How often are children being physically restrained or secluded in your school district? Which schools report the most incidents? Who is being subjected to these practices, and what questions should we be asking about the numbers?
Families, educators, advocates, and policymakers need access to this information. But finding it, understanding it, and using it should not require expertise in analyzing large government datasets.
I am excited to share the Alliance Against Seclusion and Restraint’s new Restraint & Seclusion Data Explorer. This web-based application makes the latest available federal data, covering the 2023–24 school year, explorable by state, school district, and individual school.
Our goal is to help people move from a national number to a better understanding of what schools in their own communities are reporting, and use that information to ask questions and advocate for change.
What is the Civil Rights Data Collection?
The U.S. Department of Education’s Office for Civil Rights, administers the Civil Rights Data Collection (CRDC). The collection gathers information about educational opportunities and barriers in public schools, including enrollment, access to courses and staff, school discipline, and restraint and seclusion. It supports OCR’s work to identify discrimination and enforce federal civil rights laws. You can learn more on OCR’s CRDC overview page.
The CRDC is the only nationwide federal dataset documenting physical restraint and seclusion in public schools. It is an essential resource for understanding these practices across the country, although it provides a historical picture rather than a real-time account of what is happening today.
Our Data Explorer displays 406,794 reported incidents for the 2023–24 school year:
- 263,376 incidents of physical restraint
- 16,037 incidents of mechanical restraint
- 127,381 incidents of seclusion
These are incident counts. A child subjected to repeated restraint or seclusion may appear in those counts many times.
Behind those numbers are children, families, educators, and school communities. Making the information easier to explore can help us identify where closer examination and stronger support and oversight are needed.
Why we built our own explorer
OCR already provides a public data website, including extremely limited visualizations for restraint and seclusion. However, I believe its presentation falls short of what families, schools, journalists, and advocates need.
On OCR’s state profile pages, restraint and seclusion appear within a broader “School Climate” section. Users select among separate measures for mechanical restraint, physical restraint, and seclusion, divided into IDEA and non-IDEA student categories, and then choose demographic and chart options. The site offers a few useful functions, including chart downloads and demographic information, but does not make the most important data readily accessible.
For someone trying to understand how frequently a district uses these practices, that fragmented presentation has limited value. A chart showing how many students experienced physical restraint answers a different question from how many restraint incidents occurred. Families need to understand both and more. The data set is very rich, but the OCR presentation of it is not.
We built our explorer to bring those questions together and make the path from national information to local detail more direct.
From the national picture to your child’s school
The homepage provides national totals, definitions of the practices, an interactive state map, and separate searches for states, districts, and schools.
It also displays lists of the states, districts, and schools with the largest reported incident totals. Readers can switch those lists among total incidents, restraint only, and seclusion only, then open a profile to examine the details.
State pages provide summaries and searches for districts and schools within the state. District pages bring together district totals and school-level information. Individual school pages allow readers to examine the reporting closest to their own school.
The profiles provide several layers of information:
- Incident totals, with physical restraint, mechanical restraint, and seclusion distinguished in detailed tables.
- Reported student counts, helping readers examine the difference between the number of incidents and the number of students subjected to a practice.
- Incident breakdowns by disability status, including students without disabilities, students served under IDEA with an IEP, and students with Section 504 plans.
- Race and ethnicity breakdowns for students subjected to physical restraint and seclusion, with detailed male and female counts.
- Gender charts on district and school pages, alongside tables that make the underlying counts visible.
For example, the Lincoln Public Schools profile in Nebraska displays 4,786 reported seclusion incidents and 532 students secluded during the 2023–24 school year. Seeing those figures together helps readers recognize that the incident total includes repeated use. The aggregate data do not tell us how those incidents were distributed among individual students, but they give us a reason to ask.
That is the value of making the details accessible. A national total can draw attention to an issue. Local information can help a community begin a specific conversation.
Understanding what the numbers can, and cannot tell us
The depth of the explorer also makes careful interpretation essential.
The race charts describe students subjected to a practice, while the disability-status tables describe incidents. Those measures should not be treated as interchangeable. Student counts across different practices may overlap, so adding them together does not necessarily produce a count of unique children.
Likewise, the rankings show raw reported totals. They do not account for differences in enrollment, student populations, or reporting quality. A larger total alone does not establish that an individual child faces a greater risk in one district than another. Examining disproportionate impact also requires comparing the affected students with the relevant enrollment population.
The explorer helps identify patterns and questions. Further investigation supplies the context needed to understand them.
Underreporting remains a serious concern
We must also be clear that better visualization cannot repair incomplete source data.
In its June 2019 letter to the Department of Education, the U.S. Government Accountability Office identified serious inaccuracies in restraint and seclusion reporting. Reviewing the 2015–16 collection, GAO found that 70 percent of districts reported no restraint or seclusion incidents. It found substantial evidence that nine of the 30 largest districts had incorrectly reported zeros despite having incidents or lacking the necessary data.
GAO recommended clearer reporting instructions, follow-up with districts, stronger oversight of missing-data plans, and disclosure of known data problems. Its current recommendation tracker marks all four recommendations as implemented.
Those findings concern an earlier collection; they do not quantify underreporting in 2023–24. Nevertheless, they provide a documented reason for caution. Significant underreporting remains a concern, and the current totals may substantially understate actual use.
A reported zero should prompt questions when it conflicts with family accounts or local records. It should not automatically end the inquiry.
Our commitment to accuracy
We have done our best to make this information accurate, accessible, and useful. We also want readers to understand the limitations of the data and verify information when using it to inform research or advocacy.
The Alliance Against Seclusion and Restraint provides this data for informational and research purposes only. While every effort has been made to ensure accuracy, the organization makes no warranties regarding completeness or current validity of the data. Users are encouraged to verify any information directly with the U.S. Department of Education Office for Civil Rights.
OCR and the Department of Education must do better
Collecting and publishing civil rights data is a critical responsibility. Making those data understandable and usable is part of that responsibility.
Families should be able to find reported incidents and student counts together, understand the definitions, examine demographic patterns, and see clear explanations of missing or questionable information. Researchers and advocates should have practical ways to compare schools and districts with appropriate context.
OCR and the Department of Education should invest in stronger data validation and public tools designed around those needs. A small nonprofit can help make the information more accessible, but federal agencies must ensure that the public can meaningfully use the data they collect.
The quality and accessibility of this information matter because it is intended to help protect students’ civil rights.
Turning information into action
My hope is that families will use the data explorer to prepare questions for school meetings. Educators and school leaders can examine reported practices and consider what changes in training, staffing, environments, and student support are needed. Journalists, researchers, and advocates can identify patterns that warrant further investigation.
Useful questions include:
- How do the federal figures compare with state incident reports and family experiences?
- What explains repeated restraint or seclusion?
- How is the school or district checking the accuracy of its reporting?
- What is being done to prevent crises and reduce reliance on restrictive practices?
I encourage you to explore the application, look up your state, district, or school, and share it with others who could benefit.
Our larger goal remains reducing and eliminating restraint and seclusion and advancing trauma-informed, neuroscience-aligned, neurodiversity-affirming, relationship-driven, and collaborative approaches.
Accessible information can help us ask better questions and demand meaningful change. The purpose is to build schools where children are safe, understood, and supported—and educators have the resources to make that possible.
We can and must do better.
