On June 21st, the U.S. Department of Education released the latest findings from the National Assessment of Educational Progress (NAEP) Long-Term Trends assessment of math performance among 13-years-olds. It added to mounting evidence that the alarming decline in middle school math scores is not bottoming out. Instead, the downward trend seems to be accelerating, especially for our most vulnerable students.

This is a bright red flag for our education system. It should trigger more than just soul-searching. It demands action! It also begs the question: If our current strategies aren't working, what will? This article will (1) briefly explore the implications of the NAEP findings and (2) discuss strategies proposed by STEM educators from around the country to help address this challenge.

The Implications

Source: NAEP Long-Term Trend Assessment Results: Reading and Mathematics: Trend in NAEP long-term trend reading and mathematics average scores for 13-year-old students, NCES.

The NAEP LTT is a 45-minute test given to 13-year-olds across the USA. Nationally, scores dropped by more than 3 percent, the largest drop ever recorded.

Our education system is not recovering from the pandemic. There are so many stresses on the system that it is damaging our ability to teach and our students' ability to learn. Recovery is not going to happen quickly and it's not going to happen on its own. Collectively, educators and administrators need to commit to and implement new strategies in order to meet this challenge.


Source: NAEP Long-Term Trend Assessment Results: Reading and Mathematics: Changes in NAEP long-term trend reading and mathematics average scores for 13-year-old students, by race/ethnicity: 2012, 2020, and 2023, NCES.

There was already a significant gap in scores among Black, Hispanic, and White students. The pandemic has widened that gap. For example: The drop in scores for Black students was more than double the drop for White students. In addition, average scores declined by 14 points for students already in the lowest 10th percentile a steeper decline than any other group.

Equity in education is an even more urgent challenge than it was before the pandemic. It is critical to adopt strategies that not only support all learners, but also specifically meet the needs of historically marginalized learners.

Response Strategies: Find ways to reduce daily headaches for teachers

It's obvious that teacher burnout and exiting from the calling of education are factors in the sliding math scores. There are many ways to boost teacher retention, none of them easy. What we see, unfortunately, is that many of the tactics used are temporary measures. To make for durable change, we have to actually change a teacher's job description. We have to find the things that drive teachers crazy day in and day out and tackle those.

Here are examples of things that math teachers wish their school and district leaders would do to make their jobs easier:

● Avoid mandates that teachers do more tasks without taking away any existing tasks.

● Avoid mandates that teachers do tasks they are not trained or supported to do.

● If you ask teachers to measure things, make sure they have the tools to effectively do so.

● Avoid layering on new tools that require extra work to manage on top of existing classroom management.

● If you ask teachers to use tools or resources, make sure that they support collaboration among teachers and support staff.

● Don't leave it to math teachers to address student behavioral and emotional lags caused by the pandemic. Many students are “stuck” in an emotional state that does not correspond with their age. This has a direct impact on daily math instruction and teacher stress levels.

● Give teachers a chance to voice their priorities and solutions to instructional inequity.

Integrate learning supports and scaffolds with Tier 1 instruction

The data show that learning supports and scaffolds benefit all learners, not just special populations. Consider implementing supports and scaffolds as part of universal design, rather than as accommodations for specific sub-groups. Every student can benefit from research-based strategies such as visual aids, kinesthetic learning, writing scaffolds, text narration, and making real-world connections.

Resources: The American Rescue Plan of 2021 allocates a minimum of 20 percent of district ESSER allocations for the implementation of evidence-based interventions that address the disproportionate impact of COVID-19 on underrepresented student subgroups. The deadline to allocate these funds was extended to Sept. 30, 2024.

Integrate content and language for all learners

It has long been established that multilingual learners learn best when able to draw connections between academic language and content knowledge, and when these are taught together. But did you know that all learners benefit from timely content-language integration?

Our team identified the 425 most critical middle school math concepts and found that mastering each concept requires several of the six communication modes: listening, speaking, reading, writing, viewing, and representing. Each of the six modes is critical to math modeling, reasoning, and communication.

Resources: The WIDA English Language Development Standards Framework provides standards for content and language integration in both math and science. In addition, the Council of Great City Schools recently released its new Framework for Foundational Literacy Skills Instruction, a helpful resource for emerging multilingual learners of all ages.

Ask the hard questions of your EdTech providers

Education technology is often a low-cost, highly scalable way to address instructional needs. But not all of the research done by EdTech providers is equal. For example: Many of the research claims made by EdTech providers are based on Pre/Post tests. These studies lack a control group, so you cannot know whether the gains claimed were a result of the intervention or another factor. In a controlled study, on the other hand, you compare a treatment group against a control group and look at the differences between the two groups. It also matters how similar these two groups are — otherwise it may not be an apples-to-apples comparison.

Other things to look out for in research claims:

● Who conducted the study? Was it an unbiased third party with no affiliation with the provider?

● What was used to measure results? Was it a third-party assessment that is unaffiliated with the provider, or was it the provider's own test?

● How many students were involved in the study? Was it a handful of classrooms, or were the results spread across hundreds or thousands of classrooms?

Resources: It is incumbent upon administrators to ask, "What kind of research did you do?" and not to accept claims at face value. The NewSchools Venture Fund has created a resource to help you differentiate from among different levels of research done by edtech providers: the Research Continuum.

About the author

Ben Grimley is CEO and Co-Founder of Speak Agent, Inc.