A Comparative Study Of Inequality Index In Public Transport Equity

Aiming at the problem that the evaluation method of public transport equity lacks valid reasons,225 traffic zones of Haining City,Zhejiang Province,were divided with administrative jurisdiction.Accessibility to bus stop was used as a resource index to measure the level of public transport supply distributed to a traffic zone.Bonferroni index,Gini coefficient,Theil L index,Theil T index and Zenga index were compared from three dimensions,including distribution characteristics of accessibility,sensitivity and decomposition property of inequality index.Firstly,distribution functions were used to fit accessibility data,and the differences between fitting inequality value and actual value of 5 inequality index were compared.Then,sensitivity of 5 inequality index were calculated before and after 5% accessibility data was removed in high,medium and lower level.Finally,overall inequality was decomposed into inequality between urban and rural and within urban and rural.Contributions of inequality between urban and rural to the overall for 5 methods was compared.For differences between fitting inequality value and actual value,results shown that Theil L>Theil T>Gini>Bonferroni>Zenga,meaning the importance of missing accessibility data was Theil L>Theil T>Gini>Bonferroni>Zenga.For the sensitivity of inequality index,we have Theil T ? Theil L>Gini>Bonferroni>Zenga at high accessibility level,Theil T>Theil L>Gini ?Bonferroni ?Zenga at middle and Theil L >Theil T>Gini?Bonferroni?Zenga at low shown that the sensitivity is significantly different between inequality indexes;Under Zenga index,Dagum decomposition,Theil L index,Theil T index and Bonferroni index,the contribution of inequality between urban and rural were 71.01%,53.81%,30.53%,28.96% and 6.12%.The extent to which decomposition methods attach weight to difference between urban and rural varies.In a word,inequality indexes show significant difference in 3 aspects and have different applicability,so it is impossible to arbitrarily choose without reason and necessary to demonstrate the rationality of inequality index according to the research purpose.