The Shrinking School Week

Effects of a four-day schedule on student achievement


How much face time do students and teachers need to keep pace with expectations for learning? It’s an urgent question during a pandemic that has kept many students out of school buildings for more than a year. The importance of school attendance has divided communities across the country, as they weigh the potential risks of in-person instruction with those of prolonged separation from the school environment.

We can find at least some answers in the experiences of schools that have adopted four-day school weeks, typically as a cost-cutting move. I studied the academic performance of nearly 700,000 students in Oregon, where more than 100 schools in school districts facing budget shortfalls and attendance problems opted to cut instructional time instead of raising taxes or laying off teachers. My study looks at student test scores in reading and math over a 15-year period to see what happens when schools switch to a four-day week.

I find clear negative consequences for student learning when schools adopt four-day schedules. Although many schools start class earlier or end later during the four days they are in session, overall weekly time in school decreases by three to four hours. My analysis finds that, as a result of those reductions, math scores decrease by 6 percent of a standard deviation and reading scores decrease by 4 percent of a standard deviation. These impacts are comparable to those associated with other cost-saving measures, such as increasing class sizes and cutting student-support programs.

When a local community can or should open school buildings during a pandemic is a political decision, and whether schools can effectively educate students with live remote instruction is an open question. But the impact of decreased instructional time on student learning is not. These results show that when students receive less than a full-time school schedule, learning slows.

Figure 1: Half of U.S. States Have at Least One Four-Day School

Cutting Costs by Canceling (Some) Class

The earliest known use of a four-day school week dates back to the 1930s in South Dakota. Today, most four-day schools are in the rural North and West. Their numbers have increased dramatically over the past two decades, from 257 schools in 1999 to more than 1,600 in 2019. Half of all U.S. states have at least one four-day district or school (see Figure 1).

Many school districts offer remedial or enrichment services on the day off, while others effectively create a three-day weekend for all students. When I studied the activities of 552 four-day school districts that provided information about the off day, about half were fully closed to students and staff on the off day. About 30 percent offered either remedial or enrichment programming to students, ranging from teacher office hours to field trips to very structured off-day programs. About one-quarter of the districts that did not offer student programs on the off day provided teacher professional development.

Nationwide, U.S. schools with four-day schedules offer an average of 148 school days, well below the 175–180 average typically provided under a traditional five-day schedule. Many districts lengthen each school day in order to meet their state’s minimum requirements for instructional hours. Four-day schools average seven hours and 45 minutes of instruction each day, while five-day schools have shorter days that average six hours and 54 minutes. The end result is a loss of three to four hours of instructional time each week.

Most often, four-day schedules are adopted in response to a budget crunch, with communities opting to cut back on school hours rather than lay off teachers or let class sizes grow. However, in some cases, districts have made the switch to cut down on commuting time and cite improved teacher retention and student attendance as the main rationales. Regardless, the cost savings are relatively small. In a recent paper, I found a reduction of 1 to 2 percent in per-pupil operating expenses at districts that switched to a four-day schedule.

What happens when districts decrease the amount of time students spend in school? School attendance has a major influence on multiple dimensions of child development and family life. Schools provide academic instruction, which promotes knowledge and skill accumulation, and they influence social-emotional development as well. Students also gain access to school meals and physical education on campus, which may promote their overall health and well-being. And for some students, school can be a safe haven from instability, adult-sized responsibilities, or other challenges at home.

Less time in school also can pose stark challenges for families. The lack of school on a weekday is difficult for working parents, as the pandemic experiences of families and broad exodus of women from the workforce have made clear. A 2019 study found that four-day school weeks were associated with declines in workforce participation for women overall, but not for men or single mothers. In addition, children without supervision on the off day may engage in unproductive or risky behaviors—a 2018 study in Colorado found that juvenile crime jumped by almost 20 percent when schools switched to four-day weeks. Other studies have found greater marijuana use and a higher prevalence of bullying and sexual activity among students attending schools with four-day weeks.

A shorter week may also change the quality of some school-level educational inputs. If, for example, teachers generally find the schedule appealing, as a 2018 study of Missouri teachers found, districts could attract a larger and potentially higher-quality pool of candidates and be more likely to retain talented staff. But the longer school day and potential loss of weekly subject-specific instructional time could also present instructional challenges. If teachers do not use the longer school day effectively or modify their courses to align with the new composition of instructional time, student learning may decline.

Assessing Impact in Oregon

In this study, I examine the impact of four-day school weeks on academic achievement in math and reading. My analysis focuses on students in Oregon, where approximately one in 10 schools follow such a schedule. I explore three main questions: How does the four-day school week affect student achievement? How large are the achievement returns to instructional time—both overall and in specific subjects? And how effective is the four-day school week as a cost-savings approach?

To answer these questions, I look at performance on annual statewide tests in Oregon for 690,804 students in grades 3 through 8, from 2004–05 to 2018–19. The student-level data also includes characteristics such as sex, race, free-and reduced-price lunch eligibility status, English as a second language program participation, and special education or gifted status. I also look at student absences and the percentage of days missed due to disciplinary incidents.

The total number of Oregon schools with a four-day school week increased from 108 in 2005 to a peak of 156 in 2014, before falling to 137 by 2019. For both four- and five-day schools, I calculate weekly time in class based on the start and end time of the school day, adjusting for early dismissals. Based on this data collection, 80 percent of the schools have identifiable weekly time in school information.

Overall, students in four-day schools have lower standardized math and reading test scores compared to students in five-day schools. In math, about 61 percent of students at four-day schools pass annual tests compared to about 65 percent at five-day schools. In reading, about 68 percent of four-day students pass compared to about 71 percent at five-day schools. These differences amount to about 7 to 10 more students passing annual tests at the average-size school.

There are some other key differences between four-day and five-day schools that could be contributing to the differences observed in achievement, however. Four-day schools have larger shares of low-income students, at 57 percent compared to 50 percent at five-day schools. White students make up 79 percent of enrollment at four-day schools and 65 percent of enrollment in five-day schools. Because they are predominately rural, four-day schools also have much smaller average student enrollments, at 578 students compared to 3,817 students at five-day schools.


Impact on test scores. The clear differences between four-day and five-day schools suggest that simply comparing their achievement levels may not provide an accurate picture of the effect of the schedule change alone. To provide a better view, I look at how the achievement of students in specific districts changes when they shift from a five-day to a four-day week and compare those changes to contemporaneous trends in achievement of other districts that did not make the change.

Students earn lower math and reading scores on standardized tests after their schools switch to a four-day schedule. Overall, average math scores decrease by 5.9 percent of a standard deviation in math as a result of the switch to the four-day school week, while reading scores decrease by 4.2 percent of a standard deviation. That is nearly one-third the size of the impact of having a larger class size, and equal to losing 40 minutes of reading instruction and about an hour of math instruction each week.

I also look at how the schedule change affected performance of student groups. Math scores for special-education students improve by 2.6 percent of a standard deviation after switching to a four-day week; reading scores do not change. For English learners, reading scores fall by 4.1 percent of a standard deviation, but math scores held steady. One potential explanation for these findings is that while individualized education plans may help special education students supplement math instruction at home, weekend learning loss may be exacerbated for English learners if English is not the primary language spoken in their home environment.

In considering students by age group, I find that negative math achievement effects are most prominent in 7th and 8th grades. The negative reading achievement effects are more consistent across grades than math, but the largest negative impacts are also found in 8th grade. One potential explanation for these larger negative impacts in later grades may be that parents have more difficulty helping students with more advanced math and reading coursework at home. The earlier start times at four-day schools may also negatively impact adolescent students, leading to larger negative achievement effects in these later grades (see “Rise and Shine,” research, Summer 2019).

I also track student test scores over time to better understand how switching to a four-day schedule affects achievement. There is a noticeable drop in test scores immediately after a school switches to a four-day schedule: in the year of implementation, math scores decline by 6.8 percent of a standard deviation and reading scores decline by 3.7 percent of a standard deviation.

After that initial dip, test-score performance tends to improve in subsequent years. This suggests that achievement losses ameliorate a few years after adoption of the four-day school week, but tells little of whether this is a feature of all four-day school week adoption (for example, students becoming more acclimated to the new school schedule) or driven by transitory four-day schools returning to the five-day schedule. I examine both possibilities and find that the lasting impacts of the four-day week are minimal for schools that eventually switch back to a five-day schedule. By contrast, the negative impacts for schools that permanently adopt four-day weeks appear to increase in magnitude in each subsequent year (see Figure 2). Four years after switching to a four-day week, students’ math scores fall by 8.8 percent of a standard deviation and reading scores fall 10.4 percent of a standard deviations compared to the year before adoption.

For a nation concerned about the long-lasting impact of school closures during a pandemic, these results should be reassuring. They suggest that briefly limiting students’ time in school, in this case through a four-day school week, may result in a short-term negative achievement shock for students, but has no lasting detrimental impacts on their achievement. Continuing shorter school schedules, however, could have lasting negative effects.

Figure 2: Drops in Test Scores After Switching to Four-Day Schedule

Returns to time in school. A shift to a four-day week typically produces sudden changes in the amount of instructional time students receive both overall and in specific subjects. The size of these time changes can vary, however, based on a local district’s schedule and whether it opts to offer remedial or enrichment services on the day off. This makes it possible to use the changes as a natural experiment to study how differences in instructional time influence student learning.

I find that changes in the total amount of time that schools are open to students do affect student achievement, helping to explain why students attending some four-day schools experience larger negative impacts than others. In particular, a one-hour increase in weekly time in school increases math achievement by about 1.8 percent of a standard deviation. The effects on reading are smaller, at 0.8 percent of a standard deviation.

A more interesting metric to examine, however, is subject-specific instructional time. Earlier studies by Victor Lavy and by Maria Cattaneo, Chantal Oggenfuss, and Stefan Wolter have found that a one-hour increase in weekly subject-specific instructional time boosts achievement in that subject by 6.0 percent of a standard deviation, but this is an average figure across all subjects. I used survey data on the allocation of time to different subjects in Oregon schools to calculate the returns to instructional time separately for math and reading. These calculations suggest that increasing weekly instructional time in math by one hour boosts achievement by 11.5 percent of a standard deviation. A one-hour increase in weekly reading instruction improves reading achievement by 2.5 percent of a standard deviation. We would therefore expect the learning lost due to a reduction in instructional time to be greater in math than in reading, a pattern that is evident in emerging research on American students’ achievement in the wake of the pandemic.

Cost savings. Do the cost savings from switching to a four-day school week provide a sufficient tradeoff for these losses in student achievement? Many cost-cutting interventions are associated with declines in student achievement, and a recent study by C. Kirabo Jackson, Cora Wigger, and Heyu Xiong found a $1,000 reduction in per-pupil spending reduces average test scores in math and reading by 3.9 percent of a standard deviation (see “The Costs of Cutting School Spending,” research, Fall 2020).

The average savings from switching to a four-day school week in Oregon are $350 per student, or about a 2 percent reduction in expenditures. Based on my achievement results, I find that cutting costs by $1,000 per pupil through shrinking the school-week schedule yields an achievement loss of between 10 percent and 19 percent of a standard deviation. By comparison, other research has found that when schools cut $1,000 in spending through increasing class sizes, achievement falls by 12 percent of a standard deviation. Spending cuts achieved through closing schools are associated with declines of up to 20 percent of a standard deviation. In short, this trade-off is in line with or better than some other cost-cutting interventions, but worse than what would be expected from a reduction in general expenditures.

A Cautionary Tale

A four-day school week that reduces instructional time has a negative and statistically significant impact on student learning. Evidence from 15 years of test scores across Oregon show that student achievement drops when schools switch to a four-day schedule, and that those negative trends continue so long as five-day schedules are not restored. These detrimental achievement effects appear largely driven by reductions in weekly time in school, which decreases by three to four hours.

As a cost-cutting move, adopting a four-day school week is comparable to other program cuts and presents a viable option for financially-troubled school districts to consider. But switching to a four-day schedule is likely to have implications beyond just cost savings and achievement. School attendance is an important opportunity for students’ development overall, and it provides a safe daytime activity that benefits working parents. Diminished exposure to school-based counseling and health services, school meal programs, and other supports could also negatively affect child physical health and social-emotional development. Less time in school may mean more time for risky behaviors or exposure to unsafe conditions at home or in the community. Thus, it is critical for future research to examine these outcomes before making determinations regarding the overall efficacy of this school schedule.

Policymakers in several states are pushing back against the four-day school week, which suggests they are aware of some of its negative ramifications. In Oklahoma, for example, a new state law sets minimum quality standards for school districts to implement or continue a four-day schedule, which advocates estimate more than 90 percent of current four-day districts will not be able to meet. In New Mexico, where four-day weeks are popular but under pressure, public debate has included concerns about the experiences of low-income, working families.

Still, it seems likely that four-day weeks may well continue to grow in popularity, even if cost savings is not the motivating factor. The economic trends as we emerge from a prolonged global pandemic are not yet certain, and immediate budget cuts at schools appear unlikely in the wake of the $170 billion in federal education aid under the 2021 American Rescue Plan. But teachers, students, and families also have now experienced radically different learning schedules due to school closures and broad adoption of hybrid schedules, which mimic the part-time nature of the four-day school week. That may stoke interest in a four-day schedule and put pressure on local school boards to consider it. These findings suggest that they proceed with caution.

They also point the way to an expanded research agenda regarding four-day schools: not only how reduced attendance and instructional time affect academics, but the impact of what students are doing on their extra day “off.” If students and educators want to explore school calendars outside of the typical five-day-a-week schedule, we need to know how to structure flex time to enhance and extend in-school learning. Otherwise, we risk compounding the learning losses students have already sustained in the wake of Covid-19.

Paul N. Thompson is associate professor of economics at Oregon State University and a research affiliate at the Institute for Labor Economics.

This article appeared in the Summer 2021 issue of Education Next. Suggested citation format:

Thompson, P.N. (2021). The Shrinking School Week: Effects of a four-day schedule on student achievement. Education Next, 21(3), 60-67.

Last Updated


Notify Me When Education Next

Posts a Big Story

Business + Editorial Office

Program on Education Policy and Governance
Harvard Kennedy School
79 JFK Street, Cambridge, MA 02138
Phone (617) 496-5488
Fax (617) 496-4428

For subscription service to the printed journal
Phone (617) 496-5488

Copyright © 2024 President & Fellows of Harvard College