“Battle for the thermostat” is how Tom Chang and Agne Kajackaite began the title of their 2019 paper on an experiment run with 543 students in Berlin. Session by session the room was held at a different temperature, from roughly 16 °C up to roughly 33 °C, and the students worked through a maths task and a verbal one. When the room was warmer, the women solved more of them and the men slightly fewer; each additional degree brought the women a 1.76% rise in correct maths answers, while the men's decline was slight and the statistics could not separate it from chance. The authors trace the gain mostly to women handing in more answers with no rise in their error rate, and they read that, at least partly, as extra effort.
The title is apt; the limits are these. These were students, each session lasted about an hour, what changed was the air of an entire room rather than the spot where anyone sat, and on a test of cognitive reflection the temperature made no difference whatever. Still, within that room, warmth moved performance in maths and in the verbal task, which is reason enough to take the office chill seriously.
Read as a whole, the research suggests that women and men want nearly identical settings; the gap opens in how badly each of them takes it when the air at their desk wanders from that number. My reading, and it is the argument of this piece, follows from there: some share of that chill is settled on the floor plan, by what a desk sits beside, whether glazing, a ceiling outlet or a cold slab underfoot, and by the freedom its occupant has, or lacks, to change their own conditions without lobbying the whole team.
Olesen's table: 25.4 °C for men, 25.8 °C for women
The claim at the centre of this debate came from Maastricht in 2015. Writing in Nature Climate Change, Boris Kingma and Wouter van Marken Lichtenbelt contended that rules for indoor climate lean on a comfort model built in the 1960s, whose standard figures for metabolic rate were modelled on an average man and may overstate a woman's, at most, by 35%; buildings, on that logic, could be inherently wasteful when it comes to keeping women comfortable. Its evidence was a sample of sixteen women doing sedentary work, with no men measured, and that gap is exactly what Bjarne Olesen, of Denmark's DTU, took issue with the same year in the REHVA Journal.
Olesen went back to the studies on which comfort standards such as ISO 7730 and ASHRAE 55 were built, where as many women as men were recruited, sat at the same sedentary activity and wore the same light outfit. Pooling Nevins (1966) and Fanger, preferences averaged 25.4 °C among 488 men and 25.8 °C among 488 women. Fanger and Langkilde, in 1975, recorded 25.0 °C against 25.1 °C. As for real offices, Olesen notes that women may often be found colder than men in summer under air conditioning, and puts that mostly down to dress: the women dress for the weather, the men keep their jacket and tie.
A wider survey came from Finland's VTT research centre. Sami Karjalainen, in a review published by the journal Indoor Air (2012), found that most studies detect no significant gap between the sexes in neutral temperature, the setting at which someone feels neither warm nor cool. Well over half of the laboratory and field work nonetheless has women more dissatisfied than men in the same conditions, and by his pooled estimate women are 1.74 times likelier to voice discontent with the thermal environment. Women, he concludes, react more strongly than men when conditions depart from the ideal, particularly as the room cools, and, on average, need more personal control over their conditions and more room to adapt.
India's number, and what it cannot reach
In 2019 India wrote a starting point into the room air conditioner itself. On 30 October 2019 the Central Government, with the Bureau of Energy Efficiency, notified fresh energy performance standards for room air conditioners, and on 6 January 2020 the Ministry of Power reported that a 24 °C default setting had been made compulsory from 1 January 2020. According to the notification, as the ministry quoted it, star-labelled multi-stage capacity, unitary and split units rated one to five stars, with cooling capacity up to 10,465 watts, manufactured, commercially purchased or sold in India, were to ensure a default setting of twenty-four degrees Celsius.
It was framed as an energy standard, and a default is only where a unit starts. Placed next to Olesen's table, 24 °C lies 1.4 to 1.8 degrees under what those lightly dressed, seated volunteers preferred. The trouble is that a default is a single reading for a whole room, while Karjalainen locates the difference between women and men in departures from the optimum, and departures happen at particular chairs.
Four places the display does not see
ISO 7730 draws the distinction that a layout needs. Besides a mean vote on comfort for the body as a whole, it lists four sources of local discomfort: draught; a vertical temperature gradient between head and feet; a floor that feels too warm or too cool; and radiant temperature asymmetry, meaning a body trading heat unevenly with the surfaces flanking it. They work as a checklist for a plan before any name goes on a seat. Put a desk straight under the discharge of a cassette unit and its user sits in a draught. Set one flush against a large glazed wall and the person there trades heat with the glass on one flank and with the room on the other. Place one over a cold floor and the feet live in a different climate from the head.
The quarrel over the remote begins, as I read it, as a question about a seat.
On a crowded floor some of these seats may still have to be filled; the point is to know which ones before the order goes in. An office table arrives in quantity and lands in one spot at a time; drawing the layout while the order is still being specified keeps that decision in hand.
Control is the other half. If those most bothered by cool air need room to adjust, a workstation & open plan area can hold more than one climate: seats set back from the glass and the outlets for those who feel the cool, the exposed ones for those who do not mind them, and the freedom to swap.
I cannot cite a study for the benefit I would put first, so it stays a hope of mine: that the remote stops travelling from desk to desk.
Five figures, each with its blind spot
25.4 °C and 25.8 °C: what 488 men and 488 women preferred, dressed alike and equally occupied. Silent on the draught at any given desk.
35%: the most by which standard metabolic values may overstate a woman's, according to the Maastricht paper. Drawn from sixteen women, with no men measured alongside.
1.74: women's greater likelihood of reporting thermal discontent in Karjalainen's review. A measure of complaint, with nothing to say about preferred settings.
24 °C: the default required on star-labelled room air conditioners from 1 January 2020. It says nothing about any particular desk.
4: the local discomfort factors named in ISO 7730. Each can be hunted down on a drawing.
The argument between the Maastricht pair and their critic at DTU can stay open; the drawing cannot wait for it. Mark every outlet, every stretch of glazing and every cold slab on the plan, and let the seating chart follow those marks rather than the display on the wall.
Quick answers
Why do women report more thermal discontent at work than men?
Most studies find neutral temperature statistically alike for the two sexes; women, though, respond more sharply to departures from the ideal, especially once the air cools, and report thermal discontent 1.74 times as readily (Karjalainen, Indoor Air, 2012). Olesen (REHVA Journal, 2015) attributes the summer gap seen in air-conditioned offices mostly to clothing.
Were comfort standards like ISO 7730 built around men?
The studies behind ISO 7730 and ASHRAE 55 recruited as many women as men; dressed alike and doing the same seated work, men preferred 25.4 °C and women 25.8 °C on average (as tabulated by Bjarne Olesen, 2015). In Nature Climate Change (2015) the Maastricht authors point out that the model's standard metabolic figures were modelled on an average man.
What did India's 2020 rule on a 24 °C air-conditioner default require?
Under energy performance standards notified on 30 October 2019, star-labelled room air conditioners up to 10,465 watts of cooling capacity made, bought or sold in India had to come with a 24 °C default setting from 1 January 2020, according to the Ministry of Power. It set where a unit started, for the room as a whole.
How can a layout reduce the office chill?
Before assigning seats, go through the plan looking for each discomfort factor ISO 7730 lists at local level (draught, a vertical gradient between head and feet, a floor that runs hot or cold, radiant asymmetry), and plan areas with different conditions so that people who feel the cold can choose where to sit.