How do ICE’s public dashboards handle duplicates and data revisions, and what FOIA releases fill those gaps?
Executive summary
ICE’s public dashboards are promoted as open-data tools, but public records and litigation show important gaps: independent FOIA-driven releases have supplied the granular, line-level data—often with anonymized unique identifiers—that ICE dashboards either omit or do not document how they deduplicate or version records [1] [2]. Organizations such as the Deportation Data Project and litigants like the ACLU and TRAC have used FOIA to extract arrest, detainer and detention files that reveal revision patterns and duplicate-key issues that the dashboards do not transparently explain [3] [4] [5].
1. What the official dashboards publish — and what they don’t say
ICE hosts FOIA resources and an electronic library and publishes datasets on official portals, but the agency’s public materials focus on top-line statistics and compliance with FOIA process rules rather than metadata about deduplication or revision procedures, leaving the dashboards’ internal record-linkage mechanics undocumented in the public-facing pages [1] [6]. Data catalogs associated with ICE and DHS list various datasets but do not, in the materials cited, provide a machine-readable provenance trail that explains how the agency removes duplicate person-level records or retroactively updates counts when records are corrected [7].
2. How independent FOIA releases reveal duplicates and revisions
When outside groups force the production of raw exports, those files often show multiple entries for the same person across systems and timestamps that document updates — the pattern that explains why published dashboard totals can shift after FOIA-produced spreadsheets are reconciled [3] [4]. The Deportation Data Project’s FOIA-driven release of arrests, detainer requests and detentions through October 15, 2025 provided the underlying rows for analysis and allowed researchers to see repeated I-247 detainer events and date-stamped changes that are not plainly visible in the dashboards [3] [4].
3. Legal rulings and anonymized identifiers changed the game
Courts have compelled ICE to produce line-level datasets with substitute unique identifiers in lieu of A‑Numbers, a decision that forced the agency to reveal the movement of cases across systems without exposing sensitive IDs — and that court-driven transparency makes it possible to identify duplicate records and follow revisions across datasets while preserving privacy [2] [8]. The Second Circuit’s reasoning means FOIA recipients can receive spreadsheets where ICE’s internal keys are replaced with anonymized unique IDs, enabling deduplication and longitudinal matching by third parties [2].
4. FOIA litigation and watchdog pressure filled gaps left by dashboards
TRAC, the FOIA Project and other transparency litigants have repeatedly challenged ICE’s withholding and abrupt shifts in disclosure practices; those suits documented that ICE sometimes declared key data “unavailable” or withheld fields needed to determine whether a detainer led to custody or deportation, forcing public-interest FOIA requests that restored access to line-level details [5] [9] [10]. These FOIA-driven releases augmented and in some cases corrected what the dashboards reported by supplying the granular logs needed to reconcile counts and explain revisions.
5. Remaining blind spots and methodological uncertainty
Public sources demonstrate that FOIA releases and litigation can recover the raw rows and anonymized keys necessary to detect duplicates and reconstruction of revisions, but the available ICE pages and dataset catalogs cited do not publish a formal, public methodology that states how dashboards deduplicate, reconcile multiple source systems, or handle retroactive record edits; therefore conclusions about the dashboards’ exact internal deduplication algorithms rest on FOIA-supplied exports and external analysts’ reconciliation work rather than an ICE-authored provenance document in the cited materials [1] [7] [11].
6. Where reporters and researchers should look next
The best evidence for how duplicates and revisions actually affect public counts comes from FOIA-obtained spreadsheets and court-ordered productions (including anonymized-identifier tables) released by groups such as the Deportation Data Project and those secured via ACLU litigation; those releases let independent teams trace duplicates, follow revision timestamps, and reconcile dashboard fluctuations in ways the official portal does not yet permit [3] [4] [2]. For methodological transparency, watchdogs continue to press ICE through FOIA and litigation to publish system-of-record tables and reconciliation logic so future users won’t have to reverse-engineer the dashboards from FOIA dumps [11] [10].