The Air At The Back Of The Classroom

AA classroom is among the most densely occupied rooms anyone will ever sit in. Thirty children in sixty square metres breathe out a great deal of air in the course of an afternoon, and in a building sealed for warmth that breath accumulates. Carbon dioxide is the usual measure of the problem, not because the gas is itself dangerous at the levels found in schools, but because it rises and falls with ventilation and is cheap to record. Guidance across most of Europe treats 1,000 parts per million as the point at which a room ought to be aired; a survey of 340 primary schools published in 2019 reported a mean afternoon reading of 1,450 ppm, with a tenth of the rooms above 2,500. A second contaminant enters from outside. Fine particulate matter, PM2.5, comes mainly from traffic, and the schools most exposed to it are precisely those on the roads where opening a window is the only practical way of clearing the carbon dioxide. Campaigners speak of the two together as a single crisis of school air. The researchers who measure them are markedly less willing to join them up.

BRenata Villiers, an environmental epidemiologist at the Ashfield Centre for Urban Health, has spent nine years asking what the readings do to learning. Between 2017 and 2022 her group placed continuous monitors in 128 classrooms across four cities and set the record against the end-of-year assessments sat in those same rooms: 41,000 pupil-years in all. A rise of 400 ppm in the average afternoon reading went with a fall of about 1.5 per cent in mathematics marks, and with no measurable movement in reading marks. The effect is small, and Villiers is the first to say so. What makes it worth attention, on her account, is not its size but its distribution: the badly ventilated rooms were not scattered at random but concentrated in the oldest buildings, so that the same children met the same air every afternoon for years on end. A small daily penalty, repeated without interruption, is not the same object as a small penalty.

CAn association is not a cause, and the obvious objection — that old buildings are poor in many other ways too — troubled her enough to send her looking for an accident. She found one in 2018, when a regional authority bought portable filtration units for 46 schools in anticipation of an industrial fire that never happened, and installed them on a timetable that had nothing whatever to do with the schools' results. Mathematics marks in the treated schools rose by 0.06 of a standard deviation over the following year; reading marks did not move. The gain, however, was confined to schools within 400 metres of a road carrying more than 20,000 vehicles a day; in the quieter schools the units made no measurable difference at all. The units captured particles and left carbon dioxide untouched, which suggests that where the air is merely stale such machines are beside the point.

DPiotr Nowak, an education economist at the Larkmoor Institute, accepts the measurements and disputes what is drawn from them. A filtration unit, he argues, is never fairly judged against doing nothing; it is judged against whatever the same money would otherwise have bought. Fitting and running the units for a decade costs roughly £3,400 a classroom, and he calculates that the identical sum spent on extra teaching hours delivers gains in mathematics between four and six times larger in the trials he has reviewed. Nowak volunteers two limitations. The teaching trials he draws on were run in schools that had applied to take part, and applicants may be schools where additional teaching happens to work unusually well; and his comparison assumes the two kinds of spending do not act on each other, when a class that is alert may also be a class that is easier to teach.

EThe argument now has a statute behind it. Since 2023 the education authorities of one European region have required a carbon-dioxide monitor in every classroom and an annual publication of the readings, and three further jurisdictions have drafted comparable measures. Nowak, whose figures are the more often quoted by campaigners, treats the duty to publish as the most useful part of such laws, because a school obliged to record its air cannot afterwards dispute what that air was. Villiers is here the more sceptical of the two. A monitor, she observes, reports ventilation and nothing besides, and a school beside a dual carriageway can bring its published figure down by opening the windows while raising indoors the concentration of the very contaminant the 2018 units had been installed to capture. Both dislike the habit, common among suppliers, of quoting reductions in carbon dioxide as though they were reductions in harm.

FWhat is emerging is a division of the problem rather than a verdict on it. Where the dominant contaminant is the children themselves, in schools set well back from traffic, ventilation is cheap, immediate and sufficient, and the monitors the regulators favour measure exactly the right thing. Where the dominant contaminant arrives from outside, the two remedies pull against one another, and a building must be sealed and filtered rather than aired. Which of the two cases is the more common depends on how a country's schools happen to be sited, and neither researcher claims to have counted. What both refuse is the framing that has governed the public argument — that school air either matters or does not — when the answer plainly turns on the address.