Over the last 50 years research has consistently shown cognitive ability tests are the single best predictors of performance in particular areas:
– General Reasoning Ability is the best predictor of managerial performance
– Verbal and Numerical Reasoning Ability predict administrative skills
– Diagrammatic thinking predicts information technology aptitude
– Mechanical reasoning predicts craft performance
– Perceptual Speed and Accuracy (Clerical Checking) predicts filing and secretarial aptitude
Research has shown that the relationship between personality and performance is more complex and multidimensional than in the cognitive area. For example, we cannot say that one individual has more personality than another and that this affects performance. Typically we need to link observations from three or four personality traits together to predict performance in a given domain. For example:
» Telephone Salespeople tend to be…High Conscientiousness, High Extraversion, Low Agreeableness
» Air Traffic Controllerstend to be…High Conscientiousness, Low Emotionality, Low Agreeableness
» Innovative team members tend to be…High Openess to Experience, Low Conscientiousness
» Team Leaders tend to be…High Conscientiousness, High Extraversion,
Selection research also needs to take into account the capacity of individuals to deceive themselves or their assessors in a self- report situation.
Complements to Cognitive and Personality Assessment
Cumulative validity research indicates that work samples (seeing a person performing a core job function) is the best measure of competence however this is not always possible. Assessment centres are the next best predictor. This should not surprise us as an assessment centre is a multiple method approach incorporating interviews, ability tests, personality questionnaires and high fidelity job simulations. However, as you can imagine this approach is often prohibitively expensive and time consuming.
Best practice approaches support the use of interviews, as well managed interviews can be mapped to a wide range of job components. Interviews however are relatively expensive (consider interviewer time) compared to group or on line testing (material costs from €5 per candidate). Cognitive ability tests have proven to have good predictive validity in research however they only assess a small range of job performance criteria. While ability tests enable us to look at differences in people’s capability to learn they tell us nothing of a person’s capacity to work with others, motivation or interests. Personality questionnaires can also be inexpensive but, we must always remember they are not observations of behaviour but merely a self report on what the person says they do.