American election results at the precinct level

Abstract

We describe the creation and quality assurance of a dataset containing nearly all available precinct-level election results from the 2016, 2018, and 2020 American elections. Precincts are the smallest level of election administration, and election results at this granularity are needed to address many important questions. However, election results are individually reported by each state with little standardization or data quality assurance. We have collected, cleaned, and standardized precinct-level election results from every available race above the very local level in almost every state across the last three national election years. Our data include nearly every candidate for president, US Congress, governor, or state legislator, and hundreds of thousands of precinct-level results for judicial races, other statewide races, and even local races and ballot initiatives. In this article we describe the process of finding this information and standardizing it. Then we aggregate the precinct-level results up to geographies that have official totals, and show that our totals never differ from the official nationwide data by more than 0.457%.

American local government elections database

“…The foundation for our data on election returns is previous work on mayoral elections 5 , 14 , 15 , county legislative elections 16 , sheriff elections 18 , 23 , prosecutor elections 24 , 25 , the MIT Election and Data Science’s Lab’s data on recent elections 26 , and the California statewide election database 27 . We built upon these datasets using several approaches.…”

“… pres_pctD_08 Presidential vote shares based on precinct-level data on the 2008 presidential vote 38 . pres_pctD_16 Presidential vote shares based on precinct-level data on the 2016 presidential vote 26 , 39 . pres_pctD_20 Presidential vote shares based on precinct-level data on the 2020 presidential vote 26 , 40 .…”

LocalView, a database of public meetings for the study of local politics and policy-making in the United States

“…The few large-scale datasets that do exist on local governments, which are generally government releases such as the US Census of Governments for municipalities or the Common Core for school districts from the National Center for Education Statistics, may contain structural and administrative characteristics but are generally insufficient for scholars interested in topics like policy-making and deliberations. A recent explosion of datasets in the social sciences has led to unprecedented, large-scale study of U.S. politics, elections, and policy-making at the national [5][6][7][8] and state levels 9,10 . Meanwhile, most contemporary studies of local policy-making rely primarily on case studies or small sets of individual places 11,12 , lab experiments 13 , or have required extensive (and expensive) manual data collection [14][15][16][17] .…”

The Economics of Partisan Gerrymandering

“…Data and Empirical Model. Our data are the precinct-level returns for the US House elections in 2016, 2018, and 2020, which were recently standardized and made freely available by Baltz et al (2022). For each precinct n and election t ∈ {2016, 2018, 2020}, we observe the total two-party vote k nt and the share of the twoparty vote for the Republican candidate v nt .…”

“…39 Precinct-level returns for 2012 and 2014 have been compiled by Ansolabehere, Palmer, and Lee (2014) but are less complete and less standardized than the Baltz et al (2022) data we use, which only cover 2016, 2018, and 2020. 40 While it is not relevant for determining the qualitative form of optimal districting, we can also estimate the distribution F of voter types s. At the country-level, the mean estimate of F (calculated as w) is very close to 0, and the standard deviation estimate of F (calculated as n,t k nt (w nt − w t ) 2 / n,t k nt ) is 0.63.…”