Measuring Regional Inequality: A Comparison of Coefficient of Variation and Hoover Concentration Index

Many measures, particularly the coefficient of variation (CV) and Hoover concentration index (CI), have been used in the study of regional inequality. However, opposite results are often obtained from different quantitative measures which are supposedly showing the same pattern of regional inequality. This paper employs Jiangsu province in P.R. China as a case study to explain how two measures, CV and CI, can produce completely contrasting results. This has been achieved by explicating the original CV and CI, and by appropriately modifying these indices. Two major reasons leading to different trends exhibited by the CV and CI for the same data set have been identified. They are the population weights attached to areal units and the absolute deviation from the mean in the original data. The findings serve as a note of caution on the employment of measures to unravel regional inequality. Inappropriate use of indices might lead to invalid conclusion about differential in regional development.