A Closer Look at Women and Data
Every year on March 8, Canada and the rest of the world celebrate International Women's Day. The theme for 2023 was Every Woman Counts. This was a reminder that all women of all ages and from all walks of life have a place in every aspect of Canadian society, including in the economic, social and democratic spheres.
However, women are still underrepresented in the data we collect, and where data on women is available, it shows that they continue to be disproportionately disadvantaged in social and economic spaces.
A closer look at the data
It's important to ensure that women and their experiences are represented in experimentation, design and consultations. Far too often, women (and transgender and non-binary individuals) are simply left out of research. This is concerning because data is (or should be) how we assess needs, design products and allocate resources.
Caroline Criado Perez is the author of Invisible Women: Exposing Data Bias in a World Designed for Men. She points out a few shocking statistics across different industries:
- There are cars that are 71% less safe for women than men, because they've been designed using a 50th-percentile male crash-test dummy.
- There is voice-recognition technology that is 70% less likely to accurately understand women than men, because many algorithms are trained on 70% male data sets.
- There is evidence that some personal protective equipment may have put women at risk because it was designed for men's bodies, potentially exposing women to viruses such as COVID-19.
According to Perez, "we are living in a world that has been designed for men because for the most part, we haven't been collecting data on women. This is the gender data gap." It seems like it should be obvious to include women in studies and tests, because it's easy to see how the lack of data and inclusivity can affect products, results, and—in some cases—safety. However, there is a persistent and substantial under-representation of women in data collection, analysis, and decision-making processes. It's a huge issue that can lead to biased and incomplete data sets, which, in turn, can result in unfair decisions and policies.
The gender data gap
A large reason why the data gap for women exists is because for many organizations, sex- and gender-sensitive data is a new phenomenon (and that's if they collect it at all).
In 2019, WHO's Global Health Statistics were disaggregated by sex for the first time. Similarly, Statistics Canada released new sex at birth and gender variables and classifications in 2018, making census data on transgender and non-binary people available for the first time in April 2022.
This is critical because sex and gender can have a huge impact on research findings. For example, studies using disaggregated data have shown that women are 50% to 70% more likely to have an adverse reaction to prescription drugs than men. Additionally, some studies have shown that hormone therapy, such as gender-affirming therapy, increases one's risk of adverse drug effects.
Without disaggregated data initiatives like these, it's difficult to identify sex and gender inequalities and allocate resources accordingly. The data prove that a problem exists and needs to be addressed.
A note on sex and gender:
In Canada, we use sex to refer to biological sex (female, male, or intersex) while gender refers to one's gender identity (woman, man, non-binary, etc.). However, in an international or historical context, data will often conflate the two or use the terms interchangeably. It's important to note that sex-disaggregated data is not a perfect proxy for gender-disaggregated data, and vice versa.
For example, some health disparities can be based on biological sex characteristics, while others can be attributed to socio-economic factors that are more likely correlated with gender. WHO's website says they disaggregated data by sex, and so we used their terminology when talking about their work. Otherwise, we're using “sex- and gender-sensitive data” to indicate data about either or both.
Here is a quick peek at some other inequalities revealed by critical sex- and gender-sensitive research:
- In Canada, women still earn less than their male counterparts, with a gender pay gap of around 9% as of 2022.
- In 2022, women accounted for 79% of domestic abuse victims in Canada.
- In 2015, women spent an average of 3.9 hours per day on unpaid labour (such as cooking, cleaning, and childcare), which is 1.5 hours more than men, who spent an average of 2.4 hours.
- In 2022, women represented 30% of elected federal politicians, and 10% to 49% of provincial politicians. It was also found that men who are elected tend to spend an average of 3 more years than women in their job.
The numbers paint a clear picture. It's evident that although we've been making some progress, there's still a lot of work to be done on the gender equality front. Now imagine if we took this a step further, adding variables such as race, sexual orientation, age, culture or disability. The data would expose even greater inequalities and show which are the most vulnerable groups, revealing exactly where resources are needed the most. For example, for every dollar that a non-racialized woman makes, a racialized woman will only make 68 cents.
What we can do
With a focus on women (#EveryWomanCounts), these statistics bring awareness to the importance of inclusivity and diversity. And it's not just about sex and gender; as public servants, we have a responsibility to undertake comprehensive analyses that consider the intersection of multiple factors, including ethnicity, age, (dis)ability, and geography.
Working for the public service provides us with an extraordinary opportunity to influence services and industries under public purview, especially those related to the health and safety of people in Canada. It's important that we do what we can to mitigate bias intentionally and incorporate a range of perspectives (using approaches such as Gender-based Analysis Plus) to ensure the result works for the diverse population found in Canada.
Resources
