Below you will find links to our sources and additional information about Crosscheck. Please contact us if you have additional sources, and excuse omissions as we build out the site.
For important analysis on the prevalence of double voting, and also a source of detailed data for Iowa's 2012 and 2014 Crosscheck lists.
There are more than two million cases in a national voter file in which 2012 vote records share a common first name, last name, and date of birth. We develop a probabilistic birthdate model to estimate how many of these cases represent the same person voting twice. If voter files are a completely accurate account of who voted, we estimate about 0.02% of the votes cast in 2012 were double votes. An audit of poll books, however, suggests that many of these apparent double votes represent measurement error when recording turnout in voter files. Nevertheless, concerns about double voting have led many states to participate in the Interstate Crosscheck Program, which promotes purging registration records that share a common name and date of birth. We find their proposed purging strategy would eliminate about 200 registrations used to cast legitimate votes for every one registration used to cast a double vote.
For analysis of Crosscheck in general and the discriminatory aspect of Crosscheck in particular, see The GOPs Stealth War against Voters, his twitter feed (@Greg_Palast), and much more.
And a tip of the hat to Mr. Palast, without whom few outside the Kansas Secretary of State's office would have ever known about Crosscheck.
"Conclusion: This study, using actual data, has demonstrated the problem of using name-DOB combinations to uniquely identify a person. What is, however, surprising is that despite all that was said, it’s still possible to uniquely identify a substantial portion of the population. In fact, almost 92% of the people in our database are uniquely identifiable solely by their names and dates of birth and so one might want to ask the question why bother with the remaining 8.3% who aren’t uniquely identifiable (by names and DOB.) Before one can answer the question, one must ask what purpose accurate identification serves, and additionally, what is the cost associated with misidentification."
For details on how Crosscheck (and ERIC) work at the state level, and for a great example of appropriate transparency and accountability by an organization entrusted with the sacred power to vote on behalf of millions.
The Electronic Registration Information Center (ERIC) is a non-profit organization with the sole mission of assisting states to improve the accuracy of America’s voter rolls and increase access to voter registration for all eligible citizens. ERIC is governed and managed by states who choose to join, and was formed in 2012 with assistance from The Pew Charitable Trusts.
For an understanding of how common sharing a name and birthdate is.
Introductory statistics classes are commonly presented with the Birthday Problem: the surprisingly high probability that two students in the class share the same birthday. Here, we review and extend the Birthday Problem to find the probability that two persons in a given group share an exact birthdate and in a related calculation, the expected number of matching birthdates in a group of a certain size. These thorny calculations are simplified in a simulation framework. We then apply the Birthdate Problem to a pressing legal and public policy debate concerning allegations of widespread double voting and/or multiple registration. We show that these allegations are inflated by not appropriately accounting for the Birthdate Problem, and discuss the implications of the Birthdate Problem for the debate over double voting and the means to address this perceived fraud.
Washington Post: This anti-fraud program gets it wrong over 99% of the time.