Should adolescents start school later?

An adolescent’s body clock, or circadian rhythm, is generally set to a later schedule than that of older and younger people – teenagers’ bodies tell them to stay up later and get up later, too. This means it can be challenging when they need to perform academically in the early hours of the morning. Various pilot programs have found that later secondary school start times may help students get more sleep, feel less depressed, attend school more, and achieve higher grades. For example, when one school changed its start time from the already relatively late 8:50am to 10am, 12% more students made “good academic progress,” while the average number of student sick days dropped from 15 to 11.

Yet chronotypes – our natural preferences around sleeping and waking patterns – differ between adolescents, just as they do between adults. Night owls prefer to sleep later, while morning larks like to go to sleep earlier. The teenagers most likely to underperform academically when school starts early are those predisposed to being night owls. Morning larks, on the other hand, tend to do best. Starting school later might therefore be helpful for night owls – but unhelpful, or even detrimental, for morning larks.

What’s the impact of different school start times?

But there is a problem with the evidence: Most of the studies looking to disentangle the effect of student chronotypes from school start times haven’t used a randomized design.

“Kids – or parents, at least – usually pick the school with the timing they prefer,” says Guadalupe Rodriguez Ferrante of the Laboratory of Neuroscience at the Universidad Torcuato di Tella in Buenos Aires, Argentina, who recently co-authored a study on adolescent chronotypes and school timings. “If you are a night owl, maybe you’re going to choose an afternoon school. That leads to a bias in the study’s results.” In that case, there might be a smaller effect of chronotype on grades and attendance than if students had been randomly assigned to a school timing.

Among other things, this makes it difficult to determine whether night owls are underperforming academically because their chronotype is at odds with school timings – or if there’s something intrinsic about a later chronotype that predisposes them to underperform.

This is what Rodriguez Ferrante examined at a secondary school in Buenos Aires that is unusual in offering three different school start times. The 259 students in the study were randomly assigned to start the school day at 7:45am, 12:40pm, or 5:20pm. Chronotypes were assessed with a questionnaire, and the resulting data from both when students were in their first year, and their fifth year, were analyzed.

“Most of the studies looking to disentangle the effect of student chronotypes from school start times haven’t used a randomized design.”

It was especially interesting to study chronotypes and school timings in Buenos Aires, Rodriguez Ferrante says, because “people in Argentina tend to have really, really late chronotypes. We have dinner at 10pm,” she says. “But the average start time for a secondary school is 7:30 or 7:45am.”

Many of her team’s findings support what other studies have found. On average, students attending later classes had better academic performance. But this changed as students got older, and differed by subject. For example, first-year students, aged 13-14, who attended afternoon classes had better average grades in math than those attending morning or evening classes – but in the fifth and final year, at age 17-18, those who attended the evening classes achieved higher math grades than their peers in the earlier classes. This likely reflects a shift in circadian rhythms during adolescence, Rodriguez Ferrante and her co-authors write.

“On average, students attending later classes had better academic performance.”

Interestingly, the pattern was not the same for language classes. Grades were higher in languages than in math, with fewer differences between students of the different start-time cohorts, but determining whether there was in fact an effect would require studying more students.

As expected, night owls who had morning classes performed worse in math than their morning lark peers. Students were also more likely to be promoted to the next grade at the end of the year if their chronotype matched their school shift.

Among the older students, the night owls performed worse at math than morning larks in the same classes, no matter the time of day. This suggests that it isn’t just a mismatch between chronotype and school timing that has a negative effect on performance (although, Rodriguez Ferrante notes, the difference was the greatest in the morning school shift). Rather, something else may be going on.

Why do night owls perform worse than morning larks?

The night owls’ poorer performance could stem from simply sleeping less. On school days, fifth-year students in the morning cohort slept less than 6 hours a night, on average. But around 16% of students slept for just 4½ hours, or even less. While the researchers haven’t yet determined whether members of the latter group were more likely to be night owls, people with later chronotypes often do sleep less overall, Rodriguez Ferrante says.

Although the study was randomized, the results may have been affected by certain biases, Rodriguez Ferrante points out. The widely held cultural belief that someone who goes to bed late and wakes up late is lazy, for example, might lead teachers to inadvertently give morning students better grades than evening students. Or students might see themselves in a certain light – night owls might think that there’s no point trying hard because they’re lazy.

“For the vast majority of adolescents, school is simply starting too early.”

“It’s really difficult to disentangle these effects,” Rodriguez Ferrante says. “We need to keep studying the association between chronotype, school timing, and academic performance. I’m eager to find out what mechanisms underly the effect of chronotype performance. Is it about how students see themselves? What is the link?”

Yet the takeaway is clear, according to Rodriguez Ferrante: For the vast majority of adolescents, school is simply starting too early. “It’s not easy to change. You need to reach politicians,” she says. “And that means collecting as much data as we can so that we can really have a positive influence on the life of adolescents.”