Definition
The Cellular Contactability Score predicts a person's likelihood of answering their cellular phone, and the Landline Contactability Score predicts a person's likelihood of answering their landline phone.
Technical Details
The models were built using phone responses from Murmuration’s 2024 poll. Survey respondents who answered their phone were labeled as targets while those who did not were labeled as non-targets. For the Cellular Contactability Score, we called approximately 200,000 individuals on their cellular phones, and approximately 6% of individuals answered their cellular phones and were labeled as targets. For the Landline Contactability Score, we called approximately 470,000 individuals on their landline phones, and approximately 4% of individuals answered their landline phones and were labeled as targets.
The models were 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 a voter will answer either their cellular or landline phone. We validated the model’s accuracy using a held-out test set of 20% of the original survey sample. Individuals in the top 20% of scores are 70% more likely to answer a landline phone and 34% more likely to answer a cellular phone compared to the average voter.
Use Cases
We recommend using the Contactability Scores to refine a universe built upon other scores, such as turnout and issue scores. The Contactability Scores can narrow down a segment by filtering out individuals with the lowest scores, who are therefore least likely to answer the phone. By adjusting the score threshold to match outreach capacity, partners can connect with the most responsive contacts and make better use of their resources. Even when using the Contactability Scores, partners should not expect that everyone within their target universe will respond, as response rates are still relatively low even for voters at the high end of the range. We recommend using the Cellular Contactability Score when you’re contacting cellular numbers and the Landline Contactability Score for landline numbers. Below are a few examples for how partners could use these scores:
- GOTV Issue-based Outreach: For GOTV efforts on specific initiatives, partners can start with voters who have mid-range turnout scores and high values of the issue scores that more closely align with the ballot initiative or issue at hand (e.g., the Democracy Protection Score, Education Voter Score, etc.). Then, they can apply a Contactability Score to further refine this universe by targeting those with the highest scores until they have a universe that matches their outreach capacity.
- Persuasion Efforts: For persuasion campaigns, partners can begin with a universe of voters who have high turnout scores. Then, add a Contactability Score to narrow down the segment to only those most likely to respond to direct conversations, ensuring resources are effectively utilized. They should make sure the team running the phone bank has the necessary guidance about how to talk about the issues they care about so that their phone bankers are able to be as effective as possible.
- General Audience Filtering: Partners that have created a universe for any kind of phone outreach that includes more people than they are capable of reaching out to can use these scores to focus their efforts. By filtering on only those with the highest Contactability Scores, partners can reduce the overall size of their universe and ensure that those who remain are the most likely to respond to landline or wireless phone calls.
Note: The top 20% of each Contactability Score is slightly biased towards certain racial demographics. For example, White voters are overrepresented in a universe made up of those with the highest Contactability Scores. These scores can help increase efficiency when trying to conduct outreach by phone, but they shouldn’t be used to narrow down universes for other types of outreach.
Targeting Table
The table below shows the score values associated with each decile to help you more easily target using the Contactability Scores nationally. Note: these score cutoffs may be different in your local districts.
| To target the top… | Set the minimum Cellular Contactability Score value as… | Set the minimum Landline Contactability Score value as… |
| 10% | 29 | 26 |
| 20% | 26 | 20 |
| 30% | 25 | 18 |
| 40% | 23 | 16 |
| 50% | 22 | 15 |
| 60% | 21 | 14 |
| 70% | 20 | 13 |
| 80% | 18 | 12 |
| 90% | 16 | 11 |
| 100% | 2 | 2 |