What is L2?
L2 is the largest and longest-serving non-partisan national voter file provider. L2 provides tens of thousands of campaigns and advocacy organizations as well as universities and media organizations with access to part or all of our enhanced national voter file. L2’s national voter file contains approximately 200 Million registered voters. L2 collects the data from each of the 5o states and Washington D.C. at a minimum of twice per year, often much more frequently. Each file is put through L2’s 53 stage hygiene and append processes which means that every file is run against the National Change of Address System (NCOA) and death index matching as well has matched against multiple national consumer sources primarily from credit bureau based data. Once those and many other processes take place L2 will release an individual state voter file providing eligible organizations with the best look at their state/district voter file available including all current voters and cleaned of those that have recently moved out of state or died. In addition to standard voter attributes including party ID, voter history, age etc…L2 appends hundreds of fields of consumer/demographic/contact information including cell phones, landlines and email addresses. This 200 Million record and approximately 700 field database makes up the L2 national voter file and is loaded into our campaign user interface, “DataMapping”. L2 DataMapping makes it possible to instantly pull counts, purchase and download specific or large selections of voters.
What data go into L2’s 2020 General “Return Ballot” posts?
Early and absentee voting has occurred for decades and each state and in some cases, counties have different rules on early voting as well as the reporting of those data to the public and qualified organizations. L2 receives files from a great many states, this year we’re planning on 41 the largest number of early states ever, providing us with a list of those specific individuals that have requested, been issued, or returned a mail-in ballot or voted early in person. Every state reports different information at different cadences, in these posts we’re focusing specifically on those ballots which have been returned or where a voter has voted early in person as a single aggregated number. L2 matches those data back to the state and national voter file which allows us not only to see the number of voters that have participated, which is widely available, but we go a step further and match back those known early voters back to our rich file voter/demographic/consumer insights for each voter.
Line-by-Line through a post:
“FL Return Ballots: 1,863,732”: This number is based on the number of returned ballots provided by the state in our most recent update of that state’s early return ballots, typically no more than 24 hours prior to the post. It’s important to note that while we are aggressively updating, in many states same-day registrations cannot be tracked and those individuals are not counted or displayed in the post.
“29% R / 51% D / 18.9% NP”: Based on the total return ballots above we look at those individuals that are registered with the Republican or Democratic Party. We also look at those “NP” or non-partisan voters that are either unaffiliated with a major party. In 13 states L2 models party identification based on surveys, demographic information and previous primary voting history and those data are shown. In Florida, in the example below, party ID is done by registration so the data are not modeled.
“A: 18-29 5.7% / 30-39 6.7% / 40-49 7.9% / 50-64 24% / 65+ 55%”: This field is the early/absentee turnout broken out by age ranges. L2 has preset age ranges in our DataMapping system and we use those to provide readers/viewers with a quick look at the age groups showing up early to vote. Not included in the total are the number of voters for whom we don’t have an age. In almost every state age is part of the voter file leaving only a small % or less than a % of voters not represented.
“Asian 1.7% / White 66% / Hispanic 16% / AfAm 12%”: L2 works with some of the top academic demographic researchers in America and utilizes complex naming tables and census block data to assign likely individual and broad ethnic categories to 93%+ of voters nationwide. The “broad ethnic groups” above are the % of the early/absentee voters that fall into each of those categories.
“HS Diploma 20% / Bach Degree 21% / Grad Degree+ 12%”: Using a combination of known survey data and modeled data L2 is able to assign a likely or extremely likely education completed flag to 61% of voters nationwide. These percentages are those voters who have voted early/absentee and are likely or extremely likely to have completed High School only, a Bachelor’s Degree, or Graduate degree or higher.
“1st Time Voting/New Regs: 12%”: This field provides a percentage of voters that are voting for the first time since registering. This combines those new registrants that have previously not been elidgable to vote due to age or other restrictions and those that have been eligible but have chosen not to vote.
If you have any questions about L2’s data or methodology or if you’re a researcher, member of the media, work for a political campaign or advocacy organization and would like to request a login to DataMapping please email: paul.westcott@L2political.com