The Alignment index and its application to voting at the United Nations General Assembly

Abstract We propose the Alignment index as a new statistic to measure the similarity between the voting patterns of two nations. Motivated by the analysis of roll-call votes in the United Nations General Assembly, where issues are approved by large majorities, the Alignment index corrects the bias of a previous measure, called Similarity index, that in those cases is biased towards positive values, overstating similarities. One advantage of the Alignment index is that the value of zero is a clear-cut separation between aligned (similar in the broad sense) and misaligned (dissimilar) nations. One can take advantage of this property applying the Clique Partitioning model to find the voting blocs of the Assembly. In this way, the evolution during the years of the most important blocs, such as the Western societies, the Soviet bloc, the Global South, and so on, can be traced and discussed.