Skip to content
Data on detention and arrests in the U.S. immigration system from FY 1960 - FY 2019 (September)
Branch: master
Clone or download
Fetching latest commit…
Cannot retrieve the latest commit at this time.
Type Name Latest commit message Commit time
Failed to load latest commit information.

M               M   /_  __/ /  ___                       
M M           M M    / / / _ \/ -_)                      
M M M       M M M   /_/_/_//_/\__/         __        ____
M M M M   M M M M     /  |/  /__ ________ / /  ___ _/ / /
M M M M M M M M M    / /|_/ / _ `/ __(_-</ _ \/ _ `/ / /
M M M M M M M M M   /_/__/_/\_,_/_/ /___/_//_/\_,_/_/_/  
M M M M M M M M M     / _ \_______    (_)__ ____/ /_     
M M M M M M M M M    / ___/ __/ _ \  / / -_) __/ __/     
M M M M M M M M M   /_/  /_/  \___/_/ /\__/\__/\__/     

By Andrew R. Calderon


The United States immigrant detention system is the largest in the world. A network of hundreds of private prisons, county jails and federal government facilities hold tens of thousands of people daily.

Detainees are not being punished for a crime, they're being held in "civil detention," awaiting a hearing with an immigration judge. Thousands have alleged abuses and deteriorating conditions while inside these facilities.

Using documentary storytelling and data visualization, "Detained," published on September 24, 2019, tells the origin story of the immigration detention system Donald Trump inherited and his transformation of it.

Included in this repo are the detention, apprehensions, appropriations and locations data that The Marshall Project used to show how the detention population has grown over the past four decades.

If you'd like to use the data, please read on to understand its provenance.

Data Sources and Methods

To compile this data, we dug through decades of government files dating back to the 1960s, when the work of the Department of Homeland Security (DHS) and Immigration and Customs Enforcement (ICE) were done by a single agency called the Immigration and Naturalization Service.


The bulk of the data came from the House Committee on Appropriations budget hearing documents. We also used reports and documents held in the United States Citizenship and Immigration Services library, in addition to enforcement reports published online by the Department of Homeland Security and research from the Congressional Research Service. The apprehensions data came from the Customs and Border Protection website, where they publish data tables and arrest reports.

We published original source material on DocumentCloud. If you navigate to any of the documents, you will see public notes that guide you to the place where we pulled data.

We also used detention facility data that the National Immigrant Justice Center got in response to a FOIA request to ICE. We included here a subset of that data with the names, addresses, areas of responsibility (AOR), years of operation (DD/MM/YYYY) and longitudes and latitudes for all immigrant detention facilities, from calendar year 1978 through November of calendar year 2017.


Detention, Appropriations and Apprehensions

We were able to gather an almost complete set of data describing the growth of the detention system. For the years 1997 to 2000, we were only able to acquire partial data.

It is important to note that we calculated the Average Daily Population (ADP) from years 1979 to 1993 using formulas in the INS statistical yearbook that were then confirmed by the Department of Homeland Security.

For those years, ADP was calculated by dividing the total number of days that all detainees spent in detention in a given fiscal year by the number of days in that fiscal year. These were referred to by the INS as "mandays" or "detention days." For leap years, we divided by 366. For the rest, we divided by 365.

From 1994 and 2000, we took ADP from a Congressional Research Service report that cited the DHS and the INS as their sources. Given that this document was published later than the House Committee on Appropriations budget hearing reports, we relied on the most recent data. The data from 2001 to the present was sourced directly from ICE.


We determined longitudes and latitudes of ICE detention facilities using the Google Maps Geocoding API, and manually checked that geographic data, making corrections wherever the Google API didn't accurately geolocate a facility.

Please contact us if you have any questions.


It takes a village to raise a data story. This one benefitted from the scrutiny, creativity and talents of:

  • Emily Kassie, Director of Visual Projects, at The Marshall Project
  • Andrew R. Calderon, Data Reporter, at The Marshall Project
  • Weihua Li, Data Reporting Fellow, at The Marshall Project
  • Tom Meagher, Managing Editor for Digital and Data, at The Marshall Project

Questions, Comments?

See a way we can improve this repo? We love feedback!

Contact us at: or

If you wish to receive our roundups of the best criminal justice news from around the web, please subscribe to our daily newsletter.

You can’t perform that action at this time.