This paper adopts both one-dimensional and multi-dimensional cluster analysis to analyze China's HDI data for 1982, 1995, 1999, and 2003, and to classify China's provinces into four tiers based on the three basic developmental aspects embedded in HDI. The classifications by cluster analysis depends on the observations' similarities with respect to clustering variable(s), rather than subjectively predetermined threshold values adopted by major international organizations, like UNDP, to make their classification. By investigating the multi-dimensional clustering results, this paper describes the features of each of the four tiers based on their characteristics within the three developmental aspects, explores the patterns of cross-tier disparities, and tracks the evolutions from a historical perspective. Also found is that, in China, the overall regional disparities have been increasingly attributable to regional economic disparities.