As the largest school improvement effort in U.S. history matured, so did the energetic transplants. Wanting kids of their own and more sustainable jobs, many moved back home, bringing their experiences to bear in classrooms in places where life is easier.

Patrick Dobard has been there the whole time. He was born in New Orleans, grew up in the city and cut his teeth there as a teacher. In the years before Hurricane Katrina, he worked for the Louisiana Department of Education, trying to figure out how to address the decay of New Orleans’ schools — beloved but crumbling, scandal-ridden and some of the lowest-performing in the country. 

In 2012, Dobard became head of the Recovery School District, the state agency that took control of most of the city’s public schools in the wake of the flood and steered their overhaul. There, he oversaw the district’s conversion to the nation’s first all-charter school system, as well as the return of the schools to the control of the local school board starting in 2016.

In 2017, he became CEO of New Schools for New Orleans, the district’s nonprofit partner. Today, he is a partner at the City Fund, which helps districts engineer their own school turnarounds. 

What follows is a conversation in which Dobard reflects on the first two decades of the largest — and most controversial — school improvement effort in U.S. history, and outlines his hopes for the next 20 years. 

This interview has been edited for length and clarity.

Among education wonks, discussions about public education in New Orleans can feel like a Rorschach Test. Some people are laser-focused on academics, some on policy innovations that might transfer to other school systems and still others on privatization conspiracy theories. Locally, though, one of the most enduring conversations involves popular perceptions of a takeover of a Black-led school system by white outsiders who valued test scores more than the city’s culture. This is true — but also not the capital-T truth. Can we start there?             

I started as a classroom teacher in New Orleans in 1989 at Gregory High School, which was a junior high school, grades 7 through 9, located in what at the time was a pretty middle- to upper-middle-class neighborhood, Gentilly. Most high schools then started at 10th grade. Junior highs were extremely important as athletic and band pipelines. 

The workforce for the most part at Gregory represented how the workforce across the city’s schools looked at that time: predominantly Black middle- to upper-middle-class individuals teaching kids. 

The schools that had really stringent enrollment criteria, like what was then named Lusher [now the Willow School] and Ben Franklin High School, more white students went to those schools. They had majority white staff.

By 1999, the state had created an accountability system under which a large number of the schools in the city — I believe about 60% — were identified as academically unacceptable. At that point, I was working at the state Department of Education. 

There was a lot of corruption in the air in New Orleans. There were conversations about the district being bankrupt. There was a Federal Bureau of Investigations [probe] going on. 

In 2003, the state created an agency called the Recovery School District. In the spring of 2005, the initial plan was for it to take over about three schools in New Orleans. We didn’t really know what that was going to look like. But then Katrina hit Aug. 29, 2005, and all those plans were put on hold. 

We were getting phone calls weeks and months after Katrina from teachers trying to get a hold of their teaching certificates so they could work other places. There was just mass displacement of teachers.

And that’s where I think people’s knowledge of what happened gets told in different ways. The district was bankrupt. There wasn’t a way for it to pay teachers, so it was forced to lay them off. The state didn’t step in to try to offset that. So it was the district that had to fire the teachers and not the state. 

People were setting up the modular trailers and all the things people did after Katrina to get their lives back on track. There were schools — mostly the selective-enrollment schools that I mentioned earlier — where people were able to come back a little bit more quickly. Those schools were bringing teachers back and hiring.

The Recovery School District was trying to recruit back New Orleans teachers, but a lot of the Black middle class didn’t have the ability or the wherewithal to come back to New Orleans those first few years. 

Once the state decided that they were going to try the chartering model, the RSD started to recruit from Teach For America. A large number of TFA folks from all around the country wanted to come help. A number of those individuals were white. They didn’t look like the kids that ultimately were in front of them in schools.

I have no inside knowledge of this, but I don’t think TFA leadership at that time had lots of conversations strategically thinking about race. They were just functioning the way they normally functioned. 

Data shows having kids in front of teachers that are really strong and that look like them helps make for much better educational progress for young people. Once John White came in [as state superintendent in 2012], we were hearing a lot about the lack of diversity in the teaching force. 

John and I talked to TFA leadership about diversifying. While the number of Blacks and other minorities in TFA increased, it didn’t match the number of teachers of color that were there before the storm. 

Seven or eight years ago, I brought that awareness to the teacher work that we were doing at New Schools for New Orleans [where Dobard was then CEO]. We wanted to help schools build a corps of teachers that reflected the kids in the classrooms. Schools just took that on. They owned up to where they fell short, and then they actively recruited to make sure those numbers improved. 

I don’t think anyone had intentional ill will. It was a series of unfortunate circumstances that folks were reacting to after Katrina: the bankruptcy of the district and the uncertainty of people coming back to the city at different times, which was skewed to more white and affluent Blacks coming more quickly than others. 

And yet the narrative persists. 

Yeah, the narrative persists. I think what’s been missing is no one from the Orleans Parish School Board who was part of the decision to lay off the teachers to my knowledge has ever publicly acknowledged the firings. Or apologized for having to do that. Or for what transpired in the years before 2005, with the FBI having to be there and how bad it was. 

There’s no closure. There should be a moment of healing. The times that the district leadership over the years was approached, nobody really wanted to acknowledge it that way. Some felt like it was on the state for not stepping in. So maybe the state leadership at that time should apologize.

It feels like in terms of the school governance experiment, you’re at an inflection point. There’s continued improvement, but there are also tensions over whether an accountability-driven system is still something the community is lined up behind.

I think the governance contract is working well. Around 2016, I was leading the work on behalf of the Recovery School District. School leaders and advocates like the Urban League, New Schools for New Orleans and a number of other groups were working together on what the unification structure would look like. A number of legislators at the time helped to codify in law the governance construct that was created over the years as Act 91. 

It’s no longer an experiment. It’s really how schools function in New Orleans. The district has fully embraced it, maintained it and actually improved upon it. It’s a different role than any other school board in the country, but it’s also an extremely important one that’s proven to have spurred tremendous gains.

Yes, there are some people who would like to see the district go back to the way it functioned prior to Katrina. But I don’t think those individuals are but a small minority who are for the most part consistent and persistent in their viewpoint. 

When things are working well in New Orleans, you don’t hear from people. Nobody’s going to say, “Hey, the governance structure of this school that my kids have been at for the last 15 years and my grandchildren have gone through is great.” When I was state superintendent, I used to ask kids and parents, “Do you know if this school is run by the Recovery School District or by Orleans Parish?” And they’d just look at me, like, “No.” 

The average citizen, what they want to know is, is my child getting a good education? Do we have good extracurricular activities? Does the transportation work well? 

I’ll tell you one quick story. Dana Peterson, who now is the CEO of New Schools for New Orleans, was involved in some sessions with parents. He had this parent who was just railing against charter schools. Like, “We need to go back to what was before.” 

Dana was like, “Do you realize your kid’s in a charter school?” The parent was at a loss. She was happy with where her child was, but she was indoctrinated that charter schools are bad. I think for some of the critics, that’s what it is. 

If you don’t really know what the governance construct is, at the end of the day it’s all about who’s working with the kids. There’s more proximity when it’s a charter network, with the urge to improve all the time so they can continue to have the privilege to teach kids. Versus a bureaucracy that’s expected to do it but that doesn’t have strong accountability to make sure it’s not year-after-year failure.

And that’s what it was before 2005. 

Reams have been written about the rapid academic improvements, the all-charter model and, more recently, the racial upheaval. What about the fact that you had to rebuild — as in, rebuild the buildings — 85% of the schools?   

Prior to Katrina, the buildings said to our kids and families, “We don’t care about you.” Those dilapidated buildings are no longer there. That’s something worth celebrating. 

Being a young boy that grew up in New Orleans, I’m extremely proud of the facilities. The price tag came to a little over $4 billion, if I’m remembering correctly. About a billion came from the disadvantaged business enterprise program that we started at the Recovery School District.

My first months as superintendent, there was an article in The Times-Picayune where they followed me as I rode the bus the first day of school with some kids to see what their experience was. I met the kids across the street from Dooky Chase’s Restaurant on Orleans Avenue. 

I was talking to the kids and a mom, and I pointed to Dooky Chase and right behind Dooky Chase there was a school. It was [what is now] Phyllis Wheatley Community School. There was a debate whether or not to rebuild it. 

I said, “We’re going to rebuild this school right here. And if we did, would you have your kid come here?” She’s like, “Absolutely I would.” We demolished the building and built new.

I’m extremely proud of that because my father was a part-time electrician and I could remember when I was a young boy, him getting work with one of his friends who was a subcontractor on large jobs. As I got older, I understood how important it was to be able to have subcontractors that were often minorities to work on large construction projects.

When I took over the RSD, New Orleans didn’t have a disadvantaged business program. And once I realized what that was, I felt it was important that we try to implement one. I was told that state agencies couldn’t have a DBE component. 

But we had legal take a look into it, and they advised me that the law was silent. It didn’t say whether you could or you couldn’t. I felt like it was important for us to at least try. And if we were challenged in court, we would see what a judge would say.

Once we launched it, I received maybe one or two emails from, like, the carpenters union saying we couldn’t do that. But they never followed up. We generated over $10 billion of revenue for local businesses and I’m extremely proud of that.

To this day, every now and then if I look at the placards on the front of a school I see my name and those of the folks on my team that helped make that happen. It brings me a great sense of pride and joy to know we played a small part in making sure that we have facilities where, when children and families walk in, they say to themselves, “This district really cares about me.” 

Do you have a wish for the next 20 years? 

I wish that we continue to build upon the foundation that’s been laid. I would love for us to eliminate all D-rated schools, to have a true system of good-to-excellent schools. 

That we would have a much more robust early childhood system where 5-year-olds and 6-year-olds are entering kindergarten and first grade on grade level. Today, we still have kids that enter all grades not reading on grade level, so schools have to keep almost starting over with kids. 

The last thing that I would hope for is that the school board in New Orleans more vocally embrace the structure that’s been created. To be unambiguous about its power in being a manager of a system of schools, versus being a traditional school district. That they fully understand and embrace that role and lead the way on that evolution. 

If you had a do-over, what would you change?

I would have intentionally built the next generation of leadership to think more about the system as a whole and to prepare for inevitable transitions — everything that we did 15 or so years ago. We had this rare confluence of strong leadership at almost every level: the schools, the state.

We leveraged everything in our power. For every metric evaluated by external entities, the growth was really powerful. We virtually eliminated F-rated schools. The on-time graduation rate is hovering close to about 80%. It was about 54% around 2005.

I wish we would have started to build that next cadre of leadership in real time. But it’s hard to be in the midst of something so unique and try to think 15, 20 years ahead. I was literally trying to think eight hours, one week, one month ahead. Things were just always coming at us, and we were constantly building and adjusting. 

We get these great leaders, and they do great work — almost like a meteorite or something that comes and then goes away just as quickly. People move on. 

The work is so hard. This is generational work.

Disclosure: The City Fund provides financial support to ����Ӱ��.

One is the overwhelming flood of unilateral, executive action. This has included the haphazard dismantling of the federal Department of Education, on-and-off-again suspensions of various education-related spending — from research grants and leftover COVID dollars to afterschool and summer programs — and aggressive use of civil rights laws to target diversity, equity and inclusion programs and transgender athletes.

The other is the persistence of pandemic learning loss and the absence of political interest or will to do anything about it. The from the National Assessment of Educational Progress, released less than two weeks after Inauguration Day, showed only modest improvements in math scores and continued declines in reading achievement, to the lowest rates recorded on this series of exams. Similar trends have been confirmed by vendors.

What if these two developments are related?

This suggestion is at the core of the argument I make in my new book, . The same electoral processes that gave the nation President Donald Trump — and the same democratic justifications his administration now offers to defend his norm-shattering actions — represent some of the biggest impediments to solving today’s most pressing educational challenges.

First, consider the intellectual and philosophical similarities. Trump’s aggressive approach to presidential power is based on the , a controversial scholarly thesis that escaped from the legal academy into the real world. It views a strong presidency as a democratic antidote to unaccountable, deep-state bureaucrats and “elites,” and condemns any effort to block or undermine executive authority by unelected officials as inherently illegitimate.

The downplaying or of learning loss also has roots in once-fringe academic ideas — accounts that dismiss standardized testing as outgrowth of toxic “neoliberalism.” These scholarly treatises feature their own mustache-twirling villains — corporate reformers and billionaire philanthropists, a different kind of elite — and a restoration of truly democratic control of public education.

When federal courts stepped in to block some of the Trump administration’s executive actions, the White House “unelected judges” and Vice President J.D. Vance that their intervention represented “an effort by the courts to quite literally overturn the will of the American people.” 

Defenders of chronically underperforming schools have lodged nearly identical complaints against efforts to fix them.

After Louisiana officials took over the undeniably corrupt and woefully run New Orleans school district in the wake of Hurricane Katrina and turned almost all of it over to charter schools, academic achievement and attainment . For years, however, critics attacked the new system as “.”

More recently, Houston’s state-appointed superintendent has produced turnaround less than two years after sidelining the district’s elected school board. As one local journalist : “The news from the Houston Independent School District at the end of the 2025 school year is wonderful for Houston, wonderful for Texas and wonderful for the nation.”

Rather than celebrate these improvements, however, opponents continue to demand “” to the “undemocratic seizure of Houston’s public schools” and warn that “these kinds of undemocratic seizures of power aren’t just a Houston problem — they’re a threat to every public school.”

Last November, the Houston teachers union helped persuade city voters to the school district’s $4.4 billion bond proposal — not because union leaders disagreed about the need to upgrade aging school buildings, but because that any investments must be delayed until “we have steady, trusted and accountable elected leaders running the district once more.”

Fortunately, many Americans appear to reject the Trump administration position that presidential power is unlimited and any intervention to rein it in illegitimate — suggests that public approval of his performance is underwater almost across the board. Voters seem to understand that there is much more to democracy than mere elections and that electoral control can be a double-edged sword. That unelected judges with lifetime appointments (and ) play an essential role, too, in providing restraint against the popular passions of the moment. And that groups without the right to vote, whether non-citizen immigrants or underage children, require special protections, because their interests are especially likely to be overlooked in the political process.

Similar dynamics apply at the local level. They help explain why much of the dysfunction that plagues schools is hard-wired into a governance model that puts the short-term political agendas of adults ahead of the educational interests of students that local schools serve. The argument that responsiveness to voters is the only outcome that matters and any intervention limiting the authority of elected officials is undemocratic did not begin during the Trump administration — and similar claims have long been used to insulate chronically underperforming school systems from meaningful accountability and reform.

In 1997, political commentator Fareed Zakaria published a article in Foreign Affairs warning about the rise of what he called “illiberal democracy.” Too many, he warned, made the mistake of equating “democracy” with elections. But democracy is much more than that, he argued. It is “a political system marked not only by free and fair elections, but also by the rule of law, a separation of powers and the protection of basic liberties of speech, assembly, religion and property.”

At the time, he was focused mostly on newly democratizing countries in the developing world, but the distinction seems equally relevant in America today.

Zakaria’s insight is also relevant for understanding the problem with education governance. If “illiberal democracy” is characterized by fetishizing electoral procedures at the expense of substantive performance, and if it prioritizes popular participation and (adult) majority rule over minority (and student) interests, public education has long had an illiberal democracy problem. And America’s schoolchildren have paid the price.

Local school systems straining to comply with the state’s sweeping Blueprint for Maryland’s Future have had to report to both the Maryland Department of Education and the Blueprint’s Accountability and Implementation Board, a setup creating confusion “since the get-go.”

Now, more than three years into the process, the two agencies said they are working on a memorandum of understanding that could make things a bit smoother for all concerned.

Alex Reese, chief of staff with the Maryland State Department of Education (MSDE), did not tell the state Board of Education on Thursday how long it would take to finalize an agreement, but he said a memorandum is in the works.

Get stories like this delivered straight to your inbox. Sign up for ����Ӱ�� Newsletter

State law requires the seven-member AIB to oversee the 10-year plan and approve any Blueprint documents submitted by the state’s 24 school systems and other state agencies that craft elements of the Blueprint.

The law also requires the department to provide technical assistance and lend expertise on education policy. The AIB and state Board of Education also hold and approve certain policies associated with the multibillion-dollar Blueprint plan.

Reese said “MSDE will be fully owning Blueprint implementation. We feel good about that as practitioners. We really do feel like we possess that expertise to be able to be poised to fully implement the Blueprint.”

An AIB spokesperson confirmed in an email Thursday evening an agreement is being worked on with the department.

“AIB and MSDE attorneys are working together on an MOU [memorandum of understanding] relating to the agencies’ respective roles and duties,” the spokesperson said.

“There is not currently a timeline confirmed for finalizing it. Because it is an MOU directly between the AIB and MSDE, there would be no need for General Assembly approval,” the email said.

In a quick summation to the state board Thursday, Reese said certain processes will remain the same such as the Blueprint board providing instructions to school systems on what is required in each Blueprint plan. It will continue “interagency collaboration” with agencies such as the state , which focuses on two of the Blueprint’s five pillars, or priorities – hiring and retaining high-quality and diverse teachers, and preparing students for college and technical careers.

The news was welcomed by school leaders, educators and advocates who have expressed frustration over the process of implementing the comprehensive education reform plan.

“One of the biggest complaints, if not the biggest, has been the lack of clarity and final guidance and where we get questions answered. We’ve got to run every decision by both entities [MSDE and AIB],” said Mary Pat Fannon, executive director of the Public School Superintendents’ Association of Maryland.

The association released a 12-page document in that outlined proposals to help improve the plan. One of those recommendations was clearing up the relationship between the two agencies.

“Restructuring and clarifying the relationship of the MSDE and AIB would be very beneficial in the implementation of the Blueprint. This change would clarify roles and responsibilities, and establish clear guidance to the LEAs [local education agencies, or school systems] that they are governed by the procedures and processes promulgated by the MSDE and the State Board,” the December report said.

“Somebody’s got to be the point. Somebody’s got to be the team captain on certain things,” Fannon said.  “Otherwise, it’s just completely frustrating.”

“We are happy they are doing this. This is all going to help in implementation when these guys are 100% clear with us,” Fannon said of the work on an MOU.

Sen. Mary Beth Carozza (R-Lower Shore) was also pleased by the discussions, which she said would help improve the process at the local and state levels. But the senator hopes an agreement can be reached before the 2025-26 school year begins in the fall.

“I would like to think they would make every effort to use the time between now and [when] school starts to give as much clarity to the roles and responsibilities, since it will only have a positive impact at the local level,” Carozza said. “That would be my expectation to keep that on track and to keep it moving.”

is part of States Newsroom, a nonprofit news network supported by grants and a coalition of donors as a 501c(3) public charity. Maryland Matters maintains editorial independence. Contact Editor Steve Crane for questions: editor@marylandmatters.org.

is at a historic low. and the rapid expansion of private school choice are causing widespread declines in enrollment and . The need to improve and expand educational opportunities is more urgent than ever.

A proven transformation strategy exists. Denver Public Schools’s pre-pandemic approach to educational improvement student achievement. The system was facing many of the districts across the country are now struggling with: financial instability, low performance and declining enrollment. It required rethinking and redesigning the very role of a school district. And it worked. 

The latest from the University of Colorado Denver’s Center for Education Policy Analysis provides clear, empirical evidence that Denver’s transformation of its public schools between 2008 and 2019 caused large gains in student achievement citywide, including for children the district has historically failed to serve. The study shows that significant, sustained, systemwide improvements in learning are possible, including in large school systems with high levels of student need. 

When Denver launched its reforms in 2007, it was in the bottom 5th percentile of all districts in the state. Its four-year graduation rate was 39%. After a decade of reform, the district rose to above the 60th percentile and raised its graduation rate above 70%.

The new study found that students overall who experienced two years of reform received the equivalent of six months to two years of additional schooling. Over five years, those benefits jumped to between 18 and three years of additional learning. Black and Latino students received at least an additional two years, and English learners got an average of an extra 18 months in math. 

By evaluating the improvements of students who started school in the district before the reforms began and continued to attend while they were in place, we demonstrate that these improvements were not due to changes in demographics, but are the direct results of the reforms. Their annual effects grew as they were implemented over time and more widely. Each year of reform produced larger effects than the year before. The longer students were enrolled in the district during the reforms, the more they benefited. 

These effects on student learning were significant and to those resulting from post-Hurricane Katrina school improvements in New Orleans.

So how did the district do it?  

Denver fundamentally altered the role and structure of its local school system, shifting away from a rigid, centralized institutional approach toward a more flexible and responsive model based on adaptation, differentiation and continuous improvement. 

Under the of now-Sen. Michael Bennet and former Superintendent Tom Boasberg, the district confronted decades of low performance with a shift to a portfolio model to encourage choice for families, empowerment for educators and accountability for performance.

Every year from 2008 until 2019, the district evaluated all schools, issued public requests for proposals for new schools from internal and external providers, and implemented a process for intervening in persistently low-performing schools through closures, replacements and district-led school turnarounds.

The reforms implemented in Denver mark that an elected school board voluntarily relinquished the exclusive power to operate schools within its boundaries while maintaining its authority to govern all schools in the district. In doing so, the district rejected the traditional model of singularity in favor of one built for multiplicity.

Denver’s portfolio district strategy included:

• A systemwide focus on ensuring for all students through a common enrollment system while allowing for flexibility and innovation in how education is delivered;
  
• A framework for public reporting and accountability with a common set of performance metrics for all schools;
  
• A consistent enrollment and expulsion system;
  
• A strategic focus on attracting and developing effective teachers and school leaders by on new and and holding them accountable for .
  
• A flexible funding model that allocates dollars based on student need;
  
• An annual process for evaluating schools, intervening in persistently low performers with internal and external partners, closing schools when necessary and replacing them with new ones;
  
• Support for , between traditional public schools and public charter schools and decentralized authority with shared responsibility between school and district leaders for the success of all students.

Perhaps the most important lesson from Denver’s transformation is about the possibility of improvement and the need for leadership. The district still faces many challenges, but public education there is dramatically better in terms of measurable student learning because of its decade of reform. 

Today, as school systems nationwide grapple with challenges similar to those of the Mile High City in the early 2000s, Denver’s experience offers a proven alternative to the traditional model. Denver’s approach remains one of the most controversial, consequential, comprehensive and longest-lasting school reform initiatives in U.S. history. Its success illustrates that with vision, commitment and a willingness to innovate, districts can create dynamic educational opportunities that enable all children to learn more and do better academically.

Transformative change does not come easily, but Denver’s example shows that dramatic improvement is possible.

Historically, the majority of this funding was only available to students who were low income, had special needs or attended poorly performing public schools. 

Over the past three years, that’s shifted: Today, at least 33 states offer private school choice programs, and of those 12 are “universal,” meaning any student, regardless of income or need, can apply for government funding to subsidize private, religious and — in some cases — home schools. 

Comprehensive analysis of the scale of these initiatives and their implications — both for students and state budgets — has been sparse. But a released earlier this month by , a research think tank based at Georgetown’s School of Public Policy, looks to change that. 

Policy Director Liz Cohen and analyst Bella DiMarco studied the evolution of established or emerging universal programs during the 2023-24 school year across eight states: Arizona, Arkansas, Florida, Iowa, Indiana, Ohio, Oklahoma, and West Virginia. 

Their research comes on the eve of an election where school choice measures are on the ballot in three additional states and when disagreement continues to spark over whether these programs give freedom and choice to families who have been historically locked out of private schooling or are part of a larger movement meant to undermine and defund public schools. 

FutureEd’s major finding about how universal choice has played out so far? “Policy design really matters,” Cohen said, in an interview with ����Ӱ��.

While all of the studied programs are universal in that anyone can apply, whether families end up actually receiving money, how much they receive and what accountability measures the participating schools are held to varies greatly state by state. 

They calculated that in total, 569,000 students received subsidies across these states, representing 55% of the students attending private schools with public funding and costing taxpayers an estimated $4 billion. About 40% of the nation’s 50 million elementary and secondary students are now eligible.

Here are five key takeaways.

“Universal” is not necessarily universal, and no two states’ policies look the same. 

“We talk about [universal programs] as such a monolithic thing,” said DiMarco. “I expected there to be more similarities between the programs and to see more similarities in the data. But that just wasn’t necessarily the case.” 

In Ohio for example, families receive funding on a sliding scale based on need, private schools can’t charge low-income families more than what they receive from the state and participating private schools must use the same graduation requirements.

On the other end of the spectrum, in Florida and Arizona no student who applies for funding is turned down and participating private schools don’t need to be accredited. 

“If you listen to the sort of politically charged descriptions of these initiatives you get one fairly stilted perspective— both from proponents and opponents of these,” said Nat Malkus, the deputy director of education policy at the right-leaning American Enterprise Institute. “And when you look at them piece by piece, there’s a good bit of daylight between the arrangements from one city to the next.”

But there are a few overarching themes, some of which shouldn’t come as a surprise.

All states give participating families similar amounts of money, with the average award amount coming in at around $7,000, which is approximately 90% to 100% of state per-pupil funding. 

Most states require some sort of accountability testing — but not all. And most of the students who received the funding across all eight states were already attending private schools.

For example in Arkansas, 64% of students who received funds through the Education Freedom Act in its first year, the 2023-24 school year, were already enrolled in private schools. The majority were students with disabilities. 

“So much of the attention in general has been paid to the fact that the majority of kids are already in private school,” said Cohen. “But that’s actually the expected outcome if you are giving money to kids to go to private school, and anyone can get it.”

She said the bigger question moving forward is examining if that pattern will persist beyond the first wave of funding.

Josh Cowen, education policy expert and author of said he doesn’t anticipate the demographics of participating students to shift much over time, meaning he isn’t expecting an exodus of low-income students from struggling public schools to private school alternatives..

“Put me down for projecting that the next version of this [report] is going to find something very similar and even more stark… [because] no policy that isn’t directly targeted toward at-risk children or families, will remain primarily benefiting at-risk children or families.“

The income level of participating families is murkier than people think: Well-to-do families are signing up, but so are more modest ones.

While these programs continue to serve predominantly lower- and middle-income families, the researchers found that participation among higher-income families increased last year, in every state where eligibility expanded and data was available.

“One of the big sort of headlines you keep seeing around these programs is that it’s all affluent families,” said Cohen. “And I just think the nuance to that is that that’s not actually accurate.”

While it’s true that there are many more affluent families than in previous means-tested programs, there are still significant numbers of lower-income families who are entering these programs. She pointed to Florida where 30% of families participating are low income. 

DiMarco said they saw a lot of middle-income families taking advantage of the funds who were “sort of just above the line” under previous, means-tested programs.

Impacts of funding on state budgets remain unclear.

Because the majority of families who took advantage of this funding were not coming from public schools — and therefore not bringing their per pupil public funding with them — these subsidies represent a new state-level cost.

“They’re new expenses,” said Cohen, “which could ultimately down the road — if state lawmakers don’t really think this through — end up [putting] states in a position where they have to say, ‘We’re not going to build this highway … because we have to pay the bill on this private school choice thing.’”

Goals of the programs are rarely — if ever — clearly stated, making accountability tricky. 

Some states, like Arizona and Oklahoma, have no standardized testing requirements or other performance metrics, making it, “nearly impossible to gauge how much learning is taking place under the state’s private school choice programs,” according to the report.

Other states do have more stringent requirements, although Florida is the only state the researchers studied which has mandated funding to evaluate academic performance of participating students.

“The step it feels like a lot of these states skipped is identifying a clear goal for the program and then a clear metric of how you’ll know if you achieved your goal,” said Cohen. “And without stating those things up front, what are we even trying to measure?”

Malkus sees more of an effort to track student outcomes, though he emphasized additional data would help parents make better-informed choices. 

“I don’t think the testing requirements are as strict as some people would like them,” he said, “but the idea that there’s zero accountability for these isn’t true either. It’s somewhere in the messy middle.”

Things are different in the Houston Independent School District. One year into the state intervention in the district, our results show meaningful growth for students and schools because we’ve embraced wholesale systemic transformation.  

This progress is the result of some innovative practices, but also strategies that education leaders throughout the country already know. The biggest challenge has never been knowing which policies work, but rather summoning the will to do what is right for children, even when the politics are treacherous.

Twenty years ago, leaders willing to transform school systems and the lives of their students boldly dotted the national map. But as the politics got tougher, too many educators and elected officials learned the unfortunate lesson that doing what is right for kids doesn’t always lead to reelection or a contract extension. Today, most leaders don’t dare utter the words “transformation” or “accountability,” and school systems take a piecemeal approach or abandon any effort to improve instructional quality because it is easier on adults. Never mind that honest-to-God transformation of our school systems is exactly what students need. 

In Houston, we’ve embraced transformational change with our New Education System model.  What sets it apart from many previous reforms is its wholesale systemic approach. Unlike piecemeal reforms that often falter due to a lack of coherence and sustained focus, NES aims to build a new system from the ground up, ensuring that all components work in unison toward the same goals.

In particular, it changes how students are taught and how we approach training and professional development for educators. The model combines instruction with real-time feedback through mini-assessments, ensuring that students who are behind get the extra help they need. At the same time, those who are ahead are continually challenged with multi-disciplinary projects that focus on writing over multiple-choice questions. This allows students to demonstrate critical and creative thinking. 

Effective teaching is at the core. The district has invested heavily in training principals to be instructional leaders. They conduct spot observations to better understand how to support teachers and then coach them to help improve their instruction. A new, rigorous evaluation system for principals gauges how well they achieve these goals. 

We have also established leadership academies to train and develop aspiring teachers and principals. This is essential as we work to tackle a shortage of educators. The district hopes to have at least 70 graduates from the Principal Leadership Academy by May 2025.

In addition, NES provides instruction that helps equip young people with the knowledge and skills they will need in a rapidly evolving job market and world. Amid advancements in artificial intelligence and shifts in the economy, we are expanding our offerings of electives with a particular focus on AI. One semester-long course for 11th and 12th graders provides an introduction to generative AI and how to use it safely and ethically. In addition, the district has launched a FutureReady Cohort that trains 200 school staffers on how to use the technology and begin integrating it into lesson plans. For instance, a teacher might explain a couple of methods to solve a math problem to a high school class, then use generative AI to present additional approaches to prompt a discussion.

We also want to impart and foster critical thinking skills, as these are and will continue to be essential for millions of jobs. The district has introduced a curriculum known as the Art of Thinking at all NES schools. It includes lessons on how to combine facts with logic, understand bias and correlation, and devise questions that could be asked in specific scenarios. 

After just one year of implementing NES at an initial set of 85 schools, the district has seen significant . In grades 4 and 6, students at these schools improved their reading scores by 8 and 10 percentage points, respectively, while high schoolers jumped 10 points in algebra and 7 in English.

Coalition for Advancing Student Excellence Houston called these gains .

Because our professional development and focus on better instruction extends to all schools, our districtwide transformation efforts have had effects even outside the NES model. Across the district, preliminary state accountability data shows that the number of A- and B-rated schools increased by 82%, from 93 in 2023 to 170 in 2024. Meanwhile, the number of NES schools achieving an A or B rating skyrocketed 480%, from 11 in 2023 to 53 in 2024, and the number earning a D or F declined from 121 to only 41 during that time period.

None of this has been easy. Even when everyone in a community knows things must improve, systemic change can trigger fierce pushback. But no matter how hard it is, students deserve a quality public education. It is up to superintendents, school board members and elected leaders who have a say over education to deliver it. District leaders can’t afford to abandon large-scale reform to the past. In Houston, we are seeing incredible results. We hope other districts will find inspiration to pick up the tools of transformation and join us.

If I told you there was one education reform that had the potential to address all these problems at once, you might think I was crazy. But shifting away from the one-classroom, one-teacher model in favor of a team-based approach, with different roles and responsibilities for various team members, has all these benefits and more. 

How can schools realize this potential? To find out, I spoke with leaders of three team-based teaching models — Kristan Van Hook from the (TAP), Bryan Hassel from and Brent Maddin from Arizona State University’s . Collectively, they have helped hundreds of schools transition away from one-classroom, one-teacher staffing plans.

Under the traditional approach, teachers are solely and fully responsible for what goes on in their classroom. As an example, if a school has 100 fifth graders, they are divided into four classes of 25 kids, each with its own teacher. But as Maddin points out, from a student’s perspective, this division inevitably creates something of a lottery. If one of those teachers is a beginner and one of them is a highly regarded veteran, well, take a number.  

This model can also be isolating from the teacher’s perspective. There’s no time to huddle with colleagues, and most schools have only instructional coaches for the entire staff. Districts do employ nearly a million paraprofessionals to assist teachers and smooth over the cracks somewhat, but the system still puts one teacher in charge of one class of kids, and the job doesn’t change much from year to year. Regardless of whether they’re a rookie or a veteran, the teacher will be in charge of the same number of kids. If they want to earn more money, they need to earn a master’s degree or step out of the classroom and into a leadership role. 

A team-based approach is different. Instead of each teacher being responsible for one class of kids for the full day, teams composed of teachers, paraprofessionals and instructional coaches share responsibilities for a larger group of students. 

Depending on the school, these groups are typically led by a master teacher who receives extra compensation for leading mini-teams of three to eight people. (In their original for Opportunity Culture, Emily Ayscue Hassel and Bryan Hassel called this “extending the reach” of the best teachers.) Hassel says the average teacher leader in Opportunity Culture schools receives an extra 20% in pay.

This approach has benefits for students and teachers. A student might spend some time receiving direct instruction from a master teacher, then work in small groups, or practice a skill by themselves. 

From a teacher’s perspective, the teams replace isolation with collaboration and make the job more sustainable. Team members work together to identify student needs and plan instruction. This approach helps beginners transition into the classroom and gives team members time to compare notes, which can also help if the district is pursuing some large initiative such as adopting . 

Most importantly, the team-based staffing models can boost student outcomes. External of the Opportunity Culture’s multi-classroom teacher roles found that it helped typical educators raise their performance from the 50^th to the 77^th percentile. A last year of the TAP system found that it boosted high school graduation rates by 3.8 percentage points. Those gains also reduced criminal activity and led to fewer individuals relying on welfare between the ages of 18 and 22.  

Teachers also seem to appreciate working in teams. One of TAP schools found a teacher retention rate of 94%, far surpassing national averages. In Opportunity Culture schools last year, 94% of teacher leaders a positive impact on staff collaboration, and 96% said the team approach helped improve learning.  And teachers working in schools felt more supported, had equal or lower turnover rates and were more likely to recommend teaching as a career compared with those other schools in their district. 

Team-based staffing could also help address other problems schools have been facing, such as high rates of at the same time have been hard to find. This is partly because the traditional approach has no built-in redundancy: If a teacher gets sick, there is no one on staff to cover the class. Even something as mundane as going to the bathroom can be a problem. In contrast, having teams creates built-in flexibilities enabling teachers to cover for each other on a daily and hour-by-hour basis. 

It also helps schools with the challenge of dealing with declining student enrollment. Maddin notes that schools that operate with fixed staffing ratios have a hard time navigating those declines — and the potential for staff layoffs — while in a team-based approach, staffing levels aren’t as strictly tied to student head counts. 

Transitioning to a team-based staffing system is not a simple affair. Potential obstacles include money, teacher buy-in and like licensure rules or evaluation policies. 

Van Hook says leadership stipends are typically the most expensive part of the switch, but most places already have similar people in place — they’re just not integrating them fully. Van Hook, Hassel and Maddin all say it’s possible to switch to a team-based staffing model without adding any ongoing costs. 

There are some transition costs, but accelerants like the federal and state-based policies like North Carolina’s program can help more schools make the switch. 

When asked about the biggest challenges and risks, all three leaders expressed concern about a “light touch” version of this work. Maddin says he can tell if a school’s culture hasn’t changed much if teachers don’t have sufficient planning time or feel responsible only for their same 25 kids, if student schedules look exactly the same every day or if a school needs to hire a substitute when a teacher has an early morning doctor’s appointment. 

In other words, shifting away from the one-classroom, one-teacher model requires fundamental changes. But a true team-based approach offers a wide variety of benefits for both educators and students. 

When A Nation at Risk was published in 1983, I was twelve years old. I was growing up in Los Angeles as one of three birth children in a diverse family of fourteen, and I had witnessed firsthand the unacceptable circumstances my adopted brothers and sisters experienced in a broken child welfare system. It was clear to me that our education system was similarly broken, and I could plainly tell that succeeding in school was much easier for those of us who held privilege.

As the child of two dedicated civil servants who are avowed Democrats, I never would have guessed that a report coming out of the Reagan administration would help spur a transformative movement that would tap into my activist, social justice upbringing and provide a path for me to play a small part in moving the needle on education equity.

We’ve come a long way since 1983. Over these many years, my friends and colleagues in the school reform movement and I have led hard-fought progress—as well as committed some glaring missteps. These years of experience, coupled with a deep and growing body of research, have yielded invaluable lessons for how we, as a nation, can come ever closer to delivering on the promise of education excellence for all kids.

Educational leaders have used these decades to explore an entire continuum from tweaking and refining existing systems to innovating and disrupting the status quo. The solutions are out there. The talent is out there. Even the money is out there. So what is holding us back?

In this chapter I will share reflections from our attempt to answer this question in Newark, where I served as superintendent from 2011 to 2015. Building on the lessons and experience from nearly twenty years of reform, we launched a number of ambitious but commonsense reforms, culminating in the “One Newark” plan. Our goal was to engage the entire community and its leaders to address the educational needs of every child in every neighborhood in Newark. It was a bold and controversial effort that aimed to bring the best of what educators and policymakers had learned about driving change through top-down systems reform, bottom-up community demand, innovative labor approaches, and new school models.

Much has been written about the personalities and political drama in Newark during that time, but almost nothing has been written about the actual playbook and results. More importantly, I believe the reform efforts there still serve as useful examples to illustrate what is possible, what we should correct, and where we should go next.

Examination of this period in Newark will raise critical questions about what policymakers and community leaders at all levels should do to foster a system holistic and flexible enough that it would address the needs of all children, especially those our current systems of education and social service have historically and consistently failed the most. I will outline four steps policymakers can take to catalyze change and center the students and families who face the most challenges.

The road to reform in Newark

I was appointed superintendent of Newark Public Schools (NPS) in 2011 by then governor Chris Christie and the state’s education commissioner at the time, Chris Cerf. While most school districts have a local board charged with hiring a superintendent, NPS had lost that authority back in 1995, when the state took control of the district.

New Jersey’s constitution has uniquely strong language compelling the state to intervene in the event of chronic failure. As far back as the 1967 rebellion in Newark, there had been growing political support for the state to step in and address the city’s abysmal student achievement, despite vehement objections from local elected officials.

I was the fourth in a line of state-appointed district leaders to arrive with a mandate for reform and improvement to Newark. But my arrival came with an even greater sense of urgency. About nine months before my first day, multibillionaire Mark Zuckerberg famously announced on The Oprah Winfrey Show, seated next to Governor Christie and Cory Booker, then mayor of Newark, that he was making an unprecedented one-time gift of $100 million to the city’s school reform efforts. Just prior to this, the governor had also announced his decision not to renew my predecessor’s contract, sending a clear signal that change had not moved quickly or decisively enough in Newark.

It was clear that Newark school improvement was a signature political issue for Christie, and the pressure to raise achievement and improve efficiency was intense. On a trip to Washington, DC, earlier that year, he proclaimed to national media that “in the city of Newark, we are spending $24,000 per pupil, and public money for an absolutely disgraceful public education system—one that should embarrass our entire state.”

Indeed, as I arrived in Newark, 39 percent of students who entered the system failed to graduate, and only 40 percent of third-graders could read and write at grade level. Enrollment was plummeting. The district’s nearly forty thousand6 students and one hundred schools made it the largest in the state, with the majority of students living below the poverty level. The schools themselves were a physical manifestation of the deterioration and decay. I still shudder at the memory of rats floating in basements amid boilers dating back to the 1950s and of Lafayette Elementary School (built when Abraham Lincoln was president), which had actual mushrooms growing in the cafeteria after Hurricane Irene.

Local politicians and families had grown impatient. For the five years prior to my arrival as superintendent, many elected leaders had become early adopters of a growing national charter school movement that aimed to free individual schools and networks of schools from government red tape and allow them autonomy to innovate and build excellent schools. These supporters included Booker (then a young councilman), school board member Shavar Jeffries (who now heads the charter school behemoth KIPP Foundation), and state senator Teresa Ruiz, among other notable local leaders. Charters weren’t the only new option—other school models, such as magnet high schools (often with entrance requirements) and partner-run small high schools, had gained momentum too.

Some of these schools had notable evidence of improving achievement for Newark students, and it was understandable that they were gaining strong support from local leaders, influential funders, and certainly the families of the nearly 5,500 students who were enrolled in them. Indeed, Zuckerberg’s gift earmarked a significant amount of funds to growing the charter and small-school footprint. A plan between Christie and Booker that was leaked to the press around this time (causing uproar in the community) revealed that a central focus was to build more charter schools and open new small schools as quickly as possible.

It was clear that the most impactful efforts at improving schools in Newark were actually working around the very system they were trying to improve. And in New Jersey, these new schools were funded on a per-pupil basis; in other words, the money followed the child out of the traditional system and into the public charter system. Logically, this made sense. But in practice, this proliferation of competitors to district-run schools was creating unintended consequences that few wanted to discuss.

During the interview process to become superintendent, though, I was very eager to discuss it. What was the long-term plan to ensure that students who remained in traditional district schools benefited from the cash infusion as much as those who were lucky enough to win school enrollment lotteries? What about the school closures and layoffs that would be inevitable when the footprint of traditional schools shrank? Would charter and new-school growth help bring excitement and excellence to our poorest neighborhoods, or would it give some kids better schools but make the conditions in their communities worse? Wouldn’t the district end up serving more students who required the highest levels of support? No one seemed to have answers for these tough questions.

I had questions for the anticharter folks too. How could you blame families for exercising choice given the abysmal conditions of many traditional schools? Why were thousands of families on waiting lists? What policies—labor and otherwise—were holding traditional schools back from succeeding? Why hadn’t those policies been advanced? Were community leaders ready to push the bold reforms necessary for traditional schools to compete with charters? And what should we do about the chronically failing, profoundly underenrolled schools? The answers to these questions were complicated and generally had only to do with adult politics, not what’s best for kids.

My background and the questions I asked made a lot of people very uncomfortable. Some saw me as “antireform” because I wasn’t “charter-friendly” enough, especially with the eyes of the nation now on Newark. At the same time, others felt I was in cahoots with the “privatizers” trying to kill public education by supporting new school models. In hindsight, it is easy to see that my story was just an example of a growing and deeply polarized national debate on whether and how we can radically improve the quality of education for low-income families and students of color at scale. Meanwhile, from day one, our team wasn’t focused on winning favors or promoting specific ideologies. We were focused on great schools for all kids in every neighborhood, by any means necessary.

Building a ‘system of great schools’

Given the perilous state of the city’s schools, the unrealistic expectations around quick achievement gains, and the pressure from ideologues on all sides, many speculated that the superintendent role wasn’t doable. But I was inspired by the scale and the ferocious commitment of many leaders in the community.

I interviewed with a large committee and countless stakeholders, who spent hours debating diverse theories about what to do to improve schools. I was also moved by the unusual personal and political alliance formed by Christie, a popular Republican governor in a Democratic state, and Booker, a popular mayor and rising Democratic star, to do something bold. Something was in the air, and it felt like transformation was possible for children and families who had been failed by public systems for generations. I was convinced, perhaps naively, that if we could harness the debates and emotion around what to do, we could lift up a whole city. I remember thinking, There are one hundred schools—we can do this!

I assembled a diverse team of exemplary senior leaders—some known and trusted from within Newark, some with a track record of results elsewhere, and some with a lot of promise who were ready to step up. When I say “we” in describing the work in Newark, it is not simply rhetorical. I cannot take sole credit for anything we accomplished in those four years; it was this stunning team that acted as both the architects and the engineers behind the policies and practices we implemented, tirelessly, on behalf of Newark families. I wish I could list them all here by name.

It was time for us to craft our own plan for Newark’s schools. We started with the theory that the unit of change was the school itself and embraced the idea that what we were building was what my former boss, then New York City Schools chancellor Joel Klein, called “a system of great schools,” not a “great school system.” This was a subtle but profound distinction, because it meant we were seeking to ensure that there were one hundred excellent schools serving every child in every neighborhood—regardless of governance structure.

In the long term, we knew that Newark required a citywide master plan that would account for every school building and every child. But in the immediate term, we had obvious and dire issues to address in the district’s own schools.

First, we needed to set a unifying goal for the district: every child would be college ready. That’s right, college, not just career—because we believed that choice of higher education should be up to the student, not simply determined by the inadequacy of their preparation, and because Newark families were demanding this.

While college readiness is an obvious educational goal for affluent families and communities, in Newark it was far from obvious that this was attainable or even desirable for most students. But to families, it was obvious. In poll after poll, focus group after focus group, they told us very clearly: they wanted their children to graduate college ready. Moreover, they believed that “career ready” was a euphemism for low expectations. Families felt that academic excellence was a passport out of poverty.

Most parents were with us from day one. The challenge was the well-meaning funders and other influencers who wanted to muddy the waters and talk about everything except whether students could read, write, and do math at grade level.

To make a case for action, we shared baseline achievement data and started to create a sense of urgency around this critical goal. Looking unflinchingly at this data wasn’t easy, and it made many educators and leaders uncomfortable to acknowledge so blatantly how poorly our schools were preparing our students to succeed in life. When we started sharing actual data about proficiency rates and the number of young people earning diplomas indicative of their mastery of hard content, we started to encounter real pushback, both within and outside the school system. This was a theme I became increasingly familiar with: often what families say they want can be quite different from what those who speak for them are willing to stand for.

Ensuring ‘four-ingredient schools’

With our North Star established, we rolled up our sleeves to improve the district, school by school. By this point in US education reform, it had been nearly thirty years since the release of A Nation at Risk, and there was a large and growing body of research and evidence about high-performing schools in high-poverty neighborhoods. Combined with our team’s years of on-the-ground school transformation experience, we zeroed in on four basic ingredients that every high-quality school possessed: people, content, culture, and conditions.

To extend the cooking analogy, we knew each ingredient could have a different flavor profile in a different environment. The exact mix of ingredients varies based on context as well. Getting all four ingredients right in a school building is a huge challenge on a good day and nearly impossible the rest of the time. And yet the ingredients themselves are relatively uncontroversial—even commonsense—and should be the central focus of systems leaders and policymakers.

Our aim: ensure that every NPS school was a four-ingredient school so that we could make steady progress toward college readiness for all. Our philosophy: focus on what works regardless of ideology, which often led to “third-way” solutions—combining the best of seemingly disparate views or forging a new path to transcend old, binary thinking. Our mantra: implementation matters.

People

It’s critical to have the right people in the right seats, from the leadership team to the teachers to mental health professionals to custodial staff. No matter their role, every adult in the building must be equipped with the right mindsets and skill sets to uphold the mission and goals of the school.

Education reform birthed a broad array of nonprofits and policies focused specifically on teacher quality, notably Teach for America (TFA) and TNTP (formerly the New Teacher Project), the latter with its landmark Widget Effect report. We know intuitively the power that a great teacher has, and a growing body of research reinforced this belief, showing us that teachers are the most significant in-school factor determining a child’s level of achievement. Further, the most significant factor in getting great teachers in every classroom is the quality of the principal. Meanwhile, author Amanda Ripley showed us that in Korea, master teachers are treated like rock stars, which is hardly the case here in the United States.

Partly because of my experience working on the talent side and the emerging research about the strong correlation between school leaders and student achievement, we focused on leadership from day one in Newark. I’ve never been to a great school with a mediocre principal, and I have never been to a failing school with a terrific principal (except perhaps at the very beginning of a turnaround). Within two years, we had replaced nearly one-quarter of our principals through aggressive recruiting and selection, giving preference to Newarkers and leaders who not only knew instruction but thought of themselves as community organizers and change agents.

We took a page from New York City’s playbook, enacting a “mutual consent” policy that allowed principals to select teachers aligned to their school’s goals, as opposed to having them force-placed according to seniority. We adopted a new teacher evaluation system that required more evidence-based classroom observations and feedback, taking a page from Chancellor Michelle Rhee and her team’s work in the District of Columbia. We trained and empowered principals to hold teachers accountable when they failed to uphold expectations—and we had their back when teachers needed to be removed. We created “career ladders” for teachers to become leaders without having to leave the classroom, taking a page out of the Baltimore playbook.

As was a consistent theme with our approach in Newark, pursuing and advancing third-way ideas had us making people on all sides of the issues uncomfortable. Many states at this time were starting to use quantitative test score data in teacher evaluations, and New Jersey was eager to follow suit. However, my team and I felt that the science for such “value-added” approaches didn’t hold up when it came to determining the effectiveness of individual teachers. We took a lot of flak from hardline education reformers, who had become fixated on using test scores as a shortcut to accountability and who worried that our questioning the use of test scores in teacher evaluations would water down reform efforts more broadly since Newark was such a high-profile example. But not only did we feel that using the value-added approach in teacher evaluations would be unfair to teachers, we also knew that including such a poison pill in our new evaluation plan would create a backlash that could sabotage the entire effort.

To help noncharter schools accelerate the “people” ingredient, we negotiated what was widely considered an ambitious contract with Newark teachers. We were able to find agreement with both the local and national teachers’ unions on contentious issues such as freezing pay for teachers who were ineffective; on providing bonuses for high performers working in hard-to-staff subjects; on expediting firing for the small number of teachers caught doing egregious things; and on finding pathways for individual schools to innovate outside the four corners of the contract. We also asked the state to grant us a waiver from traditional tenure laws so that we could consider quality alongside tenure when making decisions about whom to retain as we set about the necessary downsizing.

Despite agreeing to key labor reforms after more than two hundred hours at the bargaining table, some in the Newark Teachers Union and their national affiliate, the American Federation of Teachers, vociferously advocated against them within weeks of the contract being ratified by an overwhelming majority of teachers. Both groups had a long track record of preserving some of the sacred cows of labor negotiations with teachers: seniority-based placement, infallibility of teachers with tenure (regardless of what they do), and resistance to any form of accountability—no matter how nuanced. Meanwhile, we found many of these ideas to be popular among everyday teachers, who told us the quality of the teacher in the classroom next door is a factor in whether or not they want to stay at a school. In fact, research shows teachers want access to high-quality curricula, comprehensive assessment systems, and the ability to collaborate with colleagues. As the granddaughter of a union organizer and as a strong believer in collective bargaining, I was pushing largely because I believed then — and still believe now—that teachers’ unions need to evolve to become part of the solution or they will become obsolete.

We also had to completely restructure and reimagine the central office to be in service to schools and families. This required breaking senior leaders into new teams and inviting them to clearly articulate how they would enable the four school-level ingredients. It also meant crafting clear plans with goals aligned with good management and coaching—not simply doing what had always been done. Many rose to the occasion and embraced the opportunity for clarity and coaching. Some didn’t. This was another necessary and politically fraught task. Newark Public Schools was one of the biggest employers in the city, so many staff on the central team were connected to a local politician, vendor, or influencer who had their back if their job was in jeopardy, regardless of whether or not the job was necessary or they were doing a good job.

Content

A high-quality school needs high-quality and culturally competent curricula. It also needs frameworks, protocols, and data that drive great instruction and continuous improvement. As computer programmers like to say, “Garbage in, garbage out.” This “content” ingredient is all about replacing that “garbage” content with engaging and carefully crafted content.

I started in Newark about a year after the Common Core State Standards had become a force nationally and the same month that New Jersey adopted a version of them. It was good timing, because I have long believed in fewer, clearer, more rigorous standards, as opposed to the laundry list I was handed as a young teacher in California. Common Core gave us an unambiguous and evidence-based target. It also served as a catalyst to scrutinize our curricula with a more rigorous lens.

The research here is undeniable; high-quality instructional materials are critical to ensuring that students are truly internalizing difficult content. Historically, though, we had all underinvested in this area in the early reforms after A Nation at Risk. A lot of us made the mistake of keeping a hyperfocus on teachers, which indulged an assumption that if we had a perfect person in every classroom, they could invent brilliant content from scratch. Fortunately, the introduction of common standards forced the issue and led to game-changing work by leaders like John King and his team in New York to develop what would become the EngageNY curriculum. Many education advisors and publishers have caught on, and there are far more “good enough” options out there, but there are still not enough.

We also were informed by systemic approaches to “managed instruction” like efforts in Charlotte-Mecklenburg Schools, the Success for All initiative in Los Angeles, and Newark’s own NorthStar Academies (a charter network). These reforms, often literally telling teachers what to teach and how to teach minute to minute, yielded some impressive results. But they also predated important reforms around teacher quality. Our team aimed to blend the best of both by making “good enough” choices about curricula, creating spaces for teachers to use existing scope and sequences and lesson plans but also building the capacity of teachers to use data and knowledge of their classrooms to adjust.

High-quality instructional materials are an ingredient that is hard to get right when you are working only at the school or small network level. Scale is your friend. These decisions are better made at a system level, where content experts can dedicate the necessary time to addressing academic needs and cultural contexts, as well as coherence and alignment between the plethora of different curricula and assessments. It is also the area that, at the time in Newark, brought the most consensus. We did “teach-ins” for administrators, educators, influencers, and families who all really seemed to get and support the mandate for good, rigorous content that was consistent across the city.

As I write this, the country has been embroiled in a resurgence of culture wars around what we teach in our classrooms. It’s unfortunate, considering that this content ingredient was actually a rare point of local and national consensus not that long ago. The current political climate adds an unwelcome layer of complexity for systems leaders to battle.

Culture

We know from research that schools with intentionally curated environments characterized by high standards alongside high support produce better student outcomes. Students learn healthy habits, and the school community has well-established values and expectations. Norms and protocols prevent incidents, and when incidents happen, adults minimize shame and exclusion to keep students learning. In these schools, there is joy and choice.

From day one in Newark, we focused on the seminal research work and promising practices that had emerged, connecting how kids feel, how adults feel, and student outcomes. Years after comparing student achievement results to staff, student, and family survey responses, researchers Tony Bryk and Barbara Schneider found that the schools with high levels of trust were far more likely to get beat-the-odds results than their counterparts. Economists like Ron Ferguson and social policy experts like Christopher Jenks found a direct correlation between adult expectations, student surveys, and student outcomes. Though controversial, it is also no surprise to me that recent research has shown student surveys can predict student achievement as well as teacher evaluations. When young people say things like, “My teacher doesn’t stop until I get it” or, “My teacher believes I can understand anything,” or, “When my teacher asks me how I am doing, I believe s/he wants to know,” we know those students will do better. Jeannie Oakes’s Keeping Track showed very explicitly how adult biases and expectations can have a negative impact on student achievement.

Relatedly, an area where I have seen some of the greatest challenges for adults in establishing and preserving culture is in response to conflict and disruptive incidents. How we handle student discipline, struggle, and conflict is where adult biases show up the most. Black students are four times as likely to be suspended than their White peers, often for the same behaviors. Black students make up one-third of school-based arrests, though they are only 16 percent of the student population. Moreover, nationwide, Black boys are almost twice as likely to experience corporal punishment as their White peers, and Black girls are about three times as likely. Adults rate Black girls as “less innocent” than their White peers, with damning implications. Students with disabilities are far more likely to receive exclusionary discipline for subjective things like “disrespect” and “insubordination.” This is a problem not only from an equity and justice lens but also from a student achievement standpoint. Often students who need the most support and time on task are being excluded the most. Students can’t learn when they feel shame and helplessness. So it is no surprise to me that data shows that the relationship between the discipline gap and achievement is more than correlative—it is also causal.

The research squared with our team’s lived experience in Newark. Traditional schools like
Chancellor Avenue Elementary School and Sussex Avenue Elementary School, charter schools

like KIPP Spark Academy and North Star Academy’s Alexander Street Elementary School, and partner schools like Bard Early College got this ingredient right, and their results showed it. Adults, kids, families, and community partners rallied together around a common vision and values—and shared expectations and norms at the school reinforced them. Their results showed how critical it is to build a collective culture of high expectations and high support.

For these reasons, we hired administrators who showed skill in building culture and partnering with families. We created an entire central-office team focused on student well-being and discipline and charged them with building the capacity of schools to create environments where all kids would thrive and to respond to incidents skillfully. We reinvented the role of school resource officer and hosted weeklong restorative practice institutes that brought together student leaders, administrators, families, teachers, and police officers.

We made progress, but admittedly, the playbook on culture is harder to run for many reasons. Too often, discussions about what student culture should feel like are preachy, ideological, or theoretical—devoid of practical, research-based, promising practices. Building culture is far from a paint-by-numbers task. Cultures that are simply cheap forms of imitation, are inauthentic, or are misaligned to the needs of a particular community don’t create the conditions for achievement. Frequently, we think about individual tactics for establishing and preserving culture, such as specific expectations or restorative circles, but not about how they all fit together, and this leads to cultures that feel disjointed and incoherent. Effective cultures don’t feel the same in every school, but they do share key components. This is nuanced and hard to teach to administrators. At a systems level approaches to this ingredient often devolve into compliance lists and checklists. Further, the culture work requires us to surface and address adult biases about what kids can accomplish and what is considered “dangerous” behavior, and this can cause real discomfort and resistance.

Conditions

This ingredient is all about strong operations and infrastructure. The building is clean, well equipped, and well run. The trains run on time. You have the facilities, management structures, and funds to support learning, and you have the funding, supplies, and technology to support all of the other ingredients of a high-quality school.

While setting the right culture creates a social and emotional environment where both students and adults can thrive, it is important to simultaneously address the physical environment and the day-to-day operations. It may not be as compelling or sexy as the other ingredients, but none of the other ingredients of a strong school or system can succeed if we don’t address the conditions in which our children learn and our teachers teach. In Newark, we had a lot of work to do on this ingredient.

When I started in Newark, Malcolm X Shabazz High School had a river running through its fourth floor on rainy days. Many schools didn’t have air conditioning, in a city where average temperatures reach above a humid ninety degrees for months. Some schools weren’t even wired for internet access, and only a few had laptops to check out to students for the day.

Local leaders openly talked about a “rolling start” at the beginning of the school year, which referred to the fact that it took weeks to sort out the basics: enrollment, special education schedules and services, buses, and even books. Honestly, I had never heard of a system where instruction didn’t start on day one.

And while I’ve never bought into the idea that market forces are a panacea for improving school quality (and research suggests I’m right), you can bet my team took note when our colleagues at KIPP would make an overnight run to the hardware store for window A/C units to survive a Newark heat wave, while we were forced to navigate a maze of vendor regulations and nepotistic relationships just to open a window. Trust me, there’s an entire chapter to be written on reforming procurement and purchase ordering alone.

Some of these intolerable conditions were due to bad public policy and some were because of poor management. My team and I would say we could tell if a school was getting results by how visitors were greeted at the door (if at all) and how quickly families could get the answer to whatever they were asking. When organizations are well run, their primary constituents— in our case, students, families, and community members—are at the forefront of everyone’s mind. This service orientation is shared by noninstructional staff members, from the custodial staff to the crossing guard to the budget team. All are invested in the mission, goals, and shared values. We created school operations managers (similar to how some charter networks like KIPP have created business operations managers) to attend to the operational needs of the school. At the time, this got me in trouble with the administrators’ union (because I was seen as encroaching on district administrator roles and jobs). Even today our approach to operations is considered innovative, which just shows how little we prioritize the conditions in our schools.

Getting the conditions for achievement at every school in Newark was an expensive and backbreaking task, and progress was excruciatingly slow. Bureaucracy at every level—local, state, and federal—made our vision of goal-based budgeting, twenty-first-century facilities, nutritional food, and high-quality customer service feel nearly impossible on some days, even with exemplary people working on it. When academics and policymakers talk about disrupting systemic inequity, rethinking our entire system of education, or rising to the grave challenges initially posed by A Nation at Risk, they’d do well to spend a lot more time talking about basic operations. We need an entire movement to break the bureaucracy that is crippling school infrastructure.

The One Newark plan

While establishing a focus on college readiness and building four-ingredient schools was our primary focus right out of the gate, we knew we had to make progress on a citywide plan that addressed the schools beyond our purview. Looking at the full picture in Newark, you saw that everyone—local early childhood providers, the district, third-party school operators, private schools, individually run charters, and large charter networks—was doing their own thing, and the unintended consequences of this lack of coordination were becoming more evident and unsustainable every day.

From our earliest school visits, we could see that the poorest neighborhood schools were emptying out and becoming concentrated with the highest-need students and the lowestquality staff. Historic buildings were crumbling while new facilities were being built, sometimes down the street or downtown. Supply and demand were misaligned: for example, the number of magnet high school seats requiring a certain level of academic attainment far exceeded the number of eighth-graders prepared to meet them. The diversity and variety of school models wasn’t materializing; with all the new schools, we weren’t actually providing a lot of choice, just more flavors of “no excuses” ice cream at the elementary level and a bunch of run-of-the-mill high schools.

Meanwhile, every year, including my first, our district had to cut about $50 million. While there was certainly a lot of bloated bureaucracy to streamline, more than 80 percent of that money was wrapped up in people. Newark Public Schools employs many Newarkers in a city with double the national poverty rate.

With every set of data that we unearthed and every school that I visited, the pit in my stomach grew as I saw the magnitude of the challenges. Keeping open too many schools without enough students meant that thousands of Newark’s students were attending schools that were crumbling, not just physically but educationally. We were spending more money per pupil than almost every other district in the United States, with terrible results. And there was no denying that the rapid growth of the charter movement further complicated the problem.

Past district leaders—and many city leaders—viewed charters as “the enemy.” Families able to navigate the charter lotteries were fleeing the traditional system, and thousands more were on waiting lists. Who could blame them, given that many charters were radically outperforming traditional schools? As money followed students out of the traditional schools and into the charters, the available resources to revive these schools or attract talent were being stretched increasingly thin—and the trend was likely to continue. Charter leaders planned to grow the sector from serving 5 percent of students to serving more than 40 percent, which would mean $250 million of funds leaving the district with no quick or painless way to shrink infrastructure or union-guaranteed jobs. Seeing that dozens of district schools were dying a slow death, with some of the city’s neediest students trapped in them, I knew something had to be done—and soon.

As a city, we had to ask ourselves: “Is it even possible for every child in Newark to have access to a school that meets their needs? Even those children facing the longest odds?” Note that this is a fundamentally different question than “Can we get some kids in this community access to great schools?” That framing suggests we are not responsible for all of the children in our community, only those whom our school model can accommodate. It is also a fundamentally different question than “Can we build more great schools?” That question ignores the community context within which the school exists, and it fails to address very real and sometimes serious consequences when we focus on building some great schools and letting others fail. We were seeing that play out in Newark already.

Our team had no choice but to stare down these questions, which led us to some unconventional and controversial answers. The first thing we had to do: try to rise above political arguments rooted in ideology and self-interest about what type of school models should exist. There were about a hundred schools in Newark. We knew we would get to excellence more quickly if we had a variety of governance structures: traditional, charter, magnet, partner run, and hybrid. But we also knew we couldn’t simply let a thousand flowers bloom and allow others to die, especially when those vulnerable schools were serving our students with the highest need. We also knew that the community deserved excellence citywide.

Just as always, our team was guided by national promising practices and research. Only this time, we were limited by a dearth of examples in which systems leaders thought about the entire community. So we pored over our own data: student enrollment trends across governance models, overall city population trends, facilities assessments, and (of course) student outcomes. We fanned out and hosted more than a hundred community-based meetings with faith-based leaders, nonprofit executives, families in struggling schools, families in highperforming schools, charter advocates, charter operators, private schools, local funders, elected officials, union leaders, and early childhood providers. We began to socialize the idea that we needed one citywide plan across governance structures, as well as the harsh reality that the district’s footprint had to shrink. We wanted to find a way to preserve the best of the new-schools movement while also addressing some of the unacceptable consequences of its uncoordinated growth.

This process—over the course of about a year—led to a comprehensive plan we called One Newark. The plan was more than just a collection of policies and tactics. It was carefully architected with a clear and accessible framework to communicate honestly and transparently with an extremely broad array of stakeholders and get buy-in across the city. We knew that the best plan in the world would mean nothing if we didn’t tell a coherent story that motivated change.

The plan opened with three core values to drive our collective decision-making: equity, excellence, and efficiency:

• Excellence: We must ensure that every child in every neighborhood has access to a “four-ingredient” school as quickly as possible and that no kid is in a failing school.

• Equity: We must ensure that all students—including those who are facing the longest odds—are on the pathway to college and a twenty-first-century career.

• Efficiency: We must ensure that every possible dollar is invested in staff and priorities that make a positive difference for all students.

It was followed with seven focus areas, which we mapped onto the letters in the word “success” to send a clear signal of optimism and affirmation:

• (S)ystemwide Accountability: Envision and publish a standardized approach to tracking school and system success and progress across all schools (district, charter, and provider run).

• (U)niversal Enrollment: Launch a straightforward, user-friendly enrollment system that empowers our students and families to choose the school (charter, partner, or district) that best meets the student’s needs.

• (C)itywide Facilities and Technology Revolution: Create a bold plan to operationalize twenty-first-century learning environments for all students, ensuring no vacant buildings.

• (C)ommon Core Mastery and PARCC Readiness: Lead the nation in the number of students living below the poverty level (especially those currently struggling) who make progress toward Common Core mastery and readiness for PARCC, the standardized test aligned to those standards.

• (E)quity and Access for All Students: Increase the number of high-quality seats for all students, especially those currently in low-performing schools.

• (S)hared Vision for Excellent Schools: Cultivate demand for one hundred excellent schools and the groundswell of support for the changes necessary to get there.

• (S)ystemic Conditions for Success: Radically transform the district itself to ensure that it is a high-performing organization for years to come.

While the backbone of the overall picture and the building blocks of the plan were emerging in the spring of 2013, we didn’t feel we had enough operational capacity or community momentum to implement the plan that fall. Instead, we continued to engage, discuss, and refine the plan’s tenets with diverse stakeholders. We launched headlong into implementation in the winter of 2013–14.

We knew from the start that we needed a fair way to compare school quality across model types to meet the goals of our citywide vision of excellence for all. I’d seen firsthand how smart ways of tracking progress could drive good change in New York City under former mayor Mike Bloomberg, and we knew of years of research by the Consortium for Policy Research in Education and others about the correlation between accountability and student outcomes.

We started publishing “family-friendly” snapshots — across both district and charter schools — so that community members could see how their schools were doing in comparison to schools with similar populations. We looked at overall proficiency but also at growth, critical in a city like Newark with low rates across the board. We also compared schools with similar student populations to one another. There was no question that students at schools like Shabazz High School and South Street Elementary School were in much higher need than students at Science Park High School and Ann Street School. Success was possible, of course, but harder—and we needed a fair accountability system to make decisions and create the right incentives.

We created a simple red, yellow, and green system so that the community could see the landscape clearly. “Red schools” were low-proficiency, low-growth schools. Green were high proficiency and high growth (e.g., we didn’t want selective high schools to recruit “proficient” students and fail to grow them to “highly proficient”). Yellow schools were “on the move” (low proficiency, high growth) or “to watch” (high proficiency, low growth). The color-coding was clear and intuitive, and many in the community started talking about “no red schools.”

We placed an emphasis on transparent data about how schools were doing with students in poverty, students with disabilities, and English learners. We created standard measures — across district and charter schools—to report on student retention. Up to this point, accountability systems implemented under the No Child Left Behind Act of 2001 focused almost exclusively on proficiency, which unfortunately incentivized schools to enroll and retain students who were likely to be successful and to subtly counsel out (e.g., “you are not a good fit for this school”) or explicitly push out (e.g., through suspensions and expulsions) students who were harder to serve. We couldn’t afford to make the same mistake.

People from all sides fought us on this level of transparency — the unions, some charter schools (which weren’t obligated to share their data with us), and some funders who worried we were reducing children to numbers. But many families and policymakers embraced the information. There’s no perfect system, but there was no way to make a citywide plan without a decent measure of school quality.

We performed detailed enrollment analysis and defined the need for a common definition of a “minimum viable school.” From a funding standpoint, schools with fewer than five hundred students are hard to sustain with a staffing model that ensures things like appropriate class size, electives, teacher preparation times, and staff to attend to running operations. Newark had a lot of “red” schools that were also not financially viable, and many of them were in the poorest neighborhoods.

We also looked at demand data—who was applying to charters and from what neighborhoods, who was seeking new small high schools and from what neighborhoods, and which neighborhoods were growing and which were shrinking.

The picture was becoming increasingly clear: the need for a course correction was long overdue. We had traditional schools where 80 percent of families were on charter school waiting lists, but the district’s resistance to collaboration and the charters’ insistence on growing only one grade level each year meant large-scale closures and consolidations were inevitable.

The district had too many elementary schools overall, due to a population decrease, neighborhood shifts, and charter growth. We didn’t have enough early learning centers to meet the increased demand. We had too many selective high schools. Most of the new small high schools being incubated downtown were serving families from other wards, while iconic and historic high schools were emptying out. Overall, we had too many old buildings that were crumbling due to underinvestment and age, and some of them simply weren’t fixable. At one point, the district was paying more than $1 million just for scaffolding on vacant buildings that were never going to reopen. The picture was bleak. We had to make some hard decisions.

We decided to be radically transparent about our findings and the implications in a proposed ward-by-ward plan. Some charters should take over existing schools with high demand, keep families who opted in, and keep the buildings and the school name, instead of simply continuing to build new schools one grade at a time. Some elementary schools needed to convert to early learning centers. Some small high schools that were performing well needed to move into our comprehensive high schools, and some underperforming partner-run high schools needed to close. Magnets had to change their enrollment process. And some buildings had to be shut—some condemned, some repurposed, and some sold, potentially to charters.

Another anchor of the One Newark plan was ensuring that every family had equal access to choice. Both psychologically and practically, it didn’t make sense for one-third of families to get what they wanted and the rest to get what was left over. For starters, this dynamic was creating an almost civil war–like atmosphere, with charter and noncharter families pitted against each other and magnet and nonmagnet families screaming at each other in meetings. Also, one goal of establishing high-performing schools in high-poverty neighborhoods is to feed the groundswell of belief that kids can achieve. Newark’s choice system was helping create a self-fulfilling prophecy of failure in the noncharter schools.

We had to find a way for the idea of choice to lift all boats, but it wasn’t happening—and it can’t happen without good public policy and collective action. I’ve had many school choice advocates dispute this. Some ideologues will have you believe that the mere presence of competition somehow magically raises everyone’s game. It certainly didn’t happen that way in Newark, nor in the dozens of systems I have worked in since.

This is where universal enrollment came into play. Cities like New Orleans and Denver had implemented systems where families could access a common application instead of having to apply to and navigate multiple lotteries.43 We built on what they had learned (even flying in officials from both systems to participate in community panels) and took it a step further. We envisioned and implemented a system where all schools—charter and traditional—marketed themselves on the same timeline, using citywide approaches, and alongside our common accountability system. All families could access the system and apply to all schools. An algorithm gave preference to kids in the neighborhood, followed by kids in poverty, then kids with disabilities, and then everyone else at random.

It was a game changer. Now all schools were required to think about how to market themselves and own their quality, or lack thereof. By year two, more than three-quarters of the families of kindergartners and ninth-graders were using the system. At one point, we opened a family support center to help families exercise choice. We had actually planned for a soft launch, but word got out and more than a thousand families showed up on the first day, and the situation almost devolved into chaos. While our critics crowed about our operational failure—and it was indeed a failure—it also showed how much family demand there was for choice and quality. This is one of the hundreds of examples I’ve had throughout my career that defies the ridiculous stereotype that poor families don’t care about education.

The universal enrollment system may have been hardest on some members of Newark’s political elite who were used to the benefits afforded to them in an unfair, transactional system. I recall one meeting in which a prominent official—previously a supporter of mine—yelled, “You made a liar out of me! I told my cousin I could get her kid into this school!”

We had other “lift all boats” strategies. In partnership with the Newark Trust for Education, we created shared campus grants so that charter and district schools in the same building were incentivized to envision projects that helped their students and staff collaborate on schoolwide and community improvement projects. We asked charter teams to lead professional development for some of our turnaround schools on things like comprehensive approaches to instruction where many of them had more robust practices than we did. We created a collective action team of special educators across district and charter schools to help share and promote promising practices.

The plan meant a lot of changes for a lot of people. Some shuttered buildings were historic, and even though it was clear these buildings needed to have a divestment plan, community elders who remembered their heyday didn’t want to hear it, understandably. Many charter leaders and their supporters dug in their heels on their model of growing slowly and where they wanted to grow according to optimal facilities, regardless of the consequences. The idea of small schools within high schools—which had been successfully implemented in New York at scale—was new to Newark and, therefore, scary for many who had found their pet school to support. Some local and national funders who were excited about ribbon cuttings and smaller projects simply didn’t want to get involved in the far messier project of citywide progress.

Our team knew that the tenets of the plan were bold, unconventional, and controversial and that the politics were going to be tough to navigate. Choice, charters, labor reforms, and teacher excellence polled well. Laying off Newarkers and teachers and “closing” traditional schools or turning them over to highly successful charters were wildly unpopular. But to have the plan succeed citywide, you couldn’t have one without the other.

To add a deeper degree of difficulty, while the plan was emerging and leading up to the official launch, we suffered a series of seismic political blows at the worst possible moment. In September 2013, the Bridgegate scandal broke and increasingly sidelined Governor Christie. My team went from coordinating with his team and political allies in Newark (he had a lot)

on a weekly basis to going months with virtually no communication. Shortly thereafter, then senator Frank Lautenberg tragically passed away. Mayor Booker, who had also been an active and strong supporter of the plan and was working hard to build momentum around it, announced he was running for that US Senate seat. This not only effectively took him off the field from a local political standpoint, but also created a scenario where he needed the support of local officials and union leaders who opposed many parts of the plan. His announcement also spurred the need for an earlier-than-expected mayoral election where the leading candidates spent considerable time spewing hatred about charters and about me personally (although backstage and publicly, they had previously supported both). Shortly thereafter, Commissioner Cerf resigned. To use a sports analogy: the entire offensive line left the field.

The overall approach was comprehensive, and it had to be to ensure that none of our kids were trapped in failing schools, the district didn’t go bankrupt, communities weren’t living with vacant buildings, and the city was on a path to success. I described the plan to author Dale Russakoff as “three-dimensional chess” in an effort to convey why all the pieces had to happen at one time and couldn’t be phased. There were too many interdependent parts to a very complex system, and the stakes couldn’t be higher. Unfortunately, in her book about Newark, The Prize, which went on to become a bestseller, this quote fed an inaccurate portrayal of me as a top-down, cold technocrat—a narrative that was taking shape across much of the media coverage about our work in Newark. It couldn’t have been further from the truth—the emotional pieces of what needed to happen were not lost on me or the team. I lived with my husband and baby son in Newark and had conversations with neighbors in grocery stores and local watering holes on a daily basis. It all felt so heavy, but also necessary.

Results and lessons

During my tenure and the subsequent years under Cerf, our district teams improved outcomes for students in every neighborhood and every age group—from early childhood to high school.

In early childhood, we secured a $7 million Head Start grant48 (only the second district in the country to do so) to add more than one thousand early childhood seats. We brought early childhood standards to life and sounded the alarm to focus on the importance of high-quality early learning. Newark went from having fewer than half of our residents eligible for free early childhood programs (which was most families) to enrolling nearly 90 percent.

In 2015, the Center on Reinventing Public Education named Newark as the number-one district in the country for high-poverty, high-performance elementary schools that beat the odds. By 2019, more than one-third of Black students attended schools that exceeded the state average, compared with 10 percent in 2011. A study conducted by Harvard University showed across-the-board increases in reading and initially slow but then impressive gains in math. The number of good schools and schools “on the move” grew every year due to our district-run turnaround approach, charter conversion schools, and some outright closures and consolidations. Newark was among the top four cities in the country for student outcomes of Black students living in poverty.

The citywide graduation rate rose 14 points, closing the gap with the state average by 7 percentage points—with almost double the percentage of students graduating having passed the state exit exam. About 87 percent of Newark graduates returned for a second college term, far exceeding national averages given the high poverty rates.

And we saw signs that the overall community—despite the political rancor we encountered — was starting to believe in the “system of great schools.” For the first time in decades, student enrollment was increasing overall in Newark, as was the population of the city.

Our labor agreement, too, was a long-term success. More than a decade later, most of these terms still exist in the contract today, and an independent study of the agreement found that the “new evaluation system is perceived as valid, accurate, fair, and useful.” This suggests that its durability is not just due to luck and that the approach could and should be replicated elsewhere.

Similarly, universal enrollment still exists in a modified form to this day and is used by nearly 20 percent of families. I believe managed school choice was starting to play a role in dissipating the city’s deeply concentrated poverty.

Despite these significant accomplishments, we knew from day one that we would not succeed in Newark unless everyone, from the grandmother at a profoundly struggling school to the dad at a magnet school to the aunt at a charter school, believed things could be different — better for everyone, for Newarkers. We knew we had to build a completely new normal and that some of that work involved helping the entire community see that it didn’t have to accept the failed status quo. We exerted a lot of effort that, in the end, fell short of generating the kind of collective momentum we needed. The reasons are complicated but instructive.

Because we felt responsible for every child in Newark, we engaged all families, charter and district, with equal vigor. This was a good and mission-aligned approach, but it was almost impossible to execute, given the tensions (both perceived and very real) inherent in growing the charter footprint. The conundrum is perfectly exemplified by the mother who called in to ask me a question on-air during a local NPR show. She had just dropped off her kids at North Star Academy Charter School, she said, because she needed them to have access to excellence. At the same time, she was on her way to my office to picket against me on behalf of her nephew, who had lost his job as a school aide due to the smaller footprint of the district.

Our strategy all along was to be up front about failure and embrace accountability. Again, while our radical transparency seemed like a good idea on its face, it turned out that a lot of people don’t want to hear their school is failing—no matter how carefully crafted the message. Also, while some community members were grateful that someone was “finally telling the truth” (an actual quote from a community meeting I led at a failing school), others were understandably angry. Our team was on the receiving end of the grief and loss that result from telling the patient they have stage IV cancer when someone should have caught it years ago.

We were lucky to have a popular Republican governor in Christie and a Democratic mayor in Booker, who teamed up to create a real mandate for change and put a laser focus on what’s best for students. This was a tremendous asset (and seems unthinkable in today’s environment) but also a challenge. Some Newarkers resented the involvement of the state (particularly in managing the school system directly) and, by extension, me. And local officials fought even harder to exert influence, sometimes over off-the-mark things, to show they were relevant.

We prioritized students who were at the back of the line. Our universal enrollment system gave preference to students from the poorest neighborhoods and those with disabilities. We revamped the magnet school admissions process to look at multiple factors for student admissions at the central office. These were good decisions for children, families, and equity, but it also put us in the crosshairs of power brokers who were used to getting what they wanted and considered coveted seats theirs to give out. They also had access to the biggest microphones and would use them to mobilize the community against our efforts.

Some charter school operators and their supporters mobilized their constituents in opposition to these citywide efforts as well. They wanted to grow where they wanted to grow, not necessarily in alignment with supply-and-demand patterns or the overall plan. Many (not all) were content to crow about how much better they were than the district without digging into what percentage of “high-need” students they were serving—conveniently avoiding an apples-to-apples comparison that was much more complicated. They enjoyed promises from politicians and funders that were out of alignment with the One Newark collective plan. Many liked running their own lotteries because they had more control over admissions; some would say things like “we don’t offer that kind of special education program.” Also, many had legitimate concerns about turning over enrollment processes to a district that had been underperforming and had actively sought to extinguish them for many years.

Charters weren’t the only group stuck in their own goals and plans—and at least most of their concerns were in service of building quality schools. School-based partners and vendors, local nonprofits, funders, and other leaders all had their individual projects, schools, and pet issues. The incentives to keep doing one’s own thing were profound. I was stuck in a daily loop of explicit and often threatening demands to support individual agendas—many of them having nothing to do with what was best for individual neighborhoods and schools, let alone the collective. A local reporter continually nagged me about shoring up my “natural allies.” I remember wondering who they were. Breaking up monoplies and pursuing third-way ideas is a lonely endeavor, particularly in cities like Newark, with its transactional, machine politics.

Well-resourced forces of opposition spent a considerable amount of money spreading misinformation and actively attacking me personally. They made expensive sandwich boards, posters, and fliers with my face on it. In one image, the word “Liar” was printed as if it were carved into my forehead and dripping blood. It was an open secret that they hired a full-time blogger to write stories about our work and about me personally (some with twists of the truth and some with outright lies). The blog was well formatted, looked like a real newspaper, and generally contained kernels of truth that were leaked from inside. Ads were purchased to place those stories in actual newspapers so that they looked like real news. Canvassers were hired to distribute leaflets about false school closures, and social media stalkers posted where my family and I were eating dinner.

As important as we knew collective buy-in was to our success and as much time as we invested in it, our team was ultimately not successful in creating a groundswell quickly enough. We certainly had moments, but not enough. The One Newark plan should have been envisioned before the unintended consequences were at our doorstep. Maybe that would have given us more time. Surely, I was the wrong messenger: a White woman from out of town who represented the system. One Newark could have been a third-party entity with representatives from various sectors and a trusted, local leader. I thought this all along but failed to get stakeholders to agree and execute quickly enough. Meanwhile, we had to balance the budget and ensure quality in education.

I also clearly made mistakes. My messages were not straightforward and sticky enough. This work, as you can see, is complex and multifaceted, and I could have paid more attention to how to ensure good, proactive, community-friendly communication. I did not lead our team in good enough ways, small and large, to predict and combat misinformation that was rampant and that got even worse as social media exploded during my tenure. The forces for the status quo were organized and mobilized, and we were caught flat-footed. I didn’t manage the flow of information with nearly enough precision, let alone attend to building my own brand. I made a classic mistake that many leaders have made before me: I presumed that if I did good work and led with authenticity, people would support progress.

More critically, I poured valuable energy into the community without focusing as much as I should have on the community influencers closest to our work at the school level. Since then, I’ve developed a more sophisticated understanding of how to see the community in relation to the system of schools. In figure 1, the center is the school, and the next level out is the families and students (red ring). The next ring is influencers (orange ring)—folks connected to the school who have direct influence on that specific school. The next ring is community-wide partners (yellow ring)—community-based agencies and other city agencies like police and child welfare. And the next ring out (green ring) is elected officials and power brokers—for instance, pastors of large congregations, thought leaders, and community-based organizations serving the city.

We knew it was critical to focus on our families and students, and we knew it was a tremendous amount of our work to build collective action focused on them. I give us high marks for our dogged and strategic work on the red ring. But in retrospect, we spent far too much time with folks in the outermost ring—the political and power class—and not enough with those in the orange and yellow rings. It wasn’t until nearer the end of my tenure that we started to create a database for each individual school’s yellow ring. We also made a much more concerted effort to know the civil servants at all of our partner agencies. I came to realize a hard lesson—that while the politicians and power brokers confidently spoke for the community, they were often after a political win: a contract, a coveted spot in a school, a policy, or a job for a family member or friend. I wish I could take precious minutes I spent with those in the green ring and reinvest them in the yellow and orange rings.

The painful but informative experiences I had in Newark, along with a long career since then of working with systems leaders across the country, have convinced me that collective action is the missing link for change at the systems and community levels. Sadly, it is also the element I find most commonly reduced to uninspired bumper stickers and is wholly disconnected from the powerful and real work of reform. We talk about “community engagement” without honestly defining who the community is. We talk about consensus when real and hard calls have to be made every day about managing access to scarce resources, coveted high-quality seats, and community-based jobs. Many of those decisions can be in direct tension with building a system of great schools in every neighborhood by any means necessary. We interchange concepts of true grassroots organizing with community engagement and sidestep the obvious truth that power brokers and special interest groups have an organized, well-resourced, and often outsized influence on speaking for the community.

Recommendations for systems leaders

While the work we did in Newark has been treated by many in education reform as a critical case study for emerging superintendents and district leaders, it is often told as a cautionary tale of doing too much and ignoring community engagement. What is often lost are the lessons for building a successful “systems of schools.” Fundamentally, the story of the One Newark plan is a story of a district seeking to break out of the shell of its narrow school footprint and hold itself accountable for the educational futures of all its city’s children. The story of Newark should push all of us to define the role of the “system” and why it is so critical and yet so difficult to fulfill that mandate for an entire community.

Since A Nation at Risk was published, we’ve had important attempts at systemic reform that focused on specific pieces of the puzzle. Sadly, many of those efforts focused narrowly on individual “ingredients” (to use my earlier analogy) but not on the whole recipe, or the whole system.

We’ve had a lot of reforms focused on “community engagement” approaches, like the Annenberg Challenge. But those initiatives failed to address the fundamentals of building better schools and didn’t wrestle with the extremely tricky work of defining community that is illustrated in the Newark example.

We have also seen many efforts to build better individual schools; most of the charter and “high-quality schools”’ movement has been about that: creating individual proof points without thinking about the layer above the school, let alone the community. New Orleans shows the profound limitations of this strategy, as the city still struggles to figure out the role of the system after watching a bunch of individual charter operators solve some problems and create new and complicated ones. The Newark case study illuminates that while this approach can be vital for building “four-ingredient” schools, it will always be insufficient for establishing a holistic system of great schools.

We must focus on creating systems of great schools—not great school systems, not individual schools alone, not one piece of a puzzle, not some simplistic version of community engagement. We need to get clear on the roles of leaders at the systems level. In short: the system should manage the incentives, policies, guardrails, and resources to ensure that every child has access to a four-ingredient school by doing four things.

1. Enable ‘four-ingredient’ schools

As discussed above, we have promising practices when it comes to ensuring a gamechanging principal in every school and an excellent educator in every classroom. We know the impact of high-quality instructional materials that are culturally competent. We have proven research on the importance of school culture and handling discipline. We know what conditions have to be in place to enable achievement. Systems leaders should set direction and advocate; procure best-in-class materials; set policy to incentivize districts, schools, and charter management organizations to implement what we know works; and sanction practices antithetical to student progress. As one example, when Mississippi focused on the science of reading, providing best-in-class materials, training, and a way of measuring progress, kids across the state started reading at unprecedented levels. As another example, Nebraska adopted high-quality instructional materials statewide and provided options but also high-quality implementation support. This drove impressive gains in student outcomes.

Too often, cities select a superintendent or states select a commissioner because they are zealously focused on one single ingredient. Sometimes cities and states simply won’t touch an ingredient because they don’t want to fight with the union or other interest groups.

Policymakers have a responsibility to ensure that schools can obtain and mix best-in-class ingredients more quickly—trying to do so one school at a time doesn’t make sense. A lot of this work happens at the district and network levels, but leaders at all levels must put people in place who understand and are committed to all four ingredients.

2. Ensure quality and equity

Our current system of districts versus charters sadly guarantees that many kids—particularly those with the most challenges—are left behind. Policymakers and community leaders should be held accountable for allowing kids and families to fall through the cracks.

Leaders need to step up and raise their hands for being accountable for all kids to access high-quality schools. They need to embrace good enough ways of measuring what that means—in terms of what students are learning and how they are feeling. Accountability systems need to help families hold schools to a standard of excellence for all kids, including those who consistently fail in all kinds of schools. These accountability systems should be family friendly and public. And they should explicitly shine a light on where inequities show up: fewer Black students having access to AP course and magnet programs, special education students in segregated classrooms with abysmally low student outcomes, inequitable criminalization of student behavior, and kids living in concentrated poverty too often getting the lowest-quality staff. Our new accountability systems should correct for some of the mistakes we made before, from focusing only on proficiency and meaningless graduation rates to treating growth, college-readiness, and retention as critical outcome measures.

Effective accountability systems have incentives that inspire schools and communities to step up. As one recent example, schools that were designated as failing were more than twice as likely to make big gains than those that weren’t. Researchers surmised that this is because the label drove resources and supports where they needed to go—and rallied communities to do better. Good accountability systems drive decisions, sometimes hard ones, about redesigning schools, radically changing who runs them and how they are run, and even closing them.

3. Break up bureaucracy

A fundamental way to clear a runway for accelerated school improvement is to actively tear down past practices and federal, state, and local policies that block individual schools from innovating. As one example, school finance formulas are unnecessarily complicated and opaque. Most states have an even worse and more complicated approach to funding facilities and infrastructure. We need more of a “whiteboard” approach than one that tweaks decades of dysfunction.

Traditional schools will never be able to “compete” with charters if we don’t actively tear down the unnecessary bureaucracy in existing schools. As one example, when Klein was New York City Schools chancellor, he had an entire team dedicated to creating one-stop compliance and communication approaches for principals in New York City over multiple years. They shrank thousands and thousands of pages of sometimes competing “mustdos” (most of them having nothing to do with serving students better) and still felt there was more to do.

Policymakers and community leaders need to wake up every day wondering what they can do to ensure that people running schools have the time to do the right thing as opposed to managing byzantine policies and procedures from competing departments. We certainly need oversight and compliance, but it must be streamlined and reexamined every year.

4. Create cross-system and community-based solutions

The students who face the most challenges have generally been failed by multiple systems. I have a term for this: students whom systems have failed the most (SSFMs). Statistically, they are likely to be students of color. Too often they are labeled “special populations” and further marginalized out of classrooms and into separate and unequal programs. Very few schools of any governance structure meet the needs of these young people. Schools—and the systems in which they operate—are consistently failing 20 percent of their most vulnerable students.

Often these students and their families are connected to multiple systems: child welfare, public housing, homeless services, juvenile justice, criminal justice, immigration services, family services, and food programs. Some examples: nearly 90 percent of the juvenile justice population were in foster care at some point in their lives. About one-third of thirteen-toseventeen year-olds experience some sort of homelessness. These students struggle tremendously in school. In other words, we take the kids who—through no choice of their own—have been failed by one system and then fail them in another.

In order to truly reverse patterns for students that systems have failed the most, we need crossagency and community-based solutions with school success at the core. Neighborhood-based collaboratives, like the Harlem Children’s Zone, have produced promising results that we started to scale during Arne Duncan’s tenure as secretary of education. I’ve had the pleasure of working with teams creating memorandums of agreement between disparate agencies—the DA’s office, probation, public housing, and school systems—to share data and create common family support plans for young people and families connected to multiple systems. We need more out-of-the-box ideas to aggregate services and help students who are the most vulnerable succeed.

Conclusion

The insights and recommendations I’ve shared above are not based on any specific ideology. They were developed out of necessity and refined through years of application and practice across a wide variety of settings—from New York to California and many places in between, in both districts and charter networks, in small-school communities, and in the largest cities and states.

It may seem like a lot to tackle, and indeed it is. But if we are to truly transform our systems at scale, we can’t simply cling to one specific ingredient or hew to a single governance ideology. The surest way to avoid bias and ensure a holistic strategy is to zoom out to the communitylevel goal. Make the community—not just one school, network, neighborhood, or district—the unit of change.

When I arrived in Newark, we started by asking how to ensure one hundred excellent schools to educate every child—including and especially those who are typically left out of the conversations about excellence. Moreover, how could we do it as quickly as possible (because kids only have one third grade)? Many education reformers at the time were looking for silver bullets—debating whether charters are better than traditional schools or what makes the perfect teacher or curriculum. Our approach of stepping back and asking what the shared goal was for the entire community led us down a fundamentally different path. That path was inherently third-way and therefore had us at odds with hard-line choice advocates, status quo defenders, and other rigid ideologues. But it also kept us focused on the community level and (on our good days) prioritizing those members of the community closest to the center of the circle.

Our efforts in Newark stood on the shoulders of emerging research and promising practices from around the country. Our team’s focus on talent at every level was inspired by work in New York City, Baltimore, and Washington, DC, and in organizations like the New Teacher Project and New Leaders for New Schools. We took lessons from Student Achievement Partners, New York State, Charlotte-Mecklenburg Schools, and high-performing charters when it came to focusing on high-quality instructional materials. Our focus on high expectations and high support cultures was informed by research on achievement motivation, the intersection of adult expectations and student outcomes, and Chicago’s work on measuring student and family satisfaction. Our citywide plan built on work from the new small schools
movement in Chicago, New York, and Denver and universal enrollment in New Orleans and Denver. Throughout, we aimed to take the best of what was working elsewhere and ensure that it met the unique needs of our comprehensive citywide plan in Newark.

When leaders make the commitment to put community at the forefront of their work, they will encounter a number of challenges. In closing, I’ll offer four guidelines to help us move forward.

First, think about the system of schools, not the school system. As I said before, this sounds like merely a semantic difference, but don’t be fooled. Anytime we find ourselves putting aside what the broader community needs and instead focusing on what the district or network needs, we may achieve some short-term gains for the school system but often at the cost of the long-term goals for the community’s system of schools. Continually reminding ourselves of this important distinction can go a long way toward preserving our focus.

Second, embrace a better, more honest definition of community. It is critical that the bounds of the community feel authentic to the members of that community—geographically, culturally, and politically. It is often convenient for politicians and others in positions of power to leave out members of the community to allow special interest groups to frame the conversation. In the planning and execution of systemic work, we have to put at the core of the work the very members who have been consistently failed by the system itself. When we are honest about who the real stakeholders are in our community and have clear priorities, the opportunity for real systemic change becomes possible.

Third, reject the idea that we have to start from scratch. In this chapter, I have identified many policies and practices that worked, some of which have been tossed out because of politics. The past five years have been a time of extreme polarization in all areas of public discourse, with education as no exception. An emerging playbook was beginning, but it has been all but obliterated. We need innovation for sure, and we also need a clear-eyed assessment of what didn’t work, but we aren’t starting from scratch. Kids don’t have time for us to reinvent the wheel.

Fourth, accept that this is messy and that revolutions are never quiet or fast. As much as I’ve tried to codify my lessons and experiences into an actionable, coherent framework, nothing gets around the fact that transformative and disruptive systemic change will never be quick or tidy. This is not outpatient laser surgery that leaves no scar; this is chest-opening, quadruple bypass surgery with a lot of risks and long-term effects. But like a good doctor’s goals, our charge is to help the patient lead a long, healthy life. We also cannot expect immediate results. Real change takes time. The ideas and epiphanies in this chapter I share humbly and with tremendous gratitude to the countless friends, colleagues, and mentors in this sector who helped shape my beliefs about this work. It’s been more than a decade since I arrived in Newark and forty years since A Nation at Risk. My hope is that we’ve all gained a bit of useful perspective and are ready to roll up our sleeves and put the lessons we’ve learned into action.

See the full Hoover Institution initiative: . 

Moderated by Stephen Bowen, executive director of the Hoover Education Success initiative, the conversation features Hoover Visiting Fellows Corey DeAngelis, Anna Egalite, and John Singleton. Watch the full conversation, and check out our past coverage from School Choice Week: 

• 40 Years After ‘A Nation At Risk,’ How School Choice Policies Are Reshaping American Education

• Next Wave of Microschool Founders are More Diverse, Less Likely to be Educators

• New Analysis Finds Charter School Sector Still Has Plenty of Room to Grow

• Seizing on Parents’ Frustration, GOP Governors Push for Education Savings Accounts

• The New Face of Homeschooling: Less Religious and Conservative, More Focused on Quality

The past forty years witnessed several significant initiatives in K–12 education policy in the United States. These include school finance reforms to ensure equity and adequacy of funding as well as the adoption of school accountability policies (exemplified by the No Child Left Behind Act) linking measures of student success with sanctions and rewards for teachers and schools. A major development was likewise school choice: public policies that provide families with increased access to schooling options other than a neighborhood-assigned public school.

Although school choice policies in the United States take several forms in practice—this chapter surveys findings from public school choice (e.g., “open enrollment”), charter school, and private school voucher programs—the school choice movement is unified around a common logic and set of policy goals. Its intellectual heritage typically is traced to economist Milton Friedman’s argument that, while government rightly should fund primary and secondary education, a robust system where substantially more private providers also produce and deliver that education and where the funding a school (public or private) receives is directly linked to its enrollment could be more efficient and more equitable (Friedman 1955). As with the debates over school finance and school accountability, a central issue is the low productivity of public schools in the United States in comparison with other nations. School choice policies distinctly diagnose this problem as a symptom of monopoly power. Poorly performing public schools are the logical product of weak incentives to supply a high-quality education in this view. Introducing choice and therefore competition—to K–12 education follows as a remedy.

Watch: A Special Hoover Institution Panel on the State of School Choice Policies

This broadly economic rationale connects with two other motivations for the introduction and expansion of school choice policies. The first can be described as a pragmatic appeal that has historically proven important: given the entrenched power of the public education bureaucracy in the United States (e.g., the political influence of public teachers’ unions), reform from without—as opposed to from within—has often proved more feasible. In other words, rather than attempting to reform public education from the “top down,” school choice policies advance reform in a more decentralized, “bottom-up” fashion. The other motivation recognizes that the students who are most underserved by public education in the United States are disproportionately disadvantaged. In particular, it is families of low socioeconomic status who otherwise effectively lack school choice; they often do not have the resources to pay for private school tuition or to move to affluent suburban neighborhoods with good public schools. From this perspective, school choice policies stand to level the playing field for low-income and minority students both by improving access to high-quality schools and by spurring improvement of public schools.

This chapter surveys the present landscape of school choice policies in the United States and assesses what global experience so far suggests about their potential for reforming public education. This experience includes both several decades of state- and district-level choice policies in the United States as well as the experience of countries where choice has been introduced (e.g., Sweden and India) or has long been a feature of the national education system (e.g., Chile). A major point of emphasis is that the potential of school choice policies as a large-scale education reform rests in large part on the degree to which they improve the quality of education in traditional public schools. Such learning spillovers are likely to depend on how school choice programs are designed and may or may not be positive.

The chapter then summarizes key elements of the major school choice policies present across the United States, with a focus on the incentives that funding and accountability provisions may or may not create. These policies include public school choice options, such as magnet schools and open enrollment policies, as well as charter schools, which are operated independently but are publicly funded and tuition-free. While vouchers (and voucher-like programs) that defray or reduce the cost of tuition at private schools have attracted renewed policy interest recently, a much greater share of current public funding for school choice flows to charter schools: more than ten times as many students attended a charter school in fall 2019 as used a voucher to attend a private school. Charter schools educated around 6 percent of all students. Survey numbers also suggest that around as many students “chose” their public school (as opposed to attending an assigned one).

This leads into a critical appraisal of the empirical literature to date. The review focuses primarily on the evidence regarding the effects of school choice programs on public school students. It is argued that a consensus, built on the findings from both domestic and international programs, supports the view that school choice policies usually—but not necessarily—have positive spillover effects and that they do so by causing public schools to raise their quality in the face of competition. The chapter then highlights several important gaps in current knowledge. These include the need to understand the sources and significance of heterogeneity in impacts across different settings and programs and for greater evidence of effects on long-run student success (such as education attainment). There is also a need for work that considers the implications of the growing reality in the United States where several different school choice policies can be present at the same time.

The last sections of the chapter reflect on takeaways for school choice policy moving forward and on challenges of evidence-based policymaking in a post–COVID 19 future that is likely to feature greater school choice. One conclusion is that policymakers should consider and leverage incentives that stand to intensify competition. Specific applications include providing public funding for transportation to nonpublic schools; producing and supplying families with information about schools and their impacts; and targeting vouchers to those students who are underserved by public education.

Orienting the focus toward education quality in the aggregate

How should school choice policies be evaluated as successful or not? The perspective adopted in this chapter is that their impact on the quality of education, understood as the level of skills that students acquire in schools, should be the foremost focus. An important point is that this focus accommodates a broad conception of skills that encompass cognitive and other (noncognitive) aptitudes; it is not reducible to academic achievement. While measurement raises crucial questions returned to later, skill production in schools as a central focus has two advantages. The first is that it serves to summarize in terms of student outcomes the multitude of individual elements that constitute a schooling environment —for example, the quality of the teachers, the culture and leadership, the curriculum, and so on. The second advantage is from the perspective of policy: it provides a common basis for comparing the effectiveness of school choice policies with other major K–12 education reforms.

One way the question is often framed of whether school choice policies are successful or not is from the perspective of parents and caregivers: “Will sending my student to a charter school be better than sending them to their assigned public schools? Will using a voucher program to send my student to a private school be better than sending them to their assigned public schools?” This framing connects with a mature empirical literature and body of evidence that asks whether charter and private school alternatives provide higher education quality than traditional public schools. A fundamental empirical challenge for research is self-selection—the fact that the students who choose to use a voucher or switch to a charter school are not randomly drawn from the population—and scholarly work has advanced several approaches to address this. Important findings in the literature include that certain charter schools (many aligned with “No Excuses” practices) generate dramatic improvements in students’ academic achievement and whether they go on to college (e.g., Angrist et al. 2016; Dobbie and Fryer 2020); that recent national data sources show learning gains of charter school students exceed those of traditional public school students (Shakeel and Peterson 2021; CREDO 2023); and that voucher recipients in statewide programs in several US states perform lower than comparable traditional public school students on state standardized tests (e.g., Mills and Wolf 2017; Abdulkadiroğlu et al. 2018).

From a policy perspective, however, some of the attention to whether school choice alternatives are higher quality is misplaced. This is because the promise of school choice policies as large-scale education reforms rests in a fundamental way on how those policies impact students who choose to attend traditional public schools. In theory, school choice policies can increase education quality in the aggregate by creating incentives for public schools to raise their productivity. Analogized to a “tide that lifts all boats” (Hoxby 2003), this competitive response prospectively redounds to the benefit of students who remain in public schools. The key point, emphasized in the next section’s survey of the school choice landscape in the United States, is that those students who participate in choice programs (i.e., use a voucher program to attend a private school or choose to attend a charter school) are a minority of all students. As a consequence, even small spillover effects on the comparatively larger share of students who remain in public school will matter more in the aggregate. A numerical example helps illustrate this point: if 10 percent of students used a voucher to attend a private school and their learning was consequently advanced by six weeks each, education quality on average would be negligibly improved if there was zero impact on the 90 percent of public school students. While the voucher program in this example might pass a cost-benefit analysis, the point is that it would be difficult to justify the attention of policymakers concerned with developing and implementing policies that address the systemic failures of K–12 public education in the United States.

A growing literature, drawing on evidence from programs in the United States as well as international findings, tests for competitive effects of school choice policies. School choice policies can have negative spillovers, too, however, and a central empirical question concerns effects on resources and student composition in public schools and separating those effects from changes in public school productivity. A major empirical challenge for this branch of the literature is that the public schools that are exposed to competition are likely not random. The review of the empirical literature below summarizes the major findings about spillovers of school choice policies.

The school choice policy landscape in the United States

There are various policies that create or expand access to school options across the United States. This chapter provides a brief overview of three broad categories: (1) policies that provide choice among public schools (e.g., open enrollment and magnet programs); (2) charter school laws that allow for the creation and operation of publicly funded and tuition-free schools by independent organizations; and (3) voucher and voucher-like programs that defray the cost of tuition at private schools for qualified students.

Public school choice

Public school choice policies offer families choice among public schools. One example is magnet schools, which are typically district run, feature specialized curricula, and often have selective admissions. Students must apply to be considered for admission. Examples include “exam” schools such as Stuyvesant in New York City, Thomas Jefferson High School for Science and Technology in Fairfax, Virginia, and many schools of the arts around the country. Another example of public school choice is open enrollment policies intended to provide a process whereby students can transfer between public schools. Most states allow school districts to voluntarily accept students from other districts, and twenty three have policies in place that make such interdistrict transfers mandatory (Education Commission of the States 2017). Transfers within a school district are mandatory, subject to limitations, in nineteen states.

Choice mechanisms are a kind of open enrollment policy that facilitates within-district public school choice. Such mechanisms allow students to submit ranked lists of school preferences, which are then aggregated (and can be combined with preferences identified by schools and specialized programs and other criteria) to assign students to schools via an algorithm. An example is Denver’s SchoolChoice process, first put in place in 2012, where families (most with students in either kindergarten, sixth, or ninth grade) rank up to twelve schools, and matching priorities include neighborhood zones and siblings. School districts often pair choice mechanisms with guides for families to understanding the ranking and allocation process and that detail aspects and attributes of the various schools and programs available. While public school choice mechanisms expand choice options beyond charter or private schools—and introducing choice into the system may be a response to the availability of charter or voucher options there is greater ambiguity about the theoretical system-level implications of magnet, specialized programs, or open enrollment policies on education quality. This is because choice between public schools does not obviously create meaningful competitive incentives: whichever public school a student chooses to attend, they (and the revenue attached to them) ultimately stay in the public system.

It is difficult to provide an exact number as to how many students in the United States “chose” their public school via a policy option (as opposed to being assigned it based on residence). Pre-2020 survey numbers from the National Center for Education Statistics (NCES) imply a figure of around three million students (about 6 percent of all public and private enrollment) (Hanson and Pugliese 2020). A number of large public school districts have at least limited school choice mechanisms in place, including New York City, Los Angeles, Indianapolis, and Charlotte-Mecklenburg (in North Carolina). These programs can take different forms. In some cases, magnet program and charter school options are integrated into the same choice mechanism as for traditional public schools.

Charter Schools

Charter schools, in contrast, are essentially privately operated public schools. Charter schools are supported by tax revenues, are regulated by government entities, do not charge tuition, and cannot have selective admission criteria. They also participate in school accountability programs (i.e., standardized testing) and can be closed by public school authorities for poor performance. At the same time, they are created, operated, and managed by independent organizations and, like private schools, have considerable autonomy when it comes to decisions over curricula, human resources, and where to locate. Charter school operators are generally nonprofit organizations (well-known examples include the Knowledge is Power Program, Rocketship, and Success Academy), though for-profit management companies operate in several states.

How (and at what level) charter schools are funded has long been a major point of debate. This is in part because most per-pupil funding for charter school students moves out of the public school system to the charter school the student enrolls in. Thus, the fiscal implications of school choice can be immediate for public schools in the case of charter school competition. From an economic perspective, this means that the presence of charter schools in principle creates a meaningful incentive for districts to retain students (though public school leaders may not act on those incentives). In several states, however, these incentives are muted by “hold harmless” provisions that at least temporarily offset the full financial impact of enrollment lost to charter schools. Public school leadership, at both district and state levels, has at times been hostile to charter school growth in the face of fiscal impacts, and charter schools are a frequent target of public school teachers’ unions. At the same time, the amount of funding is also typically disparate—charter schools generally receive less revenue per pupil than is spent on a student in a traditional public school—and this disparity has itself been (and continues to be) a focus of policy attention.

About 3.4 million students in the United States attended a charter school during 2019–20. This equates to more than 6 percent of all school enrollment and reflects steady growth; charter schools enrolled about 4 percent of students in 2010–11. The average charter school student is much less likely to be White and more likely to be economically disadvantaged than the average public school student (NCES 2022). Although all but four states have charter school laws in place, charter school penetration varies considerably across places. In several states, more than one in ten non–private school attendees attended a charter school in 2019, including Washington, DC (45 percent), Arizona (18 percent), Colorado (14 percent), Louisiana (12 percent), and Florida (11 percent) (NCES 2022). Note that these numbers mask major within-state heterogeneity: for example, about 25 percent of Miami-Dade non–private school students attend a charter school. In New Orleans, virtually all students who do not attend private schools attend charter schools.

Vouchers

The last variety of school choice programs highlighted in brief here are those that defray the cost of tuition at private schools for eligible students. These programs are generically referred to as “voucher” policies but can take several different forms across the United States. These forms include prototypical government-funded programs that cut families a check; tax credit scholarships—which deputize nonprofits to receive tax-advantaged contributions and provide scholarships to students; and education savings accounts. Education savings accounts instead allow families to allocate a given sum of government funds across education expenses, including private school tuition.

Voucher programs have historically been means tested, whereby eligibility is restricted to students in families whose income does not exceed some threshold. A number of programs are targeted specifically to students with disabilities, and a few restrict or expand eligibility to students assigned to persistently low-performing public schools. Similar to the revenue disparity between traditional public schools and charter schools, effective voucher amounts are typically well below per-pupil expenditures in public schools and often do not cover the average sticker price at private schools. For example, Ohio’s means-tested statewide vouchers are worth $5,500 at the elementary and middle school levels for 2022–23. Other policy parameters concern accountability provisions of voucher policies. At issue are typically
three aspects of the programs: (1) whether participating private schools may apply their own admission criteria or not (with seats in oversubscribed private schools allocated by lottery); (2) whether private schools can require supplemental tuition from voucher recipients; and (3) whether voucher recipients must take the same achievement exams as in public schools.

While the country’s first voucher program (which began in Milwaukee in 1990 and served nearly thirty thousand students in 2022–23) requires lotteries and allows tuition supplements only at high schools, almost all major voucher programs instituted in the United States in the past twenty years relax these provisions.

In fall 2019, about 4.7 million K–12 students were enrolled in a private school nationally (NCES 2022). This level, which is a little less than 10 percent of all students, has been relatively constant over recent. years. However, the estimated share of private school enrollees who are part of a voucher or scholarship program has grown. More than five hundred thousand students (about 1 percent of all public and private school students) attended a private school via a school choice policy in 2019—more than double the number ten years prior (EdChoice 2023). As of 2023, thirty US states (plus Washington, DC) had some kind of voucher program in place. The country’s largest voucher program, Florida’s Tax Credit Scholarship, began in 2001 and served more than one hundred thousand students as recently as 2020–21. While private school students as a whole tend to be more advantaged (and less likely to belong to an underrepresented minority) than public school students, those students who use vouchers— reflecting in part the income-based eligibility criteria for such programs—are generally less advantaged and relatively more minority (Wolf 2020).

What do we know?

What do we know about the externalities of school choice policies on the learning of students who nonetheless attend a traditional public school? This section considers the evidence on this question. The topic of spillovers juxtaposes countervailing forces. On the one hand, incentives to retain enrollments may lead public schools to respond to school choice policies by raising quality, as intended. On the other hand, expanded school choice may negatively impact students who remain in public schools through a resource or peer effect channel. If, for example, more advantaged or higher-achieving students leave public schools (and exposure to such students benefits their peers), the learning of students left behind will suffer. There is also an important concern that low information about school quality (or preferences for other attributes of schools) could incentivize responses by public schools that are unrelated to impacts on learning.

The body of rigorous evidence to date supports the conclusion that spillover effects on student learning are usually—but not necessarily—positive on net. The pieces of evidence supporting this are twofold. First, the net effect on students remaining in an assigned traditional district school is generally zero or positive; this implies, and more direct evidence affirms, that negative externalities from peer sorting are limited. Second, the patterns for net improvements in student outcomes are consistent with increased competition causing public schools to raise their productivity.

A recent review of the findings from private school choice programs in the United States highlights that most studies find positive evidence of competitive effects and that no study finds evidence of negative effects on students attending traditional public schools (Wolf 2020). Figlio and Karbownik (2016), to highlight one example, compare two sets of Ohio public school students: (1) those in traditional public schools who marginally exceeded a score threshold (based on standardized test performance) below which all students would become eligible to receive a voucher; and (2) those in public schools who marginally failed the score threshold. The latter group, who were exposed to private school competition but should otherwise be on average the same as the first group, showed greater test score growth. The body of evidence on spillovers of charter schools on public school students is more ambiguous. Older findings drawn from a variety of states suggest that test score impacts are limited. One important study, by Imberman (2011), reports negative impacts from charter schools on the achievement of students in traditional public schools. The more recent evidence is generally more positive. Studies that find improvements in public school students’ test scores include Cordes (2018) for New York City, Gilraine et al. (2021) for North Carolina, and Ridley and Terrier (2022) for Massachusetts. The New York City study compares student test scores between traditional public schools exposed to charter school competition at different times and at different distances. These studies of US programs are accompanied by results showing increases in public school student outcomes following the introduction of vouchers in 1992 in Sweden (Sandström and Bergström 2005) and, despite increases in stratification that likely reduced average student academic potential in public schools, no changes associated with entry of private schools in India (Bagde et al. 2022).

Two kinds of evidence additionally support the claim that increased competition can cause public schools to raise their quality. The first kind is from settings where competition increases but in which other responses cannot arise. Figlio and Hart (2014) is a foremost example: students could apply for Florida’s means-tested voucher one year in advance of the program launch. The authors present evidence of increases in test scores at competitively exposed public schools in even this “pure competition” year where students could not yet re-sort. Implementing the same idea, Gilraine et al. (2021) provide evidence for competition in the context of charter schools in North Carolina. The second kind of evidence is drawn from models of school choice that generate predictions for how much competitive pressure different public schools experience. Gilraine et al. (2023), for example, show (1) that demand for charter schools offering a “nontraditional” curriculum is not very sensitive to the quality of public schools; and (2) that, as would be expected, the quality of public schools does not increase following nearby entry by a “nontraditional” charter school. They further show that the test score value added of public schools does increase following entry of math and reading skills–focused charter schools. Campos and Kearns (2023) take a similar approach to argue that competition among public schools is the mechanism supporting improvement in student outcomes, including high school completion, in Los Angeles’s high school Zones of Choice. Card et al. (2010) for Ontario, California, and Neilson (2021) for Chile, respectively, are two examples that bring international evidence of this kind to bear.

What do we not yet know that we need to know?

While the evidence summarized in the previous section points to emerging consensus on the spillover effects of school choice policies, this section highlights several gaps in current knowledge. These gaps include (1) the paucity of evidence on long-run impacts, such as on wages; (2) the little attention to implications of how choice programs are designed; and (3) the need to consider markets (which may feature several overlapping choice options and programs) as the appropriate units of analysis, not individual programs and policies.

First, a major limitation of the existing evidence on the aggregate effects of school choice policies is the limited amount of work that speaks to impacts on students’ “long-run” outcomes. These outcomes include educational attainment (such as college entrance or graduation), marriage, employment, work history, and labor market earnings. This is important because long-run indicators of success are reliably better measures of human capital acquisition than test scores, which are more widely used due to their greater availability in US datasets. Long-run markers of success, for example, depend on many kinds of skills, while test scores may only reliably measure cognitive skills acquisition. A large literature recognizes that near-term impacts on math and reading scores or measures derived from them may not capture durable (and multidimensional) skill gains. An exception in the literature is studies that examine effects on postsecondary outcomes of charter school attendance. But the lack of long-term outcome data is especially relevant to the evidence on statewide voucher programs, where negative effects on test scores for voucher recipients may conflate lower education quality with private schools’ nonalignment with the public school curriculum. Further, little to no existing work estimates competitive impacts from school choice policies on measures of student success in the long run.

A second gap in the existing evidence concerns how impacts on education quality relate to how choice programs are designed and implemented. This is because how policy elements combine has implications for the incentives facing public school leaders and thus for the potential of choice to generate competition. A simple example in illustration is whether (and how much of) public funds actually follow students who switch from public schools to a school choice option like a private school via voucher or a charter school. However, other elements of choice programs are also relevant. These include whether funding and policy support a competitive threat to public schools. For example, are negative impacts of statewide voucher programs on participants partly due to high-quality private schools being insufficiently incentivized to participate? Should barriers to charter school entry be kept low, or is it better that authorizers screen applications to open new schools or expand existing schools based on proven success and limit where new charter schools can open? Answers will require more work that tackles how schools—public, private, and charters alike—make decisions. Though they have clear implications for policy, little existing evidence in the literature speaks to these questions.

This relates to a third limitation of the existing stock of knowledge, which concerns its applicability in the current policymaking environment: increasingly, K–12 education “markets” feature multiple school choice programs. It is no longer unusual for a voucher program that defrays private school tuition, several charter schools, and some kind of choice among public schools to all be options available to families in a district. One issue this creates is interpreting findings from individual choice programs: a case in point is that recent data from the Washington, DC, voucher program shows that 42 percent of the control group students attended charter schools (Dynarski et al. 2017). The bigger question raised, however, concerns how combinations of school choice programs interact at scale. How should limited resources be allocated across schools and programs? Is the marginal dollar better spent on expanding charter schooling, or on expanding vouchers, or on implementing better public school choice mechanisms, or on increasing public school quality generally? This requires a shift in focus from school choice options and policies in isolation to “education markets” as they exist and evolve.

What should be done?

This section collects several takeaways from the global experience with school choice programs to date. A first recommendation is that policymakers should recognize the central importance—and leverage the power—of incentives. The simple point is that, for choice policies to create meaningful competition, traditional public schools and districts must feel threatened with losing students and funding. One practical application of this recommendation is to transportation: a compelling argument that education funding should be directed toward providing transportation of students to charter schools or, in the case of voucher programs, to private schools is that doing so will make it easier for students to potentially leave traditional public schools; it will increase competition. A second application is providing parents with better information about schools (public, private, and charter). A large body of evidence indicates that parents may not know about available choice options and do not have accurate information about school quality. The implication is that supplying information to parents about the effectiveness of different schools at advancing student learning will strengthen incentives facing traditional public schools to increase their quality.

Given limited resources, policy decisions would additionally benefit from considering where returns are likely to be highest. This recommendation channels an animating motive for school choice programs: expanding choice for low-income families and students will yield greater improvements in education quality overall than will expanding choice for high-income families, who already have the means to choose private schools or public schools in affluent neighborhoods. But combined with the logic of incentives, this recommendation produces additional insight: because high-income families experience little real change in their ability to exercise choice, the schools under their consideration also experience limited changes in incentives. This stands in marked contrast for low-performing schools that disproportionately serve low-income students: continuing to miss the mark, when effective choice programs are in place, will risk losing enrollments. This observation has several applications. One is that policymakers could provide financial incentives to charter schools to locate in neighborhoods and areas where traditional public schools underserve students. Several pieces of evidence show that financial considerations are important for where charter schools open (and whether they survive). Long-standing eligibility criteria for voucher programs that are based on the performance of the student’s local public school or family income carry a similar logic. This implies that making vouchers and voucher-like programs universal, while perhaps appealing on fairness grounds, also may reduce their overall effectiveness. An alternative would be to scale the generosity of the voucher with family income. Policy decisions about charter school authorization (and funding) could also benefit from considering which providers or operators generate greater competitive externalities. It stands to reason that high-quality or “proven” providers likely have an advantage in this regard, and evidence, mentioned earlier, suggests that core skills–focused charter schools compete more closely with traditional public schools.

A final recommendation is just that policymakers value the potential of data and evidence to inform policy development and implementation. For example, there is no technical obstacle to creating measures of school (and teacher) quality that recognize multidimensional, durable skill development and reflect long-run success. Such measures could be built on rich evidence from test scores across many subjects, from GPAs, from attendance and discipline records, from college entry and completion data, and even from earnings in the labor market. These measures could then be used to inform parents so that they can make better decisions about schools (publicly available information about the quality of private schools is especially scarce), to target policies in effective ways, and to allocate resources. Rather, the obstacles are tragically often bureaucratic.

Looking forward

The COVID-19 global pandemic and the resulting school closures and shift in education delivery models will likely prove to be the largest single “shock” to school choice in the United States to date. This shock has a demand side and a supply side. On the demand side, widespread dissatisfaction with the responses of public school leaders to the pandemic—in particular, delays in reopening schools drove many families to seek choice alternatives. Private and charter schools were generally quicker to return to in-person instruction. This dissatisfaction is reflected in data on enrollments, which shows declines in public school enrollment (particularly among kindergartners) that were especially pronounced in districts that were slow to open buildings to in-person instruction (Dee et al. 2021). Public school enrollment declined nationally by 4 percent between fall 2019 and fall 2020, whereas charter school enrollments over the same time period increased by 7 percent (NCES 2023). Rates of students attending private schools and being homeschooled also increased (Musaddiq et al. 2022).

The durability of this increase in demand for choice—particularly as new cohorts of students come of age—will combine with policy developments on the supply side that likewise portend a future with more school choice. In the wake of the COVID-19 pandemic, policy momentum behind expanding private school voucher programs has rapidly gained steam across numerous states. An example is 2023 legislation that extended the eligibility for Florida’s Tax Credit Scholarship, mentioned earlier, to all Florida students. Previously, eligibility was means tested (though vouchers remain prioritized to lower income students under the new legislation) and was restricted to students either attending a public school or entering kindergarten or first grade. Florida is joined by Arizona, Iowa, North Carolina, South Carolina, Utah, and West Virginia in creating new or expanding existing private school voucher programs postpandemic. Like the Florida program, several of these will be universal in student eligibility, marking a significant departure from earlier voucher programs in the United States.

The fresh momentum for vouchers highlights a fundamental challenge of evidence-based policymaking: research findings typically lag policy. This is brought into relief by the fact that the recent turn to vouchers arrives amid important new findings pointing to successes from charter school choice. Moving forward, the post-pandemic policy developments raise at least two questions for policymakers and researchers alike. The first is, for effective school choice policies, how important are accountability provisions—as opposed to accountability to the “market” alone? As detailed earlier, charter schools are generally subject to accountability provisions that are typically absent for private schools in voucher programs. The second question is how to produce rigorous evidence about new voucher (and voucher like) programs as they grow. Many existing data systems do not contain reliable information about private schools (or private school teachers), may not record or track recipients of vouchers (especially those who do not at some point have contact with the public school system), and—in states where administration of state tests is not required—will not readily contain information suitable for evaluation. Producing rigorous research findings in the post-pandemic world will thus require renewed cooperation to develop and make available high-quality data as well as creativity to identify and draw from other information sources.

See the full Hoover Institution initiative: . 

Whole-child education models are those that expand the ambit of schools beyond a traditional academic focus. While a range of whole-child models have been explored since at least the Progressive Era, use of these models has expanded greatly over the past twenty years.

Because nearly all children in the United States attend public schools, it can be a tempting place to provide near-universal access to programs and resources. However, for various reasons, some families and educators are wary of a more expansive role for schools in children’s lives beyond academic training. In the Hoover Institution’s report “,” I review several examples of whole-child reforms that have become popular over the past few decades: community schools, school based health centers, wraparound service models, and social emotional learning curricula. 

While some models have proven effective at shifting child outcomes in certain settings, none have yet been proven — at large scale, using high-quality causal research methods — to be a silver bullet that can overcome the challenges many children face today in terms of improving academic outcomes. Though they may have other positive impacts on their own, without related investment in academic reforms, they are unlikely to be the panacea for the low academic performance that plagues children in the United States. Thus, at the end of my brief, I close with recommendations for policymakers to think carefully about implementation of these models in their own contexts.

• Whole-child education models are becoming better known in the United States

• Their adoption in some public schools provides an opportunity to see which models contribute to academic success.

• However, they are a part of the topic of child welfare, not the entire picture.

In the past couple of decades, there has been a renewed interest in the idea that schools should expand their ambit to address a wider range of student needs around health and well-being. Often this is described as a focus on development of the “whole child” rather than just the academic aspects of child development.

Of course, promotion of a wider ambit for schools beyond the academic sphere is at least a century old, as is the debate about whether it is optimal. The intellectual leaders of the Progressive Era, in the nineteenth century, sought to bring a broader focus to education systems than the traditional academic one. This included various ways of engaging the whole child, some of which are similar to the models covered here, particularly the social and emotional learning curricula and community school models that have skyrocketed in popularity in the past several years.

Similarly, the roots of whole-child reforms that are focused on improving children’s physical health are deeply embedded in US education history. As early as 1850, states began requiring immunizations and sometimes hosted immunization clinics in schools, where there was easy direct access to children. Also, the beginning of what we now know as the standard school nurse model began in 1902 as a pilot program aiming to insert healthcare into schools in order to improve chronic absenteeism by managing easily treatable illnesses and focusing on prevention. Each of these foreshadowed the more recent creation and rapid expansion of school-based health centers, which insert healthcare providers directly into schools with the goal of improving academic and overall well-being.

Recent decades have seen a renewal in the popularity of whole-child models. To some extent, this renewed interest is partly a backlash to what many perceived as the laser focus of the No Child Left Behind era on student test score performance. The difficult periods of the Great Recession and the COVID-19 pandemic also contributed to this shifted focus. The recent version of this movement has also been helped by increased emphasis on the complex relationships between education, health, housing, and other social dimensions across a range of academic disciplines and policy spheres.

This whole-child movement in schools has taken many forms, some of which I describe in more detail below. Across all its forms, the theory of change driving whole-child reform has two main parts. First, many students struggle academically because their basic needs are not met. Second, supporting these basic needs directly by bringing healthcare and/or social service resources into the school itself will overcome the access barriers that some children face, particularly poor children, thereby increasing their ability to thrive academically and socially.

To some extent, this theory of change pervades the entire US education system. Almost all districts in the country provide some form of nonacademic care to students through the school nurse, school counselors, or expanded offerings like universal vision screening programs. And many provide extracurricular activities or partner with community organizations in a variety of ways. What differentiates the whole-child models of reform here from the standard public school environment is the broader range of services provided and the depth of engagement between the school and community partners.

Intuitively, the first part of this theory of change makes some sense. How can a child learn if they suffer from an ongoing undiagnosed disease or disorder that prevents them from attending school regularly, concentrating in class, or participating fully in the community around them? How can a child learn if they feel isolated in a community, are surrounded by violence, and lack strong support inside and outside of school?

There is little direct causal evidence to support this theory of change, and there are plenty of anecdotes about children thriving despite incredibly challenging experiences during childhood. Yet a majority of parents would agree that children thrive most when their basic needs are met. However, as with all aspects of childrearing, there is debate about which “needs” require fulfillment for children to thrive. Furthermore, there is debate about whether schools are the best provider of health and social services to support children.

For decades, people have debated whether schools are the most effective places to solve the deep-rooted societal problems, like poverty, that leave many children with their basic needs unmet. Some people see schools as the great equalizer, holding them uniquely responsible for the achievement and well-being of all students, regardless of their backgrounds or the social forces determining those backgrounds. Others argue that systemic poverty, isolation, violence, poor health, and other ills have such a strong role that schools cannot be responsible for overcoming them.

Because nearly all children in the United States attend public schools, it can be a useful place to provide nearly universal access to programs and resources. However, for various reasons, some families are wary of a more expansive role for schools in children’s lives beyond academic training. Some have concerns about the differences between their own values and beliefs and those promoted in the school environment, as is the case with the recent backlash among social emotional learning programs. Others have concerns about whether school employees have the bandwidth and expertise to provide an expansive range of high-quality care; instead, they suggest that a focus on academic knowledge would allow school employees, like teachers, to be more impactful. Still others distrust the push for schools to focus on issues beyond academics because of concerns about greater intrusion into the private lives of families.

In “A Nation At Risk +40,” I review several examples of whole-child reforms that have become popular over the past few decades. After describing the general framework of each, I explore research into each model’s effectiveness. Most have been described as effective by the literature, but this assertion is generally based on research that is largely theoretical, comprises mixed methods, or is conducted either at a small scale or without the types of carefully constructed comparison groups that are essential for determining causal impacts. I focus on summarizing the subset of this literature that meets the Tier 1 or Tier 2 standard of the US Department of Education for strong or moderate evidence of effectiveness from either an experimental or a quasiexperimental design study (What Works Clearinghouse 2020).

Further, since many areas of research have shown patterns of effective programs in small studies that have limited effectiveness when taken to scale, I place particular emphasis on the relatively few studies that have analyzed the effectiveness of programs with large numbers of students across multiple school settings. 

While some have proven effective at shifting child outcomes in certain settings, none have yet been proven at scale, using high-quality causal research methods, to be a silver bullet that can overcome the challenges many children face today. Importantly, when looked at in total and given the scale of the existing research, the lack of conclusive evidence of a clear positive causal effect of these reforms on children’s academic achievement casts doubt on the theory underlying these reforms. Though they may have other positive impacts, on their own and without attention to academic reforms they are unlikely to be the panacea for low academic performance that plagues children in the United States. 

Thus, at “A Nation At Risk +40,” I close with recommendations for policymakers to think carefully about implementation of these models in their own contexts. . 

Maria D. Fitzpatrick is a professor of economics and public policy in the Brooks School of Public Policy at Cornell University. She is co-director of the National Data Archive on Child Abuse and Neglect, research associate at the National Bureau of Economic Research, and an affiliate in the CESifo Research Network. Her research focuses on child and family policy, particularly education. 

See the full Hoover Institution initiative: .

This chapter summarizes the history of the development of public preschool in the United States and its current role in education reform. The chapter reviews research on preschool’s effectiveness and research related to the many decisions that states need to make after deciding to invest in public preschool. Pros and cons of each decision are discussed, along with specific recommendations for state policymakers.

Since the publication of A Nation at Risk, extending school downward to include younger children has become the most expansive and deeply rooted strategy for improving achievement and reducing the achievement gap. The theory of change for public investment in preschool is that enrolling children in an educational program before they enter elementary school can help them develop the skills they need to succeed in kindergarten and can thus set children on a trajectory for school success.

Preschool programming can take many forms, which are explored in depth here, but with regard to student outcomes, the operative phrase is “high quality.” Only programs that meet high quality standards have shown long-term effects. States have many policy levers to affect quality, including program licensing standards, teacher and administrator credentialing standards along with ongoing resources to support effective teaching, child learning standards and assessments, program quality monitoring systems, and curriculum guidance.

Policies related to preschool should not be considered in isolation. If teachers in the upper grades do not build on what children learned in preschool, the benefits will fade. States can help sustain benefits by making sure that learning standards, program standards, curriculum, assessments, and teacher credentialing are aligned across preschool and the early grades.

Preschool is not a panacea. It alone will not improve achievement scores and other important child outcomes or reduce that achievement gap. But of the many school improvement initiatives this series discusses, preschool is among the most studied, and it may have the most potential to move the needle on improving student achievement.

• Preschool has become the norm in children’s educational experience, although controversies still exist about specific goals and how to provide it.

• Policymakers increasingly view preschool as a strategy for launching children on a positive trajectory for academic success and reducing the achievement gap related to economic circumstances that exists before children enter kindergarten.

• Research demonstrates long-term benefits of preschool only for high-quality programs, especially when followed by high-quality and coherent instruction in the early elementary grades.

PRESCHOOL CONTEXT

Since the publication of A Nation at Risk, efforts to improve student achievement have taken many forms. Extending school downward to include younger children has become the most expansive and deeply rooted strategy. Nearly every state in the country has made significant investments in public preschool, and preschool has become the norm in children’s education experience.

Preschool is defined as an education program provided to three-to-five-year-olds during the two years before entering kindergarten. Traditionally, preschool has been an education-focused half-day program, which does not meet working parents’ childcare needs. To address this need, preschools are increasingly offering extended-day (“wraparound”) childcare.

Given that children are learning something all day, whether that is intentional or not, the distinction between childcare and preschool is blurry on a practical level. But the distinction has policy implications. Funding for childcare and preschool often comes from different sources, with different family eligibility requirements and standards. This chapter concerns programs (or those parts of programs) that intentionally focus on promoting children’s learning and development, not caretaking.

Unlike the K–12 system, in the United States preschool has no common governance structure, and administrative oversight rests with different levels of government. Children can attend a private, tuition-based preschool; a public preschool funded by the federal government (Head Start) or the state government (state preschool); a program funded by local resources (e.g., local taxes or philanthropy); or a preschool that involves funding from several sources.

A commitment to educating children before school has its roots in the infant school, designed in the 1820s and 1830s by social reformers to offer free daycare and education emphasizing “moral habit” for the children of poor working parents. In many respects, this is the original “compensatory education” for young children, although it emphasized values and behavior, not preparing children for school success. Soon after infant schools were created for the poor, more affluent parents began sending their young children to private infant schools where the emphasis was on early enrichment and social skills rather than moral reform. But preschools as we currently know them were not widespread in the United States for either poor or more affluent children until the 1960s.

At about the same time that infant schools were introduced in the United States, kindergarten was developed by German educator Friedrich Froebel. He believed that children should be in school from a young age, and the lessons he designed for children ages three through six emphasized music, nature, stories, and play. The first kindergartens in the United States were created by German immigrants in the mid-1800s. Over time, kindergarten for five-year-olds became a grade of schooling in the United States. In 1965, eighteen states funded public kindergarten. By 2000, all states funded some sort of kindergarten, most of them free to all children who met the age qualifications. Today nearly all (about 98 percent) children attend kindergarten prior to first grade, although kindergarten is not compulsory in most states. As kindergarten became viewed as “school,” preschool for three-to-four-year-olds (sometimes referred to as prekindergarten or pre-K) took its place as the education program for the year or two before school. Preschool has become the norm in the United States, and although private programs exist, preschool, like kindergarten, is increasingly supported with public funds. The theory of change for public investment in preschool is that an education program before children enter elementary school can help them develop the skills they need to succeed in kindergarten and can thus set them on a trajectory for school success. The next section offers a brief explanation and history of private and public preschools in the United States.

PRIVATE PRESCHOOL

In the decades prior to Head Start, children younger than five years of age were mostly in private and tuition-based preschool programs affiliated with universities or with religious or other community-based organizations. Now there are many community-based nonprofit programs as well as for-profit chains. Families that enroll their children in private preschool have, on average, higher incomes than families that enroll their children in public programs.

Although academic-focused private preschools exist, most focus primarily on social development and socialization — giving children an opportunity to learn how to get along in a setting with peers — as well as general cognitive development, including communication and problem-solving skills. Specific models, such as Montessori and Reggio Emilia, are prominent in the United States. As publicly funded preschools become more available, the proportion of children in community, tuition-based programs is declining.

PUBLIC PRESCHOOL

Specific goals of public preschool vary across funding sources, states, and localities. Kindergarten readiness, however, is ubiquitous, signified by the widespread state adoption of kindergarten readiness tests for children. There is considerable research to support the belief that kindergarten readiness predicts success in school, which in turn predicts education attainment and accompanying benefits. Because some families cannot afford private programs, public funds are needed to make preschool accessible to all children.

Public preschool has had its critics. Some opposition comes from the more conservative view that the government should not be in the business of educating young children; that is parents’ responsibility. A comment in 1971 by then president Richard Nixon sums up this argument: “For the Federal Government to plunge headlong financially into supporting child development would commit the vast moral authority of the National Government to the side of communal approaches to child rearing over against the family-centered approach.” Some opposition has also come from skeptics about its effectiveness. A related concern is the isolation of low-income children in a program, especially given evidence that low-income children benefit from being in programs with more economically advantaged peers. Despite such
objections, the federal government launched the first expansive public preschool, Head Start,
in 1965; state and locally funded programs followed.

Head Start

Head Start was created in 1965 as part of then president Lyndon B. Johnson’s war on poverty, and it is one of the only Office of Economic Opportunity programs to survive. It was launched with a clear and ambitious goal: to reverse the cycle of poverty. Head Start was considered compensatory education in the sense that the education program was compensating for inauspicious home and community environments associated with poverty.

The specific program reflects beliefs about the academic and life skills required to escape a life of poverty. Unlike most other publicly funded preschool programs, Head Start is a comprehensive program; in addition to promoting cognitive and social-emotional development, it is designed to promote children’s physical and mental health with nutrition and social supports that affect their home environment. Health screenings and provisions for physical health (e.g., dental care) and parent involvement are key components. Head Start thus overlaps in approach with the whole-child reforms described by Maria D. Fitzpatrick in chapter 2 of this series.

Head Start has served an increasing number of children over its history, from fewer than two hundred thousand in the yearlong program in 1967 to over eight hundred fifty thousand in 2020. There has been ongoing tension between the desire to serve more children and at the same time increase quality, both of which add to the program’s costs. Quality has been given considerable attention in the past couple of decades. The 2007 reauthorization of Head Start included a revision of the quality standards and a 40 percent set-aside for quality enhancement (used in part to increase teacher salaries). The reauthorization also required that by 2013 half of Head Start teachers have a BA degree and assistant teachers have at least a Child Development Associates credential. 

Opposition specific to Head Start has been consistent but has come in different forms. Many conservatives opposed all of Johnson’s Great Society antipoverty programs. Critics concerned about parent choice have suggested that the money could be better spent by providing parents with vouchers to purchase their own childcare or education program for their children. There has also been opposition to the broad whole-child goals of Head Start by those who seek a narrower focus on academic skills. Among supporters of Head Start, the primary concern has been about effectiveness. Despite opposition, the program has survived and in fact grown with bipartisan support.

State and Local Preschool

During the 1980s, states began to develop preschool programs for students from low-income families. State-funded and locally funded preschool goals are typically more limited than Head Start, focusing primarily on kindergarten readiness. Few state preschool programs offer the health and social supports for children and families that Head Start offers.

A total of forty-four states, plus the District of Columbia, now provide at least some state funding for preschool programs, enrolling in 2021 about 29 percent of four-year-olds.11 A number of cities have also used various sources of funding to increase access to preschool. For example, San Antonio, Texas, increased its sales taxes; San Francisco added a rental tax, and Seattle added a property tax. Funding levels vary substantially from state to state, from a low of $420 (North Dakota) to a high of $19,228 (Washington, DC) in 2021.

There are several reasons for the growth in state and local preschool programs. First, preschool among middle-class four-year-olds was becoming the norm. Children in families that could not afford tuition-based preschool were therefore at a disadvantage when they entered kindergarten. Second, policymakers were increasingly aware of the large achievement gap related to economic circumstances that existed at kindergarten entry and persisted through school. It was clear that any effort to reduce the achievement gap needed to start early in children’s lives. Third, there were significant concerns about the inadequacy of public education and frustration from failed reform initiatives at the K–12 level. For example, disappointed by the results of $40 million granted to reform the Philadelphia School System, the Board of the Pew Foundation turned to preschool education with the hope that it would be more amenable to change.13 Fourth, there was accumulating research evidence for the short- and long-term benefits of high-quality preschool, including programs implemented at a very large scale. And finally, neuroscientists were demonstrating the significant effects of children’s early experience on the architecture of the brain, making clear that early brain development served as the foundation for future learning and development. Although funding for state and local programs ebbs and flows, the general direction has been toward growth.

A Nation at Risk, published in 1983, played a significant role in the rise in states’ interest in early education. While the report initially stimulated reforms mostly at the high school level, findings at the time related to the benefits of preschool education amplified interest in early childhood education (ECE) as a strategy to improve education outcomes. Many national reports after the publication of A Nation at Risk make clear that promoting “school readiness” became an important element in the school reform agenda. In 1987, the Council of Chief State School Officers included pre-K for low-income four-year-olds in its list of recommendations for improving high school graduation rates. In the six national goals for 2000, created at a national education summit in 1989 under then president George H. W. Bush, “All children will start school ready to learn” was listed first. The National Association of State Boards of Education recommended that elementary schools create early childhood (EC) units for children ages four to eight and develop relationships with community preschools to help coordinate the fragmented services.

In the decade following, preschool for economically disadvantaged children was also frequently included in state school reform legislation. By 1989, thirty-one states either funded their own preschool program or contributed funds to increase access to Head Start. Nearly all of these state programs were instituted in the 1980s. The programs were typically administered by state departments of education and were located in school buildings. Some states funded preschool through separate grants, and some included preschool in their school funding formulas.

Most of the publicly funded state preschool programs have targeted children from lowincome families in a mixed-delivery system, which can include public schools, Head Start, private preschools, and family childcare homes. States vary in whether they offer a program for half or full school days, although the trend has been toward more full-day programs to meet working parents’ needs.

Some of the opposition to state preschool comes from within the ECE field. Programs for infants and toddlers that cannot avail themselves of preschool funding often take a financial hit when the older children move to a public preschool. The problem stems from reimbursement rates being too low for infant and toddler programs to survive with the low child–teacher ratios required. Programs manage financially by including preschool for older children in the mix, which allows higher child–teacher ratios overall. Also, in mixed-delivery programs, private or community-based programs are sometimes at a disadvantage because they lack the resources and the ability to pay teachers the salaries that public programs offer. As a consequence, some people working in community settings view public preschool as an unfair competitor.

Universal Preschool

Most publicly funded preschool programs target children in low-income families. But since the early 1990s, interest in universal programs available to all four-year-olds has increased. This movement was encouraged in part by research finding preschool benefits for nonpoor as well as economically disadvantaged children.

Social programs that are universally available are also considered to be politically more robust than programs targeting the poor. A universal program extends preschool opportunities to the working poor—families that are economically fragile but with incomes just above the eligibility cutoff for programs targeting impoverished families—and it relieves middle-class families of the cost of private programs. It thus develops a stronger constituency. Greenstein suggests furthermore that a program that serves not only the poor but also people significantly above the poverty line and the middle class may lessen the racial imagery of the program.

One argument some advocates make for universal prekindergarten (UPK) concerns the value of mixed socioeconomic status (SES). Programs with low-income eligibility requirements isolate the poor. Research evidence shows, however, that children from low-income families make greater gains in programs with more affluent peers than in segregated programs. De facto segregation will continue, but children from low-income families are not officially segregated where universal preschool is available.

Georgia (1995), New York (1998), New Jersey (in Abbott districts serving low-income students, 1998), and Oklahoma (1998) were pioneers in the universal preschool movement, committing to making publicly funded preschool open to all children whose parents chose to enroll. Currently, seventeen states have legislation committing the state to universal preschool. Despite claims of “universality,” however, most states that have legislation to support universal preschool do not serve all four-year-olds, partly because of inadequate funding and partly because of parent choice and staff and facilities shortages. In the 2019–20 school year (the last year not impacted by the COVID-19 pandemic), only four states and Washington, DC, had more than 70 percent of their four-year-olds enrolled in state preschool.

The infrastructure for universal preschool programs varies considerably. For example, public preschool in Oklahoma operates through the public school system. In Georgia, the program has been set up with voucher-like subsidies; money from the state is given to parents who can choose from among private and public programs that have been certified by the government. In West Virginia, state funding for public preschool flows to both local education agencies (LEAs) and non-LEAs through County Offices of Education. Most states and cities that have passed legislation to expand preschool to all four-year-olds rely on a combination of funding sources, including federal Head Start and state preschool. The one exception is California, which has used state education funds to expand preschool to all four-year-olds by essentially adding a grade to elementary school (referred to as “transitional kindergarten”).

Objections specific to universal preschool focus mainly on public subsidies for middle-class and affluent families when, critics point out, they have shown they are willing to pay for preschool. Critics have also claimed that the state should not expand preschool education when so many of the country’s public schools are failing.

ISSUES AND CONTROVERSIES

Although not uncontested, preschool is deeply rooted in the education landscape; the debates are less about whether public resources should be used to support it than they are about which type of preschool the resources should support (targeted or universal, as discussed above), where the preschool should reside (in public schools, community programs, or both), and how to ensure high quality. The latter two issues are discussed next.

COMMUNITY VERSUS SCHOOL-BASED PROGRAMS

Advocates of placing preschool in public schools argue that it is the most efficient approach to offering ECE at scale and that it contributes to alignment in curricula and teaching between preschool and the early elementary grades, which can help maintain the benefits of preschool over time. Alignment with kindergarten is difficult when children come from many different preschool contexts, and the public school they enter has no authority over the standards, expectations, curricula, and other policies at the preschool level. Another argument for public schools is that they have an infrastructure, including all the back-office supports (e.g., finance, facilities, human resources) needed. Finally, advocates point out that teachers in preschools that are part of public school systems are generally more educated and better paid and have easier access to special education supports.

But public school–based preschool has its critics. Private program providers complain that they are at a financial disadvantage, lacking the purchasing power of a school district. Another concern is that a direct link to K–12 schooling will result in a focus on academics that can divert attention from the more holistic set of goals that most experts in the field of ECE champion. Many critics expect school-based programs to “push down” from higher grades an emphasis on academics using structured, didactic instruction that is developmentally inappropriate for young children. They also worry that K–12 schools are less likely to communicate with and involve parents, a central principle of Head Start and many non-school-based programs. And there is evidence that public school–based programs often have higher child-to-teacher ratios than is recommended for young children.

For practical reasons, almost every state with a public preschool program uses a mixed-delivery approach, with some classrooms in public schools and others in community-based organizations (CBOs). A mixed-delivery system expands access by using a broad array of existing programs—including Head Start agencies, childcare centers, private schools, faith-based centers, charter schools, and family childcare homes—giving families choice and supporting small businesses. It also saves the state the cost of new facilities and allows it to take advantage of existing staff, enrollment, and organizational structures. Furthermore, including private and community providers increases political support for state programs. Another advantage of including CBOs is that teachers tend to be more diverse and likely to speak the native language of dual-language learners. A study in New York City found that CBOs also offered care for longer hours on average and were more likely to provide mental health services.

On the downside, depending on how it is organized, there are many challenges associated with coordinating preschool providers that operate in very different settings with different funding structures and often fewer resources. Also, in many states, childcare and preschool systems are overseen by the state’s human services agency, which can create a disconnect between the human services and the education aspects. It is a challenge to effectively braid and distribute funding as well as maintain quality across a wide variety of settings with different administrative oversight.

Evidence also suggests differences in the quality of CBOs and public school–based preschools. Studies show that preschool teachers in public school settings are generally better educated and better paid. A study of five large-scale, mixed-delivery preschool systems (Boston, New York City, Seattle, New Jersey, and West Virginia) showed differences in quality, despite explicit steps to improve equity in quality across settings. All five systems implemented similar program standards in both public school and CBO programs. For example, all required lead teachers in both settings to have at least a BA degree, and New Jersey, Seattle, and Boston paid teachers the same in CBOs and public schools. (New York City did so in the years after the study was conducted.) Nevertheless, analyses by Weiland et al. (2022) reveal that children in CBOs, who were disproportionately children of color and children from low-income families, were taught by less educated teachers in all localities and showed lower student gains than their peers in public schools. The authors point out, however, that differences in child gains by setting were smaller in New Jersey and Seattle, the two systems in which policies for CBOs and public schools were the most equitable. The Abbott program in New Jersey stands out for its significant efforts to ensure quality in community-based programs. The state requires frequent site visits from master teachers to coach staff, gives districts tools to conduct assessments, and employs university researchers to assess classroom quality and track children’s learning.

To summarize, there are some practical reasons for mixed-delivery systems, but ensuring quality is more challenging than in a more centralized system such as public schools. If a mixed-delivery system is used, considerable attention needs to be given to supporting equity in quality, including pay equity for comparably educated teachers across settings.

DEFINING AND MEASURING QUALITY

There are three primary levers for ensuring quality: (1) state preschool program licensing standards, including teacher-to-child ratios and teacher credentialing requirements; (2) monitoring of programs that have been licensed; and (3) resources for improvement, such as teacher professional development. In many states, monitoring and improving are combined into what is referred to as a Quality Rating and Improvement System (QRIS). Programs are rated along particular dimensions and are typically offered opportunities to improve on those dimensions rated low. In some states, funding levels are based on a program’s QRIS ratings. The National Association for the Education of Young Children (NAEYC) also offers accreditation to programs that meet its high standard, although few programs go through the arduous process to receive and retain NAEYC accreditation.

Two categories of quality are included in most state preschool program licensing standards and monitoring systems. The first involves “structural” indices that can be relatively easily regulated and measured, such as requiring particular teaching credentials, teacher-to-child ratios and group size, and aspects of the physical environment considered important for safety. The second involves “process” variables having to do with the learning environment and the interactions between teachers and children and among children.

There is fair agreement on what constitutes quality regarding teachers’ general interactions with children. Research evidence indicates the value of (1) an overall classroom climate or tone that is emotionally warm, accepting, and supportive; (2) positive, proactive, and consistent classroom management practices that include more affirmation and warmth and fewer disapproving and behavioral controls; (3) educators’ positive, non-conflictual relationships with individual children; and (4) explicit modeling, teaching, and scaffolding of social-emotional skills.

There is emerging agreement as well that effective teachers need to engage in bias-free and culturally responsive teaching. Defining and measuring quality related to instruction is difficult and controversial for all education programs. But unlike K–12 education, where there is consensus on academic achievement as the primary goal, the goal of preschool is disputed. At the heart of the controversy is how much academic achievement should be a goal and, if it is, what teaching strategies are appropriate to attain it.

What Should Preschoolers Learn?

Most EC educators believe that preschool should attend to multiple domains of development— that is, the whole child. Proponents of focusing on the whole child argue that even learning basic academic skills requires self-confidence and self-regulation skills, such as paying attention, as well as the social skills required to avoid wasting learning time engaged in conflict with peers or the teacher.

On the other hand, the accountability movement, instantiated in No Child Left Behind legislation, created pressure for preschools to emphasize academic skills. School districts became focused on raising scores on standardized achievement tests, and many believed that starting early would help them achieve that goal. Current evidence that young children are able to develop foundational literacy and math skills supports advocates’ attention to academic skills, as does evidence that early literacy and math skills when children enter kindergarten are highly predictive of reading and math achievement in school.36 Although Head Start has from the beginning been committed to supporting the development of the whole child, there have been efforts to focus it more on academic preparation. The George W. Bush administration, for example, proposed shifting Head Start from a comprehensive intervention to a program that focused on language and literacy.

The debate rests to some degree on a false dichotomy. A focus on academic skills in math, literacy, science, or any other domain does not preclude attention to other dimensions of children’s development. Nonacademic dimensions of the whole child, such as self-regulation and social skills, can be developed even in the context of academic instruction. For example, math activities can involve cooperative learning opportunities that are designed to help children develop such social skills as listening, taking turns, and negotiating. Moreover, EC educators are less critical of the goal of academic achievement if other important dimensions of development are also stressed and if literacy and math are taught in what EC experts consider developmentally appropriate ways, explained next.

How Should Preschoolers Be Taught?

The debate about how preschoolers should be taught is characterized variably as play based or child centered versus teacher directed. In the extreme of play-based/child-centered programs, children choose and initiate most activities, actively exploring and manipulating concrete materials. The teacher may build on children’s activities (“How many blocks do you have in your tower?”), but they are responding to child-initiated activity. In the extreme of teacher-directed instruction, teachers lead children through rote learning exercises, such as counting and identifying letters, and children work individually on tasks that often involve worksheets. People endorsing a more child-centered approach worry about the effects of structured, teacher-directed instruction on children’s motivation and enjoyment of learning. People endorsing a more teacher-directed approach argue that children do not learn foundational academic skills through self-initiated play.

Research suggests that child-initiated play is important for children to develop general problem-solving and social skills, but the development of subject-matter skills requires more intentional teacher guidance. Research suggests that whole-child, child-centered, play-based curricula fail to produce gains in either math or literacy. In contrast, several meta-analyses of research have shown that math curricula involving teacher-led activities can have strong effects on math learning and that literacy curricula are modestly successful in boosting literacy achievement.

Academically focused instruction does not necessarily mean rote learning and worksheets. Literacy, math, and science can be taught to young children in a developmentally appropriate, engaging way. Research suggests that literacy and math are most effectively taught through intentional but playful activities in which the teacher leads children through a learning activity, but children have some discretion in how they achieve the goal of the activity. In addition to the benefits of teacher-guided, playful activities, research supports the value of a language-rich environment in which teachers listen to and engage children in conversation and small-group instruction, which is well designed to engage children’s interest in the context of a positive emotional climate.

MEASURING AND SUPPORTING QUALITY

The research is somewhat mixed on how well-structural dimensions of quality, such as teacherto-child ratios and teacher credentials, predict child outcomes and does not pinpoint specific regulations, such as determining whether a 1:12 or 1:10 teacher-to-child ratio is ideal. The evidence is also mixed on the associations between teacher credentials and quality measured by either classroom observations or child outcomes. There is some evidence that the amount of specific training in child development and ECE may matter more than the level of the credential. And there is compelling evidence that professional development (PD) can have an impact on the quality of children’s experiences. But not all PD programs are created equal. The evidence on PD suggests the value of coaches working directly with teachers, focusing on content knowledge and effective pedagogy within an identified domain. Brief workshops are ineffective in promoting lasting changes in instruction.

Classroom observations are needed to assess process quality. Those that are typically used provide valid information on general teacher–child interaction, but they do not assess the quality of literacy or math teaching. The field needs to develop assessment instruments that states can use to measure instructional quality in preschools.

CHALLENGES TO QUALITY

The most significant obstacle to quality is cost, and the critical variable is the teacher. Although research does not provide strong guidance on exactly what kind of preparation is ideal, there is good evidence that preparation is important, as is professional development. But both come with costs. Increasing the rigor of training can improve quality, but it will exacerbate the teacher supply problem without commensurate increases in salary. Professional development also requires time on the part of the teachers and funding for the people who do the PD (and ideally for the participants).

Teacher pay affects quality indirectly through its effect on the economic stress teachers experience, their classroom behavior, and ultimately, turnover.54 Because preschool teaching is typically relatively low paid, turnover is high. Turnover undermines quality because programs that invest in professional development do not get much of a return on their investment, and high turnover causes children to experience instability, which undermines the quality of the relationships they can develop with their teachers.

The field is also challenged by inequity in quality for low-income and more affluent students. A study of 1,610 preschool sites in New York City found classroom quality to be lower in sites located in poor neighborhoods and in centers serving higher percentages of Black and Latino children. Even in Georgia, a national leader in universal preschool, state preschool classrooms in low-income and high-minority communities were rated significantly lower in classroom quality.

IMPACT

Public preschool is costly, and policymakers understandably want evidence that it helps them achieve the goal of improving education outcomes. In brief, the findings indicate that preschool can but does not necessarily have both short- and long-term effects for children that also yield economic and social benefits.

The most notable long-term impact study, which has had a significant (perhaps outsized) impact on ECE policy, is the Perry Preschool Project. The program provided very high-quality preschool education to 123 three- and four-year-old African American children.

Findings based on participants and the control group long into adulthood show many longterm benefits of the program, including higher high school graduation rates, higher earnings, and lower arrest rates. Cost-benefit analyses suggest a substantial return, which economist James Heckman concluded is about 8 to 1. A substantial portion of the return comes from reduced incarcerations. The findings from this one small study have been used to convince many policymakers of the value of investing in preschool. The public programs that have been developed since, however, are not near the level of quality of Perry Preschool, which employed primarily teachers with graduate degrees in ECE and which offered a staff-to-child ratio of 1:6, teacher weekly visits to children’s homes, and parent education. It cost $21,800 per child in 2017 dollars, compared to the $5,867 spent per child on average in state preschool programs.

Results of the many studies on the impact of preschool cannot be reviewed here. Metaanalyses, however, indicate meaningful positive effects overall on school academic readiness, reductions in grade retention, and special education placement as well as small effects on social-emotional development, especially when this is a specific goal of the program.

Many studies, including a large-scale study of Head Start, show an initial benefit of preschool, but the benefit fades over the first few years of elementary school. One study in Tennessee found that the control group had higher achievement scores in fourth grade than the children who had been randomly assigned to state preschool. These findings raised questions about the quality of the Tennessee state preschool program and what early educational experiences the control group had. But along with the many studies that have documented the fading of preschool benefits in elementary school, the Tennessee study makes clear that simply implementing a large-scale preschool program does not guarantee the desired effects on children.

There are several possible explanations for the fade-out effect. One explanation, supported by research, is that kindergarten teachers often repeat material that children had already learned in preschool, allowing children who did not have preschool to catch up with those who did. Another is that teachers in the early grades do not differentiate instruction for children with varying skill levels, which impedes children’s opportunities to continue to exhibit growth. Both explanations suggest that preschool is more likely to yield long-term academic benefits if it is followed with instruction that builds on initial gains.

Although there are only a few studies of long-term effects, there is evidence that preschool implemented at scale can have significant effects into adulthood. A random assignment study in Boston found increases in high school graduation, SAT scores, and college attendance and a decrease in juvenile incarceration, especially for boys, but no detectable impact on state achievement test scores. A recent study of the Tulsa preschool likewise found effects on college enrollment. Reviews of studies that assessed long-term effects, including some studies of Head Start participants, describe increases in college enrollment and decreases in incarceration rates and teen pregnancy.

A task force created by the Brookings Institution, which included individuals with very different perspectives on the value of state preschool, worked through all of the evidence on state-funded preschool to produce a consensus summary of impact.68 Following is a summary of the task force’s conclusions:

• Studies of different groups of preschoolers often find greater improvement in learning at the end of the pre-K year for economically disadvantaged children and dual-language learners than for more advantaged and English-proficient children.

• Pre-K programs are not all equally effective. Several effectiveness factors may be at work in the most successful programs. One such factor supporting early learning is a well-implemented, evidence-based curriculum. Coaching for teachers as well as efforts to promote orderly but active classrooms may also be helpful.

• Children’s early learning trajectories depend on the quality of their learning experiences not only before and during their pre-K year but also following it. Classroom experiences early in elementary school can serve as charging stations for sustaining and amplifying pre-K learning gains. One good bet for powering up later learning is elementary school classrooms that provide individualization and differentiation in instructional content and strategies.

• Convincing evidence shows that children attending a diverse array of state and school district pre-K programs are more ready for school at the end of their pre-K year than children who do not attend pre-K. Improvements in academic areas such as literacy and numeracy are most common; the smaller number of studies of social-emotional and self-regulatory development generally show more modest improvements in those areas.

• Convincing evidence on the longer-term impacts of scaled-up pre-K programs on academic outcomes and school progress is sparse, precluding broad conclusions. The evidence that does exist often shows that improvements in learning induced prior to kindergarten are detectable during elementary school, but studies also reveal null or negative longer-term impacts for some programs.

ECONOMIC RETURN TO PRESCHOOL

Some impacts on individuals (improved school readiness, higher achievement, increased high school graduation, higher education attainment, higher earnings) bring social benefits, such as increased taxes, reduced use of welfare, and improved health. Other public-sector benefits come in the form of reduced child abuse and neglect and reduced incarceration. Public school savings are found in reduced grade retention and reduced special education use.

Because the large-scale programs are relatively recent, most cost-benefit analyses project benefits from shorter-term impacts. For example, a study of the Chicago Child-Parent Centers (CPC) used data from subjects at age twenty-six on special education use, grade retention, juvenile and adult crime, and adult earnings to project impact beyond age twenty-six, such as earnings, taxes on earnings, incarceration, depression, smoking, and substance abuse. The net economic benefits to society were estimated to reach almost $97,000 per child (in 2016 dollars), a return of nearly $11 for every $1 invested.69 Looking across cost-benefit studies, including Chicago, Tulsa, and Head Start, estimates show a more realistic return of between 2:1 and 4:1; however, such analyses are based on many assumptions for which there is not total agreement. Another caveat is that the conclusions are based on relatively highcost, high-quality preschool programs and are not likely to apply to programs of lower quality.

CONCLUSION AND RECOMMENDATIONS

Publicly supported preschool began with the goal of giving children living in poverty a “head start” in school, but over time it became a strategy for reducing the achievement gap and improving academic achievement. Will public preschool close the achievement gap? No. There are many other factors associated with poverty—including poor nutrition, crowded and unstable housing, poor medical care, and stress—that have significant effects on children’s learning. But access to high-quality preschool can give children from low-income families a fairer chance of succeeding in school, and the research on long-term impact suggests it can alter life trajectories. The operative term is “high-quality.” All of the research on longterm benefits with economic returns to individuals and society are based on programs that employed well-trained and well-supported staff.

This chapter has outlined a number of issues that need to be addressed after the decision to invest in public preschool has been made, such as whether to support a targeted or universal program, whether to implement a mixed-delivery or school-based delivery system, and how to define, measure, and ensure quality. Below are a few recommendations related to specific choices.

Clearly articulated goals are important. The goals guide program development as well as the strategies and measures used to assess program quality, support program improvement, and track progress in achieving the goals.

Achieving consensus on the goals is no easy task, because there are strong differences of opinion. But many of the differences rest on false dichotomies. Even if academic skills are the primary goal, all dimensions of children’s development play a role. Furthermore, developmentally appropriate, teacher-guided instruction can be playful, and time can be put aside for children to play freely and explore.

Quality is the primary consideration in any preschool system. Only preschool programs that meet high quality standards have shown long-term effects. States have many levers to affect policy, including program licensing standards, teacher and administrator credentialing standards, and ongoing resources to support effective teaching, child learning standards and assessments, quality monitoring systems, and curriculum guidance. These policies need to align with the articulated goals of preschool and with each other. For example, teacher credentialing requirements and ongoing supports for teachers should prepare teachers to help children achieve the state learning standards.

Of the many dimensions affecting quality that these policy levers can be used to promote, the following are particularly important:

• Adult-to-child ratios that allow adults to form close, caring relationships with children

• Preparation and ongoing support for teachers in developmentally appropriate, differentiated instruction, including developing a safe, secure, and inclusive environment for all students

• Curricula that offer sufficient and developmentally appropriate attention to language, literacy, math, and social-emotional development as well as other important dimensions of child development such as creativity and motor development

• An assessment system that can be used to guide teacher decisions at the classroom level and track children’s progress related to the standards

Depending on the approach to be used, different policy considerations apply.

• If the decision is made to use public funds only for children in low-income families:
  • Because children from low-income families learn more on average in a mixed-SES setting, avoid isolating the poor by offering hybrid programs in which some children are subsidized by the state and some families pay tuition, or implement a sliding scale.

• If universal preschool is implemented:
  • Ensure that community-based providers for infants and toddlers are reimbursed enough to cover costs when they lose older children.

• If a school-based delivery system is employed:
  • Ensure that programs are developmentally appropriate for young children and that teachers and principals have training in ECE.

• If a mixed-delivery system is employed:
  • Establish strong program standards across school-based and CBO settings, and implement other policies (e.g., teacher pay equity) to support equity in quality. Findings suggest that even in states where efforts have been made to equalize quality, extra resources may be needed for CBO programs.

What happens after preschool affects the long-term benefits of preschool. Policies related to preschool should not be considered in isolation. If teachers in the later grades do not build on what children learned in preschool, the benefits will fade. Districts need to reflect on the nature and organization of instruction through the early grades, joining the movement expanding throughout the United States to develop greater P–3 alignment (i.e., alignment from preschool to third grade). States can help by making sure that learning standards, program standards, curricula, assessments, and teacher credentialing are aligned across preschool and the early grades. They also need to ensure strong connections among state agencies overseeing preschool and K–12 education.

Streamline connections between preschool and childcare. Working parents need childcare. The different funding sources and eligibility requirements cause difficulties for both programs and parents. States need to streamline funding and requirements for preschool and childcare as much as possible, and they need to facilitate wraparound childcare for preschool programs.

The politics of pre-K can be complicated to navigate. Many constituencies are affected by decisions related to state-funded preschool, including state agencies that oversee its implementation, community-based programs that are affected by whether and how they are included, higher education where teachers are prepared, school districts, Head Start, the people who staff the programs, and parents. All policy decisions should include their voices.

Preschool is not a panacea. It alone will not improve achievement scores and other important child outcomes. But of the many school improvement initiatives this series discusses, preschool is among the most studied and may have the most potential to move the needle.

The Hoover Institutions A Nation At Risk +40 research initiative spotlights insights and analysis from experts, educators and policymakers as to what evidence shows about the broader impact of 40 years of education reform and how America’s school system has (and hasn’t) changed since the groundbreaking 1983 report. () 

Simultaneously, philanthropist Andrew Carnegie hoped to kick-start the expansion of higher education by donating $10 million to bankroll pensions for college professors. This posed a parallel dilemma: How to decide whether a scholar had put in enough time to earn the annuity?

Thus was born the wonky educational anachronism known as the Carnegie Unit, brainchild of the trustees of the . A certain number of hours spent in a high school classroom added up to a credit, the trustees decided in 1906. A set number of credits earned a diploma. So quantified, the diploma could be used as the entrance ticket to a college or university, where Carnegie Units would add up to a degree — or the right to retire.

Carnegie Units went on to become the central currency of a dizzying number of aspects of education, ranging from what subjects students are exposed to to how states allocate school funds. But it was quickly understood that while the units were good for, say, establishing whether a public school had delivered its pupils enough hours of teaching to earn its taxpayer dollars, it was not particularly helpful at signaling what a student had learned during those hours in class — now, 117 years later, better known as seat time.

Some innovators, like the leaders of the Phoenix Union High School District, are experimenting with ways to leave the Carnegie Unit in the past. Students at PXU City, a Phoenix high school without a building, can create their own personalized educational experience from a menu of 500 options, including classes at any number of high schools, college courses and job training programs throughout the city. 

The experiment came to the attention of Carnegie’s present-day leaders, who are engaged in their own effort to replace their turn-of-the-century units. The person tasked with figuring out how better to quantify what students have learned, and how the schools of the future can help them realize the historical promise of social and economic mobility, is Tim Knowles, the foundation’s president and former director of the University of Chicago Urban Education Institute. 

Knowles recently talked to Beth Hawkins about a pilot project to reimagine seat time that includes the Phoenix district, the possible benefits of freeing teachers from unit-driven bell schedules and how to transform entire school systems. This interview has been edited for length and clarity. 

Tell us about the Carnegie Unit, what it is, how it wove itself into education’s very DNA and why it’s time to step away from it. 

In 1906, when the Carnegie Foundation created the Carnegie Unit, it suggested that a college degree should be 120 credits. Today, it’s 120 credits. It’s become the bedrock currency of the educational economy. It’s infiltrated everything. It’s how we organize high schools and universities, how we think about assessment, it’s instrumental to accreditation, to who gets financial aid and who doesn’t. It defines the daily work of teachers and professors. It is the system.

What it is, fundamentally, is the conflation of time and learning. It’s the suggestion that X number of minutes equals learning. The problem is, that it basically ignores everything we’ve learned in the last 100 years about what knowledge is and how it’s acquired. We’ve had neuroscientists and cognitive psychologists and psychologists and learning scientists come along and say, “People learn through solving real problems, they learn from peers, they learn from mentors, they learn in apprenticeship, they learn from experience.”

In its time, the Carnegie Unit was an incredibly important reform because it standardized an utterly nonstandardized educational sector. But it crept into the core DNA of educational practice and didn’t evolve or adapt in the face of a significant amount of empirical knowledge about how human beings actually learn. That’s problem No. 1. 

No. 2 is that it inhibits educational innovation. Competency- or mastery-based education has existed, arguably, since and . But it’s existed at the edges. It’s never been central. We can all point to schools that are breaking the boundaries of what learning should look like, how it’s organized, how it’s structured. While those examples exist, they’re often led by extraordinary teachers and school leaders. We haven’t figured out how to take it from the margins to the mainstream. That’s a problem.

The third thing is perhaps more existential, and that is the absence of social and economic mobility in our nation. This is not to suggest that education isn’t an essential solution to addressing social and economic mobility, it’s just that it’s not nearly powerful enough an engine for doing so. The Carnegie Unit is, in my view, partly responsible for that. In the 1950s, over 90% of young people would [end up] better off than their parents. That number is basically cut in half now. We’re going in precisely the opposite direction, and underlying that are some really fundamental inequities, which are exacerbated by race and by class. 

If we want to radically increase economic and social mobility, we need to reimagine what learning is, and really take into consideration what we know about the context in which people learn. Young people need to be engaged in much more experiential, hands-on solving of real problems and applied work. 

Why is the Carnegie Foundation the organization to take this on?

It’s been a narrative in the organization and beyond for decades. One of the questions was, what are you going to do about assessment, because if you are going to take on the future of learning, then you have to be thinking about the future of assessment. That led us in the last year to a deep partnership with ETS, which is the largest assessment company on Earth. It’s very good at determining reliability and validity. 

If we believe that learning, wherever it takes place, is important, then in order for that to take root at scale, we need to persuade parents first that the learning their young people are experiencing outside the schoolhouse is valuable. And is legible to the postsecondary sector if you’re applying to college. And legible to employers if you’re going more directly into the workforce. Everybody intuitively knows there’s enormous amounts of learning there. We need tools that can validate that learning. 

We’re going to build assessments to assess the skills, not the disciplinary knowledge, that we know are predictive of success. Things like your ability to collaborate, to communicate, how hard you work, are you persistent, your creative thinking, your critical thinking. The aim with ETS would be to get to the point where every young person in America doesn’t just graduate from high school with a transcript that has grades and attendance and test scores, but a skills transcript as well.

The wonderful thing about Carnegie, it’s got this incredible responsibility to be a place that looks around the corner. It did that in 1906, for really important reasons. It created Pell grants for really important reasons. It established standards for medical schools, engineering, law schools. It’s done these things at certain times in its history that really needed to happen, because there were gaps. And it’s positioned in a way that it can take a slightly longer-term view about where we need to get.

In Phoenix, a driving factor behind the district’s decision to move away from the Carnegie Unit more quickly and widely than it had planned before the pandemic was teenagers. Students who weren’t in high school when COVID hit had no expectation of a bell schedule. 

At the heart of accelerating learning is ensuring young people are leaning in, are engaged, are inspired and are working on problems that they think are actually useful — whether it’s useful for their own trajectories, pursuing a track that is orienting them to a particular profession or sector, or more here and now. One can learn a great deal about democracy by actually practicing it, or identifying an issue that you care about and learning how civically to engage in a way that can draw attention and potentially movement regarding that issue. 

So, yes, engagement is an instrumental variable in all this, and, as you are pointing out, teenagers have already spoken. We know they’re not engaged. There have been some systems around the country that have made marked improvements in high school completion, but there are many where 50, 60 or 70% of students are biding their time. Getting through. If we’re losing 30 or 40% of our young people before they’ve even had a shot, then we’ve got to take a step back and ask what might work better, what we need to do differently. And that’s not a small incremental step, like providing double blocks in math or high-dosage tutoring. There’s something much more fundamental involved in reconsidering how we think about time and learning. 

The other thing about teenagers is that when you create opportunities which are highly engaging, and you enlist their agency in learning, you really just have to get out of the way. Versus in a set of circumstances that may feel to them far more compliance-oriented, where they’re doing seven periods for 40 minutes or 42 minutes between bells with a two-minute passing period, with seven different teachers every day who may have so many students that they can’t even learn their names until the end of October. If you turn that on its head, young people are going to rise and surprise us. 

There was a survey of high school students nationally during the pandemic, and almost to a person they said they wanted to come back to school — not surprisingly. But not all of the time. I don’t think that was a statement about not being interested in learning. I think that was a statement about not wanting only that form of learning all the time. Valuing the community that school creates, valuing the fact that there are some domains of expertise and disciplinary areas where they need to be in classrooms with amazing teachers, but also recognizing that, “Wait, I’ve been learning independently. I’ve been pursuing things I’m passionate about.”

If we could scaffold that systematically with opportunities for apprenticeships, for internships, for community embedded work, I think it’s safe to say not only would we be hewing much more closely to what empirical evidence says is the best way to learn, but we would be in a situation where the young people were much, much more interested and excited about what they were learning.

Let’s talk for a second about the obstacle that is adult time. I’ve talked to so many people in education who say, Yeah, that’s great. But we’d have to fundamentally reorganize how the adults use their time.

We would. By saying we’re taking on the Carnegie Unit, we are not saying we’re going to have eight periods a day in 10th grade and then we’re going to layer on a whole other set of things. So how we organize teacher time, and the role of the teacher, has to fundamentally shift. 

But teacher pipeline issues are real. Schools are struggling to find really exceptional people who want to spend their career teaching. Embedded in rethinking the use of adult time is the opportunity to rethink the role of teachers. There are few people who want to have the same responsibilities on the first day of their professional career as they do on the last day of their 35th year. 

If we could turn the teaching profession into something where you’re teaching in teams, for example, or where you may be teaching in the morning and then advising groups of students through the afternoon as they engage in activities in their communities and with postsecondary institutions, the job could become much more attractive and interesting to a much wider range of young people over time. Adult time is a predicament, but it’s also an opportunity.

You mentioned a skills transcript. That got me thinking about how many people you’re dangerous to if you’re successful. So many things are proxies for whether a person has a skill set. The college degree is a proxy: We assume that because you made it through this filter — which might be meaningless — you are going to be valuable to this endeavor, this institution, this company. But we don’t actually know whether you come with the requisite skills.

Right. We’ve had a very fragmented K-12-to-postsecondary-to-employment path, replete with assumptions. If you go to an elite private or highly selective public college, whether by virtue of where it is geographically or by nature of how you get in, there all kinds of assumptions that you’re going to come with these other things that we care about, in addition to whatever’s on your transcript, like your grades and your test scores. But that’s a pretty crude measure of whether you do. So there is a threat to the established pathways. 

We could credibly determine whether a young person has a set of skills wherever those skills were developed. If I’m living on the south side of Chicago, and I take my two siblings to school every morning, and then I get to my high school on time at 7:30, I do my homework, I perform well in the traditional ways, I participate in afterschool activities and I work, those are skills that that are invisible, or less visible than the proxies you were suggesting. Or whether I was lucky enough to be born in a situation where my parents were taking me to rarefied places every summer, or putting me in rarefied summer camps.

Part of the agenda is to make the education sector a more vital engine for individuals, no matter their backgrounds, to be able to succeed in a post-affirmative action world. A skills transcript would provide elite schools with a different kind of visibility on every kid. It wouldn’t have to have anything to do with race per se, but I would hope it would help make visible the skills and dispositions young people bring, even if they’re growing up in really underresourced places. 

Devil’s advocate. When the pandemic forced schools away from seat time, lots of people said, ‘Hey, wait — maybe we could just have asynchronous learning. I wouldn’t have to report to a building anymore.’ Or the variation we’re hearing a lot about now, the four-day week. In blowing up seat time, temporarily or permanently, did states just leave the barn door open?

The conditions under which they blew up seat time during the pandemic are slightly anomalous. I wouldn’t compare what we’re trying to do to that, because we’re certainly not of the view that people should be socialized in front of a laptop. But I’m sympathetic to the accountability side. Whether we have the existing system or a fundamentally transformed one, it’s going to demand that we know how young people are performing.

None of what I’ve been talking about should suggest we no longer believe in algebra or reading. There’s things that we really do think young people benefit from learning. How they learn those things is an open question. If we are in a period where people are questioning the power of our educational system and asking questions about how we might empower it further, it has to be undergirded by accountability systems that are credible. And fair. Otherwise, you’re right. It could be a slippery slope. 

What are you learning so far?

Our agenda, which we’re working in partnership with XQ on, is in short: establish proofs, create places where this is happening, build evidence for improvement. Develop policy and national discussion about transformative learning opportunities. And then think hard about the postsecondary piece. Unless the work we do in high school is relevant, legible and understandable to postsecondary, it could falter. 

There are learnings from the people who’ve been doing this for a long time, sometimes in quiet opposition to the systems in which they sit, sometimes with some support from the state within which they sit. They are there, and it’s important to recognize and acknowledge that there are educators across the country who’ve been doing this with young people from all kinds of different backgrounds.

One of the [trends] in the educational system at the moment are these things called portraits of a graduate or portraits of learners. They’re everywhere. One of the things we did with ETS was look carefully at all the ones that we could. They’re interesting, because they represent an American consensus about what the purpose of schooling is. They are really focused on skills. Often, they’ve been developed with lots of parent voice and teacher voice and student voice. Red and blue, left and right, communities, in spite of all our polarized hype, are saying there are a set of things that we want for young people. That should make us optimistic, if we can leverage that. 

The other things that that you will hear is, A) they haven’t really made a big difference, and B) we have no way of measuring the things in them. The problem is the Carnegie Unit problem. They haven’t cracked the Carnegie Unit, they haven’t cracked this architecture of learning that we’ve established. We have to do that. We want our young people to be able to think critically, and we don’t really know how to measure that. How do we measure that they’re civically involved? 

Disclosure: The XQ Institute, which has partnered with the Carnegie Foundation for the Advancement of Teaching to explore alternatives to the Carnegie Unit, provides financial support to ����Ӱ��.

The pandemic illuminated the cost of relying on this outmoded approach to education. While other industries have spent the last 30 years modernizing, K-12 schooling remained focused on an older paradigm. And when the pandemic shut communities down, other sectors continued with limited disruption while teachers were forced to scramble to bring their one-size-fits-all classrooms online.

The recent NAEP results underscore both the pandemic’s impact and the need to reimagine how education is delivered. While many innovative school communities are generating new approaches that hold promise for students and staff, change on a broader scale can be difficult to achieve, given the time and resources needed. Innovative learning models — programs schools can adopt to shape the experiences of young people, families and educators – share ideas and approaches that have the potential to positively impact student learning and development.

To help schools handle the challenges of finding, introducing and implementing new educational models, Transcend has created an . With funding from The Barr Foundation, The Bill & Melinda Gates Foundation and The Siegel Family Endowment, the free, offers educators inspiration and practical resources to help them reshape the learning experience.

The exchange features more than 40 models that span different focus areas ranging from postsecondary preparation to social-emotional skills to project-based learning. Models can be a comprehensive school design or a targeted program that aids a particular goal — introducing a new approach to teaching math, say, or even overhauling traditional school. Models represent intentional decisions around the basic elements of school, such as curriculum, culture, adult roles and scheduling, which makes them flexible to meet the needs of different educational communities. Those decisions are then reflected in the resources, structures and practices of the model that create the student experience.

Schools can explore models like that embrace blended learning, or , which focuses on whole-child development. For those who don’t know where to begin, the platform includes for finding models aligned with their unique goals, needs and interests. 

Through collaborations with communities across the country, we at Transcend have seen the difficulties educators face when looking to redesign their school. Administrators often spend time reinventing instead of adopting and adapting, and lack the capacity to start from scratch. The exchange offers the opportunity for them to see what’s possible before expending resources and to try new ways of functioning with the support of experts from schools, districts and education nonprofits, among others. Models listed on the exchange are vetted by our team, and each contains an overview as well as implementation guidance. This can take the form of free, downloadable resources or hands-on support such as school visits, one-on-one coaching or professional development.

Through widespread, community-based engagement, the exchange makes resources and ideas  accessible to all, sparking innovations based on students’ needs in a changing world. Transcend aims to continuously add to its database to offer more options for reinventing the traditional school design. Anyone who would like to nominate a model can follow this .

Disclosure: The Bill & Melinda Gates Foundation provides financial support to ����Ӱ��.

This fall, on the heels of six months interviewing “about 100 teachers in 15 cities” who were working on substantial improvement efforts, Cohen is publishing his first book, . It’s an effort to answer a question he asks at the outset: “What would it actually look like for teachers to be at the center of discussions about school transformation?” 

Change Agents comes out tomorrow (Oct. 25). I sat down with Cohen last month to talk about the book — and about what’s next for public education as leaders move past the pandemic. 

This interview has been lightly edited for length and clarity. 

����Ӱ��: The book is built around profiles of Partners in School Innovation—tell me about them and how you came across them. 

Justin Cohen: About 15 years ago I was working for the D.C. Public Schools, and we were exploring a multi-district collaborative around early identification of young people who were on track to not finish high school. It never really went anywhere, but I got to meet a man named Derek Mitchell, in Prince George’s County, and he went on to Partners in School Innovation (PSI) years later, doing improvement science and continuous improvement in schools. 

At PSI, Derek introduced a notion that you know, continuously improving an unfair system isn’t enough. put it this way: “making incremental improvements at the margins of a system originally designed to sort children by race, class and language will only make inequitable sorting more efficient.” And so Derek insisted that continuous school improvement needed to have a racial equity lens—and PSI has been working on that for the last 13 years. 

He recently reached out to me, saying, “I’m really excited about what we’ve been able to do. What about telling this story at a broader level?”

A chance to capture their approach, codify it and make it replicable in more schools.

Right. Totally. From the outset, I envisioned something like Atul Gawande’s , which took some of the lessons of improvement science and showed how to do incremental change in health care on a day-to-day basis…but for education. I was especially interested in writing something for teachers. Something for educators to pick up and read during the very limited time that they have, that’s practical, teacher-friendly, and — hopefully — inspiring without indulging in what my editor likes to call ‘toxic positivity.’ 

I particularly enjoyed the part focused on conditional reasoning — that is, the “if/then” statements at the heart of almost any attempt to shift human behavior. It’s at the core of pragmatic, realistic change thinking. Every school improvement — every self-improvement — starts with a commitment to trying something new (“if I do this…”) in the hopes of seeing different results (“then we’ll see this outcome…”). But that doesn’t mean that humans naturally start their problem-solving that way. It’s an acquired pattern of thinking. What are some tips for teachers trying to get themselves and their colleagues into thinking about improvement more constructively? 

There’s one acronym in the book. ROCI: Results-Oriented Cycles of Inquiry. The foundational idea is that you get together with a group of your peers every week for collaboration. During those meetings, you set a target: some sort of process improvement. And whatever that thing is — a 15-minute check for understanding at the end of each lesson or whatever — you’re all going to commit to doing it together. You make time in the subsequent week to watch each other try this thing out. And then, the next week, you talk it over. 

A big part of this is that it’s not top-down school reform, right? It’s not a principal calling everyone in a room and dictating The Plan. It’s ground-up, teacher-led inquiry. 

Yes. That’s the core thing that differentiates it from a lot of the last generation of reform. This is about asking teachers at the classroom level, “What do you want to try differently tomorrow?” And then, let their judgments and expertise guide next steps. The inquiry process is just as important as the resulting improvements. It’s about building that habit, building that muscle of trying something new, seeing whether it works, observing each other to give feedback on whether or not it works, and then doing more if it continues to deliver results, and stopping doing it if it doesn’t. 

I mean, I know that sounds really basic, but people spend hundreds of millions of dollars every year on gym membership they don’t use. Habit-building is hard!

You know, it’s the flip side of , this notion that If you invest heavily in early childhood education, you get better results overall for marginalized and underserved communities, because in the early years, kids’ habits haven’t been formed, their long-term trajectories remain fluid, they’re at a point of . Most people get that. But nobody talks about the flip side: Shifting adult behavior is super hard. I mean, anybody who’s ever tried to lose weight or shift their TV habits knows this. 

Yes. In the book, I take great pains to avoid a cheerleading, “it’s easy, you can do it!” kind of mentality. Because it’s actually hard, even if it’s rewarding. One of the things that really comes through is the joy people experience when they get to see the results of shifting their practice on their own terms versus shifting practice because somebody told them to or because the state, said “We’re gonna shut your school down if you don’t change.” For the last 25-30 years, it’s been all stick, no carrot. 

But in this teacher-led approach, there’s at least a sense that you can control some of the destiny, and some of what you decide to do.

Meaningful teacher agency … that’s clearly not what we, the country, have been doing. 

It’s not. And one of the key things about this, right, is that when school improvement is driven by teachers, it’s more durable. Even when funding dries up and top-down pressure for reform goes away, when improvement consultants’ PowerPoints go away … those habits don’t. Teachers don’t just stop meeting with their peers to talk through the new things they’re trying. Once they get used to that, they keep doing it. 

Also: part of giving teachers agency is about giving them the opportunity to fail. I know that failure has been rhetorically weaponized against teachers — and families and children, in some cases — like, “failure is not an option anymore.” 

And as a person who’s indulged in this language at times, I think we have to admit that that’s not helpful and is even psychologically jarring in some cases. We have to create enough room for people to try things and maybe not succeed the first time, particularly when we’re talking about institutions like schools. We have millions of teachers operating in tens of thousands of schools and districts. 

This is a good segue. Because you’re clear, in the book, that there are good reasons that we wound up in the place we’ve been in, that we tried the reforms we’ve been trying. The absence of data on student outcomes, on student achievement, meant that, for example, kids were assigned to English as a Second Language classrooms because their last names sounded Hispanic, not because they necessarily needed those services. Educational inequities and civil rights violations thrive when we don’t keep track of what kids know and can do. So maybe we need to modify the policies imposing these consequences on schools, but … can you say more about a policy agenda that leaves more room for teacher-led inquiry and improvement? That leaves room for teachers to fail?

I’m gonna do this in a roundabout way. 

I remember, after the financial crash of 2008, thinking that it’s really nice that economic policy and monetary policy has some very clear available mechanisms. A new president shows up, appoints you to the Federal Reserve Board, and when you walk into the building, figuratively speaking, there’s a big lever labeled, “Interest Rates Down—Borrowing Up.” That’s just what happens. We know how these forces work. 

But we do not have that in education. There isn’t a “Student Achievement” lever to pull when a new administration arrives in the White House. It doesn’t exist. I think we have to acknowledge that. 

I’m not allergic to accountability. In fact, I think my book offers a very deep, very intense form of accountability at the individual practitioner level — a level of accountability that is more or less ignored by today’s policy regimes. 

And look, I’m not saying this to level a judgment on the people who crafted those policies or on the people who’ve spent decades earnestly trying to implement them. But the exact measurements that those regimes insisted upon — academic tests — haven’t shown good results. And we’re not talking one or two years. We’re talking about a generation here. That’s just a fact. 

So I think we need to let go. We need to admit that this test-and-sanctions approach didn’t work. 

I mean, policymakers will have to institutionalize the work of continuous improvement at the practitioner level. Things like, at the more local level, creating time for collaboration or relaxing some of the annual test-based accountability, and creating multi-year, more robust accountability around different longitudinal measures. Think of things like civic participation, graduation, and post-secondary attainment, all the things that we know test scores were supposed to be a proxy for.

Are things moving in the right direction, then? We replaced No Child Left Behind with the Every Student Succeeds Act, which significantly weakened federal school accountability systems, but retained some transparency mechanisms whereby schools were still publishing data on students’ progress? After all, the last administration wasn’t really interested in implementing the law, and then the pandemic . 

I think a weakened and watered-down system that we know is not working … I mean, let’s just put it out of its misery. I mean, there are a lot of people who believe that fully erasing federal testing and accountability policy would return us to some educational policy Eden. I mean, the era before this one wasn’t some perfectly equitable moment in public schools. 

My view is that we need to, you know, erase the whiteboard and start over with some core principles in mind: transparency around outcomes, equity, ensuring that no school gets to go too long getting the same results over and over again without being prompted to rethink what they’re doing. 

So we still need and should want accountability. But we need to get away from these punitive regimes and focus on doing the real improvement work that we know actually works. 

You could make the case that, operationally speaking, we’re kind of moving that direction, right? There’s just so little appetite for top-down accountability right now. 

We need to think of accountability as starting with the inquiry cycle at the practitioner level. Plan to assess. Pick a target. Meet as a grade-level team to discuss the target. Watch each other try new things. See if it had an impact. Lather, rinse, repeat. Just keep doing that. 

So: I think that at each level out from the school — district, state, federal — needs to set up somewhat longer cycles of inquiry that look at whether these short-cycle returns are adding up into meaningful, long-term, equitable improvement. 

It’s going to be extremely hard. It opens up big questions of autonomy, empowerment, who decides what and where, but it beats sticking with the ineffective accountability approaches we’re currently using. 

There are 3 million teachers in the country. It’s one of the biggest professions in the country, and if we think that the education profession and the schools in general are going to get better without a deep investment in making sure those millions of people get better and better every day, we’re kidding ourselves. 

Sure. Part of the whole systemic education reform argument is that you can build policy structures that create conditions for success, that reduce the importance of individual teacher quality as a variable, right. And while I get that, as a project, it’s obviously nonsense to skip past teachers, to treat them like plug-and-play widgets. So I’m wondering, then, in rethinking teachers’ agency, in broadening their roles as agents of change … does the book have a message for their trainers? For schools of education? 

I mean, yes. Schools of education, in many cases, teach people completely wrong things about the processes of teaching and learning and the history of racial equity in this country. If you show up as a new teacher with no awareness of the history of racial exclusion here and no knowledge of why systematic inequities manifest in your community and building, and without, say, awareness of the cognitive neuroscience of how children learn how to decode, you are not prepared to be a teacher. A lot is going to need to happen on campus to prepare you to be truly ready to lead a classroom. 

I did not write this book to solve this problem, but fortunately, the cycles of teacher-led inquiry can help bring those folks up to speed. 

Those cycles provide accountability that’s about learning opportunities, right, which, I think, is an extension of the analogy above. If you sign up for a gym alone, you maybe waste that membership. But if you’re part of a running group that meets on the corner at 6 a.m. every two days, you’ve got accountability to one another—and a commitment to improvement. 

Yes. The more you collaborate, the more you can observe, the more you can unearth things. It’s all about opening “the closed door” of each American classroom. So often, teachers operate in relative isolation from each other, with no idea about what’s going on in other classrooms. 

And then, frustratingly, observation has gotten too tightly wound up with evaluation. We need to undo that. You’re not gonna like your job if every time someone shows up to observe you, it’s all negative and you’re anxious about how they’re gonna hurt you. 

If nothing else from this book gets through, I hope this does: most of the time, when you open your classroom door for a peer or a superior, it should be a rich learning experience. You should learn interesting things about your practice as an educator.

Students who attend charter schools in Newark, including English learners and those with learning disabilities, are less likely to transfer out within two years than their peers at the district’s public schools, according to a new study. Children were also significantly more likely to shed their special-needs classification while enrolled at a Newark charter, the authors note.

The study, released as this week and Education Next, offers more perspective on the long-running debate over admissions and retention in the charter sector. Critics of the publicly funded but privately operated schools that they push out kids with learning, language or behavioral challenges like suspension or expulsion.

Co-author Marcus Winters, a professor of educational leadership and policy studies at Boston University, said in an interview that he believed individual schools of all kinds “inappropriately” encouraged some students to leave. But the Newark study, along with looking at schools in Tennessee and North Carolina, has disproven the notion that charters routinely engage in the practice, he argued.

“I do think it’s fair to say that our paper … has now sufficiently debunked the myth that charter schools — at least in these areas that have been studied — are systematically pushing these students out,” Winters said.

Winters and co-author Allison Gilmour, a professor at Temple University, set out to compare enrollment trends in Newark, a city with one of the largest charter school sectors in the country. To do so, they used data from the Newark Enrolls assignment system, which allows families to select among their choice of traditional schools and approximately 70 percent of the city’s charters. (Not all local charters participate in Newark Enrolls, but those that do account for about five-sixths of charter students.)

Several variables in the Newark Enrolls formula determine which students are assigned to certain schools, including each child’s rank-ordered school preferences; individual factors prioritized by various schools (such as sibling preference); and randomized lottery numbers that are used in case a given school is overenrolled. By gathering administrative data between the 2014-15 and 2017-18 school years, Winters and Gilmour were able to compare patterns of school entrance and exit for nearly 14,000 students.

In all, children attending Newark charter schools were 22 percentage points less likely to leave that school within two years than substantially similar students who were instead assigned to traditional public schools. English learners were 16 percentage points less likely to transfer out of a charter, and students with a disability nearly 11 percentage points less likely. The difference for Hispanic students was not statistically significant.

The smaller chances of transfers may be attributable to the system’s format of ranked school preferences. In a model that controlled for families’ ranking of schools, charter students were still less likely to leave within two years, but only by about 10 percentage points; that suggests that a sizable portion of the charter school effect is simply a reflection of students attending the school they wanted to go to in the first place, Winters said.

“You might just be more willing to stick it out with a school that you originally had as a higher preference,” he said. “If you’re attending a school that you went to on purpose, you’re just less likely to leave it. And if you’re going to a charter school in a place like Newark, where several of the charter schools are among the most popular choices … you’re probably going to one of your most highly preferred schools.”

By tracking the same students over time, the study also observes gradual movement within individual subgroups. Specifically, children with a special-needs classification at charter schools are much less likely to still have an Individualized Education Program a few years later — a phenomenon that may help explain why the percentage of students receiving services is lower in charters. The effect is particularly notable for children entering charter schools between kindergarten and third grade (31 percentage points more likely to lose a special-needs classification within three years) and between grades four and six (20 points). Those findings dovetail with research pointing to similar trends in special-needs assignment at Boston charter schools.

Comprehensive examinations, including from the federal Government Accountability Office, have shown that charters generally teach smaller numbers of kids with disabilities than district schools. More recent evidence indicates that those gaps may be shrinking, though it’s unknown how the huge upheaval triggered by COVID-19 may have shuffled enrollment trends.

If the study raises doubts about the claim that charter schools consistently work to remove struggling or hard-to-teach students from their classrooms, it offers little clarity about how they approach recruitment. At least that charters in multiple states were less likely than district schools to respond to application inquiries from parents of children with severe disabilities.

Winters concluded that the population differences between sectors could arise from only one of three sources: recruitment of students, mobility of students once enrolled and (in the case of English learners and disabled students) changes in status classification such as those detected in the Newark study. More investigation was needed into how different schools attract families, he said.

“It’s clear to me, at least, that the major driver in these enrollment gaps is who’s enrolling in the first place. We need more information about the enrollment side.”