The air traffic controller (hereinafter referred to as the controller) is responsible for the command of aircraft's activities. This post has high requirements for the controller's spatial imagination, short-term memory, situational awareness and other abilities. In order to select excellent controllers with comprehensive quality, the initial air traffic controller selection system was designed. The reliability and validity of the system were tested. The validity of the system was analyzed and the test effect was verified. 1675 air traffic controller trainees were tested with the system, the total score and score data of each sub-item were collected, and the Mann-Whitney U test method was used to analyze the difference between male and female subjects in total scores and scores of each sub-item. Cronbach’s ? coefficient and Spearman-Brown coefficient were used to evaluate the reliability of the system, and its content validity was tested according to the evaluation results of experts, and exploratory factor analysis was conducted on the obtained data to verify the structural validity of the system. The cutoff point of each sub-item was obtained according to the receiver operating characteristic curve (ROC), and the difficulty of each sub-item was calculated. The results show that there are statistical differences between men and women in some items, the sub-items are well distinguished, the reliability and validity test met the requirements of measurement, the difficulty and cutoff point of each sub-item obtained from ROC curve can provide references for air traffic controller selection, and the system has a good effect in the ability test of air traffic controller trainees.