One advantage of evolutionary computation over conventional optimization and search techniques is its ability to deal with multiple objectives. Multi-objective evolutionary algorithms (MOEAs) have proved very useful in many real-world applications and the number of publications in this area has exceeded 1000. However, no one offers a simple-to-use or widely accepted method for evaluating the performance of MOEAs. This is largely due to difficulties in visualizing non-dominated solutions in a multi-objective space when the number of objectives exceeds three. In this chapter, we propose a new visualization technique that should provide better understanding of high order multi-dimensional objectives, so as to assist the design and refinement of MOEAs.