Statistical Analysis of Isoperformance Issues in Navy Flight Training
Abstract
It is important to resolve the issue of whether extra information can help assign the probability for failure of a pilot or Naval Flight Officer (NFO) in some phase of flight training. This assigned probability for failure could be based simply on empirical data gathered over some relatively long period of time. However, if scores from selection tests, personality tests, vision exams, psychomotor tests, and the like could serve as cogent information about the probability for failure, then the probability for failure could be revised upwards or downwards based on an individual's standing on these variables. In addition, it would be interesting to find out if candidates could "trade off' high scores on one class of tests for low scores on a different class of tests, but still achieve the same level of performance. Here, level of performance is defined as the probability for failure. This brings us into contact with the idea of isoperformance. In this analysis, we examine two classes of predictor variables where candidates might trade off high scores for low scores, yet still achieve the same level of performance. The first class consists of cognitive information processing variables. Scores for the final academic grade from Aviation Pre-Flight Indoctrination (API) will serve as an example of this class. The second class consists of personality variables. We will use scores from the Pilot Biographical Inventory (PBI), a subcomponent of the Aviation Selection Test Battery, as a surrogate for scores on personality tests to be administered in future research on isoperformance.
Document Details
- Document Type
- Technical Report
- Publication Date
- Jan 19, 2000
- Accession Number
- ADA529936
Entities
People
- D.j. Blower
Organizations
- Naval Aerospace Medical Research Laboratory