This paper investigated students' achievement for learning American Sign Language (ASL), using two different methods. There were two groups of samples. The first experimental group (Group A) was the game-based learning for ASL, using Kinect. The second control learning group (Group B) was the traditional face-to-face learning method, generally used in sign language skill training for students with hearing impairments. This study was separated into two phases. In Phase I: 3D trajectory matching measurement algorithm, the Euclidean distance algorithm was employed to present the similarity between teacher and student datasets. Then, Phase II: Effectiveness of the Game-Based Learning system, showed the proposed framework of the game-based learning for sign language. Moreover, knowledge of sign language together with the corresponding actions were captured from three sign language experts using the knowledge engineering method. Then, the proposed game-based system would be analysed to provide students immediate feedbacks and suggestions based on the knowledge transfer from ASL experts. In the experiment, the students (N = 31) were divided into two groups. The first group, Group A (N = 17), learnt with the game-based learning while the second group, Group B (N = 14), learnt with the traditional face to face learning method. The study result showed a significant difference (p<0.05) on the mean score of the post-tests for both Group A and Group B. It also presented that the game-based learning approach provided a better performance of ASL vocabularies than the traditional face-to-face learning approach. Finally, in the section of Discussion and Conclusion, the effectiveness and the future work opportunity of the proposed game-based learning system were discussed for the improvement of sign language actions.