Hearst Newspapers in New York, Connecticut, California and Texas collaborated on a year-long project investigating the use of restraints and seclusion on children in schools, a wide-ranging and in-depth series of reports that immediately earned the attention and concern of lawmakers. 

The project began when an education reporter in Connecticut first heard of the practice and it was quickly determined that these efforts are largely unknown to most people, says Brendan Lyons, managing editor of the Times Union, in Albany, New York.

Primarily, restraints refer to physical holds and movement restrictions used on students by educators at times when students appear to be having an emotional outburst. Seclusion can refer to isolated, sometimes padded, rooms in which a student is placed for a period of time, but the rooms might be no larger than a closet. 

While the initial work began in Connecticut, some early digging suggested the practice is more widespread than that.

“A lot of readers we talked to, unless they were intimately involved, didn’t know this happened,” says Matt Rocheleau, an editor overseeing data and investigative projects for the Times Union and Hearst Connecticut Media Group. “That drove everyone’s curiosity. We knew it was a much bigger issue than just one statistic. Having the resources Hearst does, there was a great deal of support. I think as soon as we started to research the issue, we realized we could do something bigger.”

But that led to some pretty big challenges for compiling and analyzing data, namely each state collects, analyzes and reports on the use of restraints and seclusion rooms in schools differently. 

“We pursued data through three avenues,” says Emilie Munson, a data reporter for the Times Union. “The first was going after data collected by the U.S. Department of Education through biannual civil rights data collection process. … One of the issues we found is the last time this data was collected was the 2017-2018 school year. In addition, there have been several reports underlining the flaws in this dataset, including significant levels of underreporting by districts.” 

Then, the Hearst reporters contacted each of the state governments, requesting records from state-level education agencies.

“A bit more than half of all states collect their own data through their state education department,” Munson says. “We sent public records requests to all 50 states and got whatever information they had and compiled that in a way no other reporters had done to date.” 

The third piece of the puzzle was to get data from districts across New York. In all, the team requested school-level data from 22 school districts, including the three largest districts — New York City, Albany and Buffalo — to obtain information that was more recent than what was contained in the four-year-old federal reports.

“Through those districts, we obtained five years of reports documenting incidents,” Munson said. “We ended up getting tens of thousands of pages of records back. We used machine learning software to process those records and turn them into a data set we could analyze. We had to build our own data in New York to get a look at the frequency of use of restraints and time-out rooms.” 

“Most often, these techniques are used on students with disabilities,” Munson says. “The interventions are disproportionately used on Black students and boys.” 

Lyons called the response to the report remarkable.

“I was proud of the work the team did because they drew the immediate attention of lawmakers,” he said. “I think you’ll see changes as a result.” 

The Hearst Newspapers recently finished a year-long national investigation of how schools across the U.S. routinely restrain students or keep them in seclusion. It’s All Journalism host Michael O’Connell talked to the team from the Times Union in Albany, New York about the investigation.