What do you think?
Rate this book
Calling Bullshit: The Art of Skepticism in a Data-Driven World
Bullshit isn't what it used to be. Now, two science professors give us the tools to dismantle misinformation and think clearly in a world of fake news and bad data.
It's increasingly difficult to know what's true. Misinformation, disinformation, and fake news abound. Our media environment has become hyperpartisan. Science is conducted by press release. Startup culture elevates bullshit to high art. We are fairly well equipped to spot the sort of old-school bullshit that is based in fancy rhetoric and weasel words, but most of us don't feel qualified to challenge the avalanche of new-school bullshit presented in the language of math, science, or statistics. In Calling Bullshit, Professors Carl Bergstrom and Jevin West give us a set of powerful tools to cut through the most intimidating data.
You don't need a lot of technical expertise to call out problems with data. Are the numbers or results too good or too dramatic to be true? Is the claim comparing like with like? Is it confirming your personal bias? Drawing on a deep well of expertise in statistics and computational biology, Bergstrom and West exuberantly unpack examples of selection bias and muddled data visualization, distinguish between correlation and causation, and examine the susceptibility of science to modern bullshit.
We have always needed people who call bullshit when necessary, whether within a circle of friends, a community of scholars, or the citizenry of a nation. Now that bullshit has evolved, we need to relearn the art of skepticism.
It's increasingly difficult to know what's true. Misinformation, disinformation, and fake news abound. Our media environment has become hyperpartisan. Science is conducted by press release. Startup culture elevates bullshit to high art. We are fairly well equipped to spot the sort of old-school bullshit that is based in fancy rhetoric and weasel words, but most of us don't feel qualified to challenge the avalanche of new-school bullshit presented in the language of math, science, or statistics. In Calling Bullshit, Professors Carl Bergstrom and Jevin West give us a set of powerful tools to cut through the most intimidating data.
You don't need a lot of technical expertise to call out problems with data. Are the numbers or results too good or too dramatic to be true? Is the claim comparing like with like? Is it confirming your personal bias? Drawing on a deep well of expertise in statistics and computational biology, Bergstrom and West exuberantly unpack examples of selection bias and muddled data visualization, distinguish between correlation and causation, and examine the susceptibility of science to modern bullshit.
We have always needed people who call bullshit when necessary, whether within a circle of friends, a community of scholars, or the citizenry of a nation. Now that bullshit has evolved, we need to relearn the art of skepticism.
336 pages, Hardcover
First published August 4, 2020
Ratings & Reviews
Friends & Following
Create a free account to discover what your friends think of this book!
Community Reviews
Displaying 1 - 30 of 567 reviews
November 19, 2020
I think you need to read this book. It’s not urgent, anytime over the next couple of weeks will do fine. I was thinking while I was reading this of Bad Science (which you should also read, not least since the jokes are much better), but the advantage of this book is that it is written by people who are (how do I put this in a way so as not to hurt their feelings?) relatively dull. Dull, it’s true, but systematic (or do I repeat myself?) And so, they present the seemingly endless ways we can have bullshit served up to us in ways that feel very comprehensive. And then they go about teaching us all the ways we can train ourselves to notice. (You know, “Hang on – this smells like bullshit…”)
I want to start by assuring you this isn’t the book I thought it was going to be. When I started reading it, I was expecting it to suddenly become something a little like: “One of the major problems people face in statistical analysis is in confusing the simple binomial with the chi-square test” (at this point I was expecting the book to then take a quick digression of three or so pages of impenetrable differential equations). Then I was expecting the authors to say “As you can see, and this as should be obvious to even the most boneheaded layperson, you simply can’t perform a t-test on a non-normally distributed key variable integrated Howard’s Disappointment ratio, not unless the effect size is under 0.4 and the products of the remainders have been divided by log e….”
I was so afraid that this book would turn into one of those books that I literally flicked through the pages before buying it with an expression on my face as if, at any moment, one of those pages would set off the mathematical equivalent of a landmine.
This never happened. In fact, the book expects you to know nothing of statistics at all. Better still, it leaves you with almost as much knowledge of statistics as you started with. If you are after a ‘how to statistics while winning friends and influencing people’ book, this isn’t it. Rather, this book explains that the black boxes of statistical analysis can remain mostly obscure to you, and even so, you can still spot bullshit. Spotting bullshit is described as much more of an attitude of mind, than the clever application of an obscure statistical method. It has as much to do with checking sources and thinking through the implications of what you’ve been told, than it does in finding ways to spot the right statistical analysis to run on the data. And this is a lucky thing. Not just because you might not choose the right statistical function, but rather because you probably are never going to be given the actual data anyway, just the conclusion. And as they say repeatedly throughout, there’s more than enough bullshit out there for all of us. And they’re not just talking of the over-the-top, QAnon (hey, I just ordered for a small Margaretta and now Hillary Clinton has turned up with a 7-year-old in lingerie) type bullshit. There’s lots of bullshit that sounds and looks and feels completely reasonable. And the authors here give you tools and loads of examples to spot just that.
Who amongst us is without sin? And I’m not just asking for a friend. We’ve all shared something on the internet that we regret. Especially when we realise with a rush of all-too-rare self-awareness, that the reason we posted it was because it appealed more to our prejudices than to our reason. This is inevitable. And this is also one of the things the authors repeatedly warn us we need to worry about. They quote Neil Postman saying that the person most likely to fool you is yourself. Confirmation bias is our number one, very favourite flavour of bias. So, finding ways to trip ourselves up before we start accepting as true the latest factoid that proves that all those bastards from the other side are selfish, nasty hypocrites is essential. We need to take time to pause. Although, that is easier said than done, obviously. But I’ve said it now, so, all good.
One of my favourite bits of this book – and it is clearly among the authors’ favourite bits too, since they repeat it so often – is the idea that ‘if it seems too good or too bad to be true, it probably is’. This is a strikingly useful test – but one that is insanely difficult to use. This is because it has to overcome the ‘I bloody well knew it’ response. And speaking for myself, a team of wild horses is often not enough to drag me away from a factoid that confirms what I’ve always known to be true. You might think you are holier than me on this – I just have to say that from my own experience on social media, I am going to need some pretty strong proof from you on that.
While they were quoting Postman, I think it would have been nice if they had also quoted one of his explanations for why we are drowning in quite so much bullshit. And that is that a lot of bullshit comes down to us from things that really don’t matter in our lives at all, but that we have been made to believe we are deeply interested in. For example, a recent story has it that Melania Trump has a body double and that it was this double who was out and about campaigning with Donald during the election campaign. Even if this story was 100% verifiable, hand on Bible, true, and even if tomorrow video emerged of an actress named Jane Smithers, or something, pulling on a Melania-type dress and fake boobs – what possible difference could it make to any of our lives? It would just be one more crazy thing that happened in the Trump White House. That is, in a White House that has specialised in ensuring a dozen crazy things have happened every day for four years and all before morning tea on each of those days. Even if it was true, how would you knowing that bit of truth about the fake Melania change your world?
Or take all of the recent excitement about the discovery of water on the moon. Or when so many very old celebrities die. Like when I found out Vera Lynn had died. My first reaction wasn’t “oh god, that’s terrible – we’ll meet again, don’t know where, don’t know when…” it was rather, “But, didn’t she die years ago? She must have, the journalist has just got the names stuffed up. It’s probably Doris Day who has died.” And this is all part of the reason why we are so easily fooled – the truth is that Lynn or Day, it hardly matters at all to our real lives.
If I told you that in the Andes, they’d discovered a new species of humanoid that lived only 30,000 years ago, would you be certain that was bullshit? Or what if I told you that Honeywell are working on trapped ions because they think they are likely to prove to be more effective qubits than superconducting loops? The problem is with how the media has trained us throughout our lives. It makes our swallowing bullshit virtually inevitable. Look, I’d even read The God Particle, but not even I suspected the whole world would get quite so excited when the Higgs Boson was discovered. And this is because we live in a world where people are more interested in facts than narratives, in whats over whys. Even things that are definitely ‘true’ become bullshit when we have no context with which to understand them in. I mean, if you can’t tell me what the Higgs Boson does while it is grazing in the particle zoo, maybe your knowing it ‘exists’ doesn’t really matter.
But I digress. Look, trust me, you want to read this book. It will help you find love, your children will suddenly start to look up to you, your wife will overlook your all too numerous infidelities and you’ll win the lottery.
I want to start by assuring you this isn’t the book I thought it was going to be. When I started reading it, I was expecting it to suddenly become something a little like: “One of the major problems people face in statistical analysis is in confusing the simple binomial with the chi-square test” (at this point I was expecting the book to then take a quick digression of three or so pages of impenetrable differential equations). Then I was expecting the authors to say “As you can see, and this as should be obvious to even the most boneheaded layperson, you simply can’t perform a t-test on a non-normally distributed key variable integrated Howard’s Disappointment ratio, not unless the effect size is under 0.4 and the products of the remainders have been divided by log e….”
I was so afraid that this book would turn into one of those books that I literally flicked through the pages before buying it with an expression on my face as if, at any moment, one of those pages would set off the mathematical equivalent of a landmine.
This never happened. In fact, the book expects you to know nothing of statistics at all. Better still, it leaves you with almost as much knowledge of statistics as you started with. If you are after a ‘how to statistics while winning friends and influencing people’ book, this isn’t it. Rather, this book explains that the black boxes of statistical analysis can remain mostly obscure to you, and even so, you can still spot bullshit. Spotting bullshit is described as much more of an attitude of mind, than the clever application of an obscure statistical method. It has as much to do with checking sources and thinking through the implications of what you’ve been told, than it does in finding ways to spot the right statistical analysis to run on the data. And this is a lucky thing. Not just because you might not choose the right statistical function, but rather because you probably are never going to be given the actual data anyway, just the conclusion. And as they say repeatedly throughout, there’s more than enough bullshit out there for all of us. And they’re not just talking of the over-the-top, QAnon (hey, I just ordered for a small Margaretta and now Hillary Clinton has turned up with a 7-year-old in lingerie) type bullshit. There’s lots of bullshit that sounds and looks and feels completely reasonable. And the authors here give you tools and loads of examples to spot just that.
Who amongst us is without sin? And I’m not just asking for a friend. We’ve all shared something on the internet that we regret. Especially when we realise with a rush of all-too-rare self-awareness, that the reason we posted it was because it appealed more to our prejudices than to our reason. This is inevitable. And this is also one of the things the authors repeatedly warn us we need to worry about. They quote Neil Postman saying that the person most likely to fool you is yourself. Confirmation bias is our number one, very favourite flavour of bias. So, finding ways to trip ourselves up before we start accepting as true the latest factoid that proves that all those bastards from the other side are selfish, nasty hypocrites is essential. We need to take time to pause. Although, that is easier said than done, obviously. But I’ve said it now, so, all good.
One of my favourite bits of this book – and it is clearly among the authors’ favourite bits too, since they repeat it so often – is the idea that ‘if it seems too good or too bad to be true, it probably is’. This is a strikingly useful test – but one that is insanely difficult to use. This is because it has to overcome the ‘I bloody well knew it’ response. And speaking for myself, a team of wild horses is often not enough to drag me away from a factoid that confirms what I’ve always known to be true. You might think you are holier than me on this – I just have to say that from my own experience on social media, I am going to need some pretty strong proof from you on that.
While they were quoting Postman, I think it would have been nice if they had also quoted one of his explanations for why we are drowning in quite so much bullshit. And that is that a lot of bullshit comes down to us from things that really don’t matter in our lives at all, but that we have been made to believe we are deeply interested in. For example, a recent story has it that Melania Trump has a body double and that it was this double who was out and about campaigning with Donald during the election campaign. Even if this story was 100% verifiable, hand on Bible, true, and even if tomorrow video emerged of an actress named Jane Smithers, or something, pulling on a Melania-type dress and fake boobs – what possible difference could it make to any of our lives? It would just be one more crazy thing that happened in the Trump White House. That is, in a White House that has specialised in ensuring a dozen crazy things have happened every day for four years and all before morning tea on each of those days. Even if it was true, how would you knowing that bit of truth about the fake Melania change your world?
Or take all of the recent excitement about the discovery of water on the moon. Or when so many very old celebrities die. Like when I found out Vera Lynn had died. My first reaction wasn’t “oh god, that’s terrible – we’ll meet again, don’t know where, don’t know when…” it was rather, “But, didn’t she die years ago? She must have, the journalist has just got the names stuffed up. It’s probably Doris Day who has died.” And this is all part of the reason why we are so easily fooled – the truth is that Lynn or Day, it hardly matters at all to our real lives.
If I told you that in the Andes, they’d discovered a new species of humanoid that lived only 30,000 years ago, would you be certain that was bullshit? Or what if I told you that Honeywell are working on trapped ions because they think they are likely to prove to be more effective qubits than superconducting loops? The problem is with how the media has trained us throughout our lives. It makes our swallowing bullshit virtually inevitable. Look, I’d even read The God Particle, but not even I suspected the whole world would get quite so excited when the Higgs Boson was discovered. And this is because we live in a world where people are more interested in facts than narratives, in whats over whys. Even things that are definitely ‘true’ become bullshit when we have no context with which to understand them in. I mean, if you can’t tell me what the Higgs Boson does while it is grazing in the particle zoo, maybe your knowing it ‘exists’ doesn’t really matter.
But I digress. Look, trust me, you want to read this book. It will help you find love, your children will suddenly start to look up to you, your wife will overlook your all too numerous infidelities and you’ll win the lottery.
August 12, 2020
Brandolini’s law, which states that “the amount of energy needed to refute bullshit is an order of magnitude bigger than that needed to produce it,” explains why there is so much bullshit in the world. As Uriel Fanelli put it, “an idiot can create more bullshit than you could ever hope to refute.”
So creating bullshit is easy; refuting it is hard. And it is precisely this asymmetry that explains why bullshit persists and how it can even grow over time.
So how can one hope to rid the world of increasing levels of bullshit? Since it’s easier to create bullshit than to refute it, simply refuting each new instance of bullshit seems like a losing battle. The better strategy is educational; if you can inoculate enough people against falling for bullshit in the first place, bullshit never gains enough traction to require costly efforts at refutation.
This, in essence, is the goal of the book. The authors want to immunize you against bullshit, with a focus on the quantitative variety. While it’s relatively easy to identify old-school bullshit based on flowery language and empty rhetoric, new-school bullshit is more insidious and sophisticated with its use of statistics, charts, graphs, and scientific-sounding claims. This is the bullshit that is more persuasive, harder to refute, and ultimately more dangerous.
The authors first note that while arguments based on statistical and scientific reasoning can appear intimidating, there are basic fallacies that one can look out for that do not require any advanced mathematical ability. It is rarely necessary to look into the “black box”—the authors’ term for complex equations, algorithms, or scientific processes—when the problem with bullshit is often the data that feeds into the black box. Recognizing that the data is biased or unrepresentative of the larger population, for example, is an easy method of spotting bullshit that does not require any skills in higher mathematics.
The authors then take the reader on a tour of quantitative fallacies with several examples, all explained clearly and with humor. The reader will learn how to differentiate between correlation and causation, spot biased and unrepresentative data and small sample sizes, identify selection biases in samples, understand how data can be manipulated visually, and more. The reader will also learn how to properly evaluate scientific claims, and how the anti-vaxx movement is based on a single, thoroughly-debunked scientific study that massively confuses correlation with causation, among many other problems.
One of my favorite chapters, chapter 8, has the authors calling bullshit on arguments that claim that artificial intelligence will take over the world. This has always been bullshit and likely always will be, as the authors demonstrate the limits of how machines are designed to “think.”
The book ends with a couple summary chapters on how to spot and refute bullshit, and also on the difference between calling legitimate bullshit and becoming what the authors refer to as a “well-actually guy.” Perhaps the most important point of the book is the idea that the goal of calling bullshit is not to demonstrate your intelligence and puff up your ego; it’s to counter the spread of misinformation in the world and its direct and indirect consequences.
Overall, I suppose that if the reader has a lot of experience with informal logic and spotting fallacies—particularly of a quantitative nature—then this book might not offer anything particularly new. Although even then the book is filled with interesting, updated examples and a ton of polemical humor which makes the book a fun read. If, on the other hand, the reader has limited experience with these concepts, this book is a must read as it shows how quantitative bullshit can be spotted and refuted with even the most limited of mathematical ability.
So creating bullshit is easy; refuting it is hard. And it is precisely this asymmetry that explains why bullshit persists and how it can even grow over time.
So how can one hope to rid the world of increasing levels of bullshit? Since it’s easier to create bullshit than to refute it, simply refuting each new instance of bullshit seems like a losing battle. The better strategy is educational; if you can inoculate enough people against falling for bullshit in the first place, bullshit never gains enough traction to require costly efforts at refutation.
This, in essence, is the goal of the book. The authors want to immunize you against bullshit, with a focus on the quantitative variety. While it’s relatively easy to identify old-school bullshit based on flowery language and empty rhetoric, new-school bullshit is more insidious and sophisticated with its use of statistics, charts, graphs, and scientific-sounding claims. This is the bullshit that is more persuasive, harder to refute, and ultimately more dangerous.
The authors first note that while arguments based on statistical and scientific reasoning can appear intimidating, there are basic fallacies that one can look out for that do not require any advanced mathematical ability. It is rarely necessary to look into the “black box”—the authors’ term for complex equations, algorithms, or scientific processes—when the problem with bullshit is often the data that feeds into the black box. Recognizing that the data is biased or unrepresentative of the larger population, for example, is an easy method of spotting bullshit that does not require any skills in higher mathematics.
The authors then take the reader on a tour of quantitative fallacies with several examples, all explained clearly and with humor. The reader will learn how to differentiate between correlation and causation, spot biased and unrepresentative data and small sample sizes, identify selection biases in samples, understand how data can be manipulated visually, and more. The reader will also learn how to properly evaluate scientific claims, and how the anti-vaxx movement is based on a single, thoroughly-debunked scientific study that massively confuses correlation with causation, among many other problems.
One of my favorite chapters, chapter 8, has the authors calling bullshit on arguments that claim that artificial intelligence will take over the world. This has always been bullshit and likely always will be, as the authors demonstrate the limits of how machines are designed to “think.”
The book ends with a couple summary chapters on how to spot and refute bullshit, and also on the difference between calling legitimate bullshit and becoming what the authors refer to as a “well-actually guy.” Perhaps the most important point of the book is the idea that the goal of calling bullshit is not to demonstrate your intelligence and puff up your ego; it’s to counter the spread of misinformation in the world and its direct and indirect consequences.
Overall, I suppose that if the reader has a lot of experience with informal logic and spotting fallacies—particularly of a quantitative nature—then this book might not offer anything particularly new. Although even then the book is filled with interesting, updated examples and a ton of polemical humor which makes the book a fun read. If, on the other hand, the reader has limited experience with these concepts, this book is a must read as it shows how quantitative bullshit can be spotted and refuted with even the most limited of mathematical ability.
Read
July 6, 2021Excellent read on bullshit, and specifically the modern kind that hides behind statistics or plausible-sounding claims. Full of useful examples, ways of tackling it (including 'get off the internet'), and memorable quotes. Very definitely a book that should be issued to everyone, although what it doesn't tackle is the people who are absolutely determined to believe bullshit because it suits them better than the truth. I don't know what you do about them.
Read this as a refresher against the tsunami of crap blasting into our faces on a daily basis, and give yourself a chance not to be taken in.
Read this as a refresher against the tsunami of crap blasting into our faces on a daily basis, and give yourself a chance not to be taken in.
December 28, 2020
This is a very readable and interesting, but not particularly new or revelatory book about fake news, bullshit stats, and things like p-hacking and click-baiting articles. I've read books that are better at deepdives into this stuff, but this was a good intro to how to call BS. I found particularly fascinating the sections about deceptive graphs and stats because that stuff can be tricky. However, it does feel like bullshit is the least of our problems these days. given how people just believe outright conspiracy theories now, seems like worrying about bullshit stats is a bit quaint. There's another great book out of princeton press called "the new conspiracy theories" or something like that that delves into how these new theories are just radically different than plain lies, deceptions, and bullshittery of the past. We are now in straight-faced false narratives and theories that are a wink wink nod sort of stuff that is immune to "calling bullshit."
August 6, 2020
I had high hopes for this book. It is OK. It makes some strong points about quantification and the visualization of data sets. But in so many ways, it performs the problems it critiques, but in the inverse.
Two scientists write this book. Their commentary on science is welcome. But they are attempting to understand social media, historical transformations, the changes to education, and - indeed - affirm the value of media literacy training.
Intriguingly, the entire literature on information literacy - with a nearly 100 year heritage - is missing from this book.
What we see is - sadly - the familar narrative: "trust me I'm a scientist." The supposed "black box" of science is beyond mere citizens.
That may be true. It may not be true. What is clearly true, is there are disciplines far beyond the empirical sciences that offer powerful commentary about social media, the history of information, information literacy, publishing studies and interface studies.
Instead, we must be satisfied with the quote, "Science is humanity’s greatest invention.”
Bless.
The researchers make the important point that falsehoods are often simpler than the truth. That is why they are believed. But it remains important that scientists recognize there are specialist fields - and remarkable researchers - beyond their experience, expertise and vista.
And science is not humanity's greatest invention. Knowledge is humanity's greatest invention. And it is promiscuous, mobile and agile. It will not be locked in the chains of disciplines or labels.
Two scientists write this book. Their commentary on science is welcome. But they are attempting to understand social media, historical transformations, the changes to education, and - indeed - affirm the value of media literacy training.
Intriguingly, the entire literature on information literacy - with a nearly 100 year heritage - is missing from this book.
What we see is - sadly - the familar narrative: "trust me I'm a scientist." The supposed "black box" of science is beyond mere citizens.
That may be true. It may not be true. What is clearly true, is there are disciplines far beyond the empirical sciences that offer powerful commentary about social media, the history of information, information literacy, publishing studies and interface studies.
Instead, we must be satisfied with the quote, "Science is humanity’s greatest invention.”
Bless.
The researchers make the important point that falsehoods are often simpler than the truth. That is why they are believed. But it remains important that scientists recognize there are specialist fields - and remarkable researchers - beyond their experience, expertise and vista.
And science is not humanity's greatest invention. Knowledge is humanity's greatest invention. And it is promiscuous, mobile and agile. It will not be locked in the chains of disciplines or labels.
December 13, 2022
Falsehoods fly and truth comes limping after it.
This is a very important book to read right now. I highly recommend reading it as soon as possible. What Bergstrom and his colleague accomplishes in "Calling Bullshit" is a blueprint of all the various ways in which lies, exaggerations, contextualizations and data misrepresentation flood the media sphere and have completely corrupted truth.
First principle: science is messy. As a medical doctor myself, I know that it is INCREDIBLY difficult to prove that something is true (a drug, a therapy, mortality rates ect.) It can take decades of many randomized control trials to even create a landmark study that can change medical practice. With this in mind, how could one obscure research paper, whose hypothesis has been taken whole cloth and tacked into the twitterverse possible have any bearing on scientific thought or be a representation of reality? From my own experience almost all headlines that have to do with medicine, including the coronavirus, are complete and utter bullshit. As an ICU doc myself, I could go on and on and on about the misinformation of COVID-19. But I digress.
Calling Bullshit will outline to you the myriad of ways in which you are bullshitted. You'll learn about the Wakefield debacle connecting autism to MMR vaccine in 1998 which was based on a tiny cohort and has been completely disproved. But its influence is everlasting. The Brandolini principle is true: it's hard to disprove all the bullshit that is produced. It is much easier to blast the waves with bullshit than it is to unpack the nonsense and disprove it. Bullshit is easy to produce and easy to spread.
Technology has made the bullshit problem much worse. Forget the all-seeing eye of AI and tech, if you start out with garbage training programs for the algorithm, you will get garbage out. Is it any wonder that a paper claiming to recognize criminality from a picture would produce nothing but utter bullshit if the input data was headshots of non-criminals and MUGSHOTS of convicted criminals?
The low cost of production and publication has contributed to bullshit. There is more volume and less filtering. Advertising revenue is designed to make you click, regardless of content. "This this will 'make you cry' or 'make you angry'" ect are THE most clickable headlines. They convey emotion, not fact. News has changed from what is happening to how we are FEELING. There are those that will conflate probabilistic causality with sufficient or necessary causality stating that smoking doesn't kill you because there are smokers who don't die of lung cancer (thx Mike Pence). Bullshit also constantly conflates correlation with causality--YOU SEE THIS EVERYWHERE. But correlation doesn't sell newspapers, causation does, so this is why you see headlines proscribing that "wine is good for the heart". It's all bullshit.
Does a pitcher of beer cause more beer drinking? Or is that people who are serious about drinking beer order a pitcher? Does delayed gratification cause success (famous marshmallow trial) or does having rich parents correlate with a child being able to withstand not eating a marshmallow when it's in front of their face? Claiming the average over the median is also an enormous source of bullshit, like how claiming a tax plan will reduce "average" American taxes (when it increases median taxes and benefits the wealthy). There is meaningless numerosity: "this beverage is 99% free of caffeine" HMMM, SO IS COFFEE. Commercial branding that has an air of "mathiness" somehow commands authority over their consumers. Data visualization is a huge source of bullshit. Creating "ducks" where the data doesn't fit the graph or "glass slippers" where data has been cut to fit a specified visual representation.
Disinformation relies on trusted people in your social circle spreading bullshit. The bullshit propagates because people have emotion over a headline and repost without doing any vetting whatsoever. Computer generated faces are created now as profile pics for fake accounts and they can be very convincing. Bots are in fake real people with fake identities with a very real agenda who get retweeted by the likes of The New York Times.
Scientific publications are very guilty of bullshit. Scientists and researchers have the same motivations and desire for career advancement than anyone else. They employ the "prosecutors fallacy" in which a positive result in the test may paradoxically be more likely to be an erroneous result than an actual occurrence, even if the test is very accurate. This comes into play when the size of the matching is ignored. P hacking is a HUGE source of scientific bullshit where a researcher will curate their data to get that glorious p<0.05 to claim statistical significance. Significant results are over-represented in the data with negative results being published far less because they are less sexy. This can greatly skew and entire field. The media gets a hold of this bullshit and only publish positive results which are later debunked leading to mass disillusionment in the scientific process which is indeed still very sound. When a metric starts to become a target, it ceases to be a good metric.
The good news is that you don't need to get into the black box of the stastistical analysis of research to call bullshit. Remember the great tactic: if it's too bad or good to be true, it is. Our authors here give us some very good advice: ask yourself a few questions: what's the source? how do they know this? what are they selling? Avoid confirmation bias. Look at what the data is going in and what they are spitting out. When you call bullshit, it is a performative utterance. Be accurate, be correct, be humble and be kind. Also, pick your battles, you cannot call out ALL the bullshit.
Most of all remember this: the chief source of bullshit with which you have to contend is yourself.
This is a very important book to read right now. I highly recommend reading it as soon as possible. What Bergstrom and his colleague accomplishes in "Calling Bullshit" is a blueprint of all the various ways in which lies, exaggerations, contextualizations and data misrepresentation flood the media sphere and have completely corrupted truth.
First principle: science is messy. As a medical doctor myself, I know that it is INCREDIBLY difficult to prove that something is true (a drug, a therapy, mortality rates ect.) It can take decades of many randomized control trials to even create a landmark study that can change medical practice. With this in mind, how could one obscure research paper, whose hypothesis has been taken whole cloth and tacked into the twitterverse possible have any bearing on scientific thought or be a representation of reality? From my own experience almost all headlines that have to do with medicine, including the coronavirus, are complete and utter bullshit. As an ICU doc myself, I could go on and on and on about the misinformation of COVID-19. But I digress.
Calling Bullshit will outline to you the myriad of ways in which you are bullshitted. You'll learn about the Wakefield debacle connecting autism to MMR vaccine in 1998 which was based on a tiny cohort and has been completely disproved. But its influence is everlasting. The Brandolini principle is true: it's hard to disprove all the bullshit that is produced. It is much easier to blast the waves with bullshit than it is to unpack the nonsense and disprove it. Bullshit is easy to produce and easy to spread.
Technology has made the bullshit problem much worse. Forget the all-seeing eye of AI and tech, if you start out with garbage training programs for the algorithm, you will get garbage out. Is it any wonder that a paper claiming to recognize criminality from a picture would produce nothing but utter bullshit if the input data was headshots of non-criminals and MUGSHOTS of convicted criminals?
The low cost of production and publication has contributed to bullshit. There is more volume and less filtering. Advertising revenue is designed to make you click, regardless of content. "This this will 'make you cry' or 'make you angry'" ect are THE most clickable headlines. They convey emotion, not fact. News has changed from what is happening to how we are FEELING. There are those that will conflate probabilistic causality with sufficient or necessary causality stating that smoking doesn't kill you because there are smokers who don't die of lung cancer (thx Mike Pence). Bullshit also constantly conflates correlation with causality--YOU SEE THIS EVERYWHERE. But correlation doesn't sell newspapers, causation does, so this is why you see headlines proscribing that "wine is good for the heart". It's all bullshit.
Does a pitcher of beer cause more beer drinking? Or is that people who are serious about drinking beer order a pitcher? Does delayed gratification cause success (famous marshmallow trial) or does having rich parents correlate with a child being able to withstand not eating a marshmallow when it's in front of their face? Claiming the average over the median is also an enormous source of bullshit, like how claiming a tax plan will reduce "average" American taxes (when it increases median taxes and benefits the wealthy). There is meaningless numerosity: "this beverage is 99% free of caffeine" HMMM, SO IS COFFEE. Commercial branding that has an air of "mathiness" somehow commands authority over their consumers. Data visualization is a huge source of bullshit. Creating "ducks" where the data doesn't fit the graph or "glass slippers" where data has been cut to fit a specified visual representation.
Disinformation relies on trusted people in your social circle spreading bullshit. The bullshit propagates because people have emotion over a headline and repost without doing any vetting whatsoever. Computer generated faces are created now as profile pics for fake accounts and they can be very convincing. Bots are in fake real people with fake identities with a very real agenda who get retweeted by the likes of The New York Times.
Scientific publications are very guilty of bullshit. Scientists and researchers have the same motivations and desire for career advancement than anyone else. They employ the "prosecutors fallacy" in which a positive result in the test may paradoxically be more likely to be an erroneous result than an actual occurrence, even if the test is very accurate. This comes into play when the size of the matching is ignored. P hacking is a HUGE source of scientific bullshit where a researcher will curate their data to get that glorious p<0.05 to claim statistical significance. Significant results are over-represented in the data with negative results being published far less because they are less sexy. This can greatly skew and entire field. The media gets a hold of this bullshit and only publish positive results which are later debunked leading to mass disillusionment in the scientific process which is indeed still very sound. When a metric starts to become a target, it ceases to be a good metric.
The good news is that you don't need to get into the black box of the stastistical analysis of research to call bullshit. Remember the great tactic: if it's too bad or good to be true, it is. Our authors here give us some very good advice: ask yourself a few questions: what's the source? how do they know this? what are they selling? Avoid confirmation bias. Look at what the data is going in and what they are spitting out. When you call bullshit, it is a performative utterance. Be accurate, be correct, be humble and be kind. Also, pick your battles, you cannot call out ALL the bullshit.
Most of all remember this: the chief source of bullshit with which you have to contend is yourself.
December 31, 2020
Um bom livro sobre pensamento crítico e ferramentas para detecção de bullshit. De falas distorcidas e reportagens para chamar a atenção até comunicação científica e política. De números e fatos distorcidos a análises exageradamente complicadas para esconder uma conclusão que não se condiz com a realidade.
Tem toda uma parte sobre relações de causa e efeito e testes estatísticos que outros livros como o How Not to Be Wrong: The Power of Mathematical Thinking ou o The Signal and the Noise: Why So Many Predictions Fail—But Some Don't cobrem, mas que mesmo assim é bem interessante e complementa bem o que é discutido. O que mais gostei foi a descrição de como linguagem vaga é usada para mentir sem mentir diretamente, ou seja, contar bullshit. O capítulo sobre gráficos ruins e desinformativos também é ótimo.
Um livro que seria ótimo para introdução ao pensamento crítico em qualquer tipo de curso de graduação, mas especialmente proveitoso em formações científicas.
Tem toda uma parte sobre relações de causa e efeito e testes estatísticos que outros livros como o How Not to Be Wrong: The Power of Mathematical Thinking ou o The Signal and the Noise: Why So Many Predictions Fail—But Some Don't cobrem, mas que mesmo assim é bem interessante e complementa bem o que é discutido. O que mais gostei foi a descrição de como linguagem vaga é usada para mentir sem mentir diretamente, ou seja, contar bullshit. O capítulo sobre gráficos ruins e desinformativos também é ótimo.
Um livro que seria ótimo para introdução ao pensamento crítico em qualquer tipo de curso de graduação, mas especialmente proveitoso em formações científicas.
February 26, 2021
A very useful little book that provides techniques for detecting and calling out both bullshit and lies, with a particular focus on quantitative science.
The authors (who teach a course based on this material) observe that one significant issue with science is the specialized language and insider techniques that make it impenetrable to the outsider, something that doesn’t apply so much in other fields such as advertising or politics. And precisely because of that barrier, “science-y” language has been co-opted by other disciplines intent on bullshitting.
But the authors emphasize that it often isn’t necessary to have detailed knowledge of any discipline in order to see through lies and disinformation. They point out that essentially all research involves some kind of “black box” containing procedures (such as advanced statistical methods or analytical techniques) that are accessible only to experts in the field and whose details are mostly unknowable to the non-specialist, but in the end can be reduced to:
Data -> Black Box -> Output
And indeed, for systems like AI or Machine Learning, they note that even the practitioners don’t know what’s inside the box. Thus, it’s often sufficient just to identify inconsistencies in the input or output, which would invalidate anything that happens in the black box.
So “Calling Bullshit” involves critically examining the input assumptions, and the authors provide many examples and strategies for doing that. Yes, you do need a certain level of numeracy here – knowledge of averages, order of magnitude estimates, cause and effect and the like - but it isn’t necessary, for example, to have a detailed understanding of statistics to discover that Selection Bias exists in a set of assumptions. That chapter on selection bias was one of the most entertaining – it explains why your friends really do have more friends than you do (on average, of course!)
The authors do distinguish outright lying – where the liar goes to some length to make their lie believable - from bullshitting, where the shitter doesn’t even care whether you believe them or not, but that isn’t the main point of the book. A particularly good example is Wakefield’s dangerous and fallacious vaccine-autism link.
The final chapter is about how to call (refute) bullshit and to get bullshitters to stop; I’m less convinced here, though. Be correct, be charitable, admit your own faults, be clear, be pertinent – yes, these are all necessary, rational responses, but may not be sufficient. You do have to choose your battles.
The authors (who teach a course based on this material) observe that one significant issue with science is the specialized language and insider techniques that make it impenetrable to the outsider, something that doesn’t apply so much in other fields such as advertising or politics. And precisely because of that barrier, “science-y” language has been co-opted by other disciplines intent on bullshitting.
But the authors emphasize that it often isn’t necessary to have detailed knowledge of any discipline in order to see through lies and disinformation. They point out that essentially all research involves some kind of “black box” containing procedures (such as advanced statistical methods or analytical techniques) that are accessible only to experts in the field and whose details are mostly unknowable to the non-specialist, but in the end can be reduced to:
Data -> Black Box -> Output
And indeed, for systems like AI or Machine Learning, they note that even the practitioners don’t know what’s inside the box. Thus, it’s often sufficient just to identify inconsistencies in the input or output, which would invalidate anything that happens in the black box.
So “Calling Bullshit” involves critically examining the input assumptions, and the authors provide many examples and strategies for doing that. Yes, you do need a certain level of numeracy here – knowledge of averages, order of magnitude estimates, cause and effect and the like - but it isn’t necessary, for example, to have a detailed understanding of statistics to discover that Selection Bias exists in a set of assumptions. That chapter on selection bias was one of the most entertaining – it explains why your friends really do have more friends than you do (on average, of course!)
The authors do distinguish outright lying – where the liar goes to some length to make their lie believable - from bullshitting, where the shitter doesn’t even care whether you believe them or not, but that isn’t the main point of the book. A particularly good example is Wakefield’s dangerous and fallacious vaccine-autism link.
The final chapter is about how to call (refute) bullshit and to get bullshitters to stop; I’m less convinced here, though. Be correct, be charitable, admit your own faults, be clear, be pertinent – yes, these are all necessary, rational responses, but may not be sufficient. You do have to choose your battles.
November 17, 2023
Was this book bull-shitting me??
I found it too dense - too many examples cramming too much information in.
I also found it had a negative lilt to it, making it even heavier. A lot of the stats methods I've learned before, and it was fine to read about them again, but I didn't feel like they were effectively training the reader to be able to call bullshit.
It would have been more effective to give the reader an actual headline, ask them to pause and think about the ways it could be bullshit and then the author gives a dissection on why its bullshit.
I found it too dense - too many examples cramming too much information in.
I also found it had a negative lilt to it, making it even heavier. A lot of the stats methods I've learned before, and it was fine to read about them again, but I didn't feel like they were effectively training the reader to be able to call bullshit.
It would have been more effective to give the reader an actual headline, ask them to pause and think about the ways it could be bullshit and then the author gives a dissection on why its bullshit.
May 15, 2022
Bullshit involves language, statistical figures, data graphics, and other forms of presentation intended to persuade or impress an audience by distracting, overwhelming, or intimidating them with a blatant disregard for truth, logical coherence, or what information is actually being conveyed.
The subtitle, "The Art of Scepticism in a Data-Driven World", would be a more accurate description of this book, although the authors do seem rather proud of the main title, with the word 'bullshit' featureing 12 times on the 1st page of the preface, and 7 on the 1st page of the book proper with a 'badass' thrown in for good measure. But once they've got over that, there is a lot of valuable information in here on interpreting, and the dangers of mis-interpreting, data, much of it familar ground from other similar books, but it is certainly useful to see it pulled together systematically.
Many of the examples they give aren't really 'bullshit' as per their definition, lacking the mendacious element, but if anything this makes the book more useful, since well-intended misinterpretation is, if anything, as if, if not more, dangerous, as the Covid epidemic, with science being done and interpreted in real-time and with life-and-death political decisions being made 'following the science', demonstrated. Indeed it would be fascinating to read a follow-up book or article by the authors on Covid science, information and mis-information.
One particularly valuable section focus on the issues with the p-value used in many scientific papers, for example how, while this can (with certain biases) answer a question about the probability that the null hypothesis is true, a failure of the null hypothesis is then often used to confirm one particular alternative hypothesis, which is a different question altogether.
The various websites which keep a record of Things the Daily Mail says cause or prevent cancer are perhaps the most obvious example of incorrect conclusions from pseudo-science but also scientific studies.
Here I was surprised not to see more explicit mention of Bayesian logic. And when they referred to the Prosecutor's Fallacy, while the lack of reference to the Saly Clark case reflects the US-centric nature of the book, I was surprised, particularly given that the OJ Simpson case featured elsewhere in the book, to see no reference to the 'Defense attorney's fallacy'.
It was good to see mention of mathiness, which Paul Romer has used in the context of much economic analysis (see e.g. Piketty's flawed analyses) but which here the authors use in the context of many popular business books, which often invent nonsensical formulas, such as the well-known Trust Equation (according to which, if self-orientation is zero, trust is infinite, and which adds quantities which can't easily be measured on a common scale).
As the authors themselves acknowledge, much of the book focuses on 'spotting', rather than 'calling', bullshit, as the former is needed to do the latter, indeed one of their key pieces of advice on 'calling' bullshit is to be sure of your own facts, and, for example, present plausible alternative explanations.
Their top tips on spotting bullshit are:
1. Question the source of information (who is telling me this? How do they know it? What are they trying to “sell” me?)
2. Beware of unfair comparisons
- I'd perhaps replace this with misleading/irrelevant. Interestingly the example they quote is a 2018 study that airport security trays contain more germs than toilet seats. The author's point out that the study is unfair as it only looked at germs responsible for respiratory infections, not generally. In 2020-21, after this book was written, the risk of airport security trays spreading a respiratory virus suddenly became far more relevant.
3. If it seems too good to too bad to be true…
- ... it certainly requires more skepticism.
4. Think in orders of magnitude
- in particular the authors extol the critical skill of Here the concept of Fermi estimation.
5. Avoid confirmation bias
6. Consider multiple hypothesis
- this last being perhaps the book's key insight.
They have a separate, but similar, list for online facts, but here their key advice is think (and check) more, and share less.
The section on Calling Bullshit is only at the end of the book. They define calling as “a performance utterative in which one repudiates something objectionable” (legally the term 'hereby' defining a performance utterative), but they present a well-balanced case, indeed argue against social media spats, and also highlight the need to question why we're calling something out, i.e. is it to help others (often the 'bullshitter' themselves) appreciate and learn from the mistake made, or just to make us look good: A well-actually guy has more in common with a bullshitter than with a caller of bullshit.
Overall 3.5 stars, rounded to 3.
It's definitely a very interesting summary of the topic although, as mentioned, I didn't find much new here, there were definitely some new insights and a helpful overall framework. On the negative side, the US-centricity grated at times (the assumption the reader is American). And like many such books, a summary (here e.g. reading just the last 2 pages) or a podcast/TED Talk would likely give as much value as the 300-page book.
And this last point is why I read tens of thousands of pages of literary fiction each year, but prefer my business/popular science books in bite-sized does, since in the former it is the quality of the prose that matters not the distilled ideas.
It is perhaps no surprise that the best sentence in this book reads "Many years later, as he faced the firing squad, Colonel Aureliano Buendía was to remember that distant afternoon when his father took him to discover ice." This is of course one of the best known and finest lines in English literature, although it is written not by Gabriel Garcia Marquez, to whom the authors of this book attribute the quote, but rather by Gregory Rabassa. Gabriel Garcia Marquez wrote instead "Muchos años después, frente al pelotón de fusilamiento, el coronel Aureliano Buendía había de recordar aquella tarde remota en que su padre lo llevó a conocer el hielo."
Now me making that point in the context of this book is me being a well-actually guy. But me making this point on my Goodreads reviews is actually me calling out bullshit, since one of my main literary missions is to highlight the importance of translators and their key role, which is often relegated, particularly for famous authors, to a minor footnote on the inside pages. So this just shows that in interpreting, and debunking, 'facts', context matters.
The subtitle, "The Art of Scepticism in a Data-Driven World", would be a more accurate description of this book, although the authors do seem rather proud of the main title, with the word 'bullshit' featureing 12 times on the 1st page of the preface, and 7 on the 1st page of the book proper with a 'badass' thrown in for good measure. But once they've got over that, there is a lot of valuable information in here on interpreting, and the dangers of mis-interpreting, data, much of it familar ground from other similar books, but it is certainly useful to see it pulled together systematically.
Many of the examples they give aren't really 'bullshit' as per their definition, lacking the mendacious element, but if anything this makes the book more useful, since well-intended misinterpretation is, if anything, as if, if not more, dangerous, as the Covid epidemic, with science being done and interpreted in real-time and with life-and-death political decisions being made 'following the science', demonstrated. Indeed it would be fascinating to read a follow-up book or article by the authors on Covid science, information and mis-information.
One particularly valuable section focus on the issues with the p-value used in many scientific papers, for example how, while this can (with certain biases) answer a question about the probability that the null hypothesis is true, a failure of the null hypothesis is then often used to confirm one particular alternative hypothesis, which is a different question altogether.
The various websites which keep a record of Things the Daily Mail says cause or prevent cancer are perhaps the most obvious example of incorrect conclusions from pseudo-science but also scientific studies.
Here I was surprised not to see more explicit mention of Bayesian logic. And when they referred to the Prosecutor's Fallacy, while the lack of reference to the Saly Clark case reflects the US-centric nature of the book, I was surprised, particularly given that the OJ Simpson case featured elsewhere in the book, to see no reference to the 'Defense attorney's fallacy'.
It was good to see mention of mathiness, which Paul Romer has used in the context of much economic analysis (see e.g. Piketty's flawed analyses) but which here the authors use in the context of many popular business books, which often invent nonsensical formulas, such as the well-known Trust Equation (according to which, if self-orientation is zero, trust is infinite, and which adds quantities which can't easily be measured on a common scale).
As the authors themselves acknowledge, much of the book focuses on 'spotting', rather than 'calling', bullshit, as the former is needed to do the latter, indeed one of their key pieces of advice on 'calling' bullshit is to be sure of your own facts, and, for example, present plausible alternative explanations.
Their top tips on spotting bullshit are:
1. Question the source of information (who is telling me this? How do they know it? What are they trying to “sell” me?)
2. Beware of unfair comparisons
- I'd perhaps replace this with misleading/irrelevant. Interestingly the example they quote is a 2018 study that airport security trays contain more germs than toilet seats. The author's point out that the study is unfair as it only looked at germs responsible for respiratory infections, not generally. In 2020-21, after this book was written, the risk of airport security trays spreading a respiratory virus suddenly became far more relevant.
3. If it seems too good to too bad to be true…
- ... it certainly requires more skepticism.
4. Think in orders of magnitude
- in particular the authors extol the critical skill of Here the concept of Fermi estimation.
5. Avoid confirmation bias
6. Consider multiple hypothesis
- this last being perhaps the book's key insight.
They have a separate, but similar, list for online facts, but here their key advice is think (and check) more, and share less.
The section on Calling Bullshit is only at the end of the book. They define calling as “a performance utterative in which one repudiates something objectionable” (legally the term 'hereby' defining a performance utterative), but they present a well-balanced case, indeed argue against social media spats, and also highlight the need to question why we're calling something out, i.e. is it to help others (often the 'bullshitter' themselves) appreciate and learn from the mistake made, or just to make us look good: A well-actually guy has more in common with a bullshitter than with a caller of bullshit.
Overall 3.5 stars, rounded to 3.
It's definitely a very interesting summary of the topic although, as mentioned, I didn't find much new here, there were definitely some new insights and a helpful overall framework. On the negative side, the US-centricity grated at times (the assumption the reader is American). And like many such books, a summary (here e.g. reading just the last 2 pages) or a podcast/TED Talk would likely give as much value as the 300-page book.
And this last point is why I read tens of thousands of pages of literary fiction each year, but prefer my business/popular science books in bite-sized does, since in the former it is the quality of the prose that matters not the distilled ideas.
It is perhaps no surprise that the best sentence in this book reads "Many years later, as he faced the firing squad, Colonel Aureliano Buendía was to remember that distant afternoon when his father took him to discover ice." This is of course one of the best known and finest lines in English literature, although it is written not by Gabriel Garcia Marquez, to whom the authors of this book attribute the quote, but rather by Gregory Rabassa. Gabriel Garcia Marquez wrote instead "Muchos años después, frente al pelotón de fusilamiento, el coronel Aureliano Buendía había de recordar aquella tarde remota en que su padre lo llevó a conocer el hielo."
Now me making that point in the context of this book is me being a well-actually guy. But me making this point on my Goodreads reviews is actually me calling out bullshit, since one of my main literary missions is to highlight the importance of translators and their key role, which is often relegated, particularly for famous authors, to a minor footnote on the inside pages. So this just shows that in interpreting, and debunking, 'facts', context matters.
December 19, 2020
I started this book while waiting for Abbu outside the ICU. The book ends today. So, today again I went to the hospital in front of the ICU.
I was wondering how much bullshit one person has to experience over the lifetime or even in a month. Anyway, this is a solid piece of work. Something that goes well beyond Darrell Huff's "How to lie with Statistics" and even more.
It mostly focuses on the bullshit that is presented in the form of information or anything that we tend to consume to act upon or make decisions. And sometimes we ourselves create bullshit too. My favorite part is the scetion on Technically True things. Some professions actually require this skill of generating technically true statements and they serve so many purposes.
In case anyone doesn't have the time to read the book, they can check out the authors' INFO 198 lecture series at UW iSchool. Each video is on average 5~8 minute long. So, it won't take much time.
https://youtube.com/playlist?list=PLP...
Happy reading the book or watching the series :)
I was wondering how much bullshit one person has to experience over the lifetime or even in a month. Anyway, this is a solid piece of work. Something that goes well beyond Darrell Huff's "How to lie with Statistics" and even more.
It mostly focuses on the bullshit that is presented in the form of information or anything that we tend to consume to act upon or make decisions. And sometimes we ourselves create bullshit too. My favorite part is the scetion on Technically True things. Some professions actually require this skill of generating technically true statements and they serve so many purposes.
In case anyone doesn't have the time to read the book, they can check out the authors' INFO 198 lecture series at UW iSchool. Each video is on average 5~8 minute long. So, it won't take much time.
https://youtube.com/playlist?list=PLP...
Happy reading the book or watching the series :)
February 24, 2022
Ein geniales Buch was zeigt, wie man Datenmanipulation erkennen kann. Die Autoren zeigen bekannte Beispiele wie man mit grafischen Tricks, Aussagen manipulieren kann.
Zum Beispiel kann durch reinzoomen ein Sachverhalt bestärkt oder bewusst geschwächt werden.
Das Buch ist ein gutes Handwerk, um einige dieser Techniken zu erkennen, um in Zukunft Statistiken besser bewerten zu können.
Zum Beispiel kann durch reinzoomen ein Sachverhalt bestärkt oder bewusst geschwächt werden.
Das Buch ist ein gutes Handwerk, um einige dieser Techniken zu erkennen, um in Zukunft Statistiken besser bewerten zu können.
December 11, 2021
This is an entertaining book about recognizing bullshit, researching and calling it out. Much of the book describes the various types of bullshit, and the research required to snoop out its origin. Then, a short portion of the book is about calling it out; how to call it out, and even when to call it out. The book is filled with anecdotal bullshit, and the research the author used to ferret out its origin. Much of the bullshit is unintended--it is simply a matter of passing along incompetent analyses and conclusions. When bullshit is intentional--that is simply called lying.
Selective bias is the reason for a lot of bullshit. This occurs when a survey or a statistic is unintentionally biased in the sampling population. The author describes the situation for waiting for a bus at the airport, for your particular brand of rental car. It always seems like all of the other busses pass you by, before your bus arrives. This is not a coincidence; it is a statistical rule when busses tend to get clumped together instead of arriving equally spaced in time. The author also explains why people who are dating seem to meet nice people who are unattractive, or attractive jerks. This also is not a coincidence; the book describes why this happens!
Then there is the misleading biases in data visualization. After the Florida "Stand Your Ground" law was enacted, a figure seemed to show at first glance, a drop in homicides. A close look at the vertical axis shows that it was inverted, giving the wrong impression. It turns out that the author of the figure did not intend to mislead, but used an unfortunate representation of the truth.
The book describes how scientific journal articles often go awry. It comes down to this: "When a measure becomes a target, it ceases to be a good measure." Scientific journals have a strong preference for a statistical measure called the "p-value" to be less than 0.05. As a result, scientists use the p-value to imply the probability of a hypothesis being true. They often alter the original hypothesis during analysis--absolutely a 'no-no'. Self-selection of statistical groupings that give low p-values is called "P-hacking", and is one reason why so many scientific studies are not reproducible. As a result, p-values no longer serve their original purpose.
The author gives a set of useful suggestions for spotting bullshit on the Internet:
1) Pay attention to where information comes from
2) Find multiple sources for a story
3) Dig back to the origin of a story
4) Use reverse-image lookup; very useful for fact checking
5) Beware of deep-fake media (photoshopping, etc.)
6) Use a fact-checking organization, like Politifact.com, FactCheck.org, or Snopes.com
7) Beware of the illusory truth effect; repetition does not imply truth.
8) Beware of social media: "Think more, share less"
The book contains a set of recommendations on how to call out bullshit. They are:
1) Reductio ad absurdum (very entertaining!)
2) Be memorable
3) Find counter-examples
4) Provide analogies
5) Redraw figures
Finally, the book describes when it is useless to call out bullshit and when it is a moral imperative. And, the chief take-away from the book is this: "If a story sounds too good--or too bad to be true--it probably is!"
Selective bias is the reason for a lot of bullshit. This occurs when a survey or a statistic is unintentionally biased in the sampling population. The author describes the situation for waiting for a bus at the airport, for your particular brand of rental car. It always seems like all of the other busses pass you by, before your bus arrives. This is not a coincidence; it is a statistical rule when busses tend to get clumped together instead of arriving equally spaced in time. The author also explains why people who are dating seem to meet nice people who are unattractive, or attractive jerks. This also is not a coincidence; the book describes why this happens!
Then there is the misleading biases in data visualization. After the Florida "Stand Your Ground" law was enacted, a figure seemed to show at first glance, a drop in homicides. A close look at the vertical axis shows that it was inverted, giving the wrong impression. It turns out that the author of the figure did not intend to mislead, but used an unfortunate representation of the truth.
The book describes how scientific journal articles often go awry. It comes down to this: "When a measure becomes a target, it ceases to be a good measure." Scientific journals have a strong preference for a statistical measure called the "p-value" to be less than 0.05. As a result, scientists use the p-value to imply the probability of a hypothesis being true. They often alter the original hypothesis during analysis--absolutely a 'no-no'. Self-selection of statistical groupings that give low p-values is called "P-hacking", and is one reason why so many scientific studies are not reproducible. As a result, p-values no longer serve their original purpose.
The author gives a set of useful suggestions for spotting bullshit on the Internet:
1) Pay attention to where information comes from
2) Find multiple sources for a story
3) Dig back to the origin of a story
4) Use reverse-image lookup; very useful for fact checking
5) Beware of deep-fake media (photoshopping, etc.)
6) Use a fact-checking organization, like Politifact.com, FactCheck.org, or Snopes.com
7) Beware of the illusory truth effect; repetition does not imply truth.
8) Beware of social media: "Think more, share less"
The book contains a set of recommendations on how to call out bullshit. They are:
1) Reductio ad absurdum (very entertaining!)
2) Be memorable
3) Find counter-examples
4) Provide analogies
5) Redraw figures
Finally, the book describes when it is useless to call out bullshit and when it is a moral imperative. And, the chief take-away from the book is this: "If a story sounds too good--or too bad to be true--it probably is!"
December 11, 2021
Skepticism is important, and so I applaud these professors in their mission to fight BS, and much of what they talk about is important and true. But a lot of it is esoteric trivial examples. I'm disappointed because I was looking for a book on how to beat the very dangerous bullshit threatening the world today (in areas like pandemics).
Also, I can't get around the fact that you have to go deep into any topic before you can say something meaningful about it. And so this survey of all kinds of bullshit-detecting tactics for the ordinary citizen to use doesn't seem very practical. Towards the end of the book (p.258), the authors give an example where you have to dig down and look up the primary source to sort out a question. And then in their conclusion, they say "Make sure you have the facts at hand--don't skimp on the background research--and then double-check them." I agree. But even if you can critically read one scientific article, you can't interpret its relevance without knowing the context in that field. And nobody can do that for every topic out there.
There's a chapter on causality and the authors mention smoking and cancer as a "clear-cut" causal link. But that's no explanation: just saying it's obvious should ring bullshit alarms. It would have been instructive to explain how we know that smoking causes cancer. We do know that. It is true. It can be explained to people. You can show them the overwhelming evidence. You can explain the Uncle Norbert fallacy. But that takes time. More importantly, getting citizens or even doctors to read the original science is not how the progress in tobacco control was achieved.
Similarly, the things that I've seen that are promising for fighting global warming denial involve taking people out into nature or doing experiments and hands-on demonstrations of the evidence. But I can't imagine a scalable approach for doing that with tens of millions of people. And that would not stop the endless flow of money and beautifully-crafted lies from powerful special interests.
A recurring theme in the book that troubled me is the idea that "extraordinary claims require extraordinary evidence," because the implication of this is that claims that go along with the conventional wisdom don't require special scrutiny. So if experts say the Earth is flat, or opioids aren't addictive, then it's safe to agree with either consensus opinion, but dangerous to challenge it. I would argue that all claims should be met with skepticism and all new science requires rigorous evidence. Progress in many areas, including medicine, is hampered by adherence to BS dogma. Scientists and other gatekeepers of information should always be asking "How do you know that?" -- not just when it's easy by punching down against apparent nut jobs, but also when it's hard.
Books of possible interest:
How to Lie with Statistics.
Investigating Disease Patterns: The Science of Epidemiology
Factfulness: Ten Reasons We're Wrong About the World – and Why Things Are Better Than You Think.
Getting What We Deserve: Health and Medical Care in America.
Dark Money: The Hidden History of the Billionaires Behind the Rise of the Radical Right
The Discovery of Global Warming: Revised and Expanded Edition
Merchants of Doubt: How a Handful of Scientists Obscured the Truth on Issues from Tobacco Smoke to Global Warming
Hoax: Donald Trump, Fox News, and the Dangerous Distortion of Truth
The Art of Scientific Investigation
Science & Human Values
Also, I can't get around the fact that you have to go deep into any topic before you can say something meaningful about it. And so this survey of all kinds of bullshit-detecting tactics for the ordinary citizen to use doesn't seem very practical. Towards the end of the book (p.258), the authors give an example where you have to dig down and look up the primary source to sort out a question. And then in their conclusion, they say "Make sure you have the facts at hand--don't skimp on the background research--and then double-check them." I agree. But even if you can critically read one scientific article, you can't interpret its relevance without knowing the context in that field. And nobody can do that for every topic out there.
There's a chapter on causality and the authors mention smoking and cancer as a "clear-cut" causal link. But that's no explanation: just saying it's obvious should ring bullshit alarms. It would have been instructive to explain how we know that smoking causes cancer. We do know that. It is true. It can be explained to people. You can show them the overwhelming evidence. You can explain the Uncle Norbert fallacy. But that takes time. More importantly, getting citizens or even doctors to read the original science is not how the progress in tobacco control was achieved.
Similarly, the things that I've seen that are promising for fighting global warming denial involve taking people out into nature or doing experiments and hands-on demonstrations of the evidence. But I can't imagine a scalable approach for doing that with tens of millions of people. And that would not stop the endless flow of money and beautifully-crafted lies from powerful special interests.
A recurring theme in the book that troubled me is the idea that "extraordinary claims require extraordinary evidence," because the implication of this is that claims that go along with the conventional wisdom don't require special scrutiny. So if experts say the Earth is flat, or opioids aren't addictive, then it's safe to agree with either consensus opinion, but dangerous to challenge it. I would argue that all claims should be met with skepticism and all new science requires rigorous evidence. Progress in many areas, including medicine, is hampered by adherence to BS dogma. Scientists and other gatekeepers of information should always be asking "How do you know that?" -- not just when it's easy by punching down against apparent nut jobs, but also when it's hard.
Books of possible interest:
How to Lie with Statistics.
Investigating Disease Patterns: The Science of Epidemiology
Factfulness: Ten Reasons We're Wrong About the World – and Why Things Are Better Than You Think.
Getting What We Deserve: Health and Medical Care in America.
Dark Money: The Hidden History of the Billionaires Behind the Rise of the Radical Right
The Discovery of Global Warming: Revised and Expanded Edition
Merchants of Doubt: How a Handful of Scientists Obscured the Truth on Issues from Tobacco Smoke to Global Warming
Hoax: Donald Trump, Fox News, and the Dangerous Distortion of Truth
The Art of Scientific Investigation
Science & Human Values
June 22, 2020
I was expecting a light, funny, informational read when I selected this book. I should have known better with terms in the blurb like "expertise in statistics" and "examples of selection bias!" Don't get me wrong, this is an incredibly interesting book, and there's humor. It's thorough and it is DENSE, full of graphs and equations and research examples. It was over my head at times. I had to read small portions at a time (and sometimes re-read them) but it makes a lot of sense to me how we as a society are so easily manipulated by modern bullshit. This book covers a lot of ground and draws from many academic fields: philosophy, social psychology, social science, statistics, biology, and language. At times I was overwhelmed and upset by the blatant evidence of ways people are tricked by marketing, politicians, and celebrities. Ultimately, this is a useful book. The authors give us clear explanations and examples of how to wade through bullshit. By explaining complex principles and showing the difference between causation and correlation, the authors give the rear tools to use to help navigate today's headlines, social media posts, and myths.
Many thanks to NetGalley for my advanced copy of this book in exchange for my honest review.
Many thanks to NetGalley for my advanced copy of this book in exchange for my honest review.
February 3, 2022
هنگامی که سالها پیش سرگرم خواندن کتابِ سیروسِ شمیسا ی ارجمند؛ شاهدبازی در ادبیات فارسی، بودم ناگهان انگار پرده ی سنگین و سیاهی افتاد پایین و می شد بیرون را بهتر دید. لذت عجیبی داشت، مانند یک جور نشئه کردن، درد هم داشت اما خب لذتش بعد از نشست کردن و نشت کردن بی همتا بود و هست. او دستِ مهمل باف ها رو خوب رو کرده بود. آقایانِ شاعرانِ بزرگ در بهترین حالتش با معیارهای زمانه ی ما نظرباز بودند و در زمانه ی خودشان همان چیزی بودند که نوشته بودند از عیار در عنفوان جوانی چنان که افتد و دانی. این فرهنگِ لنگ و متقلب و بیچاره ی ما بود که می خواست خودی بنماید و همه را برکشیده و والا نشان دهد بس که تهی دست بود و نان را به نرخِ گداییِ روز نشخوار کرده بود. ورنه چه بسا آنها که ستوده شده بودند در دوران های بسیار خودشان هم چنین ادعایی نداشتند و اصلن نم یخواستند که اینجور باشند که ما دلمان می خواست. حالا بماند که این لا به لا کسانی هم بوده اند که درود به شرف شان، جای هیچ گمانه ی بیراهی را نگذاشته اند که یکی شان مثلن منوچهری دامغانی که به حق ستودنی ست
شعرها همه درباره ی پسربارگی و شراب و دست به تن هم مالیدن بود و تلاش این بود که این مفاهیمِ کاملن زمینی و انسانی را قالب بزنند به عرفانِ مبتذل و بی فایده و سیر در افلاک و باقی داستانها، اما قصه همان بود که دکتر شمیسا نشان مان داده بود و این کتاب او بی گمان تشخیص و افشای چرندبافیِ چندصدساله ای بود که هنوز هم هست. در این میان هم شماری که با این فرهنگِ جعلی دکان های پر و پیمانی راه انداخته بودند کم نگذاشتند در لعن و نفرین و آزار اما خب چه می شد کرد؟ حافظ نه آن عارف شوریده حال که پسربازی بیچاره و ترس خورده بود. خود بیچاره اش هم گفته بود بس که حکومتِ ترس بود و دروغ و فریب. حکومتِ جعل و جهل بود دیگر
همینطور وقتی زنده یاد آرامش دوستدار با گفتن این مضمون که مگر می شود از فلسفه ی دینی حرف زد در حالیکه دین، اندیشیدن را ممنوع کرده است خط کلفت و سیاهی کشید روی همه ی مهملاتی که می دانیم و می شناسیم، در نبودِ آزادیسِ اندیشه، اندیشیدن چه چرندی باید باشد؟ حالا اینها را نوشتم. این دو فقره مثال را که بپرم به پایان بندی کتاب آنجا که می گوید از قولِ والتر لیپمن که: اگر جامعه ای ابزاری برای تشخیص دروغ نداشته باشد، هیچ آزادی ای هم نخواهد داشت
پس برویم پی ابزار تا بیشتر از این بوی عفونت بیمارمان نکند، این بیماریِ ناخوب شدنیِ نادانی و منفعت طلبی تاریخی و توهمِ بافرهنگ بودن را
یک جایی هم در کتاب می گوید آقا نرو بالا و خیال ورت دارد که خوبِ قصه تو هستی! انگار مانند این دیدگاه نوشتن بر کتاب ها هم باشد. حاشا حاشا
1400/11/13
شعرها همه درباره ی پسربارگی و شراب و دست به تن هم مالیدن بود و تلاش این بود که این مفاهیمِ کاملن زمینی و انسانی را قالب بزنند به عرفانِ مبتذل و بی فایده و سیر در افلاک و باقی داستانها، اما قصه همان بود که دکتر شمیسا نشان مان داده بود و این کتاب او بی گمان تشخیص و افشای چرندبافیِ چندصدساله ای بود که هنوز هم هست. در این میان هم شماری که با این فرهنگِ جعلی دکان های پر و پیمانی راه انداخته بودند کم نگذاشتند در لعن و نفرین و آزار اما خب چه می شد کرد؟ حافظ نه آن عارف شوریده حال که پسربازی بیچاره و ترس خورده بود. خود بیچاره اش هم گفته بود بس که حکومتِ ترس بود و دروغ و فریب. حکومتِ جعل و جهل بود دیگر
همینطور وقتی زنده یاد آرامش دوستدار با گفتن این مضمون که مگر می شود از فلسفه ی دینی حرف زد در حالیکه دین، اندیشیدن را ممنوع کرده است خط کلفت و سیاهی کشید روی همه ی مهملاتی که می دانیم و می شناسیم، در نبودِ آزادیسِ اندیشه، اندیشیدن چه چرندی باید باشد؟ حالا اینها را نوشتم. این دو فقره مثال را که بپرم به پایان بندی کتاب آنجا که می گوید از قولِ والتر لیپمن که: اگر جامعه ای ابزاری برای تشخیص دروغ نداشته باشد، هیچ آزادی ای هم نخواهد داشت
پس برویم پی ابزار تا بیشتر از این بوی عفونت بیمارمان نکند، این بیماریِ ناخوب شدنیِ نادانی و منفعت طلبی تاریخی و توهمِ بافرهنگ بودن را
یک جایی هم در کتاب می گوید آقا نرو بالا و خیال ورت دارد که خوبِ قصه تو هستی! انگار مانند این دیدگاه نوشتن بر کتاب ها هم باشد. حاشا حاشا
1400/11/13
December 29, 2021
"Falsehood flies, and truth comes limping after it." - Jonathan Swift
Just what the doctor ordered for the current state of the world. Super informative, enjoyable learning, brilliantly explained with lots of everyday relatable examples- highly recommend this especially considering the nonsense that we end up consuming daily on social media in the form of stats and graphics (this has especially spiked since the pandemic began!). As a researcher it is my job to be able to see through the noise in the data and to be able to filter out the bullshit, and the "Garbage in, Garbage out" mantra that we swear by is really well elucidated here. But more importantly, the authors have done a superb job of breaking down concepts into accessible content for everyone. The bits on issues with science reporting + publishing and tips for readers to identify bullshit was especially beautifully written, I thought.
Incidentally, just as I started reading the book, I was at the receiving end of a certain amount of bullshit and mansplaining as part of what was being called a "debate" - the book had a definition for every single point that was made by said person! Real world examples don't get better than this. Bullshitting has been a survival tactic by humanity for ages, made much worse now, be it by strangers on the internet or your own friends/family- it helps to be able to see through it, if not for anything else, then at least to be able to distance yourself from the source, which is what I chose to do (well aware that it's not always an option!).
Just what the doctor ordered for the current state of the world. Super informative, enjoyable learning, brilliantly explained with lots of everyday relatable examples- highly recommend this especially considering the nonsense that we end up consuming daily on social media in the form of stats and graphics (this has especially spiked since the pandemic began!). As a researcher it is my job to be able to see through the noise in the data and to be able to filter out the bullshit, and the "Garbage in, Garbage out" mantra that we swear by is really well elucidated here. But more importantly, the authors have done a superb job of breaking down concepts into accessible content for everyone. The bits on issues with science reporting + publishing and tips for readers to identify bullshit was especially beautifully written, I thought.
Incidentally, just as I started reading the book, I was at the receiving end of a certain amount of bullshit and mansplaining as part of what was being called a "debate" - the book had a definition for every single point that was made by said person! Real world examples don't get better than this. Bullshitting has been a survival tactic by humanity for ages, made much worse now, be it by strangers on the internet or your own friends/family- it helps to be able to see through it, if not for anything else, then at least to be able to distance yourself from the source, which is what I chose to do (well aware that it's not always an option!).
Read
January 3, 2023If you are a data scientist or if you are already a Carl Bergstrom fan, this book is probably not for you. You might encounter some new examples and find the humor enjoyable, but much will already be familiar since it is written for the most general audience.
September 5, 2020
Nothing that you will learn in the course of your studies will be of the slightest possible use to you [thereafter], save only this, that if you work hard and intelligently you should be able to detect when a man is talking rot, and that, in my view, is the main, if not sole, purpose of education.
— Idealist philosopher John Alexander Smith (1863–1939)
Spin. Fake News. Conspiracy theories. Lies. We are daily confronted with a stinking quagmire of misinformation, disinformation and fact-free drivel. How do we sort the truth from the lies? This is the premise of the timely new book, Calling Bullshit: The Art of Skepticism in a Data-Driven World (Allen Lane/Random House, 2020), a book that effectively acts as a field guide to the art of scepticism.
The authors are expert guides. Carl Bergstrom is a theoretical and evolutionary biologist who researches how information flows through biological and social networks. Jevin West is a data scientist who studies misinformation in science and society. Together, they teach a popular undergraduate class offered under the same name by the University of Washington.
This book, a distillation of that course, presents a mix of amusing anecdotes, timely news stories and accessible explanations of scientific and medical data. You do not need to be a professional statistician or some other sort of mathematical wizard, nor must you invest weeks into fact-checking to see through most nonsense. Instead, the authors argue, assessing the accuracy of a particular claim using basic logic, augmented (where necessary) with information that can be easily retrieved by an online search engine is sufficient to “call bullshit”.
The authors point out that creating bullshit is easier and often simpler than speaking the truth. That is why BS is accepted as truth. Italian software engineer, Alberto Brandolini perhaps said it best when he noted in 2014 that “the amount of energy required to refute bullshit is an order of magnitude bigger than [that needed] to produce it”, which explains why there is so much baloney in the world. Uriel Fanelli helpfully observed that “an idiot can create more BS than you could ever hope to refute.”
So why bother “calling bullshit”? As the authors assert, adequate bullshit detection is essential for the survival of democracy. Regardless of political ideologies, democracy has always relied on a critically thinking electorate, and this intellectual skill is more important than ever in this modern age of online information warfare. It also is critically important for proper functioning of any social group, whether it is a small group of friends or some other social group, or a professional community.
This book teaches us how to identify the various forms of new-school bullshit: how to evaluate scientific claims, to distinguish between correlation and causation, to recognize biased and unrepresentative data and small sample sizes, to identify selection biases in samples, to understand how data can be manipulated visually, and more. They also include lots of graphs and other data images so you can practice spotting screwy data representations yourself. Whether you are confused by the anti-vax movement, which grew out of a single retracted medical study, to the claim that Artificial Intelligence can infer sexual orientation from analyzing a photograph of a person’s face, there is no shortage of nutty ideas out there to contemplate and dissect.
The book ends with two empowering chapters on how to spot and refute nonsense and, more importantly, how to do so in a useful and constructive way.
Ironically, despite the authors’ assertion that proper fact-checking is essential, the book only has an alphabetized reference list for each chapter in the back of the book, leaving interested readers to scratch their collective heads as they try to deduce which statement should be attributed to which source. And although mistakes do inevitably creep in during the writing and editing process, I was surprised that the letter M in the commonly used acronym, STEM, was erroneously attributed to medicine, instead of mathematics. And yes, I was disappointed by the poor quality paper that the book was printed on.
Despite my complaints (some of which are probably beyond the authors’ control), Calling Bullshit presents a thoughtful, careful and engaging deconstruction about how to spot and disprove nonsense. It should be required reading for high school and university students as well as for any thinking person who is working to identify questionable news sources and stories, and navigate their way around social media in these weird times.
NOTE: Originally published at Forbes.com on 28 August 2020.
Also note that I read the hardback edition, which (weirdly) is not listed on goodreads, NOT THE KINDLE VERSION.
Read
October 10, 2020Currently reading this one, but noticed that no one has yet dropped the link for the professors' course at UW, which has been available online since before the book was finished: https://www.callingbullshit.org/sylla... I'd also like to add some context.
I made author CT Bergstrom's acquaintance over the early winter after finding his course online. In addition to studying disinformation, he's a bio professor and career epidemiologist, and his Twitter feed is why I was prepared earlier than nearly everyone for the pandemic, and was able to begin preparing family (which was not a simple matter at that early date, when there was still a lot of hesitation). It's not just that I had hand sanitizer, toilet paper and was ready to shelter at home: at a time when the lockdown hadn't yet begun and we had no testing and therefore didn't know the local numbers in our region, his scientific posts along with those of the colleagues he elevated like STAT News reporter Helen Branswell and Harvard epidemiologist Marc Lipsitch may have saved lives in my circle in particular, and certainly saved that of others in the wider world. You might have seen him on the BBC etc. early on when he appeared to help educate the western world when the "flatten the curve" chart first began circulating. Here's more about that: https://www.nbcnews.com/science/scien...
And so with the arrival and US federal mismanagement of the pandemic, and the associated disinformation campaigns and conspiracy theories, it turns out the release of Calling Bullshit is more timely than the authors imagined when they began their project years ago.
I hope the ideas within are widely circulated, understood and applied by readers. If you're curious, I expect that your library already has this book available for you to browse, and to see what you think.
A related text I'd also suggest (and that is included in the authors' coursework) is Carl Sagan's wonderful The Demon Haunted World, which remains a must-read and as a bonus can be casually enjoyed chapter by chapter over time. Each section is an essay that works independently and within the whole.
I made author CT Bergstrom's acquaintance over the early winter after finding his course online. In addition to studying disinformation, he's a bio professor and career epidemiologist, and his Twitter feed is why I was prepared earlier than nearly everyone for the pandemic, and was able to begin preparing family (which was not a simple matter at that early date, when there was still a lot of hesitation). It's not just that I had hand sanitizer, toilet paper and was ready to shelter at home: at a time when the lockdown hadn't yet begun and we had no testing and therefore didn't know the local numbers in our region, his scientific posts along with those of the colleagues he elevated like STAT News reporter Helen Branswell and Harvard epidemiologist Marc Lipsitch may have saved lives in my circle in particular, and certainly saved that of others in the wider world. You might have seen him on the BBC etc. early on when he appeared to help educate the western world when the "flatten the curve" chart first began circulating. Here's more about that: https://www.nbcnews.com/science/scien...
And so with the arrival and US federal mismanagement of the pandemic, and the associated disinformation campaigns and conspiracy theories, it turns out the release of Calling Bullshit is more timely than the authors imagined when they began their project years ago.
I hope the ideas within are widely circulated, understood and applied by readers. If you're curious, I expect that your library already has this book available for you to browse, and to see what you think.
A related text I'd also suggest (and that is included in the authors' coursework) is Carl Sagan's wonderful The Demon Haunted World, which remains a must-read and as a bonus can be casually enjoyed chapter by chapter over time. Each section is an essay that works independently and within the whole.
April 9, 2022
کتاب درباره ی تشخیص �� رد کردن اطلاعات غلط و شایعاتی صحبت می کنه که روزانه باهاش برخورد می کنیم و راه حل های خیلی خوبی هم پیشنهاد می کنه. نقطه ی قوت کتاب مثال های خیلی زیاد و متنوعیه که بیان می کنه: از نشر اکاذیب(!) در سطح مقالات دانشگاهی گرفته تا خرافات و شایعات عجیب و غریبی که گاهاً تو شبکه های اجتماعی محبوب می شن! نیمه ی اول کتاب خیلی جذاب و گیرا نوشته شده ولی نیمه ی دوم به نظر من تا حد زیادی تکرار همون موارد قبلی بود . و صد البته فصل آخر با نکات خیلی خوبی که در رابطه با قوانین و اخلاقیات"بحث کردن" مطرح کرد یکی از بهترین فصل های کتاب بود.
September 11, 2020
I'm giving this four stars and probably only skipping the fifth because of the pop title which undercuts the seriousness of the topic, IMO. Logic and Rhetoric have all but disappeared from educational programs when they were once mandated. Misinformation, disinformation and the manipulation of information seem more pervasive since, and I could claim a causal relationship, but would I be correct? It doesn't matter. What matters is that one understands what determines a causal relationship from a non-causal one and without supporting data my statement is just about as valid as what one gets from many usual sources of information. Bergeron and West have adapted for this book the material from a college course they taught of the same name of the same name at University of Washington in 1917. More about this can be seen at https://www.callingbullshit.org/FAQ.html
The book reads like a textbook and is both entertaining and informative. I found this a refresher for what I had already been taught but don't often enough use, but also valuable with new lessons that address data from more recent advances in communication. I wish there were more depth but there is a bibliography for further reading.
The book reads like a textbook and is both entertaining and informative. I found this a refresher for what I had already been taught but don't often enough use, but also valuable with new lessons that address data from more recent advances in communication. I wish there were more depth but there is a bibliography for further reading.
January 9, 2021
A lot of reviews waxing lyrically about this book but it's just too dense. You can easily glean all the information you need from the comments section on Goodreads...
And it started so well...hmmm. Around chapter 3 it all gets a little repetitive, stats are fired off to support the hypothesis of recognising bullshit but sadly ends up sounding like bullshit itself.
Boil it down, it isn't hard to recognise bullshit if you use common sense, I don't need citations and theroetical ramblings to do so. If you're an idiot, which would have been a better title for this book, "bullshit for idiots", and are guilable enough to believe half of what goes on, then sure. Read this book. I get a little tired of it. Not an entertaining read and more like a statistical snoozefest.
My final point, perhaps the writers should have taken a little look at their own biases before calling it out. Which undermines the intention don't you think?
And it started so well...hmmm. Around chapter 3 it all gets a little repetitive, stats are fired off to support the hypothesis of recognising bullshit but sadly ends up sounding like bullshit itself.
Boil it down, it isn't hard to recognise bullshit if you use common sense, I don't need citations and theroetical ramblings to do so. If you're an idiot, which would have been a better title for this book, "bullshit for idiots", and are guilable enough to believe half of what goes on, then sure. Read this book. I get a little tired of it. Not an entertaining read and more like a statistical snoozefest.
My final point, perhaps the writers should have taken a little look at their own biases before calling it out. Which undermines the intention don't you think?
November 7, 2022
At first I thought this was going to be a rehash of all the other books out there on cognitive biases, but it turned out to include quite a few things I haven't heard articulated very well, like how the scientific process and publishing industry work, and about AI and big data (this section was excellent). This is a book I could happily recommend to others as a primer on critical thinking and spotting, ahem, bullshit, especially on the internet. The authors did a really good job of not making it (much) about pet theories, but about general principles that can be applied to all theories. They also avoided taking political sides which, in this day and age, is amazing.
2022-11-07
Second time through. Still good, but a bit boring and simple in parts. The second half is better IMO.
2022-11-07
Second time through. Still good, but a bit boring and simple in parts. The second half is better IMO.
February 1, 2021
I won this book in a goodreads drawing.
A book that shows you how to think critically. Could be useful. Unfortunately, it relies on Snopes for some of its examples, and they've shown they can't really be trusted. "Fact checking" has become just so much...BS.
Still, some useful stuff here.
A book that shows you how to think critically. Could be useful. Unfortunately, it relies on Snopes for some of its examples, and they've shown they can't really be trusted. "Fact checking" has become just so much...BS.
Still, some useful stuff here.
February 11, 2021
Like everyone else, I believe that I can't be fooled by...anything, ever. Knowing that it's just that certainty that makes one vulnerable, I deliberately give myself little booster shots of skepticism by reading every one of these books that comes out. But despite having a longish career in medical research spent dealing with data, there's always something. These days I think my greatest vulnerability is my own experience: often I overlook the simplest grounds for calling bullshit while trying to track down that perfect point for Fisking. I don't need to look at the tables if someone is claiming to have invented cold fusion, I just have to remember the second law of thermodynamics. Vaccination denialism: do you know what the consequences are for being vulnerable to [inside horrible malady here]?
Booster shots. This is a good one. Like many others, the authors have feared to be dry or boring and in consequence are entertaining as hell. These guys have had a live audience to practice on so they are particularly clear, straightforward, and spot on.
Of course, the one I've read most recently is always the best...no, really, this one may be. Highly and indiscriminately recommended. We all need less bullshit to wade through, especially those of us who are reading while walking and might be more vulnerable by dint of just not paying attention.
Library copy
Booster shots. This is a good one. Like many others, the authors have feared to be dry or boring and in consequence are entertaining as hell. These guys have had a live audience to practice on so they are particularly clear, straightforward, and spot on.
Of course, the one I've read most recently is always the best...no, really, this one may be. Highly and indiscriminately recommended. We all need less bullshit to wade through, especially those of us who are reading while walking and might be more vulnerable by dint of just not paying attention.
Library copy
January 7, 2021
I guess this is Sagan's Demon-Haunted World for 2021? Both authors go through the most common ways that data is used to bullshit people and to create narratives. It grew out of a university course (Calling Bullshit) and goes over ways data is misrepresented, chosen, how sometimes counter-intuitively data can be interpreted, how people trying to sell you stuff or ideas can manipulate data to tell the story they want to tell,
So if you want to learn how to spot when someone tries to BS you, go for this book.
If there's one problem with this book is that the actual 'calling bullshit' part is very short, some tools are presented and some caveats described (i.e., don't be the 'well actually' guy, only call out bullshit if there's an actual problem, not to try and make yourself look smart). I guess you could write an entirely different book on tools and techniques on the discussion and public dismantling of bullshit.
As an aside:
Recently I was at a kind of launch event for a new data science unit. At the end of the teaching demo, a government representative stood up and said that a good 95% of data science graduates are not good for their purpose, they can run algorithms and analyse data but they have zero critical thinking skills, and sometimes present results that are obviously nonsense if you stop and think about it. West/Bergstrom identify this too:
This is a problem across all STEM disciplines, we've completely stopped teaching students how to be critical thinkers, and in extension young citizens often cannot evaluate their politicians' BS.
So if you want to learn how to spot when someone tries to BS you, go for this book.
If there's one problem with this book is that the actual 'calling bullshit' part is very short, some tools are presented and some caveats described (i.e., don't be the 'well actually' guy, only call out bullshit if there's an actual problem, not to try and make yourself look smart). I guess you could write an entirely different book on tools and techniques on the discussion and public dismantling of bullshit.
As an aside:
Recently I was at a kind of launch event for a new data science unit. At the end of the teaching demo, a government representative stood up and said that a good 95% of data science graduates are not good for their purpose, they can run algorithms and analyse data but they have zero critical thinking skills, and sometimes present results that are obviously nonsense if you stop and think about it. West/Bergstrom identify this too:
But this focus on facts and skills [in STEM teaching] comes at the expense of training and practice in the art of critical thinking. In the humanities and the social sciences, students are taught to smash conflicting ideas up against one another and grapple with discordant arguments. in STEM fields, students seldom are given paradoxes that they need to resolve, [...] or fallacious claims that they need to critique.
This is a problem across all STEM disciplines, we've completely stopped teaching students how to be critical thinkers, and in extension young citizens often cannot evaluate their politicians' BS.
February 20, 2021
Most of this book is stuff that I (and probably many readers) am familiar with intellectually but don't necessarily apply reflexively whenever I read the news or hear a statistic. So for me, this book was really useful in that it primed me to intentionally be on the defensive about common misrepresentations in statistics and data visualization.
By far my favorite chapter in this book was the one on selection bias; it's easy to think about selection bias when you're reading an econ paper or a clinical trial and the cohort selection is explicit, but the authors show that variants of selection bias are at the root of many other pervasive statistical curiosities (e.g., the observation that the majority of people have fewer friends than their friends do).
I also liked the framing of the whole book in the beginning of misinformation/misrepresentation becoming more common as journalism has switched to an online model. I previously associated a lot of misinformation with social media, where the masses can spread whatever news they want, even if it is incorrect. But this book makes the additional point that long-term subscriptions (e.g., a print newspaper) incentivize high-accuracy reporting that stands the test of time, while click/ad-based Internet journalism incentivizes headlines which are flashy and, often as a direct result, misleading -- even from really reputable news sources.
By far my favorite chapter in this book was the one on selection bias; it's easy to think about selection bias when you're reading an econ paper or a clinical trial and the cohort selection is explicit, but the authors show that variants of selection bias are at the root of many other pervasive statistical curiosities (e.g., the observation that the majority of people have fewer friends than their friends do).
I also liked the framing of the whole book in the beginning of misinformation/misrepresentation becoming more common as journalism has switched to an online model. I previously associated a lot of misinformation with social media, where the masses can spread whatever news they want, even if it is incorrect. But this book makes the additional point that long-term subscriptions (e.g., a print newspaper) incentivize high-accuracy reporting that stands the test of time, while click/ad-based Internet journalism incentivizes headlines which are flashy and, often as a direct result, misleading -- even from really reputable news sources.
September 8, 2021
It is a nice and to-date summary of how and why bullshit spreads, how to protect oneself from it, and finally how to protect others from it. Over the years of dealing with such bullshit, I had been able to develop most of the rules of thumb mentioned in the book but it was still nice to see it all structured in one place. Giving it 4 stars only because while the execution was flawless, there wasn't much new to learn for me
March 9, 2021
A fascinating read, great ideas from causality to statistics, from ethics to science.
After reading books of Neil Postman, I always regretted that he couldn’t see the internet and social media era to adjust his brilliant books accordingly, but with reading this book, I think the gap is filled.
After reading books of Neil Postman, I always regretted that he couldn’t see the internet and social media era to adjust his brilliant books accordingly, but with reading this book, I think the gap is filled.
Displaying 1 - 30 of 567 reviews