Public voter data poses privacy risks, analysis finds

As reported by The Register, analysis of public voter data reveals significant privacy concerns, demonstrating how readily identifiable individuals can be through cross-referencing with other datasets.

Noah M. Kenney, founder of consultancy Digital 520, conducted research using voter data from Travis County, Texas, and Robeson County, North Carolina. His findings indicate that even with some data redaction, individuals can be re-identified. For instance, combining voter records with Federal Election Commission contribution data allowed Kenney to match 52% of contributors to identifiable voter records in Texas. North Carolina's data, which includes phone numbers for many voters, further increases re-identification potential. Kenney's research highlights that simple combinations like name and ZIP code can uniquely identify a high percentage of voters.

The study also found that voter history can act as a unique identifier, and that specific data points like APO/FPO codes for military mailings can expose deployed military families. Kenney advocates for access controls, such as rate limits and identity verification, over simple data redaction, and suggests that individuals should have the option to opt out of public data sets.

Source: The Register