Review From User :
This is a book about how to spot problems with the facts you encounter, problems that may lead you to draw the wrong conclusions - critical skills that we need today since we're blasted with information in a society based on conspicuous consumption. Everyone wants our support or to sell us something & many are skilled at leveraging our inherent flaws in reasoning to this end. Reading this should be a prerequisite for posting on Facebook.
The flaws inherent in our reasoning are manifold. We're a story-loving species & try to find patterns in everything. Both of these are methods for simplifying the reasoning process by giving us hooks on which to hang all the data. We tend toward beliefs, often snap judgements based on previous experience. Once we believe in something, it is much harder to shake our thinking process into a different pattern. A newspaper headline can be a complete lie & shown so in the story that follows, but people remember the lie.
The book is broken into 3 sections. Each teaches methods for evaluating data & then contains real world examples, many taken from the current press.
Evaluating Numbers: starts off with a quote attributed to Mark Twain It ain't what you don't know that gets you into trouble. It's what you know for sure that just ain't so.
- Plausibility: There are lies, damn lies, & statistics, so the latter are a conundrum today. We're aware they can be used to lie convincingly, but they look so good to us. What we fail to consider is how they are figured & he gives some great examples of common errors in them, probability, & percentages.
- Fun With Averages: Summary statistics (the mean, median, & mode) are all useful IF they are used correctly. They can be downright deceiving if used incorrectly, though. The mean (what we usually think of as the 'average') is very sensitive to outliers while the median & mode are far less so. Again, he does a great job of putting this into real world examples of when each should be used, when avoided, & what we should watch for.
- Axis Shenanigans: Graphs are another great way to condense information in a way that makes more sense, but they can also easily be misleading by not labeling or creatively labeling the axes.
- Hijinks With How Numbers Are Reported: This section builds on the preceding sections to show how the underlying numbers can be manipulated & why it is often profitable to do so. He also mentions the fallacy of correlation versus causation, a very common trap.
- How Numbers Are Collected:"Just because there's a number on it, it doesn't mean that the number was arrived at properly." Wow. Again, great examples.
- Probabilities allow us to quantify future events and are an important aid to rational decision making. Without them, we can become seduced by anecdotes and stories. While Levitin does a great job of simplifying it, there are some terms that I needed to get very comfortable with. This is the main section where the text really helps since he gets into Fourfold tables which help make sense of the actual odds. This is not always intuitive. In fact, our brains are wired to see them incorrectly much of the time. The rest of the book refers back to this section regularly, so it's important to understand it.
Classic probability is what we typically think of, but it is based on symmetry & equal likelihood such in the case of a coin toss. All the possibilities are known & discreet, but this is often not the case in real world problems such as how well a drug works, AKA Frequentist Probability. There is also Subjective Probability, such as when someone expresses an opinion on the likelihood of a future event. To confuse the issue further, we often combine them. We also need to consider conditional probability when an event is informed by another event (e.g., most car accidents happen during rush hour) & remember that these do not work backward (Just because it is rush hour does not mean you are likely be in an accident if another condition changes, such as not driving.) but we often forget the conditions & think that way.
Evaluating Words: Language is slippery & defines how we think about things. Half truths are often the worst lies.
- How Do We Know We rely on experts, certifications, licenses, encyclopedias, and textbooks. AKA, secondhand knowledge, so we need to evaluate the the source & the claims. Who is the source & how likely is it that they're right Experts can be wrong, they're just less likely to be IN THEIR AREA OF EXPERTISE (next section) than a random person. If the claim is a good one, we should be able to evaluate the evidence & it should be well documented. The content of footnotes are especially important & should be fully explored.
- Identifying Expertise: People are generally only experts in a narrow field, especially today when everything is so complicated. My doctor's opinion on what ails my car shouldn't weigh as heavily as my mechanic's, but we're often swayed by degrees or popularity. Levitin outlines some great ways to identify snow jobs including looking at the URL domain & other handy tips for using the Internet wisely & avoiding common terminology pitfalls.
- Overlooked, Undervalued Alternative Explanations: This goes back to our beliefs & love of stories. When we're given a likely story, it's often difficult to think of another, but the preceding sections have given us great tools for spotting inaccurate statistics/probabilities, missing factors such as a control group, &/or cherry-picked data.
-CounterKnowledge...is misinformation packaged to look like fact and that some critical mass of people have begun to believe. Generally conspiracy theories & pseudoscience. They often rely on open questions & anomalies which are then whipped into a likely story.
"If you thought that science was certain - well, that is just an error on your part." - Richard Feynman
"The whole problem with the world is that fools and fanatics are always so certain of themselves, but wiser people so full of doubts." - Bertrand Russell
Evaluating the World: deals with critical thinking overall & there is some repetition of previous material, but here he pulls all the skills he's outlined together & shows them in a variety of situations.
How Science Works: He explains deduction & induction in scientific reasoning, where & how each should be used. Again, there are several excellent real-world examples & common pitfalls outlined, such as the reversal of logical statements.
Logical Fallacies: Correlation confused with causation comes up again along with common framing & prior belief issues.
Knowing What You Don't Know"There are things we know, things we are aware that we do not know, and some things we aren't even aware that we don't know." There's a fourth possibility, of course-things we know that we aren't aware we know. It seems a bit confusing at first read, but read it over a bit & it makes perfect sense. Evaluating our own knowledge is really important & sometimes that means digging back into the foundations of that knowledge & laying out the problem properly. He does a great job of sorting it all out.
Bayesian Thinking in Science & Court: Unlikely claims require more proof than likely ones, but this relies on understanding probability properly & that's something our court system often gets wrong. It's so well known that it is called The Prosecutor's Fallacy.
Four Case Studies are great examples of weighing probabilities in real world situations. In the first, he decided on how to handle cancer in his dog. This book is worth reading for this one example alone.
There is a great conclusion to sum it up. In the print version, there is also an appendix that outlines & applies Baye's Rule, & a glossary. While this is very well read & great as an audio book for the most part, I'd suggest getting a text copy since there are some tables and logic equations that were helpful to look at.
I highly recommend this to everyone. If my kids were still in the house, I'd make it required reading & I'd test them on it. We're deluged with information constantly. This is a wonderful book on how to evaluate & make sense out of the flood.
Media Size : 104 KB