Collect less. Explain every field.

What we collect

Controlled surveys may record poll answers, county, urban or rural residence, broad age band, education level, optional gender, completion duration, an anonymous response ID and a one-way invitation-code reference. Open Public Pulse responses record the selected answer, county, broad residence type, consent version, submission time and a hashed browser identifier used to limit repeat submissions.

Every geographic and profile field includes “Prefer not to answer”. The poll does not request a name, phone number, email address, exact address or GPS coordinates.

Why we collect it

Poll answers support the stated research question. County and residence type are used to assess geographic coverage and produce aggregates when the available base is large enough. Broad demographics support weighting and quality checks. Anonymous invitation and browser identifiers support duplicate control.

We do not use poll responses for advertising, individual profiling, eligibility decisions or unrelated marketing.

Storage and access

Responses are stored in the private database attached to the Caelus Polling site. Access is limited to authorised research administrators. Administrative changes, quality decisions and publication approvals create audit records.

The platform uses managed cloud infrastructure and does not claim Kenya-only physical storage. Every production study must document applicable cross-border safeguards and processor arrangements before fieldwork begins.

Retention

Public Pulse duplicate-control hashes should be deleted or irreversibly anonymised within 90 days after a pulse closes. Pseudonymous controlled-survey records should be deleted or irreversibly anonymised within 12 months after the final study release unless a documented legal or research requirement justifies longer retention.

Aggregate releases, methods records, approvals and non-identifying audit evidence may be retained to preserve the public research record.

Your choices and rights

You may select “Prefer not to answer” for county and profile questions. Where consent is the lawful basis, you may decline before submitting. After an anonymous response is separated from any identifying information, Caelus may be unable to identify which record is yours without your invitation or response reference.

Requests for access, correction, objection, withdrawal or deletion should be directed to the research contact published with the relevant study release.

Public reporting

Research outputs are published only as aggregates. Small geographic or demographic groups must be suppressed or combined when reporting them could reasonably identify a participant. Raw response records are not published or sent to the Research Copilot.