Invisible Women: Exposing Data Bias in a World Designed for Men

The preliminaries

• The gender data gap is the lack of trend data for girls and women.
  • This is the product of a way of thinking that has been around for centuries: He is the Subject, and She is the Other.
    • We proceed as if the male body and its life experiences are the gender-neutral norm.
  • This gap doesn’t just happen because it’s not collected; where it is collected, it’s not separated from male data–this is called “sex-disaggregated data.”
• These gaps have consequences.
  • This book demonstrates how often and how widely this bias crops up, and how it distorts supposedly objective data that rules our lives.

The Default Male

• Most cultures and societies are deeply male-dominated, which means that the male experience and perspective is seen as global, while the female experience–half the population–is seen as niche.
  • It’s why the English national football team’s Wikipedia page is about the men’s national football team, while the women’s is called the England women’s national football team, and why articles about women usually include words like “woman” or “female,” but articles about men don’t contain words like “man” or male, because the male sex goes without saying.
  • It even appears in language – numerous studies of language have consistently found that what is called the “generic masculine,” where you use words like “he” in a gender-neutral way, is actually read overwhelmingly male.
    • So, when the generic masculine is used, people are more likely to think a profession is male-dominated, and more likely to suggest male candidates for jobs and political appointments, and women are less likely to apply for jobs that are advertised using the generic masculine.
  • And while there is evidence that women generally accept men as role models, men rarely do the same for women. Women buy books by and about men, but men rarely buy books by and about women.
• Beyond this, whiteness and maleness can go without saying because other identities don’t really get said at all.
  • Male universality is a cause of the gender data gap: since women aren’t seen or remembered, and because male data makes up the majority of what we know, what is male comes to be seen as universal.

Can snow-clearing be sexist?

• In a town in Sweden in 2011, the government worked on a gender-equality initiative that required them to reevaluate their policies through a gendered lens. One official laughed, saying that at least “the gender people” would keep their noses out of snow-clearing.
  • At the time, as in most places, snow-clearing began with the major traffic arteries and ended with pedestrian walkways. But this affected men and women differently because men and women travel differently.
    • Women are more likely to walk and take public transport. Their travel patterns are more complicated–women do 75% of the world’s unpaid care work, which complicates their travel, between dropping children off at school, taking an elderly relative to the doctor, and buying groceries. 
    • Men are more likely to have a simple travel pattern: a twice-daily commute in and out of town.
• Snow-clearing was not gender-neutral at all, so the councilors switched the order of snow-clearing to prioritized pedestrians and public transport users, which actually saved them money, because the cost of pedestrian accidents in icy conditions was about twice the cost of winter road maintenance. 

• The original snow-clearing schedule benefitted men at the expense of women because of the gender data gap.
  • The men who originally created this schedule knew how they themselves traveled, and designed around those needs. This is a problem in the highly-male dominated profession of transport generally.
  • Engineers and planners are predominately male, and focus mostly on mobility related to employment.
    • But because women tend to walk further and longer than men for care-giving responsibilities, this marginalization and de-prioritization of non-motorized travel affects women more.
    • Additionally, men tend to travel solo, while women are encumbered by either children, elderly relatives, or shopping.
• For women to work, the city has to support their ability to work, and one of the best ways to do so is to design transport systems that enable women to do their unpaid work and get to the office on time.
  • It’s difficult to change things like fixed infrastructure (subways and trains), but with buses, routes and stops can be adjusted, placed in safer places, etc.

Gender-neutral with bathrooms

• Why are there longer lines at women’s bathrooms than men’s? In part because of the male-based design.
  • It might seem equitable to give both bathrooms the same amount of floor space.
  • But if the men’s bathroom has cubicles and urinals, the number of people who can relieve themselves at the same time is much higher per square foot of floor space in the men’s bathroom.
  • But even if there were the same number of stalls, the issue isn’t resolved because women take 2.3x longer to use the toilet.
  • Women also are the majority of the elderly and disabled, two groups that usually need more time. And they are more likely to be accompanied by children. And then there’s a quarter of the women of childbearing age who might be on their period, who need to change their tampon or pad. And beyond that, pregnancy reduces bladder capacity, which increases the need for a toilet.
• The critical need for safe access to toilets:
  • Per the UN, ⅓ women in the world cannot safely access toilets, and girls and women spend 97 billion hours a year finding a safe place to relieve themselves.
  • Human Rights Watch spoke with young girls working in tobacco fields in the US and found that they would avoid going to the bathroom during the day. They also tried not to drink anything, which increased the risk of dehydration.
  • More than half of Mumbai’s 5 million women do not have an indoor toilet, and there aren’t free public toilets for women. But there are thousands of free urinals.
    • A 2016 study found that Indian women who relieve themselves in fields are twice as likely to fact non-partner sexual violence, compared with women with a household toilet.
    • Increasing the number of toilets has a direct result in decreasing sexual assault, which also saves governments money over time.

The workplace

• The unpaid work discrepancy:
  • Although this discrepancy might be decreasing at an individual level, at a population level, even if a couple pays for domestic help, the remaining unpaid work is left to women.
    • And as women have increasingly joined the paid labor force, at a population level, men have not matched this comparative increase in unpaid work–instead, women increased their total work time.
  • In an Australian study, researchers found that while housework time is most equal for single men and women, when women start to cohabit, their housework time goes up while men’s goes down, regardless of employment status.
  • Most of the unpaid work women do is invisible, and over time, pay gaps add up.
    • Because they can’t afford to save over their working years, many women face poverty in old age. And when governments design pension schemes, they don’t account for a woman’s lower lifetime earnings.
    • Beyond this, the last two decades have seen a global shift from social insurance to individual capital accounts.
      • This structure is based on past contributions, and the number of years during which a person is expected to be accruing benefits.
      • So what this actually means is that women are penalized for having to take time out for unpaid care work, early retirement age requirements, and for living longer.
• To address feminized poverty in old age, governments should introduce policies that enable women to stay in paid work–this begins with paid maternity leave.
  • Paid maternity leave doesn’t just increase the number of women employed, but also the number of hours they work and the income they earn. It’s particularly beneficial for low-income women.
  • US universities are another example of how allegedly gender-blind policies end up discriminating against women.
    • Professors usually have seven years to get tenure, and the years between completing a PhD and receiving tenure coincide with the years women are most likely to try for a baby.
    • Some universities allow parents an extra year per child to earn tenure.
      • But it’s not gender-neutral “parents” who need this extra year; it’s mothers. While women are throwing up, going to the toilet, going to appointments, pumping, etc. during this year, men get to dedicate more time to their research.
      • Realistically, this policy gave a leg up to men at women’s expense: an analysis of assistant professors hired at the top 50 US economics departments between 1985 and 2004 found that the policies led to a 22% decline in women’s chances of gaining tenure at their first job, while men’s chances increased by 19%.

The myth of meritocracy

• For decades, no female musicians played in the New York Philharmonic Orchestra.
  • Then they employed blind auditions in the 1970s, and today female musicians make up over 45% of this orchestra’s musicians.
• For the vast majority of hiring decisions, meritocracy is an insidious myth that provides cover to institutional white male bias.
  • Studies have shown that if you believe in your own personal objectivity, or believe that you are not sexist, this actually makes you less objective and more likely to behave in a sexist way.
• Even though women have more than half of all college degrees in chemistry and math, they make up only 25% of the tech industry and 11% of its executives, and more than 40% of women leave tech companies after 10 years compared to 17% of men.
  • That the myth of meritocracy survives in the face of such statistics is testament to the power of the male default: many men in the tech industry just don’t notice how male-dominated it is.
  • And it’s also a testament to the myth’s attractiveness that tells the people who benefit from it that all their achievements come down to their own personal merit–those most likely to believe in the myth of meritocracy are young, upper-class, white Americans.
• Brilliance bias is the unconscious bias that we have that those who are brilliant are male.
  • This bias presents itself most prominently in computer science.
  • The author recounts that one high school teacher noted that multiple parents told him that their sons would be on the computer programming all night if they could, and he hadn’t run into a girl like that.
  • However, staying up all night to do something is a sign of single-mindedness and immaturity and a love of the subject. Girls may show their love for computers and computer science differently. If you think obsessive behavior is the most prominent marker for brilliance, it will typically only be young male behavior.
• This bias appears even in “objective” processes.
  • Online tech-hiring platform Gild allows employes to comb through “social data,” which is the trace that people leave behind online.
  • Per their data, going to a particular Japanese manga sit is apparently a solid predictor of strong coding, so programmers who visit this site got higher scores.
    • And yet women might not have the spare leisure time to spend hours chatting about manga online, and since most manga sites are dominated by males and have a sexist tone, many women avoid it.
  • So, if you’re not aware of how biases operate, and aren’t collecting data and producing evidence-based processes, you can blindly perpetuate old injustices.

Workplace injuries

• While serious workplace injuries decreased for men, they increased for women.
• This is linked to the gender data gap: occupational research is traditionally focused on male-dominated industries, so our knowledge of how to prevent injuries is sparse.
  • For instance, as a carer for the elderly or a cleaner, a woman can lift more in a shift than a construction worker or miner.
  • But unlike construction workers or miners, many women don’t go home to rest, but go home to an unpaid shift of cooking, scrubbing, and caring for children.
  • And in nail salons, a female-dominated industry, workers are exposed to a huge range of chemicals that are linked to cancer, miscarriages, and lung disease.
• Beyond this, every year, thousands die from work-related cancers, and although most of this research has been done on men, women are also affected.
  • Breast cancer rates in the industrialized world have risen, but the failure to research female bodies and occupations means there is insufficient data.
• Instead, for such occupational concerns, we continue to rely on data from studies done on male bodies, as if they apply to women. Specifically, Caucasian males, 25-30 years old, who weigh around 70 kg.
  • But this is a terrible comparison. Men and women have different immune systems and hormones, which plays a role in how the chemicals are absorbed.
  • Women are usually smaller than men and have thinner skin, which lowers the level of toxins they can safely be exposed to.
  • This lower tolerance threshold is compounded by women’s higher percentage of body fat, in which some chemicals can accumulate.
• And even where women work in male-dominated industries, data on their bodies were either not collected, or they were treated as “confounding factors.”
  • For instance, in the US in 2007 there were nearly 1 million farm operators, but virtually all farm tools and equipment are designed for men.
  • Women in the military are also affected by equipment designed around the male body; the same goes for poorly designed PPE.

Workplace safety

• Workplaces that are male-dominated or have male-dominated leadership are often the worst for sexual harassment.
  • The UN estimates that up to 50% of women in the EU and 80% in China have been sexually harassed at work. 
  • A 2011 US study found that the construction industry had the highest rates of sexual harassment, followed by transportation and utilities.
• But the data gap for the sexual harassment and violence women face in the workplace isn’t just about failing to research.
  • It’s also about a failure to report, which is partly due to employers not having adequate procedures to deal with these issues. This, in turn, may come from failure of leadership–which is often male-dominated–of being familiar with this type of aggression.
  • This is another reason why diversity of experience at the top is so important.

Going to the doctor

• Historically, people assumed that there wasn’t anything fundamentally different between male and female bodies other than size and reproductive function, so for many years medical education focused on the male “norm” and anything that fell outside was considered atypical or abnormal.
  • But sex differences are substantial. Researchers have found sex differences in every tissue and organ system in the human body, and in the course/severity of many diseases.
• This can lead to women being misdiagnosed and poorly treated unless their symptoms conform to that of men–this appears most dangerously in the case of heart attacks, where women are dismissed if their symptoms do not match the symptoms of males having heart attacks.
• The reality that women are not given the same level of medical attention is often brushed aside with the notion that women generally live longer.
  • Although female life expectancy is a few years longer than male life expectancy, there is no evidence that the female mortality advantage is secure. For instance, women in the US do not have more active years in their lives, and overall, women’s longevity and active years have increased at a lower rate.

GDP

• The formulation of a country’s official GDP figure is a subjective process. This subjectivity is compounded by gaps in the data that is used to compile the figures.
• There are plenty of goods and services the GDP doesn’t account for, and the decision re which ones to include is arbitrary.
  • Originally, the main purpose of GDP was to understand how much output could be created, and what consumption needed to be sacrificed, to ensure there was enough for the war effort.
  • But this excluded certain contributions like unpaid household work (cooking, cleaning, and childcare).
  • Economist Paul Stdenski said that unpaid work in the home certainly should be included in GDP, but because these principles are man-made, at a certain point there was a decision not to include these services because collecting the data would be too big a task.
• This failure to incorporate unpaid household services in the GDP is perhaps the greatest gender data gap of all.
  • Unpaid care work could account for up to 50% of the GDP in high-income countries (roughly $3.2 trillion in 2012, approximately 20% of the GDP).
  • The consequence of failing to capture this data is that women’s unpaid work is seen as a costless resource to exploit, so when countries try to limit their spending, it is often women who pay the price.
    • For example, when the UK cut public services after the 2008 financial crash, children’s center budgets lost around 82 million pounds. Realistically, these cuts are not savings–instead, they shift costs from the public sector onto women, because the work still needs to be done.
• This emphasizes the need to invest in social infrastructure.
  • This can take the form of universal childcare or paid parental level, but also things like early childhood education and high-quality formal childcare. Investment in these can actually reduce overall education costs because it lowers the level of investment required in remedial education.
  • But these types of investments are often overlooked, in part because of the gender data gap when it comes to unpaid work.
  • In the US, an investment of 2% of the GDP in the caring industries would create nearly 13 million in new jobs, compared to the 7.5 million jobs that would be created by investing the same amount in the construction sector.

Ambitious women

• Many saw Hilary Clinton’s presidential aspirations as “pathological” because she was forging into a territory that is overwhelmingly associated in people’s minds with men.
  • So, her candidacy was a norm violation, which are aversive and associated with strong negative emotion.
• People generally associate power, influence, and ambition with maleness.
  • The social downer on women who seek professional power is partly because social power (being seen as warm and caring) is women’s consolation prize for renouncing competition with men.
  • So, a woman’s social power is intrinsically incompatible with professional power.
  • Being seen as uncaring is a norm violation for women in a way that it isn’t for men. 
• On a grander scale, this means that democracy is not a level playing field–rather, it is biased against electing women.
  • This is problematic, because male and female legislators bring different perspectives to politics.
  • Beyond that, women in politics face an alarming number of threats. It stems, in part, from gender-data-gap-driven fear: certain men, who have grown up in a culture saturated by male voices and male faces, fear what they see as women taking away power and public space that is rightfully theirs.
    • This fear won’t evaporate until we fill in that gender gap and, as a result, men do not grow up seeing the public sphere as their rightful domain.

Other examples of the gender data gap

• In the 1990s, officials in Vienna found that from the age of 10, girls’ presence in public playgrounds decreased significantly.
  • It turned out that large open spaces forced girls to compete with boys for space.
    • Girls usually didn’t have the confidence to compete with boys (a nod to social conditioning), so they tended to just let the boys have the space.
  • But when they divided the parks into smaller areas, the female drop-off reversed.
• The benefits of a gender-sensitive approach to sports is important, too.
  • A city in Sweden distributed around 80 million kronor to sports clubs.
  • The funding was meant to benefit everyone equally.
    • But the majority of funding was going to organized sports, which are dominated by boys–ultimately, the city spent 15 million kronor more on boys’ than girls’ sports. This meant that girls’ sports were less funded, so girls had to pay for them privately or didn’t do sports at all.
  • When planners don’t account for gender, public spaces become male by default.
• In order to design things that actually help women, first we need the data. For example, the attempt to disperse “clean” stoves in the developing world sheds light on the need to consult women from the beginning of projects.
  • Traditional stoves, which millions of women use to cook for hours on end, is that they give off extremely toxic fumes.
  • An international entity started a “Clean Cookstoves” campaign, and when it didn’t catch on, the companies decided that women needed more education on proper usage.
  • After a while, though, people realized that the stoves increased cooking time and required more attending; this prevented women from multitasking, and increased their workloads. Nevertheless, the main recommendation was to fix the women, instead of requiring the stove designers to create a better design.

The solution to the sex and gender data gap is clear: we have to close the female representation gap. When women are involved in decision-making, in research, in knowledge production, women do not get forgotten.

• The dangers of taking public transport as a woman–either overt sexual harassment and assault, or behaviors that are not necessarily criminal, but constitute sexual menacing (catcalling, beeing leered at, sexualised slurs)–and the need for better data collection that can lead to innovative solutions
• The Plough Hypothesis – societies that had historically used the plough would be less gender equal than those that hadn’t
• How one-size-fits-all really means one-size-fits-men and how supposedly gender-neutral products–keyboards, pianos, voice-recognition software, health-monitoring systems, VR headsets, calorie counting on treadmills, car headset design, and phones–disadvantage women
• The dangers of gendered poverty and the misinformation it breeds
• The gender data gap and post-disaster relief, which is becoming all the more urgent with climate change