State test results released last fall in math and English scores since 2024, yet still lagged far behind the standard set before the COVID pandemic. Meanwhile, the National Assessment of Educational Progress, a federal exam administered biannually to hundreds of thousands of students, recently placed Oregon in the country. Adjusted for student demographics and poverty levels, it ranked 50th among states in fourth-grade math and reading, 49th in eighth-grade math, and 47th in eighth-grade reading.

Now local observers are pointing to Oregon’s relatively brief school year, as well as high rates of absenteeism, as one explanation for the dismal results. In released by the nonprofit group Stand for Children, researchers show that sizable gaps in seat time between Oregon and other states — and even larger ones separating districts within Oregon — compound over years into massive disparities in opportunities to learn. Advocates argue that loose rules governing how states report attendance data also contribute to the problem.

Sarah Pope, the executive director of Stand for Children’s Oregon affiliate, said her state was “definitely on the low end” in terms of instructional time made available to children. On average, their school year lasts 165 days (compared with 179 days in the U.S. as a whole), and students receive 9 percent fewer instructional hours than their counterparts around the country; over time, that adds up to well over one year of missed schooling.

“When we tell people that we’re 9 percent short, their eyes glaze over because people can’t imagine what that means,” Pope remarked. “But when we tell them it’s a year difference over the course of a K–12 experience, which is like graduating as a junior, they’re like, ‘Oh gosh.’”

The group’s report notes that Oregon is one of just 10 states that sets no minimum of total school days per year, allowing districts to set their own schedules so long as they hit an annual minimum of instructional hours (900 for students enrolled in kindergarten, elementary school, and middle school, and slightly more for high schoolers). In practice, the state’s average K–8 student receives 1,111 hours of designated school time each year, considerably below the national average of 1,231 hours for K–12 students. Only Maine, Nevada, and Hawaii provide less schooling.

Those figures are drawn from by Brown University economist Matthew Kraft, which also found that differences in instructional time between states can be dwarfed by those within states. By the end of elementary school, for example, Oregon students living in a district at the bottom of the state’s school time rankings receive a full 1.4 years less education than those in a district at the top. The gap explodes to nearly three years’ worth of instruction by the time those students graduate high school.

Aside from the length of the school year, the pandemic-era spike in chronic absenteeism (the percentage of students missing 10 percent or more of the school year) has further eroded the amount of time that kids spend learning. And while that trend has proven stubbornly persistent in nearly every jurisdiction, Oregon’s spike has been higher and longer-lasting than most. , a think tank based in Washington, D.C., Oregon’s rate of chronic absence reached a stunning peak of 38 percent in 2022–23, only falling to 33.5 percent by 2024–25 (compared with a national average of 22 percent the same year). 

Kraft, who on the subject before the education committee of the Oregon House of Representatives, said it was “wildly inequitable” for students in different parts of a state to enjoy vastly less time with teachers than those elsewhere. Both researchers and elected officials needed to examine the intersection of poor attendance and inadequate instructional time more closely, he continued.

“The outliers offering substantially less time have wound up with far less learning opportunities for students,” Kraft observed in an interview with ����Ӱ��. “Curriculum is built around having x amount of minutes in a day to teach math or science, and when teachers and students don’t have that, the results illustrate the negative consequences.”

‘We should not be proud’

Those consequences could be reversed with policy changes, according to Stand for Children’s analysis. 

Using existing estimates of the of and on student test scores, the authors calculated that Oregon would dramatically improve its NAEP performance by lifting statutory requirements for schooling time and cutting absenteeism to pre-COVID rates. If those conditions were both reached, they found, Oregon students enrolled in kindergarten today would move from 48th in the nation in reading to sixth-place by the beginning of high school. A somewhat smaller leap, from 49th place to 25th place, could be achieved in math scores.

As of yet, no such sweeping changes are in the offing. If anything, a combination of diminished enrollment figures — the product of both lower fertility and a COVID-era flight from public schools — has led at least some districts to consider paring the school year back further. Reynolds School District, which enrolls around 10,000 students in the suburbs east of Portland, from its school calendar in response.

During , the state was around the country where districts chose to compress learning time. It has also been one of the national leaders in popularizing the four-day school week, with operating on a truncated schedule. While sometimes popular with family and school staff, that shift often leads to a deterioration in learning and comparatively few benefits in faculty retention.

Given Oregon’s clear decline in academic achievement, Stand for Children’s Pope said that district leaders should refuse to shorten their school year at the very least. Her organization is backing the passage of , a bill that would require state authorities to report on absenteeism four times during the school year, rather than just once, as is now mandated. Such a law — scheduled for hearings before the state Senate’s education committee in the coming days — would allow for school systems to conduct earlier outreach to families when their students are at risk of becoming chronically absent.

She also supports the adoption of a more exacting definition of learning time. At the moment, she said, up to 60 of the required instructional hours can be filled through activities like professional development and parent-teacher conferences, which occur when children aren’t in school.

“Do we think it’s right that our definition of instructional time has an allowance for approximately 10 days when kids don’t have to be there? And it can count for instructional minutes?” Pope asked.

Emielle Nischik, the executive director of the Oregon School Boards Association, said that the state’s students deserved “as much time in school as students around the country. But she added that more instructional hours and better data reporting could only be gained through increases in education spending. 

“Oregon school funding adjusted for inflation has essentially been flat since the 1990s, even as Oregon and the federal government have added staffing and services requirements that cost money,” Nischik wrote in an email. “We are open to any discussion of increasing class time as long as it comes with the understanding that more days will cost more money.”

Last summer, Gov. Tina Kotek of $11.3 billion to cover the K–12 system through 2027. According to by the National Education Association, Oregon spent nearly $19,000 per pupil in daily attendance in the 2023–24 school year, ranking 20th among all states. 

Kraft compared the resource of time to that of money. While lawsuits have been won to force states and districts to spend more money on schools, no such litigation has focused on learning time as a necessary educational input. 

“That has not been the case around time, in large part because schools are following the law, and the minimums we set in many states — not all, but many — are very, very low. We should not be proud to have met these minimums.”

The flood of resources was backed by an extensive body of evidence. Dozens of studies conducted before the pandemic showed that the positive effects of tutoring were among the largest ever seen in education policy. To help a generation of young learners return to their pre-COVID trajectory, advocates argued, there appeared to be no strategy more effective than recruiting thousands of tutors to provide regular supplemental instruction. 

But a shared exclusively with ����Ӱ�� raises doubts about whether the remarkable learning gains measured in prior studies can actually be produced by the kinds of large-scale initiatives that have been launched since 2020. Released Monday, the wide-ranging overview of over 250 high-quality studies finds that as tutoring programs grow, their impact steadily shrinks. 

The findings, which are predominantly drawn from pre-COVID papers, dovetail with disappointing results of some local efforts that have been undertaken in the pandemic’s wake. They also reflect the well-acknowledged reality — observed throughout education research and the social sciences more generally — that the enormous benefits sometimes seen in highly controlled settings are seldom if ever carried over to larger populations. 

Study author Matthew Kraft, an economist at Brown University the spread of tutoring, said that the promise of the approach should not be eclipsed by the “high, and sometimes outsized, expectations” attached to it.

“We have to be realistic about how hard it is to do anything well in education out of the gate, let alone make fundamental changes to the core structures of teaching and learning,” he said.

of the boost stemming from “high-impact” tutoring, which emphasizes one-on-one or small-group instruction in large doses, have been sizable — about as much as an entire year of reading growth for elementary schoolers, and twice that seen by high school freshmen, as quantified through standardized test scores. By comparison, the advantages conferred to students in larger interventions ranged from one-third to one-half that magnitude.

University of Virginia Professor Beth Schueler, Kraft’s co-author, argued that those outcomes remained “pretty impressive,” if not the equal of what had been measured previously. 

“Even though the large-scale programs weren’t replicating the enormous effects that you find with small-scale trials, the size of the impact that we find for these more policy-relevant studies are still quite meaningful.”

Notably, the 265 studies included in Schueler and Kraft’s analysis are all built around randomized control trials, seen as the empirical gold standard in quantitative research. They were all conducted in the countries making up the , a group of wealthy, industrialized nations whose education systems are often compared against one another. 

Across the entire sample of studies, average effects from tutoring were roughly equivalent to those found in earlier research reviews. But improvements to test scores shrank substantially when the authors looked only at programs enrolling between 400 and 999 pupils; they grew smaller still when restricted to those enrolling more than 1,000. 

Robert Balfanz, a professor at Johns Hopkins University School of Education, observed that the early hype promoting tutoring as a “silver bullet” for COVID-related learning loss was destined to be deflated when school districts began leveraging them to reach thousands of struggling students. Still, he added, even high-enrollment efforts delivered important growth to children.

“This study just shows the reality that [tutoring] is a very effective intervention, but it’s going to take a lot of time and patience and learning to get it to work at scale,” said Balfanz, who has contributed to to recruit 250,000 tutors and mentors to work in schools. “Even then, scale is always going to diminish what you can do for a smaller group.”

An issue of scale

Emerging research on COVID-era tutoring initiatives has attested to the complexities facing state and district leadership. 

Kraft last month of Nashville’s program, which was established in 2021 and has grown to incorporate about 10 percent of the district’s total students. Over its first two years in operation, students’ reading performance has improved only modestly, with no corresponding gains in math. Another low-touch experiment, targeting middle schoolers in suburban Chicago, detected only a slight upturn in standardized test scores from a handful of tutoring sessions offered over Zoom.

But some advocates caution that it may be premature to measure the influence of tutoring systems that only got underway during a public health emergency. Buffeted by school closures and an uncertain budgetary picture, the initial transition to tutoring was rocky in many areas. Districts found it challenging to coordinate with families who had disengaged from schools, and an ultra-hot labor market made tutoring recruitment especially difficult.

Ashley Bencan is the chief operating officer of the , which launched as a pilot in the summer of 2021. Since then, the organization has grown to partner with 10 district and charter school partners in over 30 locations. But even buoyed by federal and state funding, Bencan said, local schools have struggled to build up tutoring systems on top of their typical organizational demands. 

Even collecting data on which students participate in tutoring — a vital step in determining whether the efforts actually work, Bencan said — can test the capacity of both school districts and state education agencies.

“If you’re juggling all the different things you have to work on to kick off the school year — reviewing data, grouping kids, filling positions — they have to meet those basic needs first, and only then think of what else they can do,” she said. “Tutoring isn’t designed to meet those basic needs, and we need to think about how we make it part of a school’s model.”

The logistical challenges of shoehorning tutoring into already-packed school schedules, finding sites where sessions can occur, and connecting families with tutors, can be considerable. Though Kraft and Schueler write that the design of successful tutoring programs can be effectively duplicated at a larger scale, they also find that implementation quality sometimes suffers in the course of expansion. Polls of district leaders that larger schools consistently saw lower participation rates from students, and only about one-sixth of principals in one survey reported that they had faced no barriers in providing tutoring.

Encouragingly, Kraft and Schueler’s analysis suggests that some program structures can withstand the pressures of scale. If the programs conducted in-person tutoring during school hours, featuring a student-tutor ratio of no more than 3:1, and met at least three times each week (along with other conditions), their effects were more robust with larger numbers of students. While the average impact for a program serving 100–399 pupils was 42 percent smaller than one serving less than 100, those employing the high-quality practices listed above saw their effects diminished by just 18 percent.

Schueler said the diminished, though still significant, effects of scaled-up tutoring may simply suggest that policymakers have underestimated both the scale of learning loss and the hurdles to manufacturing new learning assets from scratch.  

“We are finding suggestive evidence that those implementation challenges are real, and policymakers need to think about how to get that stuff right.”

Seemingly minute differences in the length of a school day or year, whether stemming from state laws or local rules governing school districts, eventually grow into colossal gaps in learning opportunities. Over the course of their K–12 careers, the authors estimate, children living in jurisdictions requiring the most time in school benefit from over two years more education than those living in areas that require the least. 

“It’s hard for me to understand why some students should have access to a 180-day school year, and others in a district down the road get two weeks less instruction,” said co-author Matthew Kraft, an economics professor at Brown University. “Why would we want that inequity baked into our system?”

The paper, published Monday in the American Education Research Journal, relies primarily on figures collected before the emergence of COVID-19. But its resonance will inevitably be heightened by the post-pandemic crisis of chronic absenteeism, one-quarter of students nationwide missed at least 10 percent of the school year in 2023–24. At the same time, owing both to budgetary challenges and popular choice, of school districts are shifting to a four-day week.

The pronounced geographic divergences in access to instructional time are largely the product of state laws. While 37 states mandate a minimum number of days in the academic year, their requirements range from 160 days in Colorado to 186 days in Kansas; among the 37 states that set a floor for instructional hours in a year, Arizona is at the bottom with 720, while Texas is at the top with 1,260.

In other words, while the average American K–12 school is in session for 179 days a year, for just under seven hours each day, local variation is much wider. 

Drawing on data from the U.S. Census’s nationally representative , the study finds that schools at the 90th percentile of instructional time nationwide offer 1.17 more hours of school each day than those at the 10th percentile. Throughout the school year, those approximately 70 minutes per day accumulate into a disparity of 196 hours of teaching, or about five and a half weeks of school annually.

Some school districts set their own requirements for time in school higher than those set by their respective states. But at the median, schools in the five states that set highest minimum amount of instructional time (Texas, Nebraska, Arkansas, Mississippi, and Alabama) are open for 133 hours more per year than those in the five states with the lowest minimums (Hawaii, Nevada, Maine, Oregon, and Rhode Island). Cumulatively, students in the first five states will receive 1.4 more years of schooling from kindergarten through the twelfth grade.

Even acknowledging those differences, some kids are actually exposed to less instructional time than their state or district stipulates. As a case study, Kraft and co-author Sarah Novicoff examine the Providence Public School District in Rhode Island, which offers 1,174 annual hours of instructional time for elementary schools and 1,215 hours for secondary schools.

After using state data to tally students’ excused and unexcused absences, teacher absences, tardies, suspensions, and a host of outside interruptions — Kraft that intercom announcements, staff pullouts, and principal “fly-bys” can disrupt the typical classroom as many as 2,000 times each year — the authors calculate that a typical Providence elementary schooler misses 16 percent of their intended instructional time. The average high schooler misses as much as 25 percent.

Novicoff, a former middle school teacher now pursuing a doctorate at Stanford, said that school staff and administrators should aim to harvest low-hanging fruit during the school day by doing everything in their power to minimize in-class disruptions.

“They can say, ‘If I want to pull a kid from that classroom, I’m going to shoot their teacher a chat message instead of banging on their door,’” she suggested. “The difference there is the degree to which students notice and are disrupted.”

Effects on achievement

But while some educators work to maximize the time available to them, others have happily embraced a shorter school week over the last few years.

Between 2019 and 2023, the number of school districts operating on four-day weeks from about 662 to almost 900, according to a count kept by Paul N. Thompson, an economics professor at Oregon State University. The switch has been especially popular in more rural districts that tend to face greater transportation challenges and welcome a simplified schedule.

But Thompson’s research shows that districts in Oregon that made the change have seen substantial losses in achievement. Although the schools compensated for the missing day by lengthening the four remaining, students still lost out on several hours of school each week. Notably, the learning losses at those schools grew the longer they stayed on a four-day week, suggesting that the effects were compounding as students lost more instructional time.

Recently, Thompson undertook of four-day weeks around the United States, again concluding that they were associated with significant declines on test scores. The academic slippage was greatest in schools that lost more instructional time, as well as those in less rural settings. 

Emily Morton, a co-author of that study and a researcher at the said she thought her findings about the benefits of more instructional time generally dovetailed with those of Kraft and Novicoff. But she recommended that, before changing the law to require more school hours and days, states should heed the example of Providence and find ways to maximize the instructional time already being provided.

“It seems wise and more cost-effective to first focus on recovering time that currently is ‘lost’ during the school day or year (through interruptions, announcements, absences, etc.),” Morton said. “I would also say we have good evidence that summer school can positively impact student achievement, particularly in math.”

Indeed, schools and educators sometimes resist when pushed to remain in session for longer. After New Mexico passed a law last year that significantly lifted the minimum number of annual instructional hours s, the state education department declared that districts by offering a 180-day year. In response, . 

Kraft suggested that state authorities consider approaches that would allow communities to opt in to longer school years or experiment with ways of increasing instruction.

“We have to be conscientious about the potential unintended consequences of increasing minimums,” he said. “It has to be done in a way that schools and districts feel supported.”

The most important task lying ahead for education leaders is reversing the tide of disengagement and absenteeism that has rocked schools the last four years. Kraft and Novicoff’s data from Providence dates back to 2016, but nationwide attendance plummeted during the era of virtual schooling and has not recovered. It is reasonable to expect that, during the 2023–24 school year, millions of absent students missed tens of millions of classroom hours. 

Jennifer Davis, a former official in the U.S. Department of Education and the co-founder of the , called chronic absenteeism a “huge problem” that schools would have to overcome to keep their students on-track for graduation. Additional resources, including community outreach navigators and alternative learning experiences, might be necessary to rebuild the connections between students and schools, she added.

“Without this,” Davis wrote in an email, “we are going to lose the COVID generation.”

The program’s effects suggest that more exposure to supplemental instruction could yield still greater benefits. But design limitations, particularly those stemming from a pronounced shortage of qualified tutor candidates, also raise the question of whether it can be offered on the scale that some advocates envision. 

The paper was released only days after U.S. Secretary of Education Miguel Cardona declared, in a speech laying out his department’s priorities for the coming year, that every student who has lost ground during the pandemic should receive 90 minutes of tutoring each week. A number of states and large districts have already established tutoring initiatives over the last year, typically underwritten by emergency relief funds from Washington.

That push has been backed by a flood of research indicating that high-quality, high-dosage tutoring can help children realize almost unheard-of learning progress. The bulk of that literature — reviewed in a 2020 meta-analysis of nearly 100 randomized controlled trials — focused on instruction that was delivered in-person, though several COVID-era papers from Europe have also shown significant academic gains resulting from online tutoring.

To further investigate the virtual approach, researchers at the University of Chicago, Brown University, and the University of California San Diego helped organize a pilot in Chicago Heights, Illinois, a city roughly 30 miles south of downtown Chicago. Partnering with , a nonprofit mentorship organization founded by college students in 2020, the research team recruited 230 volunteer instructors and paired them with local middle schoolers for regular, one-on-one sessions held over Zoom. Students enrolled in the Chicago Heights school district are overwhelmingly poor and non-white, and only about one-quarter were meeting grade-level standards in math and reading before the pandemic.

Partly owing to the challenge of recruiting volunteers, participating students were organized into three waves that began accessing their tutoring at different times over a 12-week period. On average, students in the first wave received about four hours of total tutoring, while those in the following waves received a little over two hours. In keeping with prior research findings, students in the first wave saw more learning growth (as measured by performance on standardized tests) than those who were tutored less. While the results were not large enough to be classified as statistically significant, said co-author Matthew Kraft, they provide further evidence that the achievement gains from tutoring will accelerate when delivered in higher dosages.

“There’s this explicit interpretation, which I’ve written about in my own work, that higher amounts of tutoring produce better results,” said Kraft, a professor of education at Brown. “What we do see very clearly is this difference across our higher- vs. lower-dosage waves, which were consistent with that pattern.”

Still, he acknowledged, the boost that came from a comparatively brief, low-cost tutoring pilot was “much smaller” than the average effects found in some earlier studies. That can partially be explained by the program’s design: It leveraged the volunteer labor of undergraduates — rather than highly trained teachers or paraprofessionals, who have been shown to be more effective as tutors — and carried out instruction digitally rather than in person. But the cost-efficient nature of the experiment — among its few expenses was a $50 Zoom account — may recommend its model to educational leaders wondering how to pay for tutoring once federal relief funds are depleted.

“It was a very low-cost model,” Kraft said. “That’s something that I think districts are struggling with, despite having pretty substantial amounts of federal aid to work with, at least in the short- to medium-term. They recognize that there’s going to be a financial cliff, so how do you build something you don’t have to just abandon?”

A problem of scale

Both the study’s potential and its constraints are notable at a moment when states and districts are laboring mightily to establish their own tutoring interventions. High-profile efforts in , , and have mobilized millions of dollars in public and philanthropic money in the hopes of reaching thousands of kids who lost ground during months of school closures. A from data tracker Burbio showed that roughly one-third of districts planned to spend relief funds on tutoring.

But the frenzied start-up phase necessitates filling hundreds, or even thousands, of new positions; that kind of hiring blitz would be a daunting prospect in the best of times, but it has become especially thorny under the current economic conditions. While they would undoubtedly prefer to use current or former teachers to staff their programs, a historically tight labor market has already led to shortages in core instructional positions. 

The struggle to recruit qualified tutors is “no different in kind from the staffing challenge they are experiencing for every job within a school system right now,” said Mike Magee, CEO of Chiefs for Change. “The same market forces are impacting staffing in tutoring as are impacting staffing in teaching and bus drivers, etc. It does make scaling promising efforts significantly more difficult.”

Tennessee is one of the states that has experimented most energetically with tutoring, launching a $200 million initiative last year that seeks to reach 150,000 struggling students. Schools in Elizabethton, a rural district serving around 2,600 students, are among the 79 districts that have chosen to participate in the , offering small-group tutoring to students scoring below proficient in state reading exams.

Myra Newman, Elizabethton’s assistant superintendent for instruction, said that school teachers and principals “have seen the benefits” of the tutoring, which she hopes will be evident when students sit for their spring exams. But for the moment, the district provides only one tutor (accompanied by several assistants) for each school, and tutoring sessions are offered in a three-to-one format, rather than one-to-one.

“We wouldn’t have personnel — or the funds to hire the personnel — to do one-on-one,” Newman said. “But a three-to-one ratio is pretty good because they still get some individualized attention, and some modeling for their peers.”

Magee argued that, whatever the pitfalls faced by school systems abruptly setting up their tutoring initiatives, the learning emergency presented by COVID necessitated rapid action; while some of the programs in the field may not adhere to the ideal held up in research, they can play a critical role in stemming further learning loss now.

“In a crisis like the one we’re experiencing, which is generational in its negative impact on students, you might well want to move faster and accept some amount of downsides in order to reach as many students as possible,” he said.

But Kraft, who has been a consistent and influential advocate for states devoting increased resources to tutoring, sounded a cautionary note to district leaders.

“There is a tradeoff in navigating the current climate where what is possible might not be scalable. So instead of just saying, ‘Come hell or high water, I’m going to build a huge tutoring program,’ we might be better off starting off with a small program and building it over time, in a sustainable way — rather than cutting corners or changing the model.”