The Sleeping Teenager Problem

AThat adolescents are hard to rouse in the morning ranks among the oldest of household complaints, and for most of the twentieth century the explanation on offer was a moral one: a pupil who could not leave his bed had gone to it too late. Sleep researchers now treat that account as, at best, half an answer. The difficulty, they maintain, lies less in the hours adolescents choose than in the hours their institutions have chosen for them, since the school day was fixed around the convenience of adult employment rather than around any evidence about when learning is most efficient. Since the first districts moved their opening bell in 1998, the argument has travelled out of the laboratory and into local politics, where it competes with bus contracts, sports fixtures and the shift patterns of working parents. In several countries the policy has now outrun its supporting evidence. Which mechanism actually produces the reported gains, and for whom, remains a matter of dispute.

BThe biological case begins with a shift in the timing of the circadian clock that accompanies puberty and, in most young people, reverses again by the middle of the twenties. Ingrid Novotny, who directs the chronobiology unit at the Fenwick Institute, has tracked 1,340 pupils since 2009 and reports that the evening rise of melatonin, the hormone that initiates drowsiness, arrives on average 108 minutes later at seventeen than at twelve. Because the morning bell does not move with it, the distance between the sleep the body seeks and the sleep the timetable permits widens with each year of secondary schooling. Novotny calls the resulting deficit a scheduling injury rather than a habit. The deficit she measures is not uniform, however: it exceeds two hours a week only among pupils in the final two years of school, while younger cohorts recover most of what they lose on free days.

CEvidence of a different kind comes from districts that have changed their timetable and then changed it back. Kirtwell, an authority of 42 secondary schools, delayed its opening bell from 08:00 to 08:50 in 2014, restored the earlier time in 2018 when a transport contract lapsed, and so produced something close to a reversal experiment. During the four intervening years unexplained absences fell by 19 per cent and overall attendance rose by 2.4 percentage points; both indicators drifted back to their earlier levels within eighteen months of the restoration. The effect on examination results was smaller and less tidy, amounting to roughly 0.09 of a standard deviation, a figure that only 19 of the 42 schools reproduced in their own returns. Reversal designs are rare, and this one was not planned. Even so, the symmetry of the two transitions is harder to attribute to novelty or to publicity than a single change would be.

DWhat a later start actually does to a pupil remains the most contested part of the field. Novotny attributes the improvements to alignment: the closing portion of sleep, rich in the stage associated with the consolidation of newly learnt material, is the portion an early bell removes first. Duarte Menezes, an educational psychologist at Ravensmere College, is sceptical of that reading and contends that duration alone accounts for the results, since a pupil who gains fifty minutes in bed will benefit equally whether those minutes arrive before dawn or after it. His strongest card is a group of districts in which the bell moved later while bus collection times did not, so that total sleep was unchanged; there the reported benefits disappeared. Menezes concedes that his own trial ran for eleven weeks, recruited sixty volunteers, all of them over sixteen, and scheduled their sleep under laboratory supervision, conditions that no timetable reproduces.

EObjections to later starts are rarely framed as objections to the science. Most districts run a tiered bus system in which the same vehicles serve secondary pupils before returning for primary ones, so that moving a single tier obliges the whole fleet to be rescheduled, at a cost Kirtwell put at 3.7 million in the first year. Sport, part-time employment and the supervision of younger siblings press in the same direction, and head teachers report that resistance is loudest among families whose own working hours are least flexible. A further complication is meteorological rather than financial: Astrid Nordvik has shown that above 55 degrees of latitude a later bell in midwinter removes the morning daylight that itself pulls the body clock earlier, so that during the darkest ten weeks the intervention may work against the physiology it is meant to serve. That finding has not yet been replicated outside Scandinavia.

FThe honest summary is that the evidence is stronger than its critics allow and narrower than its advocates imply. Almost every published evaluation is observational, and districts willing to absorb the disruption of a later bell tend also to be districts investing in other reforms at the same time, which makes any single cause difficult to isolate. The two explanations are, in any case, not exclusive; the most defensible reading is that duration does the greater part of the work while alignment adds a further increment, largest in those pupils who were previously sleeping least. On that reading a delay of half an hour, protected against a compensating shift in transport, would capture most of the available benefit at a fraction of the cost of a full hour. Whether districts will find even that modest sum is another question.