Flygutbildning och flygsäkerhet : studier av relationen mellan utbildningskarriär och haveribenägenhet i ett human factors-perspektiv samt en attitydundersökning
Sammanfattning: Some pilots cause accidents while others do not. Could this difference be related to training outcome? If so, an instrument for predicting flight safety ability could be developed. The training requires an effective system with efficient predictive ability related to Human Factors in terms of flight safety ability.The main objective was to study the relationships between students' results on written examinations and flight tests and flight safety ability, by comparing accident pilots with non-accident pilots. The population consists of civil pilots trained in Sweden during the period between 1971 and 1981. Their flight activities were followed until 1983. Individuals in the accident sample were self-selected by their accident. Non-accident pilots were selected by random sampling.Statistically significant differences were observed. The accident group performed worse than the non-accident group in the prediction variables, consisting of the separate subjects: regulations, meteorology, navigation, aerodynamics and aircraft/powerplant, (p<0,05 or lower), and the number of rewritten examinations, (p<O,O 1 ). Differences in the number of comments in flight tests were also found, (p<0,05). Multivariate analysis revealed that only 6 percent of the variance in the criterion variable was explained. Discriminant analysis gave correct classification for more than 60 percent of the accident group.An attitude investigation was also made among a group of original participants. No significant differences between the groups were found. One observation is that more individuals in the accident group mentioned the use of the aeroplane for travelling purposes as an important reason for acquiring a licence, while the other group mentioned hobby aspects.Parallel to the study a new training system was developed concerning objectives and completion standards. My work suggests that predictive ability is important to create. Another consequence of my work is the need for another view of knowledge including Human Factors. Changes from measuring knowledge at low taxonomic levels are required in favor of qualitative knowledge measured by process evaluations.
Denna avhandling är EVENTUELLT nedladdningsbar som PDF. Kolla denna länk för att se om den går att ladda ner.