The group also calls for a massive study of existing research to help identify best practices, noting the last such effort was undertaken 20 years ago. 

颁搁笔贰鈥檚 , released today, tracks the traditional teacher-centered approach and the reform movement, which calls for student-led learning. 

The research organization does not choose a side, but it does note that recent efforts to retool math education often call for a mix of the two. It also pays particular attention to the 鈥渟cience of math,鈥� which gained traction in the early 2020s and argues that math instruction should be guided by empirical research and cognitive science while relying more on orderly, explicit classroom instruction.

Mathematics has been a major educational concern for years with renewed attention after the pandemic. American students have , a trend that only worsened after COVID shutdowns.

鈥淭his is a national priority that needs to be answered,鈥� said Alexander Kurz, a senior fellow at CRPE and the report鈥檚 author. 鈥淚n the absence of consensus, the guide urges educators to evaluate competing claims as they arrive at their schools and to anchor decisions in a clear-eyed look at what’s actually happening in their classrooms.鈥�

The report comes as more than a dozen states have passed legislation aimed at improving math education, in some ways following the state-driven initiatives that were built around the science of reading.

Some, like New York, are calling for of the way math is taught 鈥� with little explanation of what might change 鈥� while others have already chosen a pathway forward. 

California approved a revised math framework three years ago and in November 2025 adopted 64 programs at the K-8 level to help students reach those goals. 

recently adopted new math standards for the first time in 15 years while is in the process of revamping its K-12 efforts to ensure more coherence across grades. 

last fall and to gauge how they want to boost student proficiency. 

As state-level plans unfold, individual cities are taking their own steps to strengthen student performance, meeting with mixed success. 

New York City鈥檚 efforts have proved cumbersome, with the new administration . Boston is trying its own approach with while are using traditional and non-traditional means 鈥� including math field trips 鈥斅爐o improve student engagement.聽

CRPE urges educators looking for new ideas to consider related studies critically, noting they often do not encompass a wide student group. As a result, their recommendations might not work in all cases. 

鈥淲e should always approach these studies with a healthy skepticism,鈥� said Yasemin Copur-Gencturk, who helped craft some of the CRPE report’s recommendations. 鈥淪omething that might work with a particular group of students in a particular context may not work in another situation.鈥� 

Some groups, among them, have fallen even further behind their peers in math in recent years. 

An cites that on the National Assessment of Educational Progress tests, gaps between the highest- and lowest-performing students widened significantly after 2017. Higher-scoring kids started to rebound after the pandemic in 2024, but those in the 10th and 25th percentiles suffered steep and lasting losses.聽

CRPE points out that the National Mathematics Advisory Panel, composed of 24 expert members who reviewed more than 16,000 research publications and policy reports, marked the last nationwide effort to improve student performance in the subject as detailed in its

鈥淎ddressing core Math Wars debates, the NMAP deemed the conflict between conceptual understanding and procedural fluency 鈥榤isguided鈥�, concluding that the two are mutually supportive,鈥� the report found. 鈥淭he panel also explicitly stated that high-quality research does not support exclusive reliance on either 鈥榯eacher-directed鈥� or 鈥榮tudent-centered鈥� instruction.鈥�

CRPE argues now is the time for an updated look at this high-stakes question. 

鈥淎 new national mathematics advisory effort could revisit the earlier panel鈥檚 questions while incorporating nearly two decades of new developments in mathematics education, special education, cognitive psychology, developmental science, and the learning sciences,鈥� the report states. 

鈥淚ts aim should not be to declare winners and losers in the Math Wars, but to produce clearer, more transparent guidance on what is known, what remains open to further inquiry, and where practitioners can implement confidently.鈥�

Posited by advocates as a companion to the better-known 鈥渟cience of reading,鈥� a juggernaut that has recently brought enormous change in how that subject is taught nationwide, the 鈥渟cience of math鈥� calls for a more orderly, explicit approach to classroom instruction.

Sarah Powell, associate professor at the University of Texas at Austin and one of the movement鈥檚 most vocal proponents, said the circuitous path to learning popularized in recent years might 鈥渟ound sexy鈥� but doesn鈥檛 help kids master the skills they need to succeed 鈥� at least not in math. 

鈥淪ome teachers are helping to set students up for success,鈥� she said, while others are not. 鈥淲e don鈥檛 want to waste students鈥� time.鈥�

But critics call the movement poorly researched, saying it resurrects ineffective techniques that should have been abandoned years ago. They say, too, that it won鈥檛 counter the current mathematics crisis, evidenced by a troubling cascade of low test results at the local, state and national level.

 Powell said ineffective teaching practices are among the factors that have contributed to students鈥� performance. She said many educators, in delivering their lessons, are not getting to the point fast enough, especially considering the limited time in which they must teach complex and often abstract concepts. 

Powell, who works in the department of special education, said she observed a fourth-grade teacher last school year who instructed her students, at the start of class, to find a partner and begin to solve a mathematical problem right away, with no prior instruction.

鈥淭hat can be a really intimidating strategy for a lot of students, especially if they haven’t worked on a multi-step word problem before,鈥� Powell said. 鈥淚f you put an adult in that situation, they would be really, really annoyed. And that we put students in that situation is quite unfair, particularly when they have weak or sometimes nonexistent foundations in math.鈥�

Instead, she said, teachers would be better to explain the problem first and arm students with some of the vocabulary they will need to solve it. So, a lesson on perimeter shouldn鈥檛 start with students staring at a figure on a sheet of paper, asking themselves and one another about where the perimeter might be found, she said, but with a teacher providing a definition of the term, in this case, the distance around the outside of a shape. 

鈥淲e would like to see teachers go over vocabulary that鈥檚 going to be important or say, 鈥楲et’s do a problem together,鈥欌�� she said. 鈥淏e more systematic with that than open-ended because it could be that when you do allow students to explore, you might have some students engaging in no exploration 鈥� or maybe incorrect exploration.鈥�

Some 25,000 people are members of the science of math Facebook group, where they discuss best practices, share helpful websites and offer feedback on lesson plans. Powell is currently working with around 100 fourth- and fifth-grade teachers in Austin this year. 

Nick Wasserman, associate professor of mathematics education at Columbia University鈥檚 Teachers College, said there has been much back and forth around the teaching of mathematics through the decades, as evidenced by the math wars of the 1990s.

Part of the issue, he said, is that educators sometimes prioritize different aspects of mathematics and draw differing conclusions from the same studies. But even considering these debates, research has shown students learn better when they are asked to reason and think mathematically 鈥� a core tenet of inquiry-oriented approaches.

鈥淥f course, teachers have times when they are giving direct instruction and that is not something we disagree about,鈥� he said, adding that, 鈥淏ut it鈥檚 also really important that students have times when they are the ones being asked to think and reason mathematically. Giving students tasks for them to work on on their own 鈥� without a teacher telling them how to think 鈥� is a vital component of that. A key piece is this engagement with thinking and reasoning.鈥�

Michael Greenlee, who also works at UT Austin as a professional learning specialist with the Charles A. Dana Center, said the 鈥渟cience of math鈥� should be approached with caution. 

Greenlee, who serves as a regional director for the National Council of Supervisors of Mathematics, questions the validity of the idea, adding its special education roots might not make it applicable to all students. 

He argues, too, that there is a political agenda behind it, backed with people trying to fight against the Common Core, what they have wrongly dubbed 鈥渘ew math.鈥� 

鈥淐ommon Core illustrates more explicitly what math actually is,鈥� Greenlee said. 鈥淧eople look at it as new math because it鈥檚 vocalizing and putting in words brain processes and scaffolding strategies that students should use.鈥�

Powell scoffs at the politicization of the issue and said she is in favor of the Common Core standards, put forth in 2009 and which sought to define one rigorous set of English and math skills that every student should have at the end of each grade level. 

Initially widely embraced, the standards, particularly Common Core math, later engendered intense backlash, including from those who considered it an intrusive federal mandate. It was so confusing for parents that it sent them to understand the new techniques. 

Greenlee said the 鈥渟cience of math鈥� has no relationship to 鈥渢he science of reading鈥� and is unfairly linking itself to the earlier movement. He said, too, that the 鈥渟cience of math鈥� assumes teachers are using certain methods when they are not. 

鈥淭he reality is, we don’t know what teachers are actually doing in their classrooms,鈥� he said. 鈥淚n fact, in my job going into large systems and school districts, my experience is teachers aren鈥檛 doing what that research says they should be doing: They’re still not teaching (mathematics) appropriately. They don’t have the background knowledge themselves to be able to teach it properly.鈥�

Jo Boaler, a Stanford University professor and one of the main architects of California’s recently , said many educators are exacerbated by the 鈥渟cience of math鈥� debate. 

鈥淭hey鈥檙e rolling their eyes about it,鈥� she said, adding that those who promote the notion have cherry picked data. 

She said research shows students learn better 鈥� they are more able to master terms and methods 鈥� through activity. 

鈥淏ut if you stand at the front and tell them first, it often stops them really being able to engage conceptually because their brains are just so taken up with these methods and rules, they can’t even think conceptually,鈥� she said. 

Boaler is no defender of the Common Core standards: She calls them outdated. But this 鈥渙ld is new鈥� tactic, which immediately divides kids into 鈥渞ight鈥� and 鈥渨rong鈥�, isn鈥檛 the answer, she said. 

鈥淲hy are we not teaching kids the mathematics they will need in the world they鈥檙e moving into?鈥� Boaler asked, adding students should leave high school with data literacy and the ability to build a mathematical model that would allow them to study the issues most pressing to them. 鈥淭hey’re not learning that in school. They鈥檙e just learning the calculation piece. When they鈥檙e given a real situation and asked to make a mathematical function, they can鈥檛.鈥�

This lack of knowledge places students at a disadvantage compared to those from other countries, she said. 

鈥淲hen we have been out interviewing seniors in high school recently, they had never seen a spreadsheet in their whole 12 years of school,鈥� she said. 鈥淭he U.S. is really far behind the rest of the world in teaching kids about data and data tools.鈥�

Such knowledge would have been tremendously useful during the pandemic: Much of the information shared about the virus was done using graphs and statistics, data impenetrable to those who didn鈥檛 understand them, she said. 

But Elizabeth M. Hughes, associate professor of special education at Penn State said educators do not have to pick a side in the debate. 

鈥淎 lot of times I see things on the internet that are essentially asking, 鈥楢re you Team Fact Fluency鈥� or 鈥楾eam Conceptual Understanding?鈥欌�� she said. 鈥淭he reality of it is that students need both. They need to understand the underlying constructs of addition, subtraction, multiplication, and division and be able to do them accurately with ease. Students need to be able to explore new ideas and have a solid plan to solve known problems.鈥�

And, she said, rote memorization might have been wrongly demonized: Students must be able to recall math facts with ease and accuracy. 

鈥淭hat shouldn鈥檛 be a controversial idea,鈥� she said. 

Wasserman said teachers can find success through many strategies 鈥� including a combination of guided instruction and student exploration.  

鈥淚t鈥檚 fine to give students a definition,鈥� he said, but teachers must also give them tasks 鈥渢hat force them to wrestle and grapple with some of the nuances of that concept.鈥�