ABSTRACT
Selection is one way of dealing with the problem of improving performance in stressfuP jobs. We deal here with a particular class of such jobs, viz. that of pilots. Selection of pilots has been extensively discussed in the scientific literature. A metaanalysis of a large number of studies has shown that the best single predictor is based on a job sample task (Hunter & Burke, 1995). Ability tests of various kinds are also useful but personality measures contribute little, a mere 0.1 is reported as the average validity coefficient of personality variables. Another meta-analysis (Martinussen, 1996) reported also that the best single type of measure was previous experience with an average validity of0.3. Personality yielded an average validity of only 0.1. Thus, personality measures work poorly in this situation as in so many others (Mischel, 1968), in spite of their considerable face validity.
