Gender Data gap : a real world built and designed using data that ignores the needs of half its population.
Why is there always a line for the women’s restrooms but not the men’s? Why is it so difficult for doctors to diagnose a heart attack in a woman in time? Why do women take a greater risk in automobile accidents than men?
Or did you hear the one about the entrepreneur whose idea was deemed too “niche” by investors, but whose femtech firm, Chiaro, is now on course to raise more than $100 million by 2020? Or the female sexual-dysfunction medicine that was only evaluated on two women and 23 males for its alcohol interaction? Aren’t any of them amusing to you? Perhaps it’s because they aren’t jokes.
In a society designed in the image of men, half the population, women, are systematically neglected,” she says in response to various inquiries. This is demonstrated by the startling lack of data on women’s bodies, habits, and needs.
RELEVANT SUSTAINABLE GOALS
The Gender Data Gap
The term “gender data gap” was coined by author and researcher Caroline Criado-Perez to characterise when research is used to design products for both men and women, or women alone only involved male subjects (or virtually entirely). This may result in a product that performs worse for women. There are three issues can be linked to gender data gaps :
1. Data production. In many countries,weak policy space, and legal and financial environments, are barriers to adopt gender lens in the survey design process. Methodological problems and practical challenges relating to economic gender data have received significant funding. Gender statistics are not specified in many nations’ statistics laws and policies, resulting in the sector being under-prioritized and under-funded in national budgets.
2. Data analysis. Even where data are being disaggregated by sex, in-depth analysis of that data is not always undertaken. Too often, gender data use is hampered by limited tabulation and dissemination of existing data.
Areas such as violence against women, sexual and reproductive health and rights, unpaid care, and domestic labor are all crucial to track, yet they are underfunded. Emerging topics, such as gender and poverty, gender pay inequalities, and women’s participation in decision-making, are also in danger and require a lot more analytical effort.
3. Data Dissemination. Lack of access to data and limited capacity of users t use gender statistics to inform policies. Whenever data is available but is not made accessible or shared in user-friendly formats, it is difficult to use to inform evidence-based advocacy and decision-making. When data isn’t used to drive policy and activism, demand falls, lowering the incentive to create gender statistics.
When data is not collected and separated by gender, there is no way to learn what works and what doesn’t for different groups.
AI and The Invisible Population
With the growing context-driven by big data and the usage of Artificial Intelligence (AI) as the guide for our world, the gender-data gap ideally should have garnered more urgent attention than ever before. In this situation, the gender-data gap could result in holes in the algorithms. That, however, is only half of the story. Likewise, the human experience is a source of information if data is another name for information. As a result, omitting to integrate women’s perspectives, which appear to be gender-neutral, is a formula for design bias against men, whether intentional or not.
Missing data leads to missed opportunities. Relying on data from male bodies and lifestyles to define and solve problems results not just in discomfort – it can also be unsafe. Another drawback of gender bias and data gaps in AI is that it does not just reflect them; it amplifies them. One study found that an image-recognition software trained by a deliberately-biased set of photographs ended up making stronger sexist associations. “The dataset had pictures of cooking, which were over 33 per cent more likely to involve women than men. But the algorithms trained on this dataset connected pictures of kitchens with women 68 per cent of the time.
How Do We Bridge The Gender Data Gap ?
With the growing context-driven by big data and the usage of Artificial Intelligence (AI) as the guide for our world, the gender-data gap ideally should have garnered more urgent attention than ever before. In this situation, the gender-data gap could result in holes in the algorithms. That, however, is only half of the story. Likewise, the human experience is a source of information if data is another name for information. As a result, omitting to integrate women’s perspectives, which appear to be gender-neutral, is a formula for design bias against men, whether intentional or not.
If we choose to recognize the difficulties, we can find solutions. For example, a 2016 paper on “word-embeddings” (learning techniques essential for search algorithms) described a new methodology that reduced gender stereotyping (e.g., “He is to doctor as she is to nurse“) by more than two-thirds while maintaining gender-appropriate word associations (e.g., “He is to prostate cancer as she is to ovarian cancer“). Likewise, a study on picture tagging at the University of Washington devised a novel method that reduced bias amplification by 47.5 percent. However, these are the exception rather than the rule.
The problem is that we don’t even realize we’re doing it: 9 times out of 10, even when we say “gender neutral” we are really talking about men. We don’t realize that that is what’s going on. I believe that changing that would go a long way.
Gender data go beyond sex-disaggregation. Gender data, contrary to popular belief, is not only on women or women’s issues, but also intersecting with all sectors from birth to death, from having a proper ID to getting a good education to getting a great job to using public transportation and feeling comfortable at home and in public settings. Gender data also pays special attention to the distinct realities that women, men and non-binary gender face, as well as how those realities interact with other types of personal traits.
Gender data is essential as we revisit remaining obstacles to the promise of productive and self-determined lives for women and girls globally: to measure and report on progress achieved and to design meaningful policies moving forward.
Lead image courtesy of Saulė Gimžūnaitė
The article was initially published on 4th April 2022.
You may also be interested in :
On Poverty and Privilege: It’s Not Your Fault But Here’s What To Do About it.
