Later School Start Times Add Sleep, but Rarely Raise Grades

What the research found

Parents ask whether a teenager would do better if school began an hour later. Adolescents given that hour almost certainly sleep longer, by something in the range of twenty-five to seventy-seven minutes a night across the studies that have measured it. The evidence that those minutes reach the report card is thinner, and the campaigns arguing for the change rarely mark the difference.

One hundred and twenty-seven publications on adolescent sleep and school timing were assembled into a single account by Valentina Alfonsi and four colleagues in Rome and Modena, published in the International Journal of Environmental Research and Public Health in 2020. They are explicit that this is a narrative review rather than a systematic one, appraising no study for methodological quality and pooling no results, so the well replicated findings and the shaky ones sit in the same summary.

Adolescents need roughly nine hours of sleep a night, and about three quarters report getting less than eight. The American Academy of Pediatrics recommends that middle and high schools begin no earlier than 8:30 in the morning.

The gain comes entirely from the morning. Later start times pushed rise time back by between twenty-one and sixty-one minutes in most studies reviewed, while bedtime moved in no consistent direction and most often did not move. Carskadon and colleagues watched the relationship run backwards: students who moved from an 08:25 start in ninth grade to an 07:20 start in tenth lost about twenty minutes of sleep a night and grew sleepier by day.

Nearly every study measured sleep by questionnaire or diary. Dunster and colleagues instead used wrist actigraphy across a whole district, and a one-hour delay produced nearly half an hour more sleep per night. Net gain appears once the delay exceeds sixty minutes.

Wahlstrom and colleagues reported better academic performance after a delay, though two others found no significant effect on self-reported academic outcomes and a third an effect that varied by year group. The review’s own conclusion is that there is no clear association between start times and school performance. No randomised trial has been run, and the studies reporting a benefit cannot separate extra sleep from the simpler possibility that a test taken later in the morning is taken closer to an adolescent’s best hour.

The most quoted success belongs to Kelley and colleagues, who moved a school from 08:50 to 10:00 and recorded illness absences more than halved alongside significant academic improvement. That study included no sleep measures of any kind, which leaves the mechanism unrecorded.

What it means for your child and their school

Beyond attainment, the associated changes hold together better. Daytime sleepiness fell in every study conducted except two, in-class attention improved, and reaction times were faster in later-start samples. Three studies reported better mood and positive affect, and later start times were also associated with reduced caffeine use and with a 16.5 percent reduction in vehicular accident rates.

Benefit does not always persist. Rhie and colleagues observed an early increase in sleep duration decay as the students’ own body clocks drifted later to meet the new timetable, and several samples lost their gains over time. A Singapore cohort held its gains in sleep duration, daytime alertness and mental wellbeing at one month and again at nine months, so durability appears to vary with the setting rather than to fail everywhere.

A parent raising this with a school stands on firm ground asking about sleep and on soft ground asking about results. The precise version of the request is that a delay of more than sixty minutes is what the literature associates with a real net gain in sleep, that shorter delays frequently produce nothing measurable, though Chan and colleagues did record positive effects after fifteen minutes, and that the outcome worth tracking afterwards is daytime alertness.

Attendance is where the conversation usually turns, and the numbers there are specific enough to quote. Three studies found no significant variation in overall absences after a delay, and three found a considerable reduction in marks for lateness. A school hoping the change will fill empty seats is likely to be disappointed; a school troubled by a stream of pupils arriving after registration has reason to expect a difference in the late marks.

The obstacles the review names are practical rather than scientific: the cost of rescheduling buses, traffic, overlap with neighbouring schools, and the disruption of sport and other activities that run after the last lesson.

The evidence supports an hour, taken in the morning, argued on sleep. Argued on marks, it is a case the research does not yet carry.