Previous studies suggest that 3D visualization is fundamental to design spatial cognition, and the capability to work with 3D mental or physical models and taking perspective views from a set of 2D drawings are essential parts of design education, although there is no definitive evidence that can directly support these beliefs. This dissertation focuses on the issues of how spatial capabilities correlate with design performance and whether design education can improve students' spatial capabilities. Two types of capabilities tests, i.e. spatial capability test and general reasoning test, are used, and there are 251 Georgia Tech undergraduate students involved in this research. The results of this research suggest that the correlations between design studio performance and the tested factors are more salient among female students than male students. While female students generally have lower spatial capabilities than male students in design, they can take advantage of their general reasoning capability to compensate for the lack of the other two spatial capabilities and perform well in design studio. A stepwise regression further reveals that, for the female design student group, the general reasoning capability is the only predictor for their design performance. However, no significant interaction is observed in the male design student group between tested capabilities and design studio performance. As a result, there seems to exist a threshold requirement in spatial capabilities for design major students. After passing this threshold, other factors such as domain specific skills and knowledge or self-motivation would likely to be the dominant one. Lastly, although the results show the tested capabilities are all important for design major students from different perspectives, the design education does not show any contribution in the improvement of these capabilities.