- Key issues
What's the gender data gap?
A review of Caroline Criado-Perez’ ‘Invisible Women’
By eleanor blackwood
Tue 21 July 2020
How important is data?
Data is an incredibly powerful tool: it underpins and informs the technologies that shape our lives. Our medicines, responses to world crises, transport and so much more.
Recent media coverage has been heavily focused on misuses of personal data, from Cambridge Analytica’s interference in elections to Amazon’s plan to use artificial intelligence in their hiring policy.
But there is a larger story to tell: the data used to improve and co-ordinate almost every aspect of our lives is overwhelmingly male.
Caroline Criado Perez’s 2019 study, Invisible Women, winner of the Royal Society Science book prize, brilliantly points this out. She observes that while life is significantly improving for one half of the population, as smart phones are designed for male hands, and safety equipment is designed for the male body, the same tools can cause biologically female people to experience discomfort, injury and even death as a result of this oversight – a hangover from the past with profound ramifications for today.
What is the 'gender data gap'?
The gender data gap is the central subject of Caroline Criado Perez’s Invisible Women: Exposing Data Bias in a World Built for Men (Penguin: 2019). In short, the gender data gap is the fact that the vast majority of data that we have and use to design solutions is based upon both the male body and the typical male life pattern.
Essential to understanding the gender data gap is the acknowledgement that ‘normal’ is often in fact a proxy for ‘male’. This gap emerges obviously in the context of testing or modelling, when the face that the 'average human' is assumed to be the average man.
What are the consequences of the gender data gap?
The consequences of the gap are far ranging. Some consequences may seem insignificant, such as ‘one size fits all’ clothes, where all = most men, but become far from insignificant when taken further. For example, with Personal Protective Equipment (PPE). A 2017 TUC report found that only 5% of women working in the emergency services said that PPE never hampered their work.
A consistent theme of Criado-Perez's book is that 'one size fits all' doesn't fit women, and may endanger them.
Inadequate safety equipment
One of the most important arenas in which data about the human body is implemented is product design.
Phones and clothes are casualties of the ‘one size fits all men’ oversight, but many more women are its victims when it comes to the safety measures that we all rely on. The safety measures in a car, in particular seatbelts and airbags, have been designed with the male body in mind, and crash dummies, which are used to test them, usually have the dimensions of the average man. Inevitably, they often do not do the job they are meant to on genetically female bodies, which are on average significantly smaller.
Despite the fact that men are more likely to crash their car, women are 47% more likely to suffer a serious injury, and 17% more likely to die if they are involved in a collision. Both seatbelts and airbags were designed in the 1950s – while they have gained sophistication, the fundamentals have remained unchanged – and both were designed by men without properly taking into account female dimensions.
And this remains unchanged.
Despite the fact that car crashes are the primary cause of foetal death due to maternal trauma, a seatbelt suitable for pregnant women has still not been developed. But even when they aren't pregnant, women’s smaller and lighter frames are simply not taken into account.
Under EU law, there is no regulatory safety test on cars which requires a dimensionally correct female crash dummy. There's one which requires a 95th percentile female body - i.e. one which is smaller than only 5% of women - but this only needs to be tested in the passenger seat, which assumes that women don't drive. This dummy doesn't even represent the female anatomy accurately, instead it's a scaled down male figure.
Almost all other tests require a dummy representing a 50th percentile male body, ensuring safety measures work on the average man. The sobering statistics, which render a car crash significantly more deadly for women and babies, are not being taken into account by testers or lawmakers, who persist in a mindset where safe for men means safe for all. As Criado Perez puts it, ‘the case is unanswerable.’
Companies forget to take female needs into account
The assumption that male = normal can cause male-dominated companies to ‘forget’ female-specific elements.
For example, Apple have come under scrutiny for the late arrival of a period tracker in their in-built ‘Health’ app - failing to consider that it would be a desirable addition when the app was first conceived. Women, when it comes to data, are still viewed as an ‘outlier’, an inconvenient set of statistics outside the parameters of ‘normality’, defined in masculine terms.
Office temperature, Criado-Perez argues, is another facet of everyday life long overdue a rethink.
Calculated in the 1960s, it remains based on the optimum room temperature for a forty-year-old man with a body weight of 70 kilograms. While the ideal temperature for men is around 22 degrees, women are generally most comfortable about three degrees higher. Furthermore, due to their slower metabolic rate, women often feel colder at the same temperature, leaving them to wrap up in scarves and blankets, and bring space heaters into their place of work.
As well as general discomfort, lower temperatures in the office also negatively and disproportionately impacts women’s productivity, creating an unnecessary and avoidable obstacle to women’s enjoyment of daily life, and perhaps their success in the workplace.
Healthcare: male-only cell research ♂
There are major differences between biologically female bodies and biologically male ones.
They are generally smaller, have a greater percentage of body fat, and of course, often have a menstrual cycle, which depends on hormonal fluctuations. The differences run deeper still – sex differences are found in every organ of the body, so it would seem inevitable that medical care should have different approaches for two materially different bodies. Yet even in medical care the one size fits all approach persists, as medical research mainly studies the male body:
A 2011 review of cardiovascular journals found that when sex was specified 69% of cell studies reported using only male cells. And ‘when sex was specified’ is an important caveat: a 2007 analysis of 645 cardiovascular clinical trials (a published in prominent journals) found that only 24% provided sex specific results. A 2014 analysis of five leading surgical journals found that 76% of cell studies did not specify sex and of those that did, 71% included only male cells and only 7% reported sex-based results. And again, even for diseases that are more prevalent in women, researchers can be found ‘exclusively’ studying XY cells […]
Perhaps of a more urgent concern for women’s health is the 2016 discovery of a sex difference in how male and female cells respond to oestrogen. When researchers exposed male and female cells to this hormone and then infected them with a virus, only the female cells responded to the oestrogen and fought off the virus. It’s a tantalising finding that inevitably leads to the following question: how many treatments have women missed out on because they had no effect on the male cells on which they were exclusively tested?’
(Invisible Women, p207-8)
This lack of engagement with female healthcare has meant women have suffered a deficit in provision and tailored advice for the female body.
Plus, this skewed approach may also be failing men. Criado-Perez is currently putting pressure on the government to seek out sex aggregated data to understand coronavirus, as understanding why more women are surviving this virus may be essential in finding a cure.
Healthcare: 'too complicated' or 'atypical'
Female hormonal fluctuations means that their cells are habitually seen as ‘too complicated’ for medical experiments. The data gap here converges with a healthcare deficit, in which the challenges of the female body are not confronted but swept under the carpet as an inconvenient anomaly, in the hope of crafting an illusory world of simple, comprehendible masculinity.
However, the world is not biologically male-by-design. Heart attacks serve as a prime example. Chest pain, the universally recognised symptom of a heart attack, is actually a common male symptom. Women are more likely to experience fatigue, breathlessness and nausea, yet these are categorized are ‘atypical’ symptoms.
The result? Women are 50% more likely to be misdiagnosed or sent home after presenting with these symptoms, and as a corollary, more likely than men to die after a heart attack. Criado Perez’s research finds time and time again institutionalised sexism which erases women in the very spaces that should be working for their protection.
From the male pill, or lack thereof, which had its development halted due to side effects eerily similar to the ones found for any pill women have taken for the last sixty years, to the vaginal mesh scandal, it's clear that the health industry treats women differentially: addressing the data gap will be fundamental to fixing this.
Final notes 💡
The gender data gap is the fact that most of the data that we use to alter the world around us is based on the male body and life pattern, be that making it safer, preventing disease or treating it. Criado’s Perez’s excellent book is full of examples of incomplete data that show us both the hugely dangerous and negative impact the gender data gap has on the lives of women, as well as the necessity of addressing it.
It also raises valuable questions about the male-dominated field of tech, where women tend to make up under quarter of the workforce: what questions will fail to be asked? and how will that impact software we all increasingly rely on, for healthcare, safety measures, and product design?
She ends by arguing that, in order to improve our world for all, and use data to its full potential, we must ensure the data sets we use are comprehensive, the questions we ask pertinent, and consistently challenge the assumption that normal equates to male.
We fully recommend that you read this book, to understand a different type of data!