The three goals of restorative discipline

The three goals of restorative discipline

Amstutz’s goals of restorative discipline, presented with some commentary:

  • To understand the harm and develop empathy for both the harmed and the harmer

I believe this means that the harmer should understand the harm. I’m not sure if this is based on the idea that truly understanding the harm beyond just a surface level is a potential deterrent to future acts, or if this is simply a moral imperative to understand the results of ones actions. I’m also not sure if this is saying that the role of the school leader is to develop empathy or if it’s the role of the harmer and harmed to develop empathy (though I suppose both are good).

  • To listen and respond to the needs of the person who harmed and the person who harmed

Understanding the needs can help us to identify lagging skills and unsolved problems that may be contributing to the harmed behavior. Understanding the needs of the harmed helps us to ensure that the harmed person still feels like a valued member of the community. Some traditionalists would attack this statement as it seems to suggest that the needs of the harmer are equivalent to the needs of the harmed. “Should not the needs of the harmed come first?” they might say. But this is a false dichotomy: the needs of the harmed and the harmer are not mutually exclusive and it’s possible to satisfy both (though if we’re talking about the immediate aftermath of an event, yes, I think we’d all say the needs of the harmed are more important.)

  • To encourage accountability and responsibility though personal reflection with a collaborative planning process

This is a different accountability than the one we are ordinarily accustomed to, which is better described as retribution. Some would say that a student has not been held accountable if he has not been made to suffer for his transgression. This instinct is so deeply rooted in us that it’s hard for even the most ardent supporters of restorative discipline to shake. But there are other ways to be held accountable—by paying ones “debt” to the community, making amends for the suffering caused, successfully reintegrating oneself, and orienting ones behavior toward positive contributions to the community.


What are six-year-olds really getting suspended for?

What are six-year-olds really getting suspended for?

Chalkbeat recently posted a breakdown of the behaviors most frequently resulting in suspensions for students in grades K-2, and the results are … interesting.

Nearly a third of the 801 suspensions handed out to students in kindergarten through second grade this past school year were reportedly for incidents of violence or serious physical disruption. Those infractions included: reckless behavior with substantial risk of serious injury (115 suspensions); using force or inflicting serious injury to school safety agents or other school personnel (104 suspensions); and Category I weapons possession (22 suspensions), which includes everything from slingshots to guns.

The most common suspension is for an offense that used to be categorized as horseplay. “Altercation and/or physically aggressive behavior” is the technical name of the category, and 373 suspensions were issued for it last year, 47 percent of the 2015-16 total. Until the 2012-13 school year, the education department categorized this offense as “horseplay,” though the broadness of the label makes it hard to know how it is applied in practice.

So about 3% of the suspensions resulted from weapons possession; out of about 350,000 students in grades K-2 in NYC, I suppose 22 is a relatively low number, even if it is somewhat mind-boggling that students under the age of 7 are bringing in dangerous weapons (but perhaps this reveals my ignorance of elementary schools). I’m more curious about the relationship between a 6-year-old bringing a weapon to school and the suspension. Is the idea that the punishment is so aversive that the student will never bring the weapon again? Is it not more likely that the student perhaps was not aware of the seriousness of bringing a weapon to the school environment? When you’re planning your back-to-school routines and procedures day, I’m not sure if “don’t bring knives to school” is something that most 1st-grade teachers think to review.

Maybe the real problem is that “Altercation and/or physically aggressive behavior” is a very subjective label, as the author indicates. This seems to allow a lot of “wiggle room” for teachers or administrators looking to suspend a student to twist things. But these broad labels are created for a reason–they empower teachers and school leaders to use judgment in dealing with unique situations. That gives a power to administrators that surely troubles those who worry about the potential for racial bias; I’m sure that advocates would prefer a much more specific protocol or more clearly enumerated set of behaviors that result in suspension. I’m sympathetic to that argument, but such a policy would cut both ways–it would also hamstring the more enlightened administrators out there who are seeking to mitigate the racial disparities in the effects of school discipline by using their judgment to deal with situations individually. As I’ve written before, I’m just not sure that we’re going to legislate our way out of racial bias in school-discipline by attempting to “racist-proof” our policies.

Most suspensions come from a small number of schools. Just 263 out of the 839 district schools that serve students in kindergarten through second grade issued suspensions last year. And of those 263 schools, 40 percent (or about 105 schools) only suspended one student. That means roughly 19 percent of schools are responsible for 87 percent of all K-2 suspensions, reflecting a trend that also exists among schools that serve older students.

Sounds bad, but in a way it makes re-training efforts easier if the DOE knows which schools need the most support.

The percentage of young students who get suspended is tiny, and the number of suspensions is falling rapidly. Just 587 of the city’s youngest students were suspended this past year, or less than one-quarter of one percent of all students in those grades. The total number of suspensions issued to K-2 students is down 60 percent over the past four years, a decline that began during the Bloomberg administration. (Last year, the city required that principals get approval before suspending students in grades K-3.)

My “expert” analysis: that’s good.

Some students are suspended repeatedly. Among students who got suspended last year, 26 percent received more than one suspension. The city did not provide demographic breakdowns for the data, such as race or disability status.

The writer frames this negatively, but doesn’t this actually give the lie to one of the most common claims about suspensions–that they don’t work because they don’t change behaviors and students end up getting suspended again?

74 percent of students who were suspended once were not suspended again. On the surface, this actually makes me wonder if suspensions aren’t effective after all. But there are some missing data here–does this mean only that they weren’t suspended again in the same year? That doesn’t necessarily convince me. If I’m suspending a kid in April of 2016 and he’s getting suspended again in January of 2017, I’m not convinced the punishment “worked” just because he made it through the end of the year. Additionally, we don’t know if students’ behavior changed or if administrators who had already suspended the student realized suspension might not have worked and tried another avenue. (As i feel compelled to point out constantly [and probably annoyingly], counting suspensions tracks how schools respond to maladaptive behaviors, not the existence of the behaviors themselves. Reducing suspensions may or may not mean that student behavior changed for the better.)

The semantics of writing about race and school discipline

At this point, it’s pretty well established that under current school discipline policies, African-American students are suspended and disciplined at rates disproportionate to their percentage of the overall student population.

That is a clear and indisputable fact.

Many conclusions have been drawn from this basic fact, but some of these conclusions should warrant skepticism. For example, is it really true that “a student is three times more likely to be suspended if he or she is African-American”? This seems like a basic distortion of the data. Because not all or even most African-American students are suspended, being African-American in and of itself is not a leading cause of school suspension. It would be more accurate to say that when white students and African-American students engage in the same maladaptive behaviors, the black student is more likely to be suspended (and there does seem to be some data to support this).

What about the claim that “discipline systems unfairly target African-American students”? Also not true. While African-Americans receive overly harsh punishments for the same offenses, this is de facto and not de jure. Similarly, it’s not accurate to say that the policies themselves are racist; it would be more accurate to say that the misapplication of these policies results in racially biased decisions and harms students of certain races disproportionately. These may seem like picayune distinctions for some, but the way we talk about this will ultimately determine what is done about it.

There are actually two separate arguments that are often fused in the public discussion about this: (1) school discipline policies, which rely heavily on suspensions, result in racially biased decisions that treat black students unfairly (2) school suspensions are harmful when applied to students when less aversive interventions are possible.

Lately I have seen the arguments joined together in a way that doesn’t identify what the real problem is: if you don’t believe suspensions should be used in school discipline, then you would be in favor of reducing suspensions for all students. If you believe that the implementation of school discipline policies results in racially biased consequences, then you would be in favor of changing the way in which teachers and administrators decide who gets punished in what way for which behaviors to eliminate racial bias. Or you believe in both (as I do). But the “fused” argument: black students should be suspended less, does not follow logically from the two premises, and makes it easier for the position to be attacked in vapid and inane ways. 

Figuring out what’s working in schools with low suspension rates

Figuring out what’s working in schools with low suspension rates

Here’s an interesting report out of Denver from researchers looking to uncover what schools with low suspension rates have in common. 

Nothing here will be terribly surprising; the findings definitely provide support for the continued movement away from suspensions and toward restorative practices.

Some highlights:

Younger students, more integrated. Of the 33 schools, 58 percent were elementary schools and 58 percent were traditional district-run schools (meaning they aren’t charter or innovation schools).

The schools also had fewer children of color and fewer low-income children than other DPS schools, making them more racially and economically integrated.

An average of 61 percent of students at those 33 schools were children of color, as compared to the district average of 78 percent. An average of 56 percent were eligible for federally subsidized lunches, an indicator of poverty, as opposed to the district average of 74 percent.

The schools also had fewer English language learners and students with disabilities.

Not surprising that elementary schools have fewer suspensions, as younger students typically exhibit fewer challenging behaviors (or are more easily controlled, depending on how you want to look at it). “More integrated” is a funny way to phrase the fact that the schools with fewer suspensions were whiter and richer.

Relationships matter. Many educators interviewed attributed their school’s low suspension rates to strong relationships between teachers and students. Teachers who know their students’ strengths, challenges and triggers are more likely to understand the underlying reasons for misbehavior and be in a better position to respond, they said.

The relationship between relationships and discipline is probably not as well understood as people think, but the positive effects seem to be very consistent.

Solutions, not consequences. Many of the schools use restorative practices to address misbehavior. Students who break the rules are asked to reflect on what happened, identify the harm done and come up with a plan to repair any damage, the educators said.

While that approach takes more time than meting out a suspension, the educators said it allows students to develop conflict resolution skills and an understanding of accountability.

This is a bit circular: of course the schools who don’t use suspensions as a primary strategy are going to have fewer suspensions. I’d be more interested to find out whether the restorative approaches were successful in promoting adaptive behaviors.

Keep students in the classroom. A majority of the schools abide by the philosophy that classroom teachers should be the first to respond to misbehavior and conflict — and that students should be sent home from school only as a last resort.

“You can’t just send them home because you needed to take a break,” an elementary school principal told the researchers. “When they come back, they’re going to do the same thing.”

See above.

Be aware of racial inequity and bias. While the researchers noted that many educators were uncomfortable talking about the role of race in their discipline process, educators at about a third of the 33 schools explicitly discussed their use of culturally responsive practices.

They recognize that students of color have often been marginalized and that extra efforts should be made to connect with them and their families. By understanding the impact of racism and bias, the researchers wrote that educators at those schools “took responsibility for changing student academic and discipline outcomes, rather than blaming students and families.”


Student support services are crucial. Many educators credited their school’s social workers, counselors, psychologists, interventionists and restorative practices specialists with helping put in place positive responses to student misbehavior that helps kids instead of punishing them.

Sounds a lot like what worked in the turnaround I wrote about last week.

At this point, it’s not clear who is still supporting suspension-heavy policies. Are suspension-heavy policies happening only at schools where administrators and teachers simply don’t feel they have other options (or the other options are simply too costly in terms of the capital required to shift their approach)? Or is there legitimate support out there for suspensions? I’m not quite sure what that would look or sound like, although it’s always good to remember that zero-tolerance policies were once considered forward-thinking. I do worry just a little that those with legitimate disagreements with restorative practices are being silenced because of the racial aspect (who would be willing to come out publicly in support of policies that have been shown to be racially biased?). But not that much.

How to scale a school discipline turnaround

How to scale a school discipline turnaround

An uplifting story today from Chalkbeat about how a school got itself off of the state’s persistently dangerous list.

In my 12 years as an educator, the toughest moment came exactly a year ago, when my school was designated as one of New York City’s 27 “persistently dangerous” schools.

Many of my students and their families face overwhelming odds every day, so as their principal I was distressed by the notion that my staff and I were failing in the one place where they should always feel protected and supported. Despite the clear flaws in the state’s designation of what classifies as a “disruptive” or “violent” incident, the designation presented a critical opportunity for change within our school.

From the outset, my staff and I knew this wasn’t a challenge that could be resolved with more rules and a more punitive disciplinary code. We know our students and the challenges they might face in math or English class pale in comparison to those they encounter outside of school. Some students who struggle with stress or anxiety lack the resources to channel those emotions or to express them in a constructive way.

It’s good to see a leader recognize that creating new rules and ratcheting up punishments wasn’t the answer. Schools and school leaders seem to be coming to the consensus that consequences alone–especially suspensions–don’t improve student behavior (although we don’t seem to be able to agree on precisely why).

I’m curious if the staff that undertook this change initiative was partially or entirely new. It seems unlikely that in any given school, there would be clear consensus on the idea that “this wasn’t a challenge that could be resolved with more rules and a more punitive disciplinary code,” unless this was a new staff that was recruited for that particular philosophy or onboarded that way. The other possibility is that the principal was able to change “hearts and minds” on this issue with her staff extremely rapidly–in which case we need to know more about how she did it.

Throughout this past school year, my staff worked tirelessly to put a new support structure in place. Among other measures, we implemented a school-wide, whole-class incentive system that offers students positive reinforcement for strong behavioral or academic performance, while also allowing teachers and staff to track and communicate about students’ academic and social-emotional progress. We created student support teams comprised of a guidance counselor, social worker and school staff, each team worked with teachers, parents and students in a specific age group, to implement a social-emotional curriculum.

The consistent communication and collaboration around student needs is the cornerstone of our school’s new culture. Specifically, the social-emotional curriculum creates an opportunity for students of all ages to learn about their emotions in a safe and supportive environment, and for our staff to proactively identify and address any emotional issues. We can’t expect our students to change their impulses and behavior unless we truly teach them how, using the same patience, persistence and dedication we use to teach other subjects.

I’ve written before about the drawbacks to incentive systems; I can’t unequivocally say that they’re bad–so much depends on how they’re implemented and more specifically the messaging coming from teachers and staff around the system. But I do wonder if the bigger factor behind this school’s success wasn’t the incentive system but the commitment to actually teaching students adaptive behaviors rather than simply raising expectations or increasing punishments.

I know Chalkbeat is written for a general audience, but it would be great to get more depth on the concrete steps Keville took on this turnaround. That would be the only way to know whether this kind of effort was scaleable.

Getting caught in an ahistorical trap on school discipline

Getting caught in an ahistorical trap on school discipline

One of the biggest problems I see in education reform and education policy in general is ahistoricity: policies are often enacted with little attention to the historical context of what’s been tried before. Additionally (and more relevant to this post), when new policies are enacted, previous policies are dismissed with a level of vitriol that makes it seem as if they were cooked up by evil masterminds bent on ruining kids’ lives.

Take what’s happening in the national conversation on school discipline right now: we are seeing a mass convergence of federal, state, and local governing bodies moving away from zero-tolerance, heavy-suspension policies and toward restorative practices. While I certainly support this movement and am thankful to be working in an environment wherein current policies align with my personal beliefs and experiences, I also have to chuckle a bit at the way in which we are now discussing the zero-tolerance movement.

The Improving America’s Schools Act of 1994 (from which zero-tolerance policies sprung) was passed by an overwhelming bipartisan majority and had broad public support. People thought we were finally getting on the right track! Yes, we have now witnessed two decades of the deleterious effects of this legislation, but at the time, these policies were favored by those who claimed the moral high ground much in the way that supporters of restorative practices are doing now. (Joe Louis Clark expelled half his school and was a hero who got a movie made about him!)

Copious research now shows us why suspension-heavy policies are bad for kids. But it’s unfortunately also possible that in 20 years we’ll have an equal amount of research that shows the negative effects of restorative practices on schools. I certainly hope this is not the case, and I don’t think it will be, but it would be foolish to not at least acknowledge the possibility. You won’t hear many supporters saying this, of course, and I don’t blame them for that. They are trying to lead a movement that they genuinely believe will help kids.

But it’s also important to understand that every educational movement of all time was meant to help kids. Acting as though your movement is the first to have a true ideological purity is dangerous, and not just because it makes you seem arrogant: it sets you up to believe that your movement will succeed because it is “truly” what’s in the best interest of kids. If you believe that your side is the only one who “really cares,” you’re likely to pay less attention to the details that can sink your initiative, believing the righteousness of your intentions will win the day.

I’m sure this post reads as either cynical, nihilistic, or both, but I express these concerns because I have a nagging fear that if advocates are not careful and thoughtful, restorative discipline practices–which I believe in wholeheartedly–may wind up on the ash heap of once-promising initiatives.

What if behavioral choices aren’t choices at all?

What if behavioral choices aren’t choices at all?

Traditional models of school discipline advocate for students to be punished as a result of challenging behavior based on the idea that students need to learn that there are consequences for making bad choices (or some such variant of this statement).

In fact, the idea of challenging behavior as a “choice” seems to be so deeply ingrained in the way our society thinks about discipline that it is rarely if ever questioned. To even imply that a student might not be consciously “choosing” to exhibit challenging behavior opens one up for criticism for being “soft,” “permissive,” or “easily manipulated.”

And yet, when we take the time to get to know kids who exhibit challenging behavior, we come to understand that the situation is significantly more complicated. Although we are not yet willing to say that the way in which a student behaves is completely out of his control, we also simply cannot accept that a student would rationally consider a range of adaptive and maladaptive behaviors and then rationally choose the latter. Ross Greene in Lost at School explains this with his maxim that “Kids do well when they can rather than when they want” (I wrote more about this here). In his latest book Helping Children Succeed, Paul Tough tackles the issue from a neurobiological standpoint:

Our usual intuition when children and adolescents misbehave is to assume that they’re doing so because they have rationally considered the consequences of their actions and calculated that the benefits of misbehavior outweigh the costs. And so our response is usually to try and increase the cost of misbehavior by ratcheting up the punishment they receive. But this only makes sense if a child’s poor behavior is the product of a rational cost-benefit analysis. And, in fact, one of the chief insights that the neurobiological research provides is that the behavior of young people, especially young people who have experienced significant adversity, is often under the sway of emotional and psychological and hormonal forces within them that are far from rational.

This doesn’t mean of course that teachers should excuse or ignore bad behavior in the classroom. But it does explain why harsh punishments so often prove ineffective over the long term in motivating troubled young people to succeed. And it suggests that the school-discipline programs might be more effective if they were to focus less on imposing punishment and more on creating a classroom environment in which students who lack self-regulatory capacities can find the tools and context they need to develop them.

Tough’s argument against harsh punishments like out-of-school suspensions is a variant on the idea that harsh punishments are not an effective deterrent to maladaptive behaviors because the period of “active consideration” of possible consequences that we imagine students to be entering into prior to misbehavior simply doesn’t exist.

Tough doesn’t discuss neural pathways, but to me, the presence or absence of established neural pathways seems to explain this situation well. When we engage in a certain behavior or thought pattern, we create and then expand the neural pathways associated with that. For example, a student might have trouble in kindergarten learning to raise his hand, wait to be called on, offer a response, and engage in respectful discourse. But through repeated practice and feedback, the “hand-raising” pathway becomes well-worn and easily traversed. Hand-raising becomes seemingly instinctual, and no thought or consideration is given to the act. Conversely, the idea of calling out is almost unfathomable simply because it hasn’t been done in so long. The stimulus behind calling out would have to be so strong as to effectively “blast” a new pathway (think of what it would take to divert a flowing river). Not likely to happen, although if it did begin to happen, it would become more likely the more it happened.

Now consider the opposite–a student who has created a deep and wide “calling out” pathway in his brain. Consider how hard the student would have to work to create a new pathway for the hand-raising-and-respectful-discourse habit. It’s extremely difficult–but certainly not impossible. Is this student “choosing” to call out?

In a sense, yes. But the odds aren’t even–the presence of the neural pathway for “calling out” is so well-traversed that the cognitive energy required to go against it is many times what it would take to go with it. (In some cases, the adaptive behavior has never been done or attempted, so there is literally no pathway for it rather than just a weak one.)

We don’t get upset when we encounter a river whose water is flowing through a certain course; we understand that if we wish to divert some or all of this water, we have to work to create a new path.

I realize that this seems to be a particularly behaviorist way of thinking about behavior on a blog that typically eschews behaviorist tactics. That’s something I need to reckon with, because if I’m right about the way I’m describing the relationship between neural pathways and child behavior, then there is simply no moral component to behavior at all; if we’re going to say that bad behavior isn’t a result of rational choices, then surely neither is “good” behavior: the behaviors we “do” are simply those that we’ve learned how to do and done before. I need to think about this some more.