It may come as a surprise to you that the history of statistics and data science is fraught with racism, sexism, class, race, colonialism and anti-Semitism. Even if you do not have much of a data science background, you may be aware of some of the issues. For example, Fisher’s exact test is a statistical significance test used in the analysis of contingency tables. However, Fisher was a well-known eugenicist and was steadfast in those beliefs throughout his life, disagreeing with the UNESCO efforts to wrestle with Nazi science in the post-World War 2 period. Pearson’s correlation coefficient is the measure of linear correlation between two sets of data, which was created by Karl Pearson and commonly used today. Pearson, however, was a vocal supporter of eugenics. The ‘father’ of eugenics is Francis Galton, who put forward plans to artificially produce a ‘better’ human race through regulating marriage and procreation. You may have used Galton’s statistical methods without realizing it; he developed the concept of correlation, and regression to the mean.
As a field, statistics has been quietly and metaphorically dismantling statues while divorcing statistics from its uncomfortable roots. Here in London, University College London quietly removed Galton’s and Pearson’s names from its buildings and classrooms. The Committee of Presidents of Statistical Societies renamed its annual Fisher Lecture. Additionally, the Society for the Study of Evolution did the same for its renowned Fisher Prize.
What has this topic to do with intersectionality, diversity, inclusion and equality? Intersectionality describes how discrimination against different facets of a person’s identity can overlap and impact their lives. In other words, characteristics of you – the colour of your skin, your gender, disability, age, sexual orientation and other protected characteristics all interact to affect your lived experience. These characteristics can combine to create to unequal outcomes, the impact of which cannot be attributed solely to one dimension or axis. Putting these things together, we have to be very, very careful when we think about people and their characteristics and acknowledge that these characteristics are inter-related and impact one another. There is the temptation to simplify down to binary brackets – gay or straight, black or white, male or female, rich or poor, old or young, middle class or working class and so on. As it stands now, the world of statistics and data science is dismantling its own statues, acknowledging that statistics were misused to isolate and subjugate people with one dimension to humanity’s cost.
In her 1989 work Demarginalizing the Intersection of Race and Sex, the US lawyer and civil rights advocate Kimberlé Crenshaw wrote: “Because the intersectional experience is greater than the sum of racism and sexism, any analysis that does not take intersectionality into account cannot sufficiently address the particular manner in which Black women are subordinated.” Intersectionality helps us to think more holistically about people as individuals through a multi-dimensional lens of characteristics, drivers and inhibitors.
What are these characteristics? These are subject to debate, but let’s go with the Equality Act (1980) in the UK for now, when we say that the protected characteristics are as follows:
- age
- disability
- gender reassignment
- marriage and civil partnership
- pregnancy and maternity
- race
- religion or belief
- sex
- sexual orientation
What does the data say? The data shows that, in the workplace, employees who belong to two or more underrepresented categories experience oppression. When people do not consider intersectionality, they fail to recognise the consequences leading to inequalities such as reduced access to opportunity, wage inequality, and increased sexual harassment. The World Economic Forum has a lot of data on the topic, and institutions such as the University of Bristol are spearheading initiatives to recognise intersectionality issues and topics as part of their student and employee experience, stating that, for example, the experiences of women of colour are different from those of white women.
Developing a lens for Intersectionality in Technology
In diversity and inclusion efforts, there is a tendency to focus on women at the expense of other groups. The current author did a lot of work in the past decade on ‘women in technology’ efforts, while also explaining that we need to focus on ‘diversity in technology’. I worked hard to talk about diversity, inclusion, inequality and, more recently, intersectionality as my own understanding has increased and improved – although, please note, I have a lot of learning to do still.
Women in Tech is considered a starting point, but this means that people think of it as a linear process along a single dimension and it does not include considerations of intersectionality. This error inadvertently leads to inequalities between groups. It is a privileged perspective that prioritizes one characteristic over another. At a high level, the thinking goes something along the lines that ‘we can sort out the women in tech issue… and then look at increasing racial diversity…. and then ‘other’ protected characteristics…’. and it demonstrates very privileged thinking – as if it is ok, somehow, to make these decisions for other groups. It ignores intersectionality while dialing up inequalities for other people, and we need to develop a lens for recognizing and tackling these lines of thinking before it takes root, impacting people adversely.
It is important to note that these concepts are inter-related, and that should inform how we think about each of these concepts. So, in the name of ‘diversity’, people might focus on women. That means that the other groups are not considered in this ladder, so they don’t get the chance of inclusion or equality. There are a few wrong assumptions with a sole focus on women in technology which demonstrate a saucer-deep understanding of intersetionality, equality, diversity and inclusion, and how these consequently impact people. First of all, who has the right to say that one set of characteristics takes priority over another – who thinks it is acceptable to make that decision and dictate this sequence to other people? If people are looking at ‘women in tech’ first and then ‘people of colour’ in tech, then the natural consequence is that ‘women in tech’ is only referring to white women in tech since ‘white’ is the privileged default. For more reading on default and othering, read Invisible Women by Caroline Criado Perez. (Note: I do not get any incentive for you clicking through and purchasing a book. That’s grubby and any proceeds should go to the author.)
What does intersectionality mean for the workplace and technical events?
Diversity, Equality and Inclusion efforts focus on the experiences of marginalized groups in an effort to fix collective biases and obstacles. It is about making things fairer for everyone. A few years ago, I wrote up a Decency Charter which was aimed at welcoming people into the tech community. The effort was in vain and a waste of my time, since my experience has shown that it is clearly not a welcoming place. So I did try but a lot needs to be done in the workplace and in tech community. One thing to consider is code-switching and its impact on the people.
Code-switching and ‘Bring your True Self To Work’
Focusing on ‘women in tech’ only, where it ends up referring the default white women in tech by default, hides the code-switching that other groups need to do in order to fit in. Code-switching refers to the situation where a person in a minority group reduces some of the most obvious elements that associates them with their community in order to fit into a more mainstream, default group. A member of the LGBTIQA community may avoid discussing sexuality around heterosexual people to avoid making other people uncomfortable; for example, this may mean hiding that they have a partner. For a person of colour, that might mean changing their appearance, accent or dialect when they are with white people. They may, for example, avoid referring to their background or culture so they are not perceived as different. It’s ultimately about protecting yourself so that you fit in.
The Decency Charter was aimed at making people feel welcome and reducing the need for code-switching. Focusing on ‘solving the women in tech problem’ means that we are still not acknowledging the lived experience of people of colour, or LGBTIQA+, for example. People from different backgrounds may be asked to support ‘women in tech’ as initiative, which can also make things worse for them since this may be at the expense of supporting initiatives, such as LGBTIQA+, further exacerbating inequalities.
‘Bring your True Self To Work’ is a privilege for those who have a ‘true self’ who fits in. Technical community and workplaces are so much kinder if you fit in. It’s easier to see direct hits than subtle microaggressions or judgements that are the consequences of blunt privilege.
How can you help?
Here are some ideas, and please feel free to add some more ideas to the mix
It’s not just about the numbers, KPIs and metrics.
Focusing on the number of women, for example, isn’t enough. It can make organisations and communities feel as if they have won the race. If the organization focuses on ‘women in tech’ without recognizing women of colour in technology, or LGBTIQA+ women in technology, then the organization is saying that they care about one of your characteristics that they find more palatable but not the others. It makes people feel like they are doing something and winning, while ignoring the parts that they are not comfortable with. In my own personal experience, I ask curious questions all the time of people around me and I have always found that people respond well to genuine questions where you want to do and be better while developing compassion, empathy, and introspection.
Don’t make people wait their turn at the dinner table
Continue to speak for marginalized groups, of course. As a leader in the community of the workplace, try to develop the lens of intersectionality and don’t accept the easy, saucer-deep ‘diversity lite’ way of thinking.
We talk about diversity as being invited to the dinner table, and inclusion as being invited to speak. If we ignore intersectionality, we are asking people to wait their turn at the dinner table. Let’s not do that.
Think about your impact when making choices
Where can you have the biggest impact? If white women in tech earn more than women of colour in tech roles, why are you focusing on white women? What can you do to help women of colour, and why is that not happening?
Tune in to the Data Platform Summit webinar – I’m chairing!
I am leading a panel on this topic and others. You can grab details from here and it is free to register. It takes place on 15th September at 3.30 UTC. We will talk about intersectionality and related topics. People are not one-dimensional and these can all interact to produce inequality. Until we discuss these issues, we cannot do anything positive to combat them. All these ways to be disadvantaged can interact with each other to produce even greater inequality in terms of pay, learning opportunities, mentoring opportunities, promotion and other things that people care about.
This is an important topic for preparing new leaders in the workplace and it is crucial for people’s careers. Having a single focus on one area is an easy option but it can create inequalities for someone else and that is not fair. So, for example, ‘women in tech’ often neglects the experience of women of colour and it becomes exclusive rather than inclusive, creating inequality for someone else. These are not easy topics and I am learning every day too but we cannot look at people on just one dimension. We cannot just tell people that we are going to prioritise some people because we are more comfortable with one aspect of them than another – we cannot expect people to ‘wait their turn’ to come to the dinner table. Everyone should be invited.
This is a brave topic and I’m keen to try and bring in some outside perspectives on this important topic in order to provide some interesting insights.
Conclusion
At root, we cannot exacerbate inequalities for some groups to improve the lot of one particular group. One group gains at the expense of another group, and it ignores the data and current research on intersectionality as an important component of people. I am still learning on this topic, and I’d love to hear what you think. This top;ic naturally involves listening rather than dictating, and I truly appreciate advice, thoughts or constructive criticism.
I do support women in tech – but I also support LGBTIQA+ in tech, along with other initiatives that care about the protected characteristics. I also learn from these different intersectional groups, and it is my privilege to do so. A better understanding of our fellow wanderers on life’s journey can only lead to a greater degree of empathy understanding. Even where we do not share all of the qualities that makes them them, we can live alongside each respectfully and in peace.
Notes
Note; Normally I’d give references for these facts in order to support what I am saying. However, the current author finds these views so utterly reprehensible that I am not going to link to them. You can go and find them yourself. Some of my ancestors were in Auschwitz, and fortunately survived, so this is personal for me. The material is shocking and I have read through it, and I’ve taken the decision not to share it.
Happy people don’t need to get involved in trolling or online drama and I have no need for it, so I will not respond to trolls or nonsense. Sometimes, it is rooted in an interaction of two confirmation biases. One is the Dunning Kruger effect, where people overestimate their knowledge in an area. It interacts with its corollary, where people fear something they don’t understand and they assume it is out to harm them. These two biases interact to lead people to go on the attack, and they are not in a good place. I can’t have a reasonable discussion with people who are in that place, and my thought is that I hope they get to a happier place for their sake and mine as well, of course. Before you get involved in any online drama or even trolling, I’d ask you to consider your own part in it, and perhaps ask yourself why people who haven’t seen or talked to each other much, if ever, are behaving in this way towards other people and whether you really need to put yourself in that arena. To misquote a Polish proverb: not your monkeys, not your circus, but you are the Ringmaster of your own behaviour and you can choose to leave the tent. Rise above it to have better, kinder conversations rooted in true enquiry and learning with compassion.
I’m clearly findable on social media so I look forward to hearing from you. I’m here to learn but I want to make it clear that I’m open to civil discussions. It’s a shame I have to write this, but that’s where my experiences have led me.
My motivation is to learn, and I look forward to learning from you all.