Assessment of Nontechnical Skills

Abstract

The assessment of pilots’ performances and more specifically their nontechnical skills turns out to be difficult because researchers and practitioners find that inter-rater reliability tends to be moderate to low. In this study, we proposed, consistent with field observations, that assessment is a categorization rather than a measurement issue. Categorization, even though, and especially when, based on imprecise information and assessments, can be modeled mathematically using fuzzy logic. We present the structure of the approach and provide an example from a large database of pilots providing reasons for their assessments of other pilots’ (simulator) flying performances. Considering the implications, we discuss possibilities for using the variations in performance ratings positively, as a source for learning and resilience building.