How does ICE classify nationality and citizenship in its public deportation datasets?
Executive summary
ICE’s public deportation datasets report a person’s “country of citizenship” as a primary breakdown, and DHS statistical products (which use ICE administrative records) also surface related fields such as “birth country” and “departed country” so analysts can identify third‑country removals; these fields come from ICE case records stored in the Enforcement Integrated Database (EID) and are published or reposted by intermediaries like the Deportation Data Project and OHSS [1] [2] [3]. The official classifications are administrative, not legal determinations of complex nationality status, and the published documentation warns that dual citizenship, missing values, and data errors can make “citizenship” an imperfect proxy for an individual’s nationality [4] [2].
1. What ICE actually publishes about nationality and citizenship
Public ICE and DHS tables routinely present counts “by country of citizenship” (sometimes labeled “citizenship country” or “Nationality”) and include companion columns for “birth country” and “departed country” (the country to which an individual was removed), with metadata appearing on the ICE ERO statistics pages and DHS OHSS products that the OHSS Persist/SSOR dataset is the statistical record of ICE administrative data [1] [5] [4]. The Deportation Data Project republishes raw ICE files obtained via FOIA and posts codebooks and internal agency documentation to help users interpret those fields [6] [3].
2. How “citizenship” is defined in these systems
DHS documentation describes citizenship in plain administrative terms — a “country to which a person owes allegiance and by which they are entitled to be protected,” noting that countries have different rules for granting citizenship via birth or naturalization [2]. In practice, ICE records reflect the nationality information captured at intake or in agency interactions and are tallied under the country-of-citizenship column in ERO arrest, detention, and removal tables [1] [3].
3. The data infrastructure behind the labels
The statistics derive from ICE administrative records kept in the Enforcement Integrated Database (EID), which feed the OHSS Statistical System of Record (SSOR) and the OHSS Persist dataset that DHS publishes; ICE’s ERO dashboards and monthly tables are produced from those systems [2] [5]. Independent projects — notably the Deportation Data Project — disseminate copies of these datasets, add codebooks, and annotate known idiosyncrasies to enable external analysis [6] [3].
4. Known limitations, edge cases, and explicit caveats
ICE and the Deportation Data Project explicitly warn that the “Citizenship Country” field may not capture dual or multiple nationalities, that some people may not have a recorded country of citizenship, and that errors are possible in administrative records; the recommended best practice to detect third‑country removals is to compare “Citizenship Country” with “Birth Country” and “Departed Country” [4] [2]. DHS also partitions enforcement counts by criminal‑history categories alongside country counts, which can shape public interpretation of nationality breakdowns [1].
5. How the data are used — and sometimes misread — in public reporting
News outlets and researchers commonly visualize “arrests/removals by country of citizenship” to show the national composition of enforcement, as in Newsweek’s maps that rely on ICE’s country‑of‑citizenship breakdowns [7] [8]. Analysts relying solely on that single field risk overstating national ties (for example, missing dual citizenship or third‑country repatriations) unless they cross‑check birth country, departed country, and case notes in the raw EID‑derived files reposted by the Deportation Data Project and TRAC [4] [3] [9].
6. Bottom line — what readers should take away
Official ICE/DHS deportation datasets classify people primarily by an administrative “country of citizenship” field sourced from ICE case records in the EID and published via OHSS/ERO outputs; supplementary fields (birth country, departed country) exist to disambiguate third‑country removals and catch edge cases, but dual nationality, missing data, and entry errors mean citizenship in the public tables is a blunt instrument rather than a definitive legal statement about an individual’s nationality [2] [5] [4]. Independent curators such as the Deportation Data Project and TRAC provide the raw files and documentation necessary to probe these limitations and to avoid simplistic readings of country tallies [6] [3] [10].