Each fall, as we approach the end of the season for statewide score reporting, I find it helpful to reflect on trends in how students performed across the country and lessons learned from another year of testing. This year’s results may be an early indicator that historical progress in mathematics is at risk.

Roughly 27 million students took a statewide summative assessment last spring, spending about four hours—or less than one percent of their academic year—demonstrating their mastery of grade-level learning standards and their readiness to take on the challenges of this new school year.

Statewide assessments serve an important purpose beyond helping individual students and their families gauge how well they are meeting grade-level expectations. We use state assessments not to evaluate the students but to evaluate ourselves. Are we as educators meeting the sacred charge that we have been given to prepare students for the future—to build an equitable society and face the global challenges that lie before us? Certainly, one test score cannot possibly measure such an important commitment—but statewide assessments are the best measure we have to ensure every school is doing its utmost to serve our students, and our future.

So how did we do this year? Twenty-seven million students means 27 million stories. Some students found inspiration from a great teacher, had the opportunity to engage with rich curriculum and thrived—and their test scores reflected that. Teachers like Michael Dunlea, a third-grade teacher in Tabernacle, New Jersey, who helped write both the state standards and the district’s curriculum in math, make math accessible. Mr. Dunlea won a Presidential Award for Excellence in Mathematics and Science Teaching this month. “As an early elementary teacher, I try to foster a love for mathematics by making the learning interactive, relevant, and fun,” he said.

Other students did not have access to an inspiring teacher like Mr. Dunlea or access to advanced coursework because teacher shortages in their state made it impossible for principals to hire the effective teachers these students need and deserve. A thousand different variables impact the learning opportunities students had last year—and that’s just within the school setting.

Overall, however, students generally improved last year in their English language arts proficiency and held steady or declined in math. These trends appear to hold generally for states working both with New Meridian and with Smarter Balanced—the only states that still are able to compare how well they are preparing students to meet international benchmarks for college and career readiness beyond locally developed expectations.

The continued improvement in English language arts is encouraging: Students are learning to critically read meaningful, important texts, develop a point of view, and make compelling arguments based on evidence and facts—all critical skills we need students to develop as they prepare to address the challenges of the future.

We have to ask ourselves, however, why are more students not mastering the language of mathematics and statistics to model and solve real-world problems in ways we know will be critical for addressing the challenges of an increasingly quantitative world? From coding and artificial intelligence, to medicine, business, economics, climate science and public policy—all of these fields increasingly require a mastery of mathematics and mathematical thinking to engage, define, and solve complex problems quantitatively.

As Michael Kirst, the former State Board of Education president, reviewing this year’s Smarter Balanced results in California acknowledged, “We’re doing better in English language arts than I predicted and worse in math. The problem is serious.”

The lack of progress in mathematics suggests at a minimum that we have likely harvested the “low hanging fruit” on the pathway to preparing students for college and career readiness, and that we have to deepen student engagement in more meaningful and challenging tasks if we are going develop students’ critical thinking, reasoning, and problem-solving skills to meet the more rigorous expectations of college and career ready standards.

This is especially true in mathematics. Experts cite the fact that development in mathematics builds sequentially across the grades, and gaps in foundational understanding taught in elementary grades have a large impact on students’ readiness to grapple with more advanced concepts and skills in middle school and beyond. Declining proficiency rates from grade 3 to Algebra I reflect this snowballing effect as expectations rise and students fall behind.

The importance of high-quality classroom tasks was highlighted in a recent study published by TNTP, The Opportunity Myth. Researchers observed and analyzed nearly 1,000 lessons over two years and found the majority of classroom assignments are not aligned to grade-level expectations. “Students spent more than 500 hours per school year on assignments that weren’t appropriate for their grade and with instruction that didn’t ask enough of them—the equivalent of six months of wasted class time in each core subject,” the study said. Students in lower income neighborhoods were twice as likely as those from higher income neighborhoods to spend their days working diligently on low-level tasks that neither challenge or engage them to meet grade-level standards.

To help address this epidemic of mis-aligned, low-level curriculum, New Meridian is making available exemplar classroom tasks aligned to grade-level standards and assessments to help teachers and students calibrate their expectations of what is required to master grade-level standards. This is only one piece of the puzzle we face in deepening instruction to improve student learning outcomes, but an important one. Together with trained teachers who know how to structure meaningful and engaging learning opportunities, quality tasks are at the heart of the instructional core that has been shown to accelerate and deepen student learning.