The Social Security Administration (SSA) has corrected nearly 15% of active Social Security Number records using internal agency data and information from other government sources, while concerns remain about the accuracy of some records used in the cleanup, according to an October 26 Nextgov/FCW report.

The agency said it had corrected records across approximately 330 million U.S. citizens and nearly 390 million total records. SSA said the initiative was intended to improve identity data, prevent fraud and support the administration of government programs.

As part of the effort, SSA added 17 million dates of death to records that previously lacked them, using internal agency systems and Department of State data. 

It also updated approximately 25 million records that were missing citizenship information, drawing on the Department of Homeland Security’s Systematic Alien Verification for Entitlements (SAVE) service. 

The agency said it was working to correct another 8.5 million records with inaccurate immigration status.

The cleanup has drawn criticism from the National Committee to Preserve Social Security and Medicare. In an October 23 blog post, the group warned that SAVE contains errors and has previously identified some U.S. citizens incorrectly as non-citizens. It argued that using the database to update SSA records could carry those inaccuracies into another system.

SSA referred questions about the corrections to its public announcement.

Data Quality Depends on Reliable Sources and Validation

The cleanup illustrates a central challenge for chief data officers (CDOs): correcting a large database using information from other systems can improve completeness, but it can also spread errors if the source data is unreliable.

Data matching alone does not establish that a record is accurate. Organizations need to understand where information originated, how it was verified and what happens when two sources disagree. Those controls are particularly important when records inform identity checks, eligibility decisions or public benefits.

A robust data governance program should establish accountability for source quality, record validation, correction procedures and the ability to trace changes. For a database used across government, those safeguards can help prevent a correction in one system from becoming a persistent error elsewhere.