Scan barcode
A review by rbruehlman
Invisible Women: Data Bias in a World Designed for Men by Caroline Criado Pérez
4.0
Invisible Women explores the biases that exist in data (from inaction, negligence, or oversight) that negatively impact women. Often, women's behavior, bodies, and needs are assumed to be the same as men's, when they are not, which means they are excluded from representation in research or data collected on them is not analyzed by sex for meaningful differences. The end result, Caroline Criado Pérez argues, is a world designed for men, not women.
This was a exhaustingly well-researched book and the points it made was important. I don't think about the world in a gendered lens and never have. I've only very rarely had people make comments to me about my gender in a way I found offensive or off-putting. No one has ever said to me, "Women are worse software engineers" or "you were only hired because you're a woman", or something blatant like that. So, where actually is gender discrimination in the US, I wondered? It seemed like something now only expressed by misogynistic hillbillies or something.
I think this book is a good reminder that discrimination and bias does not have to be blatant or intentional. It can be very subtle, in the implicit assumptions people make (do you think of a woman when you think of a software engineer?), or the data biases we make (testing car crashes on male dummies, when women are much shorter and designed differently). Probably none of the people testing car crashes are biased against women, and are great people. They've simply assumed the dummy they're testing with covers all the bases. But the assumption men and women are the same in car crashes is dangerous.
Actually, I think this book's data is the case I've ever seen made for diverse representation in government / work / etc. Usually, increased representation is posited as wanting to see someone like you in a given position (e.g., a woman as president, to show it can be done). This has an impact, to be clear, and I absolutely don't want to dismiss the powerfulness of that effect. But it's more than that. When you are a member of a given group, you do not necessarily think or make decisions that would positively benefit other groups. You don't experience the problem, so you don't even know it exists as a problem to be solved. If you only have, say, rich, white males in charge, they may be great at solving problems that exist for everyone, or for their demographic group ... but parking spaces close to the building for pregnant employees likely just wouldn't occur to them. Diverse representation ensures that someone experiences that problem and can solve it for other people.
I think this is a really, really important point to introduce to the conversation of bias and discrimination. It's a no-fault argument that doesn't feel alienating. If you come in and say, "Men are discriminating against us," that (understandably) puts men on the attack, as you assume the biases are intentional. Most men really want to do the right thing, and are not intentionally trying to commit bias or discrimination. If you change the conversation to, "It's important to include women because we might think of and solve issues that don't impact men as much," it's a no-fault thing that both people can work towards, without feeling under attack. It is not pointing fingers.
I think the most "oh, yikes" moment for me in the book was reading about medical trials. Women do react differently from men to medications, and it is just wild to me that people would assume they are biologically the same. It's absurd women are often excluded from medical trials because they menstruate monthly--like, hello, what do you think will happen when you release the drugs to the masses? Nobody who takes it is menstruating? That needs to be part of the clinical trial, because it's real life. If it doesn't work as well when someone is menstruating, or is even more effective, or whatever, you need to know that, because that is reality for 50% of the population!
The book did have some bad points, though. While it was exhaustingly researched, there were various points where Pérez made leaps of logic that were unsubstantiated. For instance, when discussing the negative feedback female professors get in reviews of their college courses, she claims negative comments in the reviews are "only getting worse." There was no citation to that, which was frustrating. Sometimes the book could take on a bit of an alarmist tone.
The biggest issue with the book, though, is that it just went on for too long. I got the gist halfway through, and I just didn't need the level of detail she continued to supply after that. The book's length would be justified if she had a section on tangible actions that could be taken, but the reader mostly had to intuit that. Without a "so what now?", the book simply overstayed its welcome reiterating on the same points.
This was a exhaustingly well-researched book and the points it made was important. I don't think about the world in a gendered lens and never have. I've only very rarely had people make comments to me about my gender in a way I found offensive or off-putting. No one has ever said to me, "Women are worse software engineers" or "you were only hired because you're a woman", or something blatant like that. So, where actually is gender discrimination in the US, I wondered? It seemed like something now only expressed by misogynistic hillbillies or something.
I think this book is a good reminder that discrimination and bias does not have to be blatant or intentional. It can be very subtle, in the implicit assumptions people make (do you think of a woman when you think of a software engineer?), or the data biases we make (testing car crashes on male dummies, when women are much shorter and designed differently). Probably none of the people testing car crashes are biased against women, and are great people. They've simply assumed the dummy they're testing with covers all the bases. But the assumption men and women are the same in car crashes is dangerous.
Actually, I think this book's data is the case I've ever seen made for diverse representation in government / work / etc. Usually, increased representation is posited as wanting to see someone like you in a given position (e.g., a woman as president, to show it can be done). This has an impact, to be clear, and I absolutely don't want to dismiss the powerfulness of that effect. But it's more than that. When you are a member of a given group, you do not necessarily think or make decisions that would positively benefit other groups. You don't experience the problem, so you don't even know it exists as a problem to be solved. If you only have, say, rich, white males in charge, they may be great at solving problems that exist for everyone, or for their demographic group ... but parking spaces close to the building for pregnant employees likely just wouldn't occur to them. Diverse representation ensures that someone experiences that problem and can solve it for other people.
I think this is a really, really important point to introduce to the conversation of bias and discrimination. It's a no-fault argument that doesn't feel alienating. If you come in and say, "Men are discriminating against us," that (understandably) puts men on the attack, as you assume the biases are intentional. Most men really want to do the right thing, and are not intentionally trying to commit bias or discrimination. If you change the conversation to, "It's important to include women because we might think of and solve issues that don't impact men as much," it's a no-fault thing that both people can work towards, without feeling under attack. It is not pointing fingers.
I think the most "oh, yikes" moment for me in the book was reading about medical trials. Women do react differently from men to medications, and it is just wild to me that people would assume they are biologically the same. It's absurd women are often excluded from medical trials because they menstruate monthly--like, hello, what do you think will happen when you release the drugs to the masses? Nobody who takes it is menstruating? That needs to be part of the clinical trial, because it's real life. If it doesn't work as well when someone is menstruating, or is even more effective, or whatever, you need to know that, because that is reality for 50% of the population!
The book did have some bad points, though. While it was exhaustingly researched, there were various points where Pérez made leaps of logic that were unsubstantiated. For instance, when discussing the negative feedback female professors get in reviews of their college courses, she claims negative comments in the reviews are "only getting worse." There was no citation to that, which was frustrating. Sometimes the book could take on a bit of an alarmist tone.
The biggest issue with the book, though, is that it just went on for too long. I got the gist halfway through, and I just didn't need the level of detail she continued to supply after that. The book's length would be justified if she had a section on tangible actions that could be taken, but the reader mostly had to intuit that. Without a "so what now?", the book simply overstayed its welcome reiterating on the same points.