Hundreds of elementary students arrested at U.S. schools each year

12/9/2022

I analyzed federal school-discipline data to expose that hundreds of elementary-age children are arrested each year — and that the burden falls heaviest on Black students and those with disabilities.

ETLPython
➜ view live project
Hundreds of elementary students arrested at U.S. schools each year

Drawing on the U.S. Department of Education’s 2017-18 Civil Rights Data Collection, I isolated every elementary school that reported at least one “school-related arrest,” then spent weeks confirming each figure with districts and state education officials, pruning out erroneous records.

With a cleaned, open-sourced dataset and reproducible code on GitHub, I calculated arrest totals and rates, broke them down by race and disability status, and mapped where the youngest children were most likely to be handcuffed.

The analysis revealed more than 700 arrests of kids as young as five in a single year and showed that Black students and children with disabilities bore a sharply disproportionate share — findings that sparked calls for discipline reform and gave parents, educators, and lawmakers actionable, district-level evidence of the problem.

I first reported on this issue when I was at CBS News Chicago, where we published a story about similar disparities at Chicago Public Schools. I was then able to broaden that story out to the entire country while at CBS News. And a year after the story was published, we followed up when new data was released. We found young kids were still arrested even during pandemic years when most were learning remotely.

Tools and techniques used

ETL
Extract, Transform, Load processes for data cleaning
Python
Data analysis and statistical programming language