U.S. Presidential Election Constituency Returns (1976-2020)

Dataset Summary

This dataset contains state-level constituency returns for U.S. presidential elections from 1976 to 2020, compiled by the MIT Election Data Science Lab. The dataset includes 4,287 observations across 15 variables, offering detailed insights into the voting patterns for presidential elections over four decades. The data sources include the biennially published document “Statistics of the Congressional Election” by the Clerk of the U.S. House of Representatives.

• Source: MIT Election Data Science Lab
• Time Range: 1976–2020
• Number of Variables: 15
• Number of Observations: 4,287

Dataset Structure

The dataset contains the following variables:

•	year: The year in which the election was held.
•	office: The office being contested (U.S. PRESIDENT).
•	state: The full name of the state.
•	state_po: The U.S. postal code abbreviation for the state.
•	state_fips: The FIPS code for the state.
•	state_cen: The U.S. Census state code.
•	state_ic: The ICPSR state code.
•	candidate: The name of the candidate as reported by the U.S. House Clerk.
•	party_detailed: The detailed party affiliation of the candidate (uppercase). For states that allow candidates to run on multiple party lines, vote totals for each party are reported separately.
•	party_simplified: The simplified party affiliation of the candidate, categorized as DEMOCRAT, REPUBLICAN, LIBERTARIAN, or OTHER.
•	writein: Indicates whether the vote total includes write-in candidates:
•	TRUE – Vote totals for write-in candidates
•	FALSE – Vote totals for non-write-in candidates
•	candidatevotes: The number of votes received by the candidate for the specific party.
•	Note: Some states, like Massachusetts and New York, may include votes for “blank” or “scattering” candidates, which should be treated carefully in analysis.
•	totalvotes: The total number of votes cast in the election.
•	version: The date of the last update to the dataset (format: yyyymmdd).
•	mode: The method by which the vote was cast (e.g., absentee, election day, early voting).

Usage Considerations

• In states where candidates appear on multiple party lines, the votes for each party are reported separately. When performing analysis on candidate totals, aggregation across party lines is necessary.
• For two-party vote totals, care should be taken with candidates receiving votes under multiple party labels.
• Future versions of this dataset aim to include codes for candidates endorsed by major parties, regardless of the party label under which votes were cast.
• Some vote counts may include “blank” or “scattering” votes, especially in states like Massachusetts and New York. Users should account for this when analyzing votes cast for candidates.

Source

@data{DVN/42MVDX_2017,
author = {MIT Election Data and Science Lab},
publisher = {Harvard Dataverse},
title = {{U.S. President 1976–2020}},
UNF = {UNF:6:F0opd1IRbeYI9QyVfzglUw==},
year = {2017},
version = {V8},
doi = {10.7910/DVN/42MVDX},
url = {https://doi.org/10.7910/DVN/42MVDX}
}