Disabled Vet Benefits logoDVB
Sign Up — Free

Research Ethics & No-Reidentification Policy

Last updated: March 2026

Disabled Vet Benefits publishes anonymized research about veteran benefit trends, built entirely from aggregate statistics. This page documents the standing safeguards we apply so that no individual can ever be identified from that research. It is referenced from our Privacy Policy (see Section 3.6).

Our No-Reidentification Commitment

Disabled Vet Benefits (operated by CI Industries, LLC) commits to the following whenever aggregate or statistical data derived from user information is shared, published, or sold:

  1. Technical safeguards. Any aggregate dataset or report is built so that no data point represents fewer than 30 users (the “n≥30” threshold). Any segment that would fall below this threshold is suppressed and labeled as insufficient sample size rather than shown as a number.
  2. No re-identification, ever. DVB will never attempt to re-identify any individual from aggregate or de-identified data, for any purpose.
  3. Contractual protection. Anyone who receives DVB’s aggregate research (subscribers, sponsors, partners) is contractually prohibited from attempting to re-identify individuals from it.
  4. No inadvertent release. Aggregate reports are reviewed against the n≥30 threshold before publication or delivery, every time, as a standing step in the report-generation process — not a one-time check.

This policy applies to all current and future aggregate data products, including our aggregate insights reports, /advertise audience statistics, and the annual survey results.

The n≥30 Sample-Size Floor

The n≥30 floor is the core technical safeguard behind everything above, and it is a formally documented policy — not an informal rule of thumb. Before any figure appears in a report, dashboard, or dataset we produce, we confirm that it summarizes at least 30 distinct users. If a particular slice of the data — for example, a small state combined with a narrow rating band — would describe fewer than 30 people, we do not publish a number for it. Instead we suppress that cell and label it as having an insufficient sample size.

This is what makes our research genuinely anonymous rather than merely aggregated: small groups are exactly where re-identification risk lives, and the floor removes that risk by design before any human ever sees the output.

Questions

If you have questions about how we build our research or protect individual privacy, contact us at .