Citation

Martinussen M, Torjussen TM. Hum. Factors Aerosp. Safety 2004; 4(3): 233-243.

Copyright

(Copyright © 2004, Ashgate Publishing)

DOI

unavailable

PMID

unavailable

Abstract

The main purpose of this study was to examine the predictive validity of a new computer-based assessment (CBA) battery introduced in the Norwegian Air Force for pilot selection. By replacing the existing mechanical and paper- and-pencil based test versions with one based on CBA technology we intended to improve the selection process both in terms of efficiency and quality. The test battery was administered to a sample of applicants already screened on paper-and pencil tests. Approximately 10% of the applicants are finally admitted to Basic Flying School thus, the predictive validity was estimated on a highly selected sample (N=108). The criteria used in this study were pass/fail in training as well as instructors' ratings. The instructors used behaviourally anchored rating scales (BARS), which were grouped into two composite scores (pilot skills and personal skills). The mean test score based on all 13 tests was significantly correlated with pass/fail in basic training (r=.26), and several of the individual tests were significantly related to both pass/fail and the other criteria (pilot skills and personal skills). The findings are promising, indicating that the CBA approach may represent a step forward in our effort to improve the pilot selection process in the Air Force. However, the findings should be replicated on a larger sample as more candidates are selected.