Application of the local Moran index to detect electoral anomalies

Abstract

This article discusses the use of one of the methods of spatial analysis, as the local Moran index, for analyzing election results in order to identify the existence of electoral anomalies. The author conducted a study based on the results of the 2016 parliamentary elections in the context of Moscow districts for the five largest parliamentary parties:“Edinaya Rossiya”, KPRF, LDPR, “Spravedlivaya Rossiya” and “Yabloko”. Cartograms of clusters of spatial autocorrelation were obtained for each batch, according to the results of which areas with abnormal results were identified. Electorally abnormal areas were considered for which a violation of the neighborhood effect was recorded, that is, there was a difference in the voting index from neighboring areas. An analysis was also carried out and possible reasons for these differences were expressed. In conclusion, it was noted that some parties demonstrate uniformity in the territorial distribution of votes, which is why abnormal areas were not allocated. Conclusions were also drawn about the effectiveness of using this spatial analysis method to identify electoral anomalies. At the end, the general results of the work were spelled out, including also possible further ways of developing this topic.