Over the past several months, Pew collected data about the 2012 presidential election from nearly every state and the District of Columbia. We used the findings to create a snapshot of each jurisdiction, focusing on how many people voted, how long they waited to cast their ballots, how they cast them, and how many ballots were not counted. These snapshots will be released over the coming months, five at a time, and the Election Data Dispatches will take a closer look at the latest snapshots each week.
Texas reported that only 860 people voted early in the 2012 presidential election—0.01 percent of the total ballots cast in the state—according to the U.S. Election Assistance Commission’s Election Administration and Voting Survey. These numbers contrasted sharply with the 4.9 million early ballots—60 percent of all votes cast—that the state reported in 2008.
In 1987, Texas became the first state in the country to allow early voting. A 2008 study by the commission reported that “an increasing percentage of [the state’s] voters take advantage of early voting with each successive Federal election.” So what happened in 2012?
When contacted by Pew about the data, Texas election officials reported that the actual number of early ballots cast was approximately 4.9 million, or about 62 percent of the nearly 8 million votes in the state.
The discrepancy between the two figures highlights the challenge and importance of good data collection by states and localities and by the commission. Jurisdictions need consistent and reliable systems for data collection and reporting so that the public, policymakers, and election officials can better understand how elections are being run. By the same token, organizations and agencies like the commission, that compile and publish state-provided data should avoid merely repeating what is reported to them and instead verify what is occurring in the states.