How did TRAC and the New York Times analyze ICE FOIA data to reassess who was deported under Obama?
Executive summary
Two independent deep-dives — one by the Transactional Records Access Clearinghouse (TRAC) at Syracuse and one by The New York Times — used ICE records obtained through Freedom of Information Act requests to reclassify and quantify who was actually removed during the Obama years, finding that most removals were for minor offenses or immigration-specific violations rather than serious felonies [1]. Both analyses relied on case-level ICE data releases that allowed researchers to group convictions, trace enforcement pathways and challenge the long-standing public narrative of Obama as chiefly a remover of dangerous criminals [1] [2].
1. How the records were obtained and what they contained
TRAC and the Times started with ICE data that was produced in response to FOIA requests — datasets that contain case-by-case enforcement records including arrest, detainer and conviction codes and, in some releases, linked identifiers that allow researchers to follow an individual through ICE’s enforcement pipeline [2] [1]. Those FOIA-derived files were more granular than headline removal totals, enabling both teams to sort removals by conviction type, fiscal year and processing pathway instead of relying on aggregate, often agency-published counts [2] [1].
2. The analytical approach: merging, recoding and reclassifying convictions
The core of both projects was data cleaning and recoding: researchers merged ICE tables using the identifiers supplied in the FOIA release, recoded ICE’s offense fields into clearer categories (for example separating traffic infractions, immigration offenses, misdemeanor-level nonviolent crimes and serious felonies) and then calculated shares and trends over time — methods TRAC has long used in its reports and that the Times described in its analysis [2] [1]. That recoding revealed a large growth in ICE’s “criminal” deportee classifications driven not by violent or drug felonies but by traffic-related convictions and immigration offenses, a shift TRAC quantified as dramatic increases in traffic violations (+191%) and immigration offenses (+167%) in certain years [1].
3. What the numbers showed and why they mattered
Both teams reached a similar substantive conclusion: a majority of removals attributed to the Obama era involved people with no serious criminal convictions — often traffic infractions or immigration-specific offenses — and only about one-fifth of cases involved convictions for serious crimes such as drug offenses, violent felonies or sexual crimes (the Times quantified serious-crime removals at roughly 20% or about 394,000 cases) [1]. TRAC’s year-by-year breakdown highlighted concrete examples: in FY2013 the top recorded “criminal” convictions among deportees were illegal entry , DUI and traffic offenses , whereas traditional serious crimes like burglary and domestic violence were far less common among removals [1].
4. Limits, alternative interpretations and political context
The FOIA-based approach exposed both strengths and limits: while case-level ICE data lets researchers reclassify offenses and trace enforcement pathways, certain ICE releases omit or complicate some tables (the Deportation Data Project notes missing removal/encounter tables in some releases), meaning analysts must be transparent about gaps and the choices made when recoding offense categories [2]. Policy scholars caution that higher removal totals under Obama coexisted with shifting stated priorities (targeting criminals and recent border crossers), so reassessing who was deported does not erase higher absolute removal counts — it reframes who those people were and whether they fit the administration’s stated priorities [3]. Political agendas shape both FOIA requests and media narratives: immigrant-rights advocates spotlight overreach, enforcement proponents emphasize total removals — both benefit from the granular data but interpret it differently [1] [3].
Conclusion: a reassessment grounded in granular public records
By obtaining ICE’s case-level records via FOIA and applying transparent recoding and merge techniques, TRAC and The New York Times independently showed that many Obama-era deportations categorized as “criminal” were for low-level traffic or immigration offenses rather than serious felony convictions, challenging simple labels and reframing historical comparisons of presidential enforcement records [1] [2]. Both analyses underscore that counting deportations is a methodological exercise as much as a political one: what is counted, how offenses are classified, and which datasets are used materially change the story.