Invisible Women: Data Bias in a World Designed for Men

The assertion that 70% of those in poverty are women is a zombie statistic or “a spurious statistic that just won’t die” (255). It is especially hard to disprove such unfounded statistics when data is scarce. The current determination of gendered poverty, which compares poverty in male- versus female-headed households, makes two sketchy assumptions: first, that resources are shared equally among those in the household, and second, that there is no difference in the allocation of resources between male- and female-headed households. Yet research does not bear this out: when programs transferred income from men to women, more money was spent on children; female-headed households, even when poorer than male-headed ones, also spent more money on children; finally, the majority of poor women in one survey belonged to non-poor households. It is time to stop determining poverty at the household level.

Tax systems penalize most women. Joint tax returns typically cause women to pay a higher rate of tax on earnings and studies have shown that “joint filing disincentivizes married women from paid work altogether” (259). For example, the UK gives an incentive to a couple’s lower earner, typically a woman, to reduce hours so as to qualify for a tax credit. Such incentives contribute to the gender pay gap. Meanwhile, tax cuts tend to benefit higher-income men, while consumption taxes hurt women. When tax cuts result in reductions in social programs, women have to pick up more unpaid labor. Such systems of taxation are driving gendered poverty. Additionally, since the effects of tax systems on women are not studied, there is again a gender data gap.

Around the world, men dominate governments. For example, in democracies, women comprise only 23.5% of parliamentarians. However, studies have found that female perspectives would affect governance, with women more likely to focus on family policy, education, and care. “Decades of evidence demonstrate that the presence of women in politics makes a tangible difference to the laws that get passed” (266). In a few countries, educational investment increased as women’s participation in government did. Yet women are disadvantaged in the competition for political office. Those seeking office are dubbed overly ambitious, as Hillary Clinton was.

Power, influence, and ambition are “equated with maleness” (268). When women enter this world, they are challenging norms and will likely be depicted as cold, bossy, and uncaring, because in the public imagination, women either have professional competence or warmth. Women are penalized for this perception, while men do not pay a price for seeming cold or uncaring. Moreover, women and ethnic minorities are less likely to succeed if they engage in “diversity-valuing behavior” (271): Reminding voters of their difference invites negative stereotypes. Even when candidates speak about issues, the media and challengers accuse women of dwelling upon their gender.

The rules of elections in democracies are often biased against the election of female candidates. Although in the UK, female participation dramatically improved after the use of all-women shortlists (AWS) in internal party elections to determine general election nominees, the government has refused to extend AWS legality beyond 2030, almost certainly ensuring the future decline of female officials at the local and national levels. In the UK’s system of winner-take-all or first-past-the-post legislative elections, AWS are the only effective means to correct male bias.

Systems with proportional representation are somewhat better for female representation. In those systems, citizens vote for parties, which then send a proportion of its listed candidates to Parliament depending on what percent of the vote they won. In other words, if there are 100 seats in the Parliament and a party won 20% of the vote, the first 20 people on its list are elected. In Sweden, the candidate lists alternated male and female candidates—a system that substantially improved women’s representation.

When women are elected, they face discrimination on the job. They are excluded from patronage networks and substantive discussions held in private areas. Men interrupt women more often and with impunity—yet when a woman interrupts a man, she is often admonished. Men also make patronizing comments to female politicians. For example, David Cameron, former Prime Minister of the UK, told a Member of Parliament to “calm down, dear” (278) during policy debate. Even worse, female politicians are subjected to sexual harassment and violence to some degree in every country. According to one survey, more than one in five female parliamentarians experienced one or more acts of sexual violence.

As the percentage of female representation increases, so too do threats of violence, abuse, and discrimination. This hostility causes some women to leave politics and prevent others from entering this arena. The resultant decline in female participation increases the gender data gap and makes legislation addressing women’s needs less likely. Without structural changes, such as laws increasing penalties for threats of violence against elected officials, the exclusion of half the world’s population from governance creates a gender data gap at the “very top” (285).

Natural disasters, conflicts, chaos, and breakdown impact women more negatively than men. Hillary Clinton raised the issue of women’s rights at the United Nations Fourth Conference on Women in the midst of genocides in Bosnia and Rwanda—and was criticized for it by critics who seem to have missed the ongoing “mass and systematic rape of women” (289) in Bosnia and Rwanda.

Authorities frequently fail to include women in post-disaster planning. As a result, women’s needs and work are disregarded. In two instances, homes were rebuilt without kitchens! After devastating hurricanes in Miami and New Orleans, male planners favored business interests over people’s needs in reconstruction. Business centers were rebuilt, while nursery schools and health centers were not. In New Orleans, public housing units were demolished despite being structurally sound—destroying the homes of women who looked after neighbors and had access to buses and stores. As a result, thousands of people were displaced and, importantly, a social infrastructure was destroyed.

Although the well-meaning United Nations Resolution 1325 “urges actors [in conflicts] to increase the participation of women and incorporate gender perspectives in all United Nations peace and security efforts” (293), there is a paucity of data to determine if this resolution is being met. Only two women have served as chief negotiators. Rather, when women are included in negotiations, they are usually excluded from positions of power. In fact, the United Nations honors the requests of participants to exclude women despite the fact that the presence of women during negotiations makes it more likely that an agreement will be reached and have staying power. Women who negotiate “bring important issues to the peace-building agenda” (295), such as the accessibility of institutions and the significance of local and informal spheres.

Natural disasters, pandemics, and conflicts disproportionately impact women because in modern warfare, more civilians than combatants are likely to be killed, and domestic violence and rape increase during conflicts. For instance, in the Bosnian conflict, approximately 60,000 women were raped, while as many as 250,000 women were raped during the Rwandan genocide. Additionally, the indirect effects of war harm and sometimes kill women: Without access to birthing care, maternal deaths increase.

The spread of infectious diseases following natural disasters and conflicts usually hits women harder. During the Ebola outbreak, for example, pregnant women were at higher risk given their contact with healthcare workers, who were mainly female and decimated by the virus. In yet another example of the gender data gap, there are very few academic studies on the gendered impact of infectious diseases. Failing to account for gender during the Ebola and Swine flu outbreaks put women at higher risk of infection. Since women do the majority of care work for the sick and cleaning in hospitals, they are at higher risk of exposure.

Given that climate change has dramatically increased the number of natural disasters, women, especially poor ones, are increasingly in danger. The “higher the socio-economic status of women in a country, the lower the sex gap in deaths” (300). It is not the disaster that accounts for this disparity in deaths, but the restrictions on women’s lives. In some societies, women are not allowed to leave their homes without a man, so they must wait for a male relative to escort them away in an emergency. In India, more men than women survive earthquakes because they sleep outside on warm nights, something women are not allowed to do. In Bangladesh, women are not taught to swim and therefore are more likely to be killed by floodwaters than men. Bangladeshi storm shelters are built with only men’s needs in mind, with a lack of sex segregation and a bucket for urination. As a result, women do not go to the shelters. Even in wealthy countries, such as the United States, women faced the danger of rape and beatings when Louisiana’s Superdome served as a shelter during Hurricane Katrina. Similarly, homeless shelters do not take the concerns of women in mind. Although women are more likely to experience homelessness than men, they do not feel safe in shelters and choose to live outdoors instead. Gender-neutral facilities do not meet the needs of women: Gender-neutral does not mean gender-equal.

When women escape war and disaster, the situation can be just as dire in refugee camps. Toilets are not always sex-segregated and often are in remote locations with inadequate lighting. As a result, women have to improvise to ensure their own safety. There are multiple cases in all parts of the world of male guards abusing women in refugee camps and detention centers. Perez speculates that “perhaps no male staff should be in positions of power over vulnerable women” given the “steady stream of abuse reports from around the world” (304). There is a shocking irony in ignoring potential male violence in refugee systems: Female refugees are often already fleeing male violence, whether it be rape, forced marriage, or domestic abuse.

Perez gives an illustration of what we lose when we don’t include women: In two hours, a woman named Daina Taimina determined how to create a model for explaining hyperbolic space, a solution that “had eluded mathematicians for over a century” (310), by crocheting a model of a hyperbolic plane, where space curves away from itself at every point. Her creations are now the standard models—they benefit all of humanity. The inclusion of a woman brought a different perspective to an old problem and solved it.

Planners and developers must keep in mind three things. First, the female body can no longer be invisible, but must be accommodated in medical, technological, and architectural developments. Second, male sexual violence against women must be studied so that the world can be designed to account for it—otherwise, women’s freedom is limited, and their access to essential services, such as shelters, is at risk. Third, women do more than a fair share of unpaid care work, so data must be collected on women and their lives. The excuses for not taking these matters into account are not acceptable. Excluding half the world’s population in designs and algorithms does not serve human needs. In all spheres of life, female representation must be increased. When women are in positions of power, they are more likely to include women and they lead the way to closing the gender data gap.

COVID-19 became an official pandemic one year after the book was first published. While the disease has sex-specific effects, with men dying at a two-to-one ratio to women, there is a “continual failure to systematically collect and publish sex-disaggregated data on symptoms, infection rates, and death rates” (319). As late as September 2020, only 30% of countries were reporting sex-disaggregated data about the disease. The historic failure to study the female body also slowed the search for a cure. The female immune system accounted for the lower death rate, but little is known about that immune system given the lack of research.

Women are endangered because masks are too large, designed with the male face in mind. Additionally, the pandemic has hit women hard economically, especially women of color—many are in high-risk jobs and others have lost their jobs. With children out of school, women typically had to bear more of the burden of unpaid care. The result has been a decline in women’s productivity.