Accession Number : ADA018151

Title :   Automated Pilot Aptitude Measurement System.

Descriptive Note : Final rept. Jul 73-Jul 75,

Corporate Author : MCDONNELL DOUGLAS ASTRONAUTICS CO-EAST ST LOUIS MO

Personal Author(s) : Long,George E. ; Varney,Nicholas C.

Report Date : SEP 1975

Pagination or Media Count : 131

Abstract : This study addresses the problem of predicting success in undergraduate pilot training (UPT) through the measurement of performance on a learning sample of flight tasks administered prior to the initiation of training. The learning sample which would be used to augment existing pilot selection procedures was proposed as a means of reducing attrition during UPT. Performance measures on a five hour learning sample of flight tasks, administered with an Automated Pilot Aptitude Measurement System (APAMS) were collected from 178 candidates for UPT selected for training with current procedures. These performance measures were then compared with performance during T-41 and T-37 phases of UPT. The results indicated that performance in both phases of training could be predicted from performance on the learning sample. Grades given by instructors in T-41 training were highly correlated with performance measures on the learning sample. Also, candidates who were eliminated from both phases of training, including those eliminated for Manifestation of Apprehension (MOA), Self Initiated Elimination (SIE), as well as eliminees for Flying Training Defiency (FTD) could be discriminated by their performance on the learning sample. The results indicated that the learning sample approach could contribute substantially to existing pilot selection procedures in reducing current attrition rates in UPT. (Author)

Descriptors :   *Pilots, *Flight training, *Automatic pilots, *Aptitude tests, *Learning, Performance(Human), Feedback, Selection, Regression analysis, Aptitudes, Automation, Predictions, Training planes

Subject Categories : Humanities and History

Distribution Statement : APPROVED FOR PUBLIC RELEASE