Multi-dimension evaluation of rural development degree and its uncertainties: A comparison analysis based on three different weighting assignment methods

• Main Authors: Jian, Y. (Author), Liu, X. (Author), Liu, Z. (Author), Shi, L. (Author), Zhong, H. (Author)
• Format: Article
• Language: English
• Published: Elsevier B.V. 2021
• Subjects: China; comparative study; Comparison analysis; Development degree; entropy; Entropy methods; error analysis; Index system; Indicator indicator; Indices systems; Mean square error; Mean-square-error methods; Petroleum reservoir evaluation; Regional planning; rural development; Rural development; Rural development degree; Rural development degree (RDD); spatial distribution; Spatial distribution; spatiotemporal analysis; sustainable development; Uncertainty; uncertainty analysis; Uncertainty analysis; Weighting assignment; Weighting assignment (WA)
• Online Access: View Fulltext in Publisher

 Rural development degree (RDD) evaluation is a very valuable guidance for rural sustainable development. Earlier studies of RDD evaluatons more focusd on establishing index system, analysing spatial–temporal evolving patterns and functional differentiations. Weighting assignment (WA) method selection is a vital step for rural development degree (RDD) evaluation, but impacts of different WA methods on indicator weight determination and RDD evaluation were not well clarified. This study therefore employed three dominant WA methods, covering equal weight method, entropy method and mean square error method, along with a developed RDD evaluation index system to compare the differences of indicator weights and uncertainties of RDD evaluation. The results indicated that the spatial patterns of three WA-based RDD maps had great differences although using the same evaluation indicators. The RDD types with the largest proportion generated from different WA methods were also spatially various. Spatial distributions of RDD generated by various WA methods regions performed largely differences in central, northeastern and southwestern China. Our analyses found that the differences from industrial prosperity dimension and ecology livability dimension owing to utilizing different WA methods were largely responsible for the RDD spatial distributions in this study. This study gave some potential suggestions for WA method selection in RDD evaluation according to the data characteristics, WA method principles and application requirements. Among them, entropy method was suitable for indicator data with great dispersion degree and mean square error method would be better describe indicator differences with large indicator number. Besides, this study also underlined that WA-derived uncertainties should be paid more attentions in rural development and rural revitalization evaluation. © 2021 The Author(s)