Utilization of psychomotor screening for USAF pilot candidates: enhancing predictive validity

Subjects for this research were 320 prospective pilots who were tested on computerized versions of the Two Hand Coordination (2HC) and Complex Coordination (CC) psychomotor tests. Independent variables included five basic error measures associated with the two tests, as well as seven new variables that had not previously been utilized. Results of MANOVA and multiple regression analyses revealed that performance on the two psychomotor tests were significantly related to Undergraduate Pilot Training (UPT) outcome. A trend was observed for a prediction model based on early and total trial data to yield higher simple and multiple correlations than a model based on late trials. Regressing the basic five independent variables on UPT outcome yielded R values of 0.334, 0.271, and 0.310 for early, late, and total trial data, respectively. The predictive validity of the basic five error scores was not incrementally increased by stepping in independent variables associated with the hypotenuse of horizontal and vertical error or stick movement. A stepwise multiple regression analysis revealed that the best two-variable prediction equation included the hypotenuse of horizontal and vertical error for both psychomotor tests (R = 0.329).