摘要: The Good Stick Index validation study statistically investigated an empirically derived measure of pilot proficiency in an air combat simulator. Statistical methods, including regression and discriminant analyses, were used to eva... 展开 The Good Stick Index validation study statistically investigated an empirically derived measure of pilot proficiency in an air combat simulator. Statistical methods, including regression and discriminant analyses, were used to evaluate GSI scores as predictors of student free-engagement performances in the Tactical Air Command Air Combat Engagement Simulator (TAC ACES I) simulator training program. Statistically derived performance predictors are obtained from objectively measured parameters recorded during simulator training. The effect of inclusion of student pilot demographic data with the objective data is investigated. Edumetric and psychometric data are presented as indicators of skill development. Results of the study yield performance predictors for four groupings within each TAC ACES I class; (a) winners, (b) winners or runners-up, (c) upper-half winners, and (d) student quartile ranking. The empirically derived measure shows a probability of winner prediction of 25 percent, whereas the statistically derived optimal measure shows a probability of winner prediction of 80 percent. The reliability of the performance predictors is assessed. Potential utilization and limitations of the Good Stick Index are addressed. 收起