Use data analytics and behavioral targeting to find unregistered voters. It's easier and cheaper than you think.
VoterPAC aims to register, rally, and empower the rising American Electorate in California Central's Valley. VoterPAC and DemLabs are testing a new behavioral targeting approach to find areas likely to have unregistered voters. The approach uses technology and data from esri and GfK MRI. This blog explains:
- The demographic profile VoterPAC is using to find unregistered voters
- How different variables such as income, ethnicity, language spoken and past voting patterns are used in a search
Unregistered voter search criteria
Voter PAC chose five criteria for this project to find unregistered voters - ethnicity, income, language spoken and whether the person had been politically active (voted, signed a petition or attended a rally). Sanger in CA-22 was chosen as the town for the pilot project.
Understanding the area
"Kings, Tulare, and Kern counties are home to about 1.5 million people – represented by three Republican Congressmen Reps. Devin Nunes, Kevin McCarthy, and David Valadao. The Central Valley lags far behind in voter registration and participation. Across California, voter registration averages almost 90% of eligible voters. In the Central Valley, the average is much less (70 – 80%)." - Voter PAC
Voter PAC wanted to identify areas to target which would have likely unregistered voters. The analysis tool used allows for searched by state, congressional district, county zip code, census tract and group block.
Applying demographic and behavioral filters
You can search for areas based on a person's behavior with esri products in addition to demographic criteria such as age, income, ethnicity and gender. esri presents this information in an aggregated form to protect people's identities. This includes criteria such as:
- Did the person participate in any public activity in the last 12 months?
- Attended public meeting on town or school affairs?
- Engaged in fundraising?
- Volunteered for a charitable organization?
- Wrote or called a politician in the last 12 months?
- Attended a political rally in the last 12 months?
- Voted in a Federal, State or Local Election in the last 12 months?
- Signed a petition?
- Affiliated with Democratic or Republican party?
- Liberal or conservative outlook?
- Vote in National, State or Local Election? (Always/Sometimes/Never)
The search criteria can be adjusted with sliders to zero in the on the area with the most potential. This demonstration shows how areas with an annual income under $50,000 and a high percentage of hispanic families in Sanger with a higher political engagement was chosen. The results were used to create an infographic with the key facts.
This pro bono research project was completed in about two hours. The esri technology and data can be applied to any county in America and an annual nonprofit esri license costs about $100.
The Democratic candidate in CA-22 lost to Devin Nunes, the Republican incumbent by approximately 25,000 votes which is nearly identical to the margin of Republican versus Democrat voter registration in the district. 170,000 people voted in CA-21 in the 2020 election. This was the smallest turnout in California, where participation averaged 300,000 – 400,000 voters per congressional district. There are tens of thousands of voters to be contacted, listened to, and motivated to vote.
Voter PAC fights suppression with a Voter Outreach, Training, Education, and Registration (V.O.T.E.R) strategy. It aims to register and rally the rising American Electorate – people of color and younger voters – in California’s Central Valley.
Growth From Knowledge (GFK)
This analysis was done with an esri app that includes both esri and GFK data.
GfK has been a global leader in data and analytics, providing consumer and market insights since 1934. It operates in over 50 countries and has over 8,000 employees and 10,000 customers. GfK MRI collects data with both traditional surveys and passive measurement such as online convenience panels that track what people do, what they think, and what they say. This approach removes biases found in online convenience panels and captures the true diversity of the US consumer market.