How does ICE define and report “criminal conviction,” “pending charge,” and “immigration violator” in its public datasets?
This fact-check may be outdated. Consider refreshing it to get the most current information.
Executive summary
ICE’s public enforcement datasets classify people arrested or detained into three mutually exclusive categories at the time of an enforcement action: “Convicted Criminal” (those with a criminal conviction recorded), “Pending Criminal Charges” (those with unresolved criminal charges recorded), and “Other Immigration Violator” (those without known criminal convictions or pending criminal charges but with immigration violations) — definitions and reporting appear in ICE’s own statistics and in public data derivatives [1] [2] [3]. Public presentation and media quoting, however, sometimes bundles convictions and pending charges together or uses broader rhetoric about “criminal arrests,” producing confusion about who is actually convicted versus merely charged or subject to civil immigration violations [4] [5].
1. How ICE labels people at the time of enforcement: administrative triage, not courtroom verdicts
ICE’s systems of record mark an individual’s criminality status “at the time of the enforcement action,” classifying subjects as “Convicted Criminal,” “Pending Criminal Charges,” or “Other Immigration Violator” — language and operational definitions echoed in ICE data releases and by researchers who analyze those datasets [2] [1]. That means the “Convicted Criminal” tag reflects a recorded criminal conviction in government records at arrest or booking, while “Pending Criminal Charges” reflects unresolved charges in criminal systems; “Other Immigration Violator” denotes immigration law violations (civil violations such as unlawful presence, visa overstay, or illegal reentry) with no known criminal charges or convictions recorded at that moment [1] [3].
2. What “convicted” and “pending” do — and what they do not — imply about guilt or severity
A “Convicted Criminal” entry in ICE public datasets is an administrative fact: a conviction is recorded in the exchanges ICE uses, not an ICE legal determination about the nature or severity of the underlying offense; many convictions can be for low-level or nonviolent offenses such as traffic violations or DUI [6] [1]. “Pending Criminal Charges” means charges have been filed by prosecutors but remain unresolved, and researchers warn that counts that fold pending charges into “criminal arrests” overstate the number of people actually convicted — pending charges may later be dismissed or downgraded [7] [8].
3. “Immigration violator” as a residual category and its legal character
ICE’s “Other Immigration Violator” category captures people who have violated immigration law but who lack recorded criminal convictions or pending criminal charges — this includes visa overstays, visa-waiver program violators, and illegal entry/reentry cases that are civil (not criminal) immigration violations [1] [3]. Because immigration violations are civil matters in U.S. law, being labeled an “immigration violator” in ICE datasets does not, by itself, mean a criminal finding; the label is a function of immigration enforcement criteria rather than criminal adjudication [8].
4. Reporting choices, public messaging, and the politics of counting
DHS and ICE public statements sometimes present combined figures — for example, asserting percentages of arrests that have “criminal convictions or pending criminal charges” — which can be technically accurate but obscures distinctions between conviction, charge, and civil immigration status and thus can mislead public perception about who is actually convicted or dangerous [5] [4]. Independent analysts and advocacy groups repeatedly note that the share of detainees with no convictions is large and rising in certain enforcement periods, and that local jail handoffs and policy shifts change the mix of categories reported [9] [10] [3].
5. Data limits, comparability, and prudent interpretation
Public ICE datasets and derivative compilations are useful for trends but have constraints: they reflect snapshots “at the time of enforcement,” rely on interagency data exchanges that may record dismissed or inaccurate charges, and differ across datasets (arrests vs. detained populations vs. removals), making year-to-year or claim-to-claim comparisons tricky unless methodologies are harmonized [2] [4] [11]. Where reporting goes beyond what the datasets state — for example, equating pending charges with convictions or asserting that “most arrested are the worst of the worst” without disaggregating categories — the reader should treat the claim as a communicative framing rather than a precise reflection of conviction status [5] [7].