United Nations General Assembly Ideal Points

10.7910/DVN/LEJUQZVoeten, ErikErikVoetenGeorgetown UniversityUnited Nations General Assembly Ideal PointsHarvard Dataverse2025LawSocial SciencesUnited Nationsroll-callsInternational RelationsPolitical ScienceErik VoetenErikVoetenGeorgetown UniversityErik VoetenErikVoetenErik VoetenErikVoetenGeorgetown University2025-072015-082025-072026-04-03roll-call data142732102208165181454396902165461350321795772990808105585619327042403248text/csvapplication/x-rlang-transporttext/comma-separated-valuesapplication/x-rlang-transporttext/plainapplication/vnd.openxmlformats-officedocument.wordprocessingml.documenttext/tab-separated-valuestext/tab-separated-valuestext/csvtext/tab-separated-values38.0CC0 1.0This dataset contains ideal point estimates based on voting behavior in the United Nations General Assembly. This is the first version where the ideal point estimates are based on years rather than UNGA sessions. The reasons for this are that researchers in practice virtually always use years as the basis of analysis and that UNGA sessions increasingly spill over into the following year and are held in special emergency sessions on issues such as the Gaza and Ukraine. There are two types of ideal point estimates: • idealpointfp: ideal point estimates based only on votes on the final passage of resolutions (including failed votes). These are now available from 1946-2024 in IdealPointsJuly2025.tab. These estimates are updated most frequently as the raw data can readily be found online. • Idealpointall: ideal point estimates based on all votes, including votes on paragraphs, motions, and amendments. These votes are based on more data and thus should be more precise. One word of caution is that in some years this means that there are very large numbers of votes on a specific issue, such as the war in Gaza. The correlation between these ideal points is .9846 but there could of course still be some important differences. The data also includes idealpointlegacy, which is based on sessions (all votes). The correlation with idealpointall is .9877. There is a separate file and codebook that includes estimates for 2025. There were some really radical changes between 2024-2025. Some clearly reflect real ideological changes: most notably Trump in the US and Milei in Argentina. But voting has also become more two-dimensional. So, for instance China and Russia in 2025 seem relatively closer to the US, partially because of the US votes on the Ukraine crisis and various human righst resolutions. But in one dimension, this also means that they have come closer to the EU. When you look in two dimensions, you see more clearly that the US is splitting from its allies but not China/Russia. However, dynamic two dimensional estimates are complex and I am not quite ready to post these. Over the summer I will produce a short paper on this and updated two-dimensional estimates. For now, proceed with caution. Aside from the 2024 and 2025 final passage votes, the raw UN voting data are from the UNGA-DM Database: https://unvotes.unige.ch/ Citation: Fjelstul, Joshua, Simon Hug, and Christopher Kilby. "Decision-making in the United Nations General Assembly: A comprehensive database of resolution-related decisions." The Review of International Organizations (2025): 1-18. The ideal point estimates are based on the methodology described in: Citation: Bailey, Michael A., Anton Strezhnev, and Erik Voeten. 2017. Estimating dynamic state preferences from united nations voting data. Journal of Conflict Resolution 61 (2): 430-56.