Analysis Finds Serious Flaws With New Jersey Voter Fraud Report

For Immediate Release
Wednesday, December 14, 2005

Contact Information
Natalia Kennedy, 212 998–6736

Analysis Finds Serious Flaws With New Jersey Voter Fraud Report

New York, NY A new analysis conducted by the Brennan Center for Justice at NYU School of Law and Dr. Michael McDonald, an elections expert at George Mason University, found factual and methodological errors in a recent voter fraud report submitted to the state Attorney General in September. The report claimed to have uncovered deep problems of voter fraud in New Jersey. In its continuing effort to scrutinize the accuracy of claims of voter fraud, the Brennan Center enlisted Dr. McDonalds assistance in analyzing the reports underlying data.

The analysts found serious methodological problems with the report, echoing the problems with the notoriously flawed suspected felon purge lists in Florida in 2000 and 2004. They cautioned election officials not to rely on the flawed report and urged the Attorney General to follow the normal statutory procedures for maintaining the states voter registration rolls.

Even a cursory examination shows that the methodology used to develop the alleged fraud lists is seriously flawed, said Dr. McDonald, who analyzed the lists of allegedly illegitimate votes submitted with the report, along with the New Jersey voter file. For example, it fails to resolve inconsistent data within individual records. Moreover, by assuming that two people who share a name and birth date are the same individual, the report fails to take into account even the most basic principles of statistics.

Justin Levitt, associate counsel at the Brennan Center, noted: Our analysis shows that the report substantially overstates the problem of illegitimate votes in New Jersey. In fact, the vast majority of the accusations in the report are simply unwarranted.

We hope that New Jersey does not allow these trumped-up allegations of voter fraud to influence any purges of the states voter registration rolls or to make it harder to vote, added Wendy Weiser, associate counsel at the Brennan Center.

The preliminary analysis uncovered multiple basic flaws in the report. Those flaws include:

• unjustified claims of double voting by individuals whose records were specifically flagged on the lists submitted to the Attorney General as less likely to indicate illegal voting.
• unjustified claims of double voting most likely based on faulty processing of one countys voter registration records. Analysis of the records of Middlesex County voters shows data processing errors resulting, for example, in fathers and sons listed with the same birth date.
• unjustified claims of double voting based on the similarity of two names that most likely belong to different people. For example, W_____ S. Smith and W_____ C. Smith, with the same first and last names, but with a different middle name and in different towns, were deemed to be the same person.
• an unjustified assumption, contrary to basic statistics, that two people who share the same name and date of birth, on a list of 3.6 million voters, must be the same person. In reality, such coincidences are surprisingly common given the 417 Robert Smiths who voted in New Jersey in 2004, it is more than 90% likely that two will share exactly the same birth date.

The full analysis is available here.

The Brennan Center for Justice unites thinkers and advocates in pursuit of a vision of inclusive and effective democracy. As part of its mission to facilitate access to the ballot, the Brennan Center monitors and evaluates techniques used to assess voter eligibility, for adding names to or purging them from the voter registration rolls. For example, the Brennan Center identified fundamental flaws in Floridas infamous purge list of suspected felons in 2004. Click here for more information about the Brennan Centers work in this area.