Are Nontraditional School Calendars Harmful for High School Students?

Landon and Pope (2026) study whether changing how schools schedule instructional time—specifically longer school days paired with fewer school days—affects student achievement and teacher turnover  . They ask whether reallocating the same total annual instructional hours alters productivity. Using administrative data from over 2 million Los Angeles Unified School District students (2002–2012) and policy-driven calendar changes, they estimate causal effects. They find minimal impacts for elementary and middle school students. In contrast, high school students experienced test score gains of roughly 0.08–0.15 standard deviations in math and 0.06–0.10 in English after moving to shorter, more numerous school days. Teacher turnover also declined under the traditional schedule.

Why This Article Was Selected for The Policy Scientist

This article addresses a policy question with unusually broad relevance: whether institutional schedules—how time is structured rather than how much time is allocated—shape productivity, learning, and workforce stability  . The topic extends well beyond schooling, informing debates on workweeks, shift design, and human capital formation. It is timely given current experimentation with compressed schedules and persistent concerns about adolescent performance. The authors have contributed extensively to research on education, time use, and behavioral responses, situating this study within a mature line of inquiry. The analysis leverages a large administrative dataset covering over two million students, enabling precise estimation. The design exploits policy-driven calendar transitions using difference-in-differences with student fixed effects, a credible causal inference strategy. External validity may reasonably extend to large, urban districts operating under similar constraints, though institutional differences warrant caution.

Full Citation and Link to Article

Landon, T. J., & Pope, N. G. (2025). Schedule-driven productivity: Evidence from nontraditional school calendars.Journal of Policy Analysis and Management, 45(1), e70075. https://doi.org/10.1002/pam.70075 

Central Research Question

Landon and Pope (2026) investigate whether the way schools structure mandated instructional time influences student achievement and teacher turnover, holding constant the total number of annual instructional hours  . Rather than asking whether more time in school improves outcomes, the authors examine how reallocating fixed instructional hours across the school day and school year affects productivity. Specifically, they evaluate calendars that lengthen the school day while reducing the number of school days, as well as calendars that redistribute vacation breaks without altering daily schedules. The central question is whether these scheduling choices alter academic performance, attendance patterns, and workforce stability among teachers. This framing situates school calendars as a policy lever governing time allocation efficiency rather than resource expansion.

Previous Literature

The study builds on several strands of research concerning time use, educational production, and labor productivity. Prior work on instructional time has largely emphasized quantity—additional days, extended years, or increased classroom hours—often yielding mixed findings. A related literature evaluates later school start times, with consistent evidence that delayed starts improve adolescent outcomes through sleep-related mechanisms. Another body of research examines compressed schedules, including four-day school weeks, which typically produce small but negative achievement effects, partly attributable to reduced total instructional time. Year-round calendar studies have similarly reported modest or negligible effects when total days remain unchanged but breaks are redistributed.

Landon and Pope extend this literature by isolating the allocation dimension of time policy. Unlike studies confounded by changes in annual hours, their analysis explicitly holds yearly instructional time fixed. This distinction clarifies whether performance differences arise from scheduling design itself rather than exposure length. The paper also contributes to research on teacher labor markets, where schedule characteristics, workload intensity, and working conditions may shape retention. In synthesizing these threads, the authors address an underexplored issue: whether the temporal organization of schooling affects productivity through behavioral responses such as fatigue, attendance, or at-home study.

Data

The analysis relies on comprehensive administrative records from the Los Angeles Unified School District (LAUSD), spanning academic years 2002–03 through 2012–13  . The dataset includes over two million students enrolled in grades K–12, linked longitudinally across years. Student-level measures include standardized math and English test scores from the California State Tests, normalized within grade-year cohorts; attendance records enabling calculation of the fraction of days absent; indicators for grade repetition; and high school dropout status. The data also identify English language learner status but lack individual demographic variables such as gender or race. To compensate, the authors merge school-level demographic characteristics from the Common Core of Data, including racial composition, student–teacher ratios, and proportions eligible for free or reduced-price lunch.

Teacher-level records provide information on school assignments, years of experience, and turnover, defined as leaving a given school between consecutive years. School calendar data from the California Basic Educational Data System identify whether each school operated under a traditional or nontraditional calendar and the timing of transitions. The scale, longitudinal depth, and administrative precision of these data enable detailed estimation of both student and teacher outcomes while minimizing measurement error typical of survey-based datasets.

Methods

The authors employ a difference-in-differences framework exploiting the staggered elimination of nontraditional calendars across LAUSD schools  . During the 1990s, many schools adopted alternative calendars to address overcrowding. Declining enrollment and litigation subsequently induced the district to phase out these calendars between 2002 and 2012. This policy-driven variation generates quasi-experimental conditions in which schools transitioned at different times. The identifying assumption is that, absent calendar changes, outcomes in transitioning schools would have followed trends parallel to those in always-traditional schools.

The baseline specification includes school fixed effects and grade-by-year fixed effects, controlling for time-invariant school characteristics and common shocks. Additional controls incorporate lagged student achievement, attendance, and suspension indicators, mitigating concerns about compositional changes. Alternative models introduce student fixed effects, allowing estimation from within-student variation as individuals experience calendar transitions. Standard errors are clustered at the school level.

Recognizing potential biases from heterogeneous treatment effects, the authors implement estimators proposed by Sun and Abraham (2021) and Borusyak et al. (2024). Dynamic treatment effect models evaluate pretrends and post-transition trajectories. Additional robustness checks adjust for possible pre-existing trends using methods developed by Freyaldenhoven et al. (2021). Collectively, these techniques represent contemporary best practice in causal inference using panel data, strengthening interpretability relative to cross-sectional multivariate regression.

Findings/Size Effects

Results indicate that calendar structure exerts heterogeneous effects by student age group. For elementary and middle school students, transitioning from nontraditional to traditional calendars produces negligible changes in standardized math and English scores. Estimated effects are small and statistically indistinguishable from zero across specifications. Attendance outcomes differ modestly: elementary students exhibit reductions in absences following transitions, with declines ranging from approximately 0.2 to 0.4 percentage points, equivalent to roughly 0.36 to 0.72 fewer days absent per year. These patterns suggest limited sensitivity of younger students’ learning to scheduling design but some responsiveness in attendance behavior.

In contrast, high school students display substantively meaningful achievement gains after moving from calendars characterized by longer school days and fewer school days. Standardized math scores increase by roughly 0.08 to 0.15 standard deviations, while English scores rise by approximately 0.06 to 0.10 standard deviations. These magnitudes are comparable to widely cited estimates of improvements associated with large changes in teacher quality. Effects appear broadly distributed across the achievement spectrum, though higher-achieving students experience somewhat larger math gains.

Analyses of ancillary outcomes reveal limited changes in grade repetition or dropout probabilities, with estimates generally close to zero. Teacher outcomes show clearer effects: transitions away from longer-day, fewer-day schedules reduce turnover by about 3.3 percentage points, corresponding to an approximate 16 percent relative decline. No consistent turnover effects emerge for calendars that merely redistributed breaks without altering daily schedules.

Mechanism analyses suggest that earlier school start times and reductions in at-home study time plausibly account for observed achievement effects. Fatigue associated with longer days explains only a small fraction of the estimated impacts. Reductions in overcrowding appear insufficient to explain results. Dynamic models reveal relatively stable post-transition effects with minimal evidence of differential pretrends for most outcomes.

Conclusion

The study demonstrates that reallocating fixed instructional time across the day and year can influence academic performance and teacher retention, particularly among older students  . By isolating scheduling effects from changes in total instructional hours, the authors clarify an often-overlooked dimension of education policy: temporal design. The findings indicate that longer school days paired with fewer school days are associated with lower high school achievement and higher teacher turnover relative to traditional schedules. Younger students appear largely unaffected academically, though attendance improves modestly under traditional calendars.

Methodologically, the analysis leverages credible causal inference strategies, including fixed effects and policy-induced variation, supplemented by modern estimators addressing treatment heterogeneity. The scale and administrative quality of the dataset support precise estimation. External validity plausibly extends to large urban districts operating under comparable institutional constraints, while differences in demographics, labor contracts, or transportation systems may moderate effects elsewhere.

Overall, the article contributes to a broader understanding of how institutional time structures shape productivity. It reframes calendar policy as a question of efficiency and behavioral response rather than mere exposure, extending implications beyond education to organizational design more generally.