Assessment Center for Pilot Selection: Construct and Criterion Validity and the Impact of Assessor Type

Cette recherche a examine la validite d’un centre d’evaluation pour la selection de pilotes. Les scores de N = 1,036 participants ont ete utilises pour etudier la validite de construit. Un sous-echantillon de participants performants a ete suivi et les evaluations des pairs ont ete retenus comme mesures du critere. Les resultats demontrent une premiere evidence de validite de construit et de critere pour cet outil d’evaluation des competences interpersonnelles et liees a la performance. Par ailleurs, les resultats ont aussi montre que le type d’evaluateur (psychologue vs pilote) modere la validite predictive des scores du centre d’evaluation. Cet effet “type d’evaluateur” depend de la sorte de variables predictives. Les resultats sont discutes et des implications pratiques sont suggerees. This study examined the validity of an assessment center in pilot selection as a new field of application. Assessment center ratings of N= 1,036 applicants were used to examine the construct validity. A subsample of successful applicants was followed up and peer ratings were chosen as criterion measures. The results provide first evidence of the construct and criterion validity of this assessment center approach for rating interpersonal and performance-related skills. Furthermore the type of assessor (psychologist versus pilot) moderates the predictive validity of the assessment center ratings. This type-of-assessor effect depends on the kind of predictor variables. The results are discussed and practical implications are suggested.