July 2014 podcast: ProPublica reporting on seclusion and restraints results in playbook for reporters interested in pursuing similar story

ProPublica reporter Heather Vogell talks with Student Press Law Center Executive Director Frank LoMonte about her recent reporting into seclusion and restraint practices at schools across the country.

Frank LoMonte: Welcome to the Student Press Law Center podcast, the monthly rundown of news and developments affecting the rights of student journalists around the country. I’m Frank LoMonte, executive director of the Student Press Law Center and from to time we like to focus the podcast on especially interesting examples of investigative journalism using public records that might be inspiring to students around the country interested in following the example.

We have an especially exemplary one to talk about this month. It’s Heather Vogell’s work for ProPubica — the nonprofit, investigative news site — that took a deep dive into the world of the restraint of unruly students in America’s public schools. Hundreds of thousands of time every year, kids are being confined in locked rooms, pinned down in wrestingly holds, hog tied like calves in a rodeo, or otherwise physically restrained for acting out at school. This is a practice that would be illegal if done in a federally regulated mental institution but not at a public school, as Heather and her reporting team at ProPublica discovered.

Heather Vogell is well known to people in the education journalism field for her groundbreaking work on standardized test cheating while a reporter at the Atlanta Journal-Constitution. That reporting exposed widespread wrongdoing that led to criminal indictments against dozens of people in the Atlanta schools, including the former school superintendent, so we’re especially delighted to have Heather Vogell of ProPublica here to talk about her reporting on this subject. Thanks for being with us.

Heather Vogell: Hi. I’m happy to be here.

Frank LoMonte: Well, Heather, I guess if you could just sort of dive in with the origin of this story — how did you become aware that there was a larger problem that needed scrutiny, of students being restrained in public schools and how did you go about taking on that project?

Heather Vogell: I became aware of the issue through a phone call from somebody that I’d been in touch with for awhile, who was an advocate for parents who had children that were special education, special needs students. I got a call from her one afternoon, right as I was transitioning to take the job at ProPublica.org, and she told me about a mom she’d been in touch with who had a kindergartener who was being restrained repeatedly in the classroom. And I think even at one point the mom had seen the kid being restrained and not being allowed to get up and get a drink of water, something like that. And so I hadn’t been aware of that happening in classrooms and had also coincidently gotten an email recently about the federal legislation that was being sort of advocated for on the federal level that there had been an investigative report out recently. So, both those things came around the same time and peaked my interested and I started looking into it and found out there was a lot to write about on a national level. There had been a lot of local stories when incidents happened here or there, but nobody had really put it together recently and written a big overall national story or looked at the most recent data.

Frank LoMonte: Well, you mention that the data, and of course that’s the backbone of any investigative project like this, so can you talk about how do you went about getting a hold of what had to be a massive number of public records, and maybe some of the obstacles that you encountered on what is, undoubtedly, a sensitive subject, that a lot of schools would probably prefer not to talk about.

Heather Vogell: Yeah sure, and the data was definitely a key part of this. In reading that report to the Senate about the restraints and seclusion issues in public schools, they had mentioned that there was data being collected by the Office of Civil Rights and the U.S. Department of Education, and that that data had shown, I think in the 2009-2010 school year, that there was 66,000 uses of restraints and seclusions. And it struck me because this report had come out in February, and this was 2009-2010 data, and so I just was curious whether there was going to be an update with more recent data coming out soon, so I called the Department of Education and found out that indeed there was some new data coming out in the spring. That was terrific and I was excited about that, but, when it came out, the data had a lot of problems with it. It was very dirty data, meaning that it wasn’t something you could take the whole database at face value because there had been errors in inputting it by districts, as well as the fact that it was still clearly incomplete. Do you want me to talk a little bit more about that and how we worked through that?

Frank LoMonte: Definitely, because I think that’s something that a lot of people fall into, is that they get a hold of the data and they think they’re reporting is over and sometimes it’s really just beginning.

Heather Vogell: Yeah. I work with data a lot in my stories and I increasingly see the data not as sort of being an end point, it’s really just a beginning point, a tip sheet if you will. You’ve got to really work hard to validate what you have, one, and two, understand it in context, which can mean that it can be revealing in some ways but data can also hide some nuances in what’s happening. In this case, we weren’t getting a database that showed, faithfully, every single incidence of restraint and seclusion in the country. What it did show, was that there were a lot more instances of this happening than had ever been previously documented. It turned out that that early sample that I had looked at, that was in the report from February, was just from a sample of districts that were over 3,000 kids, and then were were some states that reported all their districts. The newer dataset, which is the one that I went after for this story, was the first time that the federal government had collected this information from every school district in the country. And despite that, there was acknowledgement that a lot of districts were still not collecting this data, including, potentially, the three largest districts in the country, which all reported zeroes. They may have policies that don’t permit restraint as often, or in most cases even, as other districts, but pretty much everywhere that I’ve seen that has been faithfully reporting, even a state like Connecticut that has a rule that says you can only use restraints in emergencies, they still report thousands of uses because you have fights, you have real things that happen. So, if the data was real from these districts, you’d expect to see at least a handful of situations with a distict that big.

So, first of all, we knew that we were dealing with incomplete data. And second of all, I started to see things in the data that set off alarm bells. Like, a district that reported that it had equals numbers of every category. For instance, it would have 11 restraints, 11 seclusions, 11 restraints for kids with disabilities, 11 restraints for kids without disabilities, it was like every category across had the same number, so clearly the statistical possibility of that happening was so low that we basically went through and excluded those.

Frank LoMonte: I’m sure this is starting to remind you of seeing people cheating on standardized tests. Right? I just bubbled in “B” for everything.

Heather Vogell: Right. It would be sort of the same thing. You sort of know it’s probably not valid. It’s possible, but unlikely. And so basically what we ended up doing to sort that out as best as we could and be as faithful as we could to the idea of wanting to present a real number, we went back to the people who had collected the data at the department of education and just had a conversation with them about what they thought the limitations were in the data. And so they shared a few tips for sorting through it, to try to catch some of the errors in data input that were in there. So we ended up excluding a good number of incidents that we thought were probably largely over-reports. Although, we probably ended up throwing out some instances that were real in the process but we knew that a lot of them weren’t real. So, that process took a few weeks just to get through that and I worked with our data editor here, Jeff Larson, to do that.

Frank LoMonte: So when you got down to actually having to look into individual cases at the institution level, and there’s some very powerful ones that are mentioned in your story, I have to imagine that you ran into objections about student privacy and student confidentiality. Can you talk about ways to deal with that?

Heather Vogell: Yeah, sure. You hit the nail right on the head. That is always an issue with reporting on schools. The confidentiality and privacy issue, this sort of formal issue that there’s a federal protection under FERPA is the acronym for it, that requires schools to keep students’ personal data, academic information, private, and there’s also sort of, I think, a moral and ethical component to it for a reporter, in that I don’t believe in thrusting a student into the limelight without their being aware of it and in complete agreement of it happening. I mean, when you have a situation with an arrest or something and the name becomes public in another way, that’s different, but you have to tread very carefully when dealing with vulnerable populations, kids in this case, and a lot of them had disabilities.

So, what we needed to do was find a situation where everybody was very comfortable with sharing the story. It turned out that original tip that I got, that that family was not comfortable with it. They were pursuing a legal avenue and their attorney wanted them to be patient and not work with the press at this point and so my lead anecdote disappeared, just a couple weeks into the reporting. So I had to figure out, “well, I still think this is really interesting and it seems to be happening a lot, so I’ve got to find somebody else.” The way I did that was to call around to some advocates for kids with disabilities. And it turns out that in every state there is something called a protection and advocacy agency and they’re all federally mandated groups that protect the rights, they’re there to advocate for disabled people and help them stand up for their rights. And a lot of the complaints about what’s happening in schools were funneling to these groups from local attorneys in different states. So, calling around to some of these guys, talking with the National Disability Rights Network, which is sort of the head of all the different protection and advocacy agencies, and getting different names and numbers, sometimes of advocates of attorneys, brought me to a few others families and frankly that happened fairly quickly. I was surprised at how easy it was to find examples that were really heartbreaking and representative of what seemed to be happening out there. And I ended up focusing on two in my story but there were many, many others that were out there, across the country.

Frank LoMonte: So you were able to gather some pretty eye popping, aggregate data, showing that there are literally, we’re talking about the magnitude of 200,000 times a year, somewhere across the country a child is being either locked away in a room or put in a restraining hold, or sometimes even literally put in cuffs or something like cuffs, and I guess the question is, were you able to, we know that there are some subset of these where as you mentioned some subsection, like breaking up a fight, where it’s necessary to physically grab one of the combatants and immobilize them for a second to keep somebody from getting more badly hurt, so some subset of those probably falls into that legitimate category, were you able to draw any conclusions about whether there is a systematic overuse of restraint when it’s not one of these imminent, physical danger situations?

Heather Vogell: I think it’s impossible for us with the data that’s out there to draw a definite conclusion like that because the data is so spotty. We don’t know how much under — we know there’s underreporting going on, we don’t know how much there is. There’s some schools that are reporting very faithfully and they appear to have very high numbers of restraints and seclusions and I think the advocates feel there are too many, that it’s a sign that their treatment plans, their work with disabled students, is not, typically with disabled students it’s not effective.

But I think that overall there is a consensus that restraints and seclusions — seclusions perhaps never, there’s a lot of people who say that there’s never a reason for those — restraints, the advocates, and a lot of of the experts say, they have no therapeutic benefit for children. There’s a great risk of harm to the children and also to staff members who are trying to implement them. This is a dangerous thing to do. You’re putting your hands on a child who’s out of control. There’s a consensus that they should be used as little as possible and what we’re seeing is that there are places that are turning to them very often and so that was sort of the most of a conclusion that we could come up with, that these are things that are supposed to be used rarely, if ever, but there are places that are using them a lot. And we don’t have a very clear, there’s not a very clear picture of everywhere. We don’t have a complete understanding of what’s happening out there and so there are sort of injuries and situations that are popping up here and there that maybe we wouldn’t have been aware of, that these places have a problem.

I look at Connecticut, they are connecting data perhaps in a more diligent way than even the federal government has been on the use of restraints and seclusions and they have a very, as I said a law that only allows it during emergencies, and what they’re doing is they’re using — and I think Florida does this too — they’re using that data to look for districts that stand out among other districts, as outliers, as having too many. And they’re going and saying, “Hey, what’s going on here? Why are you using this tool so much, is there something that we can do to help you figure out a better way — perhaps more crisis intervention training, more training in de-escalation techniques for staff or better planning with your special needs students who are at risk for these sorts of outbursts?” There’s real concrete things that they can do when they see the problem through the data. The problem is that nationwide, A, there’s not one standard nationwide and 2, the data isn’t good, so we have sort of a murky picture right now and we’re forced to rely on anecdotes more than I think would be really prudent for good policy, if that makes sense. It’s a really long answer to a simple question.

Frank LoMonte: No, no, no. And you mention that patchiness of the record keeping and one of the really interesting pieces on the ProPublica.org site — which I recommend that everybody check out — has a graphic looking at how much or how little states regulate the use of seclusion and restraint and you’ve actually got a map that people can check out so as to compare the state of regulation in their own state and you’ve got 15 states and the District of Columbia on this map that do not have any state-level regulation whatsoever, so this is really being left to the discretion of districts or of individual people at the building level. So that is one way certainly that a reporter who is interested in this subject might be able to go forward and localize it, by comparing the state of their state to others around it. What other suggestions or ideas might you have for somebody who’s interested in doing a project of this kind in their own community?

Heather Vogell: I’ll add that, it’s a really good point, our map is a really good place to start in terms of getting a sense of where your state is on regulations. The thing that I would look at to compare it to would be the U.S. Department of Education has come up with voluntary guidelines for school districts on policies on restraint and seclusion and we used some of those elements to create this map but if you wanted more details you can look at your state’s law and compare it to what the federal government says is best practices. And I think a lot of states, or most states, are still pretty far from that ideal. So, that’s a good place to look at the policies. Then you have to add the district policies on top of that. Each of these districts are either writing policies or not having a policy, either of which could be very interesting. That’s a good place to start. I always say don’t assume that policy is practice. What you’re going to need to do is find a way to get at practice, which is the hard thing. And that’s when you start talking to parents of special needs students and advocates. Looking for those complaints or concerns, teachers, people who are in the building who can talk about what’s happening. A big piece of this that I wrote a little bit about but that I think really deserves more attention, is that districts are supposed to have these fairly sophisticated approaches to handling behavior that’s challenging. They’re supposed to be using something called Positive Behavior Interventions and Supports, which is sort of a whole system of decisions you can make. That is supposed to reduce the chances of having these sorts of outbursts, basically to help teach students who are having a hard time with their behavior, help them learn how to control their behaviour better. It’s a very important foundation for them as they get older and if it’s successful it can really help them live independently and get something out of school sometimes in cases where they may not have without that help. And I think I read that only one in five districts nationwide is using it. I’m not 100 percent sure about that statistic, but it was fairly low. So look at how your school district is handling crisis intervention, I guess behavior management, challenging behavior, seeing whether they’re doing positive behavior, what kind of training they offer — is it in restraints, is it in these other areas? Those are all good places to look.

Frank LoMonte: Just by way of wrap-up in a minute or two, any sort of closing thoughts or tips. Not about this subject in particular but for a student journalist who is out there and is trying to embark on a really data-heavy project that might involve quite a consumption of time, quite a consumption of manpower. You’ve been through several of these projects now in the last few years — any advice for somebody that’s trying to dive in for the first time?

Heather Vogell: I think that the thing you want to do, again, understand the context of the data that you’re getting and analyzing, and that means probably doing your best to get to the source, to talk to the people that, usually government, work closely with the data that are either present or help with the data collection, understand the data collection really well, they know where the holes are in data collection and what it might be missing, what it might be over-reporting, how it might be misleading, and they can also help you figure out how you can use the data to help look for interesting things, anomalies, maybe haven’t come out before. That’s a good way to start.

Frank LoMonte: I would add to that, as well, that sometimes the absence of the data in itself is the story. If you have something that’s sufficiently important, especially something that involves people’s physicals safety, and government agencies haven’t bothered to count it or keep track of it, that’s a very valid question for journalists to be asking. If we don’t know the number, if we don’t know the frequency, then the question is why don’t we know that, and that question in itself can be a story that can lead to real change.

I want to thank Heather Vogell of ProPublica for being with us and sharing some tips about investigative reporting using data. ProPublica.org is the website and that’s one of many terrific investigations on this nonprofit newsite that we commend to your attention. So, are there other resources that you could recommend that journalists use, things that ProPublica offers, that might help them get started on an investigative project of this kind?

Heather Vogell: Absolutely. In this issue in particular, what we have done is we have provided a whole bunch of different tools for reporters online to get started, including, what we’ve been doing is collecting anecdotes from around the country, examples of this, restraint and seclusion, from parents, sometimes educators, and we are matching them up with actual reporters who are interested in learning about their stories and using these anecdotes. We also have the data to share that we’ve cleaned and we’ve got, of course, the information about the state by state regulatory information and also a tip sheet of, sort of a reporting recipe, for how to approach reporting on restraints and seclusions. And all these can be found on ProPublica.org.

Frank LoMonte: I’ll make sure to put links to some of those resources on the SPLC’s Twitter feed, which is just @SPLC. We try to use our social media to share ideas and tips for students interested in doing investigative reporting in their own communities. Heather, thanks for being with us and for everyone who’s listening who is involved in a public-records intensive project, Student Press Law Center is there to help. We’ve got free legal help for you. splc@splc.org is the email, the 703-807-1904 is the phone number and our attorney specializes in helping people get a hold of records from schools and colleges that are difficult to obtain. We’d love to hear from you, help you out, and we encourage you to use all of the free resources available on the SPLC.org website. Thanks for listening and we’ll talk to you next month.