We have studied a central composite design for the chiral separation of amphetamines using capillary electrophoresis. Five variables, i.e., buffer concentration, pH, chiral selector concentration, temperature, and applied voltage, were investigated. Enantiomeric resolutions as well as analysis time and generated power were established as responses for each experiment. A model of each response was obtained by multiple regression of a quadratic-degree mathematical expression. From the models, we determined the most favorable conditions for the chiral separation by optimizing the resolution and setting the other responses at threshold values. Results were compared with a previous study in which a systematic investigation of the operating parameters was carried out. In order to visualize the robustness of the method, response surfaces were drawn for the significant variables. We have concluded that experimental designs offer a rapid means of optimizing several variables and provide an efficient test for the robustness of the analytical method.