What do you think?
Rate this book
Correlation Is Not Causation: Learn How to Avoid the 5 Traps That Even Pros Fall Into
Correlation Is Not Causation. You know it and I know it, and yet we are constantly having to be reminded of it because we can’t seem to help but get it wrong. How many times have you heard someone really smart say something like ‘wow, this correlation has a p-value of 0.000001 so A must be causing B…’? It’s not our fault though – we’re only human. We seek explanation for patterns and events that happen around us, and if something defies logic, we try to find a reason why it might make sense. If something doesn’t add up, we make it up. OK, so if correlation does not necessarily imply causation, there must be a reason for that, and there must be something that is causing what we observe. That is what this book is all about. If we discover a correlation between a pair of variables there are five alternatives to one being the direct cause of the other, and we’ll unmask all five in this book. Then, once we understand each of these alternatives, we’ll formulate a plan to discover whether we have a direct causal link or whether there is some other explanation. Correlation Is Not Causation explains how to systematically test for the five most common correlation-causation pitfalls that even the pros fall into (occasionally). We’ll learn to create strategies to analyse the data and interpret the results in a way that is easy to understand. Best of all, there is no technical or statistical jargon – it is written in plain English. It is packed with visually intuitive examples and makes no assumptions about your previous experience with correlations – in short, it is perfect for beginners! Discover the world of correlation and causation. Get this book, TODAY!
- GenresScience
39 pages, Kindle Edition
Published March 22, 2018
23 people are currently reading
50 people want to read
Ratings & Reviews
Friends & Following
Create a free account to discover what your friends think of this book!
Community Reviews
Displaying 1 - 7 of 7 reviews
May 20, 2018
Readers coming to Lee Baker's Correlation Is Not Causation have probably read the earlier two books in his Bite Size Statistics series and might guess that this third volume would be more of the same. If so, they would probably get a rude introduction to the theme of this volume. For, while the first two books were short, breezy, entertaining reads that did not require any advanced mathematical knowledge, Correlation Is Not Causation may be equally short, but its second half assumes a certain level of knowledge of statistical terms and concepts well beyond that needed to enjoy the first two books.
The first half of Baker's newest book, at least, is reasonably enjoyable. Correlation Is Not Causation is a statistical axiom of which most people are aware at a certain fundamental level, namely that if A and B happen together, then it is wrong to automatically assume that A causes B to happen. Baker provides five alternative theories that might validly explain the correlation between A and B under a particular set of circumstances.
One possibility Baker mentions, which occurs in the real world quite often, is that there is actually a “third cause,” something that causes both A and B to occur. As an example, he mentions a possible correlation between people carrying matches in their pocket and people who develop lung cancer. This might well be true but it certainly doesn’t mean that carrying matches leads to lung cancer (and, thus, that not carrying them would be a preventative). Instead, people who carry matches are more likely to be smokers and smoking (the third cause) definitely contributes to lung cancer.
In the first half of Correlation Is Not Causation, Baker explains in depth five common alternate explanations for correlation and why it is often very important to know the correct explanation. As the matches/cancer example illustrates, scientists often try various treatments for diseases and knowing if a particular treatment is actually effective or merely an example of a particular fallacy is crucial.
Unfortunately for the lay reader, the second half of Correlation Is Not Causation can be somewhat rough sailing. It presumes a knowledge of advanced statistical terms that it makes no attempt to explain. A typical sentence reads: “If you did a univariate analysis between your pair of variables, such as a Pearson correlation, the result actually tells you a lot less than you think it does.” Actually, this sentence tells many readers a lot less than Baker thinks it does.
Since I would assume that most people who pick up Correlation Is Not Causation: (at least Baker offers the book for free) will not be graduate school statistics students performing their own univariate analyses, the second half of this book will probably not appeal to them very much, especially for those who enjoyed Baker’s first two books and thought they were getting more of the same. Further, those same grad school students probably won’t be hunting for a book from the author of Graphs Don’t Lie. The end result is a book that’s solidly in no-man’s land, too complex and uninteresting (at least the second half) but not meaty enough to serve as any sort of text for serious students of the subject. As a free book, there’s enough worthwhile material in the first half of the Correlation Is Not Causation to warrant a mild recommendation, but readers will find little correlation between this book and Baker’s earlier works in the series.
The first half of Baker's newest book, at least, is reasonably enjoyable. Correlation Is Not Causation is a statistical axiom of which most people are aware at a certain fundamental level, namely that if A and B happen together, then it is wrong to automatically assume that A causes B to happen. Baker provides five alternative theories that might validly explain the correlation between A and B under a particular set of circumstances.
One possibility Baker mentions, which occurs in the real world quite often, is that there is actually a “third cause,” something that causes both A and B to occur. As an example, he mentions a possible correlation between people carrying matches in their pocket and people who develop lung cancer. This might well be true but it certainly doesn’t mean that carrying matches leads to lung cancer (and, thus, that not carrying them would be a preventative). Instead, people who carry matches are more likely to be smokers and smoking (the third cause) definitely contributes to lung cancer.
In the first half of Correlation Is Not Causation, Baker explains in depth five common alternate explanations for correlation and why it is often very important to know the correct explanation. As the matches/cancer example illustrates, scientists often try various treatments for diseases and knowing if a particular treatment is actually effective or merely an example of a particular fallacy is crucial.
Unfortunately for the lay reader, the second half of Correlation Is Not Causation can be somewhat rough sailing. It presumes a knowledge of advanced statistical terms that it makes no attempt to explain. A typical sentence reads: “If you did a univariate analysis between your pair of variables, such as a Pearson correlation, the result actually tells you a lot less than you think it does.” Actually, this sentence tells many readers a lot less than Baker thinks it does.
Since I would assume that most people who pick up Correlation Is Not Causation: (at least Baker offers the book for free) will not be graduate school statistics students performing their own univariate analyses, the second half of this book will probably not appeal to them very much, especially for those who enjoyed Baker’s first two books and thought they were getting more of the same. Further, those same grad school students probably won’t be hunting for a book from the author of Graphs Don’t Lie. The end result is a book that’s solidly in no-man’s land, too complex and uninteresting (at least the second half) but not meaty enough to serve as any sort of text for serious students of the subject. As a free book, there’s enough worthwhile material in the first half of the Correlation Is Not Causation to warrant a mild recommendation, but readers will find little correlation between this book and Baker’s earlier works in the series.
June 23, 2018
Complexity made simple
Even though I have experience in the authors field - it in no way comes close to that of the author. Not being a pro I have waltzed down these paths many times by not being able to articulate "why" to management.
The author succinctly explains the 5 traps and how to identify them in a way that is logical and understandable.
Even though I have experience in the authors field - it in no way comes close to that of the author. Not being a pro I have waltzed down these paths many times by not being able to articulate "why" to management.
The author succinctly explains the 5 traps and how to identify them in a way that is logical and understandable.
January 2, 2019
Easy to understand and logical
Fast read book that cuts to the chase on how to distinguish between correlation and causation. Lee provides real world examples to help the reader understand the basic concepts.
Fast read book that cuts to the chase on how to distinguish between correlation and causation. Lee provides real world examples to help the reader understand the basic concepts.
May 22, 2020
Simple and Straightforward
The perfect book for the non-statistician interpreting statistics and data. Correlation is not causation, and you will learn five reasons why and how to test for them. A well written easy to read book.
The perfect book for the non-statistician interpreting statistics and data. Correlation is not causation, and you will learn five reasons why and how to test for them. A well written easy to read book.
November 22, 2025
excellent information
Shows why what liberals say doesn’t make a connection between variables. Just lies and deception. This the 236th book I’ve read since 2012, mostly on the deception of the drug companies, medical establishment, and our government.
Shows why what liberals say doesn’t make a connection between variables. Just lies and deception. This the 236th book I’ve read since 2012, mostly on the deception of the drug companies, medical establishment, and our government.
May 20, 2018
Readers coming to Lee Baker's Correlation Is Not Causation have probably read the earlier two books in his Bite Size Statistics series and might guess that this third volume would be more of the same. If so, they would probably get a rude introduction to the theme of this volume. For, while the first two books were short, breezy, entertaining reads that did not require any advanced mathematical knowledge, Correlation Is Not Causation may be equally short, but its second half assumes a certain level of knowledge of statistical terms and concepts well beyond that needed to enjoy the first two books.
The first half of Baker's newest book, at least, is reasonably enjoyable. Correlation Is Not Causation is a statistical axiom of which most people are aware at a certain fundamental level, namely that if A and B happen together, then it is wrong to automatically assume that A causes B to happen. Baker provides five alternative theories that might validly explain the correlation between A and B under a particular set of circumstances.
One possibility Baker mentions, which occurs in the real world quite often, is that there is actually a “third cause,” something that causes both A and B to occur. As an example, he mentions a possible correlation between people carrying matches in their pocket and people who develop lung cancer. This might well be true but it certainly doesn’t mean that carrying matches leads to lung cancer (and, thus, that not carrying them would be a preventative). Instead, people who carry matches are more likely to be smokers and smoking (the third cause) definitely contributes to lung cancer.
In the first half of Correlation Is Not Causation, Baker explains in depth five common alternate explanations for correlation and why it is often very important to know the correct explanation. As the matches/cancer example illustrates, scientists often try various treatments for diseases and knowing if a particular treatment is actually effective or merely an example of a particular fallacy is crucial.
Unfortunately for the lay reader, the second half of Correlation Is Not Causation can be somewhat rough sailing. It presumes a knowledge of advanced statistical terms that it makes no attempt to explain. A typical sentence reads: “If you did a univariate analysis between your pair of variables, such as a Pearson correlation, the result actually tells you a lot less than you think it does.” Actually, this sentence tells many readers a lot less than Baker thinks it does.
Since I would assume that most people who pick up Correlation Is Not Causation: (at least Baker offers the book for free) will not be graduate school statistics students performing their own univariate analyses, the second half of this book will probably not appeal to them very much, especially for those who enjoyed Baker’s first two books and thought they were getting more of the same. Further, those same grad school students probably won’t be hunting for a book from the author of Graphs Don’t Lie. The end result is a book that’s solidly in no-man’s land, too complex and uninteresting (at least the second half) but not meaty enough to serve as any sort of text for serious students of the subject. As a free book, there’s enough worthwhile material in the first half of the Correlation Is Not Causation to warrant a mild recommendation, but readers will find little correlation between this book and Baker’s earlier works in the series.
The first half of Baker's newest book, at least, is reasonably enjoyable. Correlation Is Not Causation is a statistical axiom of which most people are aware at a certain fundamental level, namely that if A and B happen together, then it is wrong to automatically assume that A causes B to happen. Baker provides five alternative theories that might validly explain the correlation between A and B under a particular set of circumstances.
One possibility Baker mentions, which occurs in the real world quite often, is that there is actually a “third cause,” something that causes both A and B to occur. As an example, he mentions a possible correlation between people carrying matches in their pocket and people who develop lung cancer. This might well be true but it certainly doesn’t mean that carrying matches leads to lung cancer (and, thus, that not carrying them would be a preventative). Instead, people who carry matches are more likely to be smokers and smoking (the third cause) definitely contributes to lung cancer.
In the first half of Correlation Is Not Causation, Baker explains in depth five common alternate explanations for correlation and why it is often very important to know the correct explanation. As the matches/cancer example illustrates, scientists often try various treatments for diseases and knowing if a particular treatment is actually effective or merely an example of a particular fallacy is crucial.
Unfortunately for the lay reader, the second half of Correlation Is Not Causation can be somewhat rough sailing. It presumes a knowledge of advanced statistical terms that it makes no attempt to explain. A typical sentence reads: “If you did a univariate analysis between your pair of variables, such as a Pearson correlation, the result actually tells you a lot less than you think it does.” Actually, this sentence tells many readers a lot less than Baker thinks it does.
Since I would assume that most people who pick up Correlation Is Not Causation: (at least Baker offers the book for free) will not be graduate school statistics students performing their own univariate analyses, the second half of this book will probably not appeal to them very much, especially for those who enjoyed Baker’s first two books and thought they were getting more of the same. Further, those same grad school students probably won’t be hunting for a book from the author of Graphs Don’t Lie. The end result is a book that’s solidly in no-man’s land, too complex and uninteresting (at least the second half) but not meaty enough to serve as any sort of text for serious students of the subject. As a free book, there’s enough worthwhile material in the first half of the Correlation Is Not Causation to warrant a mild recommendation, but readers will find little correlation between this book and Baker’s earlier works in the series.
August 15, 2018
Knowledge is not the same as wisdom :P
Displaying 1 - 7 of 7 reviews