Overview

U.S Customs and Border Patrol (CBP) is the largest federal law enforcement agency of the U.S Department of Homeland Security. Also the country’s primary border control organization. It holds unaccompanied children after their initial arrest. CBP is legally obligated to refer them to the Office of Refugee Resettlement’s unaccompanied alien children program within 72 hours, but in 2017, such children were being held for weeks or months in CBP custody. The cruel conditions children had to undergo in these facilities are inhumane and violate their constitutional and human right as refugees seeking asylum. Americans for Immigrant Justice published a report Do My Rights Matter? The Mistreatment of Unaccompanied Children in CBP Custody in October of 2020. The report highlights the following conditions that these children underwent which includes being held in frigid rooms, sleeping on concrete floors, are being fed frozen food, and having little to no access to medical care. The children also described how they were subjected to emotional, verbal, and physical abuse by CBP officers. The following graph below, illustrates the journey that unaccompanied children must navigate through from their apprehension to their court proceedings.

The Marshall Project, a non-partisan, nonprofit news organization focus their work around the U.S criminal justice system and exposes violations and abuses in the system to influence change. The organization published a news article 500,000 Kids, 30 Million Hours: Trump’s Vast Expansion of Child Detention. The analysis confirms that more kids were held for 72 hours or more. For this lab, I was inspired to take the data from this analysis to recreate the charts and explore the data even further.

Process

Data

Through Data is Plural’s full archives, I found datasets relatable to my field of interests revolving around U.S-Mexico border and found Child Detention that referenced the project and dataset. The Marshall Project obtained public records from the U.S Customs and Border Protection in October, 2020 for the article mentioned above that both Anna Flagg and Andrew R. Calderón worked on. The dataset has 583,808 rows/observations and a total of 11 features/columns.

The dataset includes the following features:

• Date and time the child entered CBP custody
• Date and time the child left CBP custody
• Apprehension Date
• Age group
• Hours in custody
• Gender
• Citizenship
• Border
• Field Office
• Source
• Sector

Selected tools

OpenRefine

I imported my .csv file into OpenRefine to make sure there weren’t any discrepancies within my data. For example, I found that the gender feature had four different labels “F- Female”, “M – Male”, “Female”, and “Male”. I clustered them into “Female” or “Male. There were also “NA” or “Unknown” categories and I was unsure how to interpret this so I removed them from my dataset. I attempted to find the abbreviations of the “border” feature but was not able to successfully to do. Once finished, I saved and exported my updated dataset.

Tableau Public

I opened my updated dataset from OpenRefine onto Tableau Public and used The Marshall Project’s analysis report charts as a starting point on where I’d begin my exploration. Tableau Public is a software that makes it easy for the public to create data visualizations.

Analysis

Using Color Brewer, I used a 3 class color palette and implemented this theme onto my visualizations.

My questions began with the same investigation focus that The Marshall Project took on. How often did the CBP violate the 72 hour limit imposed by 2008 Trafficking Victims Protection Reauthorization Act and 1977 settlement agreement known as Flores Act and how has that changed over the past couple of years? How many children were kept more than 72 hours? While thinking about this, I was curious to see which countries were among the highest of children seeking U.S asylum? What was the relationship between gender and country of origin? What was the largest age groups that were seeking refuge in the U.S?

As shown below, bar charts and line graphs were the optimal representations to portray the data given.

Results

Similar to the original visualizations from the report, this chart illustrates the amount of children detained by day between Jan 17, 2017 and June 19, 2020. As noted in the graph, the highest peak of children registered into CBP was May 4, 2019. From those children, half spent 72 hours or more in custody. By January 2020, you can see a spiked decrease in children taken to custody. This would have to be explored further to analyze what factors have influenced this to happen.

This side by side bar graph illustrates the amount of hours children were in custody by year in CBP detention centers. The hours are grouped by intervals of 9. I chose 9 because it is easily divisible by 72 and that is an important mark throughout this analysis. I capped the bins to 144 hours, which equates to 3 days. However, I do have to emphasize that the data registers a larger range between 180 and 27,576 (1,149 days) a child has been kept in custody but when implemented to the graph are outliers. We can see a rapid growth of children being detained for more than 72 hours.

The following line graph illustrates the top three countries where children were escaping between 2017 and 2020 by gender. As annotated in the graph, 2019’s second quarter (April-June 30) was the peak where CBP received 21,390 young refugee girls from Hondura’s and 31,161 young male refugees from Guatemala. Since the countries differ and have brought significant numbers of the opposite sex. There is space to do further research on what was occurring in these two countries and how they were affecting the children to flee their country of origin.

The following bar graph displays the total amount of the children’s age groups between 2017 and 2020. If you group 0-5 year old children, it equates to 159,897 not too far from children who are within the age group of 15-18 years old. This begs to ask how many infant caretakers are there in the CBP facility to give adequate support and care for these children while they are in custody.

Reflection

Given the current skills I have developed utilizing Tableau, I was not able to fully replicate the original charts that were used in the report. I found myself gravitating towards geography and gender a lot but couldn’t create a chart that would portray those features strongly so I decided to put that aside from publishing. Throughout the process I discovered new features such as parameters and create a calculated features, hopefully I will implement in later projects. I did duplicate hours in custody feature column and created bins for figure 3 above. As I analyzed the visualizations even further, more questions stirred that were outside the magnitude of what I had available to me.

Citation

“Trump Detained More Migrant Children At The Border For Far Longer Than We Knew.” The Marshall Project, 30 Oct. 2020, https://www.themarshallproject.org/2020/10/30/500-000-kids-30-million-hours-trump-s-vast-expansion-of-child-detention.

Paz, Rosario, et al. Do My Rights Matter? The Mistreatment Of Unaccompanied Children in CBP Custody. Americans for Immigrant Justice, Oct. 2020, p. 74. Zotero, https://aijustice.org/wp-content/uploads/2020/10/Do-My-Rights-Matter-The-Mistreatment-of-Unaccompanied-Children-in-CBP-Custody.pdf.

Related Articles

“Family Separation under the Trump Administration – a Timeline.” Southern Poverty Law Center, https://www.splcenter.org/news/2020/06/17/family-separation-under-trump-administration-timeline. Accessed Oct. 2021.

“U.S. Detention of Child Migrants.” Council on Foreign Relations, https://www.cfr.org/backgrounder/us-detention-child-migrants. Accessed Oct. 2021.