Definition
The Education Voter Score predicts how likely a person is to prioritize public education when voting in state or local elections. Individuals with higher scores are more likely to prioritize education as a key voting issue.
Technical Details
The model was built from the following survey question:
How important is public education to you when deciding who to support in a state or local election?
- Extremely important — it's one of the top issues I consider
- Very important — I care a lot about it, but it's not my top priority
- Somewhat important — I consider it, but other issues matter more
- Not very important — it rarely influences how I vote
- Not at all important — education is not a political issue I care about
Respondents who answered "Extremely important" OR "Very important" are classified as education voters. Respondents who answered "Somewhat important," "Not very important," or "Not at all important" are classified as non-education voters.
The Education Voter Score was trained on polling data from the Summer 2025 Murmuration poll. We collected 7,721 responses online from registered voters in July and August 2025. The survey used stratified random sampling to ensure national representativeness across age, race, gender, partisanship, education and geography.
The model was trained using gradient-boosted decision trees. Model features were drawn from the Atlas by Murmuration dataset, which includes demographic (age, race, gender, etc.), commercial, geographic, and vote history information for all registered voters nationally.
Scores range from 0-100, where higher scores indicate greater likelihood that education is a driving factor in voting decisions. The score represents the model's predicted probability (scaled to 0-100) that an individual would rate education as "Extremely important" or "Very important" when deciding who to support in state or local elections.
We validated the model's accuracy using a held-out set of 1,537 polling respondents (20% of the original survey sample) whose data was not used during model development. Among individuals with scores in the top 20% of the Education Voter Score, 84% are actual education voters—making them 25% more likely to prioritize education than the average voter. By comparison, voters in the top 20% of the legacy Education Voter Score were only 18% more likely to be education voters (and those in the top 20% of the big changes in education score were just 14% more likely). At the other end of the distribution, individuals in the bottom 20% of the refreshed Education Voter Score are 25% less likely than average to prioritize public education.
To ensure generalizability, we also validated the model against other survey questions where respondents listed their top 3 issue priorities in general. Among voters in the top 20% of Education Voter Scores, 26% spontaneously mentioned education as a top priority, compared to 20% across the full population and just 13% in the bottom 20%. This confirms the score captures genuine education salience beyond the training question. Voters in the top 20% of scores were also 30% more likely to list education as a top 3 priority when given space to talk about their top issues than average.
The model achieved a test set AUC of 0.651, demonstrating strong discriminative ability in identifying education voters. In held-out test data, the model correctly distinguished between education voters and non-education voters at rates significantly above baseline.
Use Cases
- School Board and Local Education Ballot Measures: Partners can use the Education Voter Score to identify and mobilize voters for school board races, education funding ballot measures, and local education policy campaigns.
- For GOTV efforts, target voters with Education Voter Scores above 70 and layer on top of this the middle range (30-70) of a turnout score relevant for your election (such as a Municipal Turnout Score, or Midterm General Turnout Score if your election is held concurrently/downballot with the midterm).
- For persuasion campaigns focused on education funding measures, you can start by filtering on those with higher turnout scores (70 and up) who are highly likely to turnout in the election, and go a bit lower on your Education Voter Scores threshold (e.g. by using Education Voter scores of 60 and above).
- Coalition Building with Education Advocacy Groups: Partners can use the Education Voter Score to identify volunteers and activists for education advocacy campaigns. Target voters with scores above 75 to recruit for school board campaigns, parent organizing efforts, or education policy coalitions. You can layer on top of this filter those voters with high values of the Community Engagement Score (30 and above), which will help to identify the voters who are more likely to attend school board meetings, or volunteer and take on leadership positions within community organizations.
Targeting Table
The table below shows the score values associated with each decile to help you more easily target using the Education Voter Score nationally. Note: these score cutoffs may be different in your local districts.
| To target the top... | Set the minimum score value as... |
| 10% | 80 |
| 20% | 77 |
| 30% | 74 |
| 40% | 71 |
| 50% | 68 |
| 60% | 65 |
| 70% | 62 |
| 80% | 58 |
| 90% | 53 |
| 100% | 36 |