perceived to be fair and valid by both recruiters and candidates, as all candidates are treated in the same way, including the amount of time to respond (although this may reduce the chances of slow writers or reflective thinkers). It removes non-job-related cognitive factors, and is visibly related to the job in question.
It has a high predictive validity of .37. to .54 and leads to less turnover of staff.
Criticism of work samples includes that they are atheoretical and related to an empiricist and Western view of the individual and work (Searle, 2003). Work samples must be carefully designed to test specific items. They give problem where more attention is paid to face validity than content validity and can also miss small, but critical factors (such as color vision for engineers).
In any concern for fairness, work samples are of particular value as they have both higher face validity and greater fairness for non-traditional candidates (Lievens and Klimosky, 2001).
Lievens F. and Klimosky, R.J. (2001) ‘Understanding the assessment centre process: where are we now’ in Cooper, C.L. and Robertson, I.T. (eds). International Review of Industrial and Organizational Psychology, vol.16, Chichester, Wiley, Ch.8.
Searle, R. (2003). Selection and Recruitment: A Critical Text, London: Palgrove Macmillan and Milton Keynes: The Open University