Tested across
Nine states
Civly Turnout Score · Research Brief
The Civly Turnout Score · Testing turnout predictions
The Civly Turnout Score improved on past voting history in our tests
A participation rate measures the share of past general elections in which someone voted. Campaigns use it to identify likely voters, but it can miss changes in who participates in primaries, municipal elections, and other low-turnout races. The Civly Turnout Score estimates each voter's probability of participating in a specific election.
Calls correct, 2024 general
78% → 84%
On a low‑turnout race
67% → 83%
New data needed
None
Limits of past voting history
Two ways a turnout estimate can miss
Ranking voters by participation in recent general elections is one way to build polling samples and outreach lists. It can be less useful in a primary or municipal race with a different electorate. A list based on that ranking can leave out people who will vote and include people who will stay home.
Miss the real voters
The people who’ll show up aren’t on your list
Sporadic and newly registered voters may rank too low for an outreach list even when they intend to vote in the upcoming election.
Target the wrong ones
62% actually vote
Much of the list you build stays home
In the low-turnout race we tested, only 62% of voters identified as likely to vote by the participation-rate rule cast a ballot. Nearly four in ten did not.
Across all voters in that race, a 2025 municipal election, the rule classified 67% correctly.
Measured against real turnout
Higher accuracy across the elections tested
We generated scores using information available before each election and compared them with recorded turnout. In the 2025 municipal race, the Civly score classified 83% of voters correctly, compared with 67% for the participation-rate rule. It was also more accurate in each of the nine states tested for the 2024 general election.
67%rule of thumb
→
83%Civly Score
Calls correct on a low‑turnout race
Accuracy measures whether each prediction, vote or no vote, matched the recorded outcome. In the 2025 municipal election, the Civly score was correct 83% of the time, compared with 67% for the participation-rate rule.
The table shows classification accuracy for the 2024 general election in each state:
| State | Rule of thumb | Civly Score |
|---|---|---|
| Kansas | 71% | 87% |
| Delaware | 76% | 86% |
| Missouri | 77% | 86% |
| North Carolina | 78% | 86% |
| Pennsylvania | 79% | 83% |
| Florida | 78% | 81% |
| District of Columbia | 78% | 80% |
| Rhode Island | 80% | 82% |
| Michigan | 85% | 87% |
The Civly score was more accurate in all nine states. It classified 84% of voters correctly on average, compared with 78% for the participation-rate rule. Predictions used information available before the election and were checked against official state turnout records.
Using the score
Use turnout estimates to plan outreach
The Turnout Score works with Civly's voter file, polling, and message testing. Campaigns can use it to weight poll results, choose voters for outreach, and compare support before and after a campaign activity.
Evaluated against official turnout in the 2024 general election across nine states and a 2025 municipal election. The comparison uses each voter's participation rate in past general elections. Both methods predicted the same number of voters as actually turned out.