Niya Kenny pulled out her cell phone and began recording.

It happened in 2015, after a classmate had refused to hand over her own cell phone during class and was being pulled from her chair by a police officer based at their school, Spring Valley High School in Columbia, S.C. When Kenny loudly protested and, like her classmate, refused to hand over her phone, she too was arrested.

The charge: disturbing a school.

Kenny's story is one of several featured in a new investigation from Education Week, " Policing America's Schools," exploring the debate over school discipline, the role of police officers in schools, and why it is that black students are arrested at disproportionately high rates in much of the country.

Education Week will roll out the results of its investigation over the next few weeks, including the tool below that allows users to explore their state and school-level data more closely.

In the 2013-2014 school year, according to a new analysis of federal data by the Education Week Research Center, black students accounted for 16 percent of students enrolled in public schools but 33 percent of arrests in those schools. In 10 states, black students' share of arrests in schools with at least one arrest exceeded their share of enrollment by at least 20 percentage points.

NPR Ed spoke with Education Week staff writer Evie Blad about the project. The interview below has been edited for clarity.

What stood out to you from the investigation's findings?

That the disparities we see in arrests and discipline referral rates are most persistent for black students across a majority of states.

Just looking at arrests in 43 states and Washington D.C., black students made up a larger share of students arrested than they did the population in schools that had at least one arrest. So, they're more heavily represented in the group of students that face contact with law enforcement in schools.

In most states, there are a couple hundred arrests. In some, the number is small enough that it doesn't take a lot to move a percentage point. But we found that, in 28 states, the share of arrested students who are black is at least 10 percentage points higher than their enrollment in schools that arrested students. And in 10 states, the gap is at least 20 percentage points. There's a pretty big difference between their representation and their share of arrested students. And the same is true for referrals to law enforcement.

What did you find in terms of a student's likelihood of being in a school with a law enforcement officer?

At both the middle school and high school level, black students are most likely to be in a school with a school-based law enforcement officer.

At the high school level, in general, officers are more common. So, 74 percent of all black students were in schools with an on-site officer, followed by 71 percent of both Hispanic and multi-racial students, 65 percent of Asian students, and 65 percent of white students.

In a middle school, it's about 59 percent of black students, and the next lowest group is Hispanic students at 49 percent, followed by white students at 47 percent. So, the disparity is a little more severe. Elementary school police officers are far less common so there's a little bit of a closer clustering of races.

Do we know whether a student who attends a school with a law enforcement officer is, for that fact alone, more likely to be arrested?

It would be hard to look at this one piece of federal data and draw that conclusion definitively because there are so many differences between schools. But other researchers have explored this. There's some research that came out in the last year that compared student arrests based on specific offenses, like vandalism, fights without a weapon — things like that — at schools that were demographically similar. They found that the rates were higher if a school had a school-based officer.

In theory, schools that have officers should be treating them like the officers you would call in off the street. So, if I have an officer and a student is doing something wrong, would I call 9-1-1 for this student's behavior? But, civil rights groups feel that schools aren't setting clear enough boundaries for officers and some of them are getting involved in routine discipline. Some [groups] also feel that the hiring of law enforcement and the push for school security are driving resources away from student supports like counselors and social workers who might be able to remedy some of these issues before they escalate.

One of the things that interested me about your reporting is the fact that some states have codified tougher penalties for being disruptive in class, which may lead to higher arrest rates for students doing things that, in many places, would not justify a call to 9-1-1.

Right. So, that's been a big focus in recent years. In 2015, one video that got a lot of headlines was of a girl being arrested in South Carolina. A school-based officer pulled her out of her chair and dragged her across the floor before arresting her. The charge that she and her classmate [Niya Kenny] were arrested under was disturbing a school, which is a South Carolina law. A lot of folks argue that it was originally written to apply to people who came in from the outside and disrupted public schools but that it's often applied to students. That law specifically limits things like obnoxious behavior.

[Kenny] is the lead plaintiff in a lawsuit challenging the disrupting a school law in South Carolina. In other states, folks argue that laws against disorderly conduct are applied pretty broadly and subjectively — both in schools and on the street and that, if you have any kind of an implicit bias, that's where it will come into play.

Interesting. Can you think of other states that have laws like this?

It's a little difficult to spell out because the laws have different names. But they're pretty broadly spread throughout the country. Some states have done a better job of issuing guidance through their departments of education to help schools put boundaries around how those laws are applied. But in many states, the complaint is that, if things are too subjective and too broad, then there's not enough consistency in how they're applied.

Where do we stand now on practices being used in schools to not only minimize arrests but suspensions too?

People tend to think of arrests as a safety issue and suspensions as a discipline issue. But I think a lot of folks would tell you that this is all kind of on a continuum of how schools a.) deal with student behavior and b.) approach their climate. How do they support students and remedy their behavior so that it doesn't continue if they do something wrong?

Copyright 2017 NPR. To see more, visit http://www.npr.org/.