What does the evidence say about... increased instructional time?
Examining 'learning loss' and the calls to increase instructional time in post-pandemic schools
Written by: Matthew Schuelka
There has been a lot of conversation and debate in education these days about the implications from the past year regarding student learning. Emerging research, such as from McKinsey and the University of Cambridge and RTI International, suggest that school closures and distance learning has led to significant ‘learning loss,’ particularly in low-resource and marginalized communities around the world. While it has become clear that the COVID-19 pandemic’s effect on schools has blown open the already-existing racial and socio-economic disparities that students carry with them, the analysis and prescription of ‘learning loss’ has fallen back into the well-worn grooves of educational reform: more instructional time. There is now a burgeoning conversation about the means in which to increase instructional time, such as through summer school in the United States. In England, there are proposals to increase summer schools, extend the school day, and offer additional tutors. (The story broke on The Guardian just last Friday that the tutoring scheme relied on out-sourcing teenage tutors from Sri Lanka that were paid only £1.57/hour, so … not off to a great start.)
The metrics that define ‘learning loss’ themselves are not universally agreed-upon, and the calls to increase instructional time are often perpetuated by politicians and educational ‘experts’ with little attention paid towards the voices of teachers and students in articulating what they actually need and want. There are some excellent pieces pushing back against the notion of ‘learning loss’ that I would encourage you to read: for example, Larry Ferlazzo in EducationWeek and John Ewing in Forbes. In short, how ‘experts’ measure ‘learning loss’ is typically done by comparing standardized test scores before and after curriculum delivery, and assumes that there is a universal level of learning attainment that is fixed to a child’s age. This is how McKinsey, for example, can calculate ‘learning loss’ down to a number of months; because they are assuming that children learn and develop homogenously and universally.
The guiding question for this week’s newsletter is this: Given that the notion of ‘learning loss’ is already a debated concept, what are we to make of the calls to increase instructional time?
Evidence on Instructional Time and Student Achievement
As mentioned in the introduction, the use of increasing instructional time as an education policy reform tool is nothing new. In fact, it’s become something like an involuntary response to supposed educational ‘crises.’ Conversely, reducing instructional time has also become a typical response when educational systems face budgetary shortfalls. There are actually 25 states in the United States where at least one school district is using a 4-day school week to cut costs, according to the National Conference of State Legislatures. Budget issues aside, the basic answer to the premise of this newsletter is that the effect of increasing instructional time has produced modest gains in student academic achievement. However, this comes with a few big caveats.
While much of the research literature does point to modest gains in student achievement when instructional time is increased, it makes a significant difference how instructional time is increased. In other words, simply adding time does not equate to an increase in educational quality. Increasing instructional time is often a preferred policy solution because it is naturally quantifiable, i.e. “we’ve added x hours and students are performing y-percentage better!” Like anything quantifiable in education, however, what is often missing in the analysis is the quality and experience of the teachers and students themselves. Simplistic correlations like “x-hours = y-percentage change” completely misses all of the complex factors that go into what makes a school and what comprises student learning. Those x-number of hours mean teachers, and curriculum, and peer-effects, and the rhythms of a child’s body, to name only a few. Increasing instructional time does not simply mean that educational outputs rise at the same rate. In fact, there is evidence that there may a ceiling or limit to how much achievement can increase proportional to increased instructional time. This is similar to the work-culture in Japan, where Japanese work more hours on average than anyone else in the world, and yet Japan has the lowest economic productivity among top-tier OECD countries. Adding instructional time may boost student test scores by a modest amount, but it does not necessarily mean that educational quality is being holistically benefitted, and that it could not be even better if more attention were focused on quality rather than quantity. It should also be noted that by increasing instructional time within the ever-finite resources available to schools, other important things may be sacrificed such as academically enriching extra-curricular activities; subjects not deemed ‘important’ or ‘measurable’ such as art, music, foreign languages; and after-school sports.
A better illustration of how increasing instructional time is not necessarily a panacea for increasing student achievement is by comparing cross-nationally. In comparing international achievement test data with data on instructional time between 52 countries, David Baker and colleagues found that there was no correlation between test scores and instructional time. In other words, the countries with the most instructional time did not necessarily have the highest achievement scores. (According to international achievement tests, which are themselves controversial.) What Baker and colleagues did find, however, was that increasing instructional time within countries did lead to modest student achievement. However, as they argue, “the impact of instructional time is so dependent on its relationship to curriculum and instructional quality as to make it trivial compared to those more complex and primary resources in the schooling process” (p. 331).
The other major caveat when looking at the evidence of increasing instructional time is that the effects of such an increase are not evenly distributed. Several studies, such as one conducted by Michael Hayes and Seth Gershenson in 2015, found that increased instructional time actually benefitted the already-high achieving students more. This is known as the ‘Matthew Effect’ (no relation to the author of this newsletter!). This effect was also found in a study conducted in Switzerland by Maria Cattaneo and colleagues– perhaps even to a greater degree – in that they found that increased instructional time also increased educational gaps between students. In other words, increasing instructional time did not close the achievement gap between students, it perpetuated it. It may be possible to use increased instructional time in a more targeted fashion to focus specifically on students that are struggling, which has been done in low-performing schools in Florida and Washington, D.C.. However, extending the school day comes at high monetary cost and these costs may be much harder to sustain and implement with quality in low-resource educational systems, as pointed out by Umut Özek, which can also perpetuate education inequalities. He also notes that increasing instructional time for some students and not others might further stigmatize and marginalize those students and schools.
The Implications
The bottom line is this: increasing instructional time does lead to modest increases in student achievement, but the costs of such (both monetarily and systemically) are probably not worth it.
Examine the ‘learning loss’ discourse more carefully, and it is easy to see that policy-makers and educational ‘experts’ are caught up in constructed comparisons. This is often how education policy discourse is deployed: comparing students with other students, comparing schools with other schools, comparing countries with countries. There always has to be a winner and a loser. However, learning is only ‘lost’ when we believe that there is a fixed amount of learning to be had. The subject of re-implementing achievement testing during this school year has ignited a whole new firestorm of criticism and debate, and will be the subject of a future EdPop newsletter.
Instead of worrying about ‘learning loss’, what I am more concerned about is the loss of educational quality, capacity, and trust. I worry about the loss and re-establishment of schools as supportive community spaces. Yong Zhao, an educational researcher working both in Australia and the United States, suggests in a recent article in the journal Prospects that we need to avoid the ‘learning loss trap’ and resist the calls for increased instructional time; resist doubling-down on remedial – or ‘catch-up’ – work focused only on reading and mathematics, which are quantified. Instead, he suggests the following:
Meet the students where they are
Treat students as individuals that have had individual responses to learning during the pandemic.Pay attention to all educational outcomes
Avoid a myopic focus on reading and mathematics that sacrifices other kinds of knowledge, learning, and educational outcomes.Engage learners as partners of change and owners of their learning
Embrace the positive skills that learners may have learned during the pandemic, such as independence, resilience, and self-determination.Keep families engaged
Connect with parents/care-givers, particularly as they have had an intimate experience with their child’s learning over the past year.Keep online/remote learning
Embrace some of the digital innovations that have worked over the past year, such as opening up classrooms as global learning spaces and expand learning beyond the classroom walls.Build back better
Reduce the risks of future educational losses during system-shocks such as a global pandemic.
Select Evidence & Resources
Simon Andersen, Maria Humlum, and Anne Nandrup, “Increasing instruction time in school does increase learning,” Proceedings of the National Academic of Science,2016: 113(27)
David Baker, Rodrigo Fabrega, Claudia Galindo, and Jacob Mishook, “Instructional time and national achievement: Cross-national evidence,” Prospects, 2004: 34(3)
Aaron Benavot, “A global study of intended instructional time and official school curricula, 1980-2000,” Background paper prepared for the Education for All Monitoring Report 2005.
Maria Cattaneo, Chantal Oggenfuss, and Stefan Wolter, “The more, the better? The impact of instructional time on student performance,” Education Economics, 2017: 25(5).
Ümmühan Yeşil Dağlı, “Effect of increased instructional time on student achievement,” Educational Review, 2018: 71
David Farbman, “The case for improving and expanding time in school: A review of key research and practice,” National Center on Time and Learning, 2015
Michael Hayes and Seth Gershenson, “What differences a day can make: Quantile regression estimates of the distribution of daily learning gains,” Economics Letters, 2016:141
Elena Meroni and Giovanni Abbiati, “How do students react to longer instruction time? Evidence from Italy,” Education Economics, 2016: 26(6)
Umut Özek, “The effects of instruction time on student outcomes,” CALDER Policy Brief No. 7-0918-1, 2018
