What is a P-Value Anyway? 34 Stories to Help You Actually Understand Statistics

by Andrew J. Vickers

What is a p-value Anyway? offers a fun introduction to the fundamental principles of statistics, presenting the essential concepts in thirty-four brief, enjoyable stories. Drawing on his experience as a medical researcher, Vickers blends insightful explanations and humor, with minimal math, to help readers understand and interpret the statistics they read every day. Describing data; Data distributions; Variation of study results: confidence intervals; Hypothesis testing; Regression and decision making; Some common statistical errors, and what they teach us For all readers interested in statistics.

Genres: Mathematics, Nonfiction, Science

224 pages, Paperback

First published January 1, 2009

Reviews

[Suzanne LaPierre · Rating 5 out of 5]

This book is designed for math-phobic folks like me who want (or need) to get a better grasp on statistics. Vickers mitigates the intimidation factor in several ways: use of humor and analogies, short chapters and pithy chunks of information. There is even a single cell cartoon heading off each chapter with a joke that summarizes the subject of that chapter. I had never taken statistics and was able to get a beginner's grasp on the topic using this book. However, I also think this would be a good book for those who want to brush up on the subject, including those who may have taken a class many years ago. The author poses questions on each chapter that may be used in a classroom or study group setting. So if you want to learn more about lead time bias, confidence intervals, errors in regression, and why not to compare p-values- start here.

[Ali Dawood · Rating 5 out of 5]

"Statistics is about people even if you can't see the tears"
لا تكفي الكلمات لوصف الكتاب
لا أري ان هناك كتاب يصلح كمقدمة أفضل منه

[Michael · Rating 4 out of 5]

Very high level but had some really good stories to teach some of the topics in an easy to understand way.

[mirela Darau · Rating 4 out of 5]

It's a funny, intuitive help for laymen. It's also a place for me, to ponder upon creative ideas for statistics' class for non-math students.
I particularly liked some questions from the discussion sections.

Samples from the book:

* "As you probably know, the hassle of organizing something rises exponentially with the number of people you have to organize, following the famous equation E=mc^2 where E is effort, m is the mean fussiness (or flakiness) of you friends, and c is the size of the crowd.

[Jennae · Rating 4 out of 5]

One of the best textbooks I have read. It made statistics fun and entertaining!

[Jen · Rating 5 out of 5]

If you’re like me you don’t think about statistics daily and avoid it when possible. however this book contains many examples of how statistics (and hypothesis testing) can be done without a major study being involved and fit into worldly questions you have every day. While I don’t think this is a casual read for anyone not involved in statistics or an actual math hater .... it is a quick almost fun read (because of the real world examples) for those trying to understand descriptive statistics.

[John · Rating 3 out of 5]

This is your corny uncle trying to talk about statistics, but he is also a real statistician.

The first time I tried to read this book I was put off by the tone but this time I made it through and by the end I thought it was very cute.

In terms of statistics I do feel like I have a slightly better intuition for things now. I would still love to actually work through some proofs etc to really understand some stuff but I feel like this has given me a good context in which to do that.

[Evan · Rating 3 out of 5]

This book is a slim, ambitious volume that would, I think, be excellent after some substantial revisions. However, I think the current version tries to teach a little too much in too little space, and I felt that my time reading it wasn't well spent as a result.

I think that writing an introductory stats textbook is entirely honorable. However, doing statistics involves doing the math, even if it's hard, or takes time to understand, or takes multiple examples. So in Vickers' many examples, he'd be helped by showing more (all) of his work, rather than just saying, "If you're interested, the p-value is 0.12."

Here's an example, p. 147, in which Vickers' criticizes a statistics-ignorant surgeon: "I tried each analysis on the statistical software drop-down menus, and that was the one that gave the smallest p-value." And then here's Vickers himself, say on p. 75: "When I plug these numbers into a sample size formula (using the typical alpha of 5% and a power of 90%), I get a total sample size of 44." What is this formula? What is alpha? What is power? Why those numbers? It's probably that Vickers mentioned them elsewhere, but this is NOT a stats textbook. It's an explanatory text, and if Vickers has to repeat himself every chapter to make his point, he should, because that's what his book's cover claims to do. Throughout the book, Vickers repeatedly does exactly what he tells us not to do.

Similarly, Vickers has done some math. It's fine if that's not in the chapter, but it should be in an appendix. After reading his chapters on power laws, I still don't know how to use his "sample size formula" to calculate a sample size, even though Vickers brags about doing it in his head at parties.

This book should have a calculation of p-values in every chapter that matters. Show the math for confidence intervals. Explain what they mean every time. Tell us why frequentist statistics is better than Bayesian statistics. Show both approaches for the same problem. Use a big data example with a sample size of 100 million. Tell me what log odds are; don't just mention them in Discussion section 19.3.

In the end, this book is worth skimming and highlighting. If you're looking for the math, you won't find it. For me, this didn't help me actually understand statistics, either.

[Venkatesh-Prasad · Rating 2 out of 5]

A book attempts to provide a lighthearted introduction to statistics. While the book covers most of the basic statistical concepts, it does not provide any formula; not even calculations. This can make the book a more involved read for readers not familiar with statistics. On the other hand, if the reader has some rudimentary knowledge of statistics, then the book helps reinforce the concepts with examples.

I suspect there may be other better introductory books.

[Vasiliskrin · Rating 4 out of 5]

This is a great book for people with some prior statistics knowledge, as it helps clarify some fundamental things about statistics. It is a good complementary textbook to introductory statistics books. I enjoyed the author's writing style and the examples he used to convey his purpose.

[Jon · Rating 5 out of 5]

Dr Vickers does a great job of making statistics much more accessible. His focus on the general and not getting bogged down with formulas is a welcome relief for someone looking to get the “big picture”. I highly recommend this to teachers and students alike.

[BradMD]

The author has a band. He does prostate cancer medical papers with others. I think he was the inventor of a decision curve analysis. "Relative risks are well known to have little value for making decisions." - page 97.

[Andrej · Rating 2 out of 5]

Básico y ameno, una lectura que introduce pequeñísimos elementos sobre estadística y probabilidad. No es un texto que recomiende.

[Rhushana tyrell · Rating 5 out of 5]

Idiot

[Jacob Roberts · Rating 3 out of 5]

A fine book that gives students another way to understand this tricky subject.

[Frank David · Rating 5 out of 5]

Highly readable, math-lite stats intro that is highly relevant readers of biotech clinical trial results.

[Victorlohk]

read chin. ver.

[Anna Amato · Rating 1 out of 5]

did i only read like one chapter? …yes but im counting it 😃

[Unwisely · Rating 3 out of 5]

I got asked to analyze some data at work. While I had statistics in college, I haven't done anything with it since, and it's, *ahem*, been a while.

I was digging around online, but finally gave up and turned to a professor I know who tries to teach math to poli sci students. After he gently explained that my "double-humped" data was "bimodal" to grownups, I asked for maybe something to read about this? He liked this book (and another one the library didn't have.)

I was not expecting it to be as thin as it is - more like magazine thickness. So it's not a huge commitment, and is *extremely* readable. Also well chunked - you don't have to do it all at once and if you aren't it tells you what to go back and read. All points in its favor.

But, man, statistics. I read the p-value chapter at least three separate times over the week I was reading through this book and if you asked me for a definition, I would still stumble (but I'd remember about the toothbrush, so I guess it wasn't a total failure).

So I think I can now go back and ask better questions, which is what I needed (but it's not comprehensive by any means!)

[Dale · Rating 4 out of 5]

This would serve as a great adjunct to an introductory or second semester course in statistics. You won't learn much (if anything) about specific statistical tests, and you won't learn any of the mathematics underlying modern statistics from this book. What you will learn is a set of ways to think about what you (should) hope to achieve from applying statistical methods to a problem. Vickers clearly believes, with reason, that statistics can be used to help us understand the world, and by understanding, improve it. So he wants to root out silly, misapplied statistics, and get statisticians and the rest of us, to do statistics right. "One hypothesis, one p-value" could almost serve as the dedication of this book. He wants us to think deeply about the problems we are working on, and distill the solutions to those problems down to specific testable hypotheses, and to then conduct the surveys, studies, or experiments that will let us evaluate the hypotheses. This is not earth-shatteringly new, but Vickers provides lots of examples where this isn't done, even (or maybe especially) in peer-reviewed publications.

[Chris · Rating 4 out of 5]

This books helps drive home basic statistics concepts but won't be a big help in actually performing statistical analysis. The book contains maybe five math equations, and some may appreciate reading about statistics without needing to digest or skip over math everywhere. However, this means the book probably will not be a helpful resource if you are trying to use statistics yourself. Instead, it's value is more for those who are trying to analyze statistics used by others (including the ability to spot misuse) and for understanding the basics of how the scientific method is applied to statistics, i.e., what questions statistics can answer and how it answers them. If you want to be able to understand the value (or lack thereof) of the statistics you read, I would recommend reading this and then passing it on to someone else. I wouldn't keep this as a reference book, and if I wanted to do statistics myself I would find a different book to use as my guide (although this one would still be helpful for study design).

[Paul · Rating 3 out of 5]

Excellent book as a compliment to beginner course-work or other texts, but not a good source in and of itself. It won't explain the "how to" of statistics, but will give you a valuable "why the #%?! am I learning this stuff" which other sources usually skip in favor of just the facts. It's presented with a good sense of humor, easy to read and digest, and full of great, short tips that are easy to remember and recall when you finally get to use statistics in a more practical, less theoretical/ academic setting.

[Craig Citro · Rating 2 out of 5]

This was a nice idea for a book, but it didn't really deliver. It tried to be a nontechnical book about technical content, but unfortunately ended up being useful to neither audience. I was expecting that the second half of the book would unfold into some equations and exercises, but I was sorely mistaken. I personally appreciated the geeky humor, but I'm sure it would only help to scare off just those readers he's ostensibly trying to draw in.

[Phiphy · Rating 5 out of 5]

It's a fantastic book about the basic concepts in statistics. As a first-step introduction, it does a good job on giving meanings to the work that statisticians are doing. Even if you've already learned some statistics, it still helps you to clarify the conceptual misunderstandings in the maths. I really recommend it to anyone who is interested in data analysis and statistics.

[Christian Crowley]

Another selection in my quest for accessible, intuitive statistics instruction.

[Vadim · Rating 4 out of 5]

Книга поможет понять и запомнить основы статистики. Заниматься не поможет, но большинству из нас это и не надо.

[Sergey Alexeev · Rating 3 out of 5]

Won't really help to understand statistics, but can arouse interest to a subject.

[Mills College Library]

519.5 V637 2010

[Carol · Rating 5 out of 5]

Good information in an interesting format. Great real-world examples.

[Larry · Rating 4 out of 5]

does help put Statistics into the real world.