Counting: How We Use Numbers to Decide What Matters

If you've read Stone's earlier book, the Policy Paradox, this book will feel familiar. It's an updated version of her numbers chapters with more examples and nuance. Stone is excellent at drawing our eyes to the things we take for granted in society and policy. The book is full of examples from multiple sectors that areas useful to drive home hey points but would also be useful in a class. The plentiful examples can start to feel a bit redundant though.

More like 3.5 stars. I enjoyed this book. It discusses the importance of understanding more details behind quantitative data and asking critical questions - what was counted, what was the motivation, what biases might be embedded. A standout quote for me: "Social issues aren't math problems. The danger of converting a social issue into a math problem lies in teaching citizens that there's one right answer and only mathematicians can find it... Sometimes the right answer to a question is to challenge it."<br /><br />I felt like it was an accessible read - perhaps a bit more simplified than I was looking for. Other books on this topic that I felt were more in-depth include: <br /> <br /><a href="https://goodreads.com/book/show/43736666.Bad_Data_Why_We_Measure_the_Wrong_Things_and_Often_Miss_the_Metrics_That_Matter" title="Bad Data Why We Measure the Wrong Things and Often Miss the Metrics That Matter by Peter Schryvers" rel="noopener">Bad Data: Why We Measure the Wrong Things and Often Miss the Metrics That Matter</a><br /><br /><a href="https://goodreads.com/book/show/41104077.Invisible_Women_Data_Bias_in_a_World_Designed_for_Men" title="Invisible Women Data Bias in a World Designed for Men by Caroline Criado Pérez" rel="noopener">Invisible Women: Data Bias in a World Designed for Men</a><br /><br /><a href="https://goodreads.com/book/show/28512671.Everybody_Lies_Big_Data__New_Data__and_What_the_Internet_Can_Tell_Us_About_Who_We_Really_Are" title="Everybody Lies Big Data, New Data, and What the Internet Can Tell Us About Who We Really Are by Seth Stephens-Davidowitz" rel="noopener">Everybody Lies: Big Data, New Data, and What the Internet Can Tell Us About Who We Really Are</a>

Look, if you're new to this idea about how numbers work in social policy, then this is a good book! Here is what I think Stone was aiming for: an accessible text that takes a subject with potentially hard edges and softens them for your lay audience. If so, okay. It's not like this subject hasn't been repeatedly and frequently addressed for the same audience for many years - perhaps the opportunity to update the basic premises is helpful. And, granted, books like Cathy O'Neill's <i>Weapons of Math Destruction</i> are really about "big data" and algorithms, even if it's pitched at the same audience Stone is aiming at, and Safiya Umoja Noble's <i>Algorithms of Oppression</i> is much more specialized, and Jerry Z. Muller's superficial (and philosophically shaky) <i>The Tyranny of Metrics</i> reads like a warmed over version of <i>Weapons of Math Destruction</i>, and that's only a handful of the recent publications... never mind the classics like Darrel Huff's <i>How to Lie with Statistics</i>, which at least Stone explicitly acknowledges and diverges from... wait, what was I saying? That Stone provides any additional wisdom in this packed field?<br /><br />Read it a different way: the field illustrates that the topic is important (I think it's absolutely crucial). That's true enough, and credit to Stone for honing in on it. This was my reason for reading the book: I like her earlier work in public policy research, I think the subject is important enough to read <b>another</b> book on it, hoping that she pitches at the public policy level, and, hell, why not give it a chance.<br /><br />A mess. Especially, one imagines, for the experts in the room. After all, the truism is well known about numbers and statistics across the board: "garbage in, garbage out." <br /><br />Early on, Stone clarifies that she is concerned with numbers and counting in social issues - ie., your Fitbit that counts your steps, or your performance metric that suggests if you have done well in a given year - and not about the numbers that scientists use to measure the physical world. This is not very true: she happily ventures into discussions about drinking water in Flint, about the metre as a unit of measurement, about nutrition as a corollary of income, and so on. But she may imagine that this disclaimer creates a sanitary cordon that allows her to speculate on the fictitious quality of numbers without countenancing the radical implications of her arguments. Let's think about pollution, for example. Stone notes that talking about "parts per million" is not good communications (a blinding insight!), but what would her thesis about the fundamental fiction of measurements and number have to say about carbon dioxide equivalents and the thesis of climate change? Would, like her discussion of water in Flint, she ridicule the technical language of scientific measurement in environmental science? (Stone has a problem with the words "reference level" despite relying on the legal and scientific judgements that depend on the same the measurement provided by said reference level, thus exposing an inability to either understand the issue or demonstrating her preference for snappy, anti-technical platitudes over the engagement with real issues.) Tough to know, but an example of an absence of judgement on the author's part.<br /><br />My sense is that Stone avoids the question of numbers in science because she has no good answers for real policy issues. Instead, she superficially glides along from example to example about the use and abuse of numbers in mostly social policy. Even here, however, she is frustratingly opaque. It is well-known that the numbers used in the motivational devices and tracking applications are essentially made-up: 10,000 steps being totally arbitrary, and the Fitbit being merely a mechanical prompt. Stone notes that the Fitbit merely prompts action, but - as is typical - never engages with this as an issue. Does Fitbit usage correlate with higher fitness levels? Are Fitbits serving their purpose? You'll never know from this book, because Stone doesn't bother to engage with the research - or, if she does, she won't tell you about it, because her point is simply that, if I can paraphrase, "when you measure something, people start to work toward that measurement outcome".<br /><br />Gee, I've never heard that before.<br /><br />And then there is the arch tone of vaguely sanctimonious, vaguely ignorant judgement, as if Stone knew she was "on the right side of history" (and would use that phrase). I don't mean to criticize her occasional focus on issues of equity and justice - though I think that others (particularly O'Neill and Noble's books referenced above, and Caroline Criado Perez's <i>Invisible Women</i>, are examples of books that do this SO MUCH better). But I <i>do</i> mean to criticize her intellectual snobbery and laziness, as demonstrated by ideas like this one:<blockquote>We acquire a sense of power by being able to make a difference in how other people treat us. Okay, some people get a sense of power from winning video games and coaxing smartphones to do their bidding, but these devices treat them to stare at their LED screens like zombies, oblivious to anyone else. Only interacting with humans can teach the social and civic skills that Konnor learned when he questioned his mother and grandmother about counting his age.</blockquote><br />Hold up, what? If someone has to tell someone, especially a well-educated and well-resourced professional that many people train self-assertion and find meaningful interactions through video games and smartphone use - especially in smartphone use! - in 2020, then I think that person needs to go back to their cozy study and think about the fact that she is more concerned about making value judgements without any evidence than she should be about lecturing the rest of us about something that others have done quite effectively. Maybe this is an overreaction on my part - but what arrogance on Stone's part. Yes, interacting with human beings is of paramount importance! But hey: technology doesn't just turn people into zombies! Come on!<br /><br />The problem isn't "garbage in, garbage out". If it were, we wouldn't have this book. Depending on your philosophical bent, the problems are about social structures of power, inequalities in access to resources, justice, and administrative decision-making, and, some would say, about the human brain and/or human nature and the complexities of modern life. Counting and numbers are also part of the problem, but I cannot recommend this book at all.

So I struggled with rating this book. I enjoyed the content but I felt like the book was too long for what I got out of it. My standard for 3 stars is “worth the time it took me to read” and for 2 stars is “would not recommend.” Neither is wholly accurate. Perhaps my struggle to find a numerical rating encapsulates the idea of the book!

The book simply expounds on the author’s thesis that numbers, or statistics, are subjective. They are based on the definition you give to categorizing whatever it is you are counting. I did not finish this book because I didn’t find it interesting enough. <br /><br />Big thanks to Netgalley and the publisher for the advanced reader copy in exchange for an honest review.

A compact review of how statistics can lie because of how we count and how counting something changes it. She shares Jill Lepore's disdain of opinion polling, which I appreciated. Her discussion of the "Fitbit effect" is really relevant during the pandemic--putting out projections of infections/deaths changes how people behave and leads to the recursive rounds of closing and reopening.

If you've read other, similar books, such as <a href="https://goodreads.com/book/show/28186015.Weapons_of_Math_Destruction_How_Big_Data_Increases_Inequality_and_Threatens_Democracy" title="Weapons of Math Destruction How Big Data Increases Inequality and Threatens Democracy by Cathy O'Neil" rel="noopener">Weapons of Math Destruction: How Big Data Increases Inequality and Threatens Democracy</a> or <a href="https://goodreads.com/book/show/51777543.Data_Feminism" title="Data Feminism by Catherine D’Ignazio" rel="noopener">Data Feminism</a> or <a href="https://goodreads.com/book/show/42527493.Race_After_Technology_Abolitionist_Tools_for_the_New_Jim_Code" title="Race After Technology Abolitionist Tools for the New Jim Code by Ruha Benjamin" rel="noopener">Race After Technology: Abolitionist Tools for the New Jim Code</a>, or other books on modern AI, predictive algorithms, and their use in society these days...this book is likely not worth reading. It's pretty good, but doesn't seem to be worth reading if you are up to speed on this topic.

Deborah Stone argues that numbers are simply a tool, that in many cases we have assigned too much power to. ⁠<br />⁠<br />Numbers are beneficial to us, absolutely! I am an engineer - I use numbers ALL THE TIME! However, something becomes a number (or a metric) based on HUMAN decisions. Numbers are tools that we decide how to use. If we assign the power of judgment to a number or an algorithm, do we really understand how those numbers were developed and what they mean?⁠<br />⁠<br />Deborah Stone does an incredible job of breaking down how algorithms are developed, how categories are used, and how numbers can be taken out of context to make "objective" decisions. This book is a quick read whose lessons you will take with you far into the future.⁠

My beef with this book is that when Stone, reasonably enough, talks about all of the reasons we ought to be careful about, skeptical of, arguments made using numbers, it seemed that the easiest conclusion for a reader to draw was to not trust numbers or the people who use them. In a dangerously anti-science culture, I worry about the consequences of making that argument. <br /><br />I don't believe that is what ms. Stone wants the take-away to be. Her training and job suggest quite the opposite. But she offers too little help on how to take precautions against being hoodwinked by numbery arguments to give me any confidence that many readers won't just throw the baby out with the bath.

The author relayed a personal narrative that a male author wrote about his experience with breastfeeding vs. bottle feeding. The personal narrative had nothing to do with counting or statistics, it didn't change the science of the matter and my husband had the opposite experience than the author. Not sure why she threw in random stories. I was hoping for a book that was more mathematical and scientific; a book that actually took a deep look into how we use numbers and what they mean. Disappointed.