Learning loss is the decline in academic knowledge and skills that occurs when students experience extended gaps in instruction. It can happen over a summer break, during prolonged school closures, or as the result of chronic absenteeism. The concept gained widespread attention during the COVID-19 pandemic, when national test scores dropped by the largest margins ever recorded, but the underlying phenomenon has been studied for decades.
How Learning Loss Works
Your brain naturally prioritizes information it uses regularly and lets infrequently accessed knowledge fade. This is why skills that require consistent practice, like math computation and second-language vocabulary, tend to erode faster during breaks from school than skills reinforced by everyday life, like basic reading comprehension. The longer a student goes without instruction, the more ground they lose. Students can lose anywhere from a third of a year to a full year of learning depending on how long the disruption lasts and what supports are available at home.
Learning loss isn’t just about forgetting specific facts. When students fall behind in foundational skills, they struggle to keep up with new material that builds on those foundations. A third grader who loses ground in multiplication doesn’t just have a gap in multiplication. By fifth grade, that gap has compounded into difficulty with fractions, division, and word problems. This cascading effect is what makes even modest periods of lost instruction so consequential over time.
What the Pandemic Revealed
The COVID-19 school closures created the largest natural experiment on learning loss in modern history. The National Assessment of Educational Progress, often called “the nation’s report card,” found that average scores for 13-year-olds dropped 4 points in reading and 9 points in math compared to the 2019-20 school year. Compared to a decade earlier, the declines were even steeper: 7 points in reading and 14 points in math.
Lower-performing students were hit hardest. In math, scores for students at the bottom of the performance distribution fell 12 to 14 points, roughly double the 6 to 8 point decline among higher-performing peers. This pattern showed up consistently across datasets: the students who could least afford to lose ground lost the most.
Why Low-Income Students Lost More
The pandemic didn’t affect all students equally, and the reasons are largely structural. High-poverty schools were more likely to shift to remote instruction, and their students lost more learning when they did. In high-poverty schools that were remote for more than half of the 2020-21 school year, students lost roughly half a year’s worth of typical academic growth. That’s a staggering gap from a single year of disrupted schooling.
One of the most telling findings came from states like Texas and Florida, where schools largely stayed open for in-person instruction. In those districts, achievement gaps between high-poverty and low-poverty schools did not widen. Where schools stayed remote, gaps widened sharply. The difference wasn’t about the students. It was about whether they had reliable internet access, a quiet place to work, adults available to help during the day, and the dozens of other supports that wealthier families could more easily provide.
How Learning Loss Is Measured
Researchers typically quantify learning loss in two ways. The first is criterion-based: comparing a student’s performance against fixed grade-level standards. Did the student master the skills expected for their grade, yes or no? The second is norm-based: comparing a student’s performance against their peers using percentile ranks.
These two approaches can tell very different stories. After the pandemic, overall performance dropped so steeply that a student scoring the same raw score in 2022 as a peer did in 2019 would land at a higher percentile, simply because everyone around them also declined. In Texas, for example, a fifth grader hitting the same math benchmark would have ranked at the 48th percentile in 2019 but the 61st percentile in 2022. The student didn’t improve. The comparison group just did worse. This is why percentile ranks alone can mask the true scale of learning loss.
The Role of Chronic Absenteeism
Learning loss doesn’t only happen during dramatic events like a pandemic. Chronic absenteeism, typically defined as missing about 14 or more school days in a year, produces a quieter but steady erosion. A study tracking students in Chicago found that chronically absent students in the middle grades scored, on average, about two months behind their peers in math by eighth grade. The effect on reading was smaller and not statistically significant, which aligns with the broader pattern: math skills are more dependent on classroom instruction and more vulnerable to missed time.
Chronic absenteeism rates spiked during and after the pandemic, compounding the problem. Students who were already losing ground from remote instruction continued losing ground by not showing up consistently once schools reopened.
Long-Term Economic Consequences
Learning loss doesn’t stay in the classroom. It follows students into the workforce. Researchers at Harvard’s Center for Education Policy Research estimated that the pandemic-era decline of 9 to 11 percentile points in math achievement, if it becomes permanent, would translate to roughly $43,800 in lost lifetime earnings per student. Across the 50 million public school students in the U.S., that adds up to over $2 trillion in total lost earnings. For context, that figure is about 10 times the $200 billion Congress allocated for pandemic school recovery.
Students in high-poverty schools that stayed remote for most of 2020-21 face a projected 5 percent decline in average career earnings based on historical relationships between test scores and income. These aren’t abstract projections. Lower academic achievement consistently predicts lower wages, higher unemployment, and reduced economic mobility across decades of labor market research.
What Helps Students Recover
The most rigorously studied recovery intervention is high-dosage tutoring, which means frequent, structured one-on-one or small-group sessions rather than occasional drop-in help. A pooled analysis of literacy tutoring studies found an average effect size of 0.35, which translates to roughly a 14-percentile-point gain for a student starting at the 50th percentile. That’s a meaningful jump, enough to recover a substantial portion of pandemic-era losses.
Frequency matters, and the ideal schedule depends on the student’s age. For preschool through first grade, four to five sessions per week outperforms three. For students in grades two through five, three sessions per week actually produces better results than four or five, possibly because older students benefit from time to practice independently between sessions. The key across all age groups is consistency. Sporadic tutoring produces sporadic results.
Districts that stayed remote the longest and served the highest-poverty populations need the most intensive recovery efforts. Federal aid was distributed with this in mind, but spending that money effectively on sustained, high-frequency academic support rather than one-time purchases remains the central challenge for schools working to close the gap.