集中连片特困地区一维、静态多维和动态多维贫困测度与比较

因中国各省之间的县级统计指标不一、县级层面贫困方面的数据较缺乏,湖南省县级层面的数据相对连续和完整,所以文章以湖南省集中连片特困地区为例,利用该特困地区37个特困县10个贫困维度的平衡面板数据,采用两阶段主成分分析方法,考察了一维、静态多维和动态多维贫困的理论和实证差异。研究发现:其一,相对于一维和静态多维贫困得分,动态多维贫困得分更能全面准确地反映特困县的贫困程度。以动态多维贫困得分为基准,一维和静态多维贫困得分都存在不小的偏差,但静态多维贫困得分的偏差要小很多。其二,湖南省武陵山区各特困县的平均贫困程度比湖南省罗霄山区各特困县的稍微严重。武陵山区中,娄底市各特困县的平均贫困程度最高、邵阳市各特困县的排第二,常德市各特困县的最低。罗霄山区郴州市各特困县的平均贫困程度要比罗霄山区株洲市各特困县的更为严重。

The Measurement and Comparison of One-Dimensional, Static and Dynamic Multidimensional Poverty in Concentrated Contiguous Poverty Areas

Due to the inconsistency of county-level statistical indicators among Chinese provinces, and the lack of county-level poverty data, the data of the county level in Hunan province are relatively continuous and complete, taking the concentrated contiguous poverty areas in Hunan province as an example, the theoretical and empirical differences of one-dimensional, static and dynamic multidimensional poverty were investigated by using the balance panel data of 10 poverty dimensions in 37 poverty-stricken counties and using two-stage principal component analysis method. The findings are as follows :(1) compared with the one-dimensional and static multidimensional poverty scores, the dynamic multidimensional poverty scores can more accurately reflect the poverty level of destitute counties. Taking the dynamic multidimensional poverty score as the benchmark, both one-dimensional and static multidimensional poverty score have a large deviation, but the deviation of static multidimensional poverty score is much smaller. (2)the average poverty level of the poverty-stricken counties in Wuling mountain area of hunan province is slightly worse than that in Luoxiao mountain area of Hunan province. In Wuling mountain area, the average poverty level of the poverty-stricken counties in Loudi city is the highest, counties in Shaoyang city is the second, counties in Changde city is the lowest. The average poverty level of the poverty-stricken counties in Chenzhou city is more serious than that in Zhuzhou city.