Last updated: Oct 04, 2024

Pennsylvania

Memos

Pennsylvania 2024 State Report

The 2020 presidential election in Pennsylvania saw a host of challenges presented by the pandemic coupled with a closely contested election in the state that had implications for nationwide results. Pennsylvania was one of the first states to experience the host of broader nationwide challenges of election skepticism, controversies regarding election administration, and new types of interactions between political parties and election administrators.

In the years between 2020 and 2024, Pennsylvania has been the site of several legal battles with important consequences for the state’s election administration, especially in the realm of voting by mail and other methods that became popular and widespread after the pandemic. Data from the 2024 primary demonstrates Pennsylvania’s continued use of mail-in voting years after the pandemic, while voter registration data reflects changes in the state’s electorate, including the implementation of Automatic Voter Registration (AVR). As election administrators look toward the 2024 general election, this report analyzes some of the administrative and legal issues that will continue to shape the election landscape in Pennsylvania and the nation.

Memo / September 12, 2024

State Resources

Department of State

Voting & Election Information

Data Visualizations

Voter Registration

These plots show the week-by-week change in the total number of registered voters in Pennsylvania, with voters split into three categories: registered Republicans, registered Democrats, and all other registered voters.

About the Data

The data are acquired from the Secretary of State's office at this website (as of September 10, 2024, they are obtained by clicking on the link "Voter registration statistics by county"). The date that the data were last updated is displayed at the top of the first sheet of the file, and the timestamps indicate that they are updated every Monday.

About our Process

We compute the net change from one week to the next by subtracting the number of registered voters of a given party in that week from the number of registered voters of that party in the previous week's file. Voters affiliated with another party or no party are combined into our Other category.

The net change that we plot is distinct from the number of new registrants of a given party. Those two numbers can appear substantively different: for example, if more Democrats register than Republicans in a given week, while in the same week more Pennsylvanians cease being registered Democrats than registered Republicans, then the net change would appear different from the raw number of new registrants. When Democrats experience a greater net decrease or a smaller net increase than Republicans, it may partly be attributable to the larger number of Democrats in Pennsylvania overall, but measuring the differential rate in attrition between registered voters of the two parties would require individual-level data; for example, from the Pennsylvania voter file.

Last updated: October 2, 2024

This graph shows the weekly net changes to Pennsylvania's voter registration totals, from mid-July through the end of September. Voters are split into Republicans, Democrats, and all others, and the net change is the week-by-week difference in the total number of voters registered in each of these categories. There has consistently been a larger net increase in Republicans than in Democrats or Other registrants, with Republicans typically netting about 1,000 to 2,000 more voters than Democrats throughout August, with the gap increasing to 4,000 or 5,000 more in the second half of September.

Previous Voter Registration Graphs

September 24, 2024

This graph shows the weekly net changes to Pennsylvania's voter registration totals, from mid-July to late September. Voters are split into Republicans, Democrats, & all others; the net change is the week-by-week difference in the total number of voters registered in each of these categories. There's consistently been a larger net increase in Republicans than in Democrats or Other, with Republicans typically netting about 1,000-2,000 more voters than Democrats in recent weeks. Republicans have ranged from adding about 3,000-7,000 net registrants every month; Democrats have ranged from losing 2,000 net voters in the second-last week of July to gaining about 5,000 in the last week of July. The number of Other registrants has increased by about 4,000 every week. In the last week the net increase in Republicans decreased from about 13,700 to about 12,600, while the net increase in Democrats declined from around 9,000 to nearly 8,000. The net increase in Other registrants rose by 7,400.

September 10, 2024

This graph shows the weekly net changes to Pennsylvania's voter registration totals, starting in mid-July and continuing through early September. Voters are split into three categories: registered Republicans, registered Democrats, and all other registered voters, and the net change is the week-by-week difference in the total number of voters registered in each of these categories. There has consistently been a larger net increase in Republicans than in Democrats or Other voters, with Republicans typically netting about 1,000 to 2,000 more voters than Democrats in recent weeks. Republicans have ranged from adding about 3,000 to 7,000 net registrants every month, while Democrats have ranged from losing 2,000 net voters in the second-last week of July to gaining about 5,000 in the last week of July, and the number of Other voters has increased by about 4,000 every week with comparatively little variation. — This graph shows the weekly *net changes* to Pennsylvania's voter registration totals, starting in mid-July and continuing through early September. Voters are split into three categories: registered Republicans, registered Democrats, and all other registered voters. The net change is the week-by-week difference in the total number of voters registered in each of these categories. There has consistently been a larger net increase in Republicans than in Democrats or Other voters, with Republicans typically netting about 1,000 to 2,000 more voters than Democrats in recent weeks. Republicans have ranged from adding about 3,000 to 7,000 net registrants every month, while Democrats have ranged from losing 2,000 net voters in the second-last week of July to gaining about 5,000 in the last week of July, and the number of Other voters has increased by about 4,000 every week with comparatively little variation.

Mail Ballot Status

About the Data

This plot describes the number of mail-in ballot applications approved in Pennsylvania, and the number of ballots that have been cast. The Pennsylvania Department of State maintains a Statewide Mail Ballot File that voters may opt into in order to receive a mail-in ballot. The File lists every application for a mail ballot, whether or not that application was approved, and whether the voter who made the application has returned their ballot.

To compute the number of Approved Applications, we use the Statewide Mail Ballot File, as of the date that we generate the plot. We subset the Mail Ballot File to only approved applications, and we count the cumulative number of applications that had been approved on or before a given day. To compute the number of Ballots Cast, we sum the cumulative number of ballots in the Statewide Mail Ballot File that are marked as having been returned by that day. Every day we update the plot up to the day before the file was released, so that we are always reporting on a day that has already finished. For example, suppose that at 9 AM on a Friday we update our script using a file that was last updated by the Pennsylvania Department of State at noon on Thursday; then Wednesday would have been the last complete day of applications logged in the file.

Note that there are strong periodic effects in when mail ballot applications are processed. Each curve representing Approved Applications periodically flattens for a few days, then increases, then flattens, and so on. This is a result of applications tending to be processed more on some days of the week than others. The figure below shows the number of applications approved by day of the week since January 1; while there is little evident connection between weekdays and the number of approved applications, very few approved applications are dated to either Saturday or Sunday. The minimal changes on these days together causes the periodic flattenings visible on our daily plots.

A bar graph showing the number of applications approved by day of the week. Saturday and Sunday are far lower than the other days, which all appear roughly evenly divided.