Predicting Training Success with the NEO-PI-R: The Use of Logistic Regression to Determine the Odds of Completing a Pilot Screening Program

摘要

The revised NEO personality inventory (NEO-PI-R) measuring the 'big five' personality domains has been used as a descriptive tool for assessing 'normal' personality attributes of Air Force pilots; yet the predictive potential of NEO-PI-R has not been realized. An opportunity to transform the NEO-PI-R into a predictive instrument has been provided by the recent collection of data from the U.S. Air Force's Enhanced Flight Screening Program. This paper describes a technique used to provide a practical and intuitive method of predicting whether a student pilot with a particular profile on the NEO-PI-R will be successful in training.