A new investigation posted at the Civitas Institute’s main website reveals some significant flaws in the accuracy of North Carolina’s voter rolls. Some highlights from the report:
We chose five counties — Buncombe, Durham, New Hanover, Pasquotank and Wake — to include in our own mailing to check on voters’ status. Our plan was to see how many voters’ letters would be returned as undeliverable, indicating they were not at the address they gave to election officials. We would mail to two sets of voters: First, those who registered to vote at the end of the regular registration time, between March 1 and April 13, 2012 (the deadline to register to vote), and, second, those voters who registered to vote during SDR.
The first surveys were mailed on April 20 and 21 to voters who registered to vote between March 1 and April 13, 2012 (the deadline to register to vote). A total of 17,531 surveys were mailed to these voters and 531 were returned to us as undeliverable — a 3.06 percent rate.
To put that in perspective, in the 90 days leading up to the 2008 General Election voter registration deadline more than 200,000 people registered to vote. So in a general election with a return rate of 3.06 percent would number approximately 6,000 verification mailings returned as “undeliverable”.
Then a total of 5,019 surveys were mailed to SDR voters. The surveys were mailed on three different dates – two dates before Election Day and the last on May 12, four days after the Primary Election. These mailings produced 365 undeliverable pieces of mail — a rate of 7.3 percent (more than twice the rate of those who registered during the normal registration process).
As this project unfolded, it was incredible to see the envelopes marked “undeliverable” flood our office. Given the highly competitive nature of many key races in North Carolina, the fact that so many registered voters with bad addresses continue to be on the rolls should present cause for concern.