Naked Statistics

This book serves such an incredibly important purpose in that it really explains the fundamentals of statistics in a way that can be appreciated and understood by the vast majority of reasonably educated laymen. Of course, that in itself begs the question: so what? What about statistics is so "important" that the average person should take notice? Well, for a lot of reasons that Mr. Wheelan explains so well its hard to add anything to it, but I'll give one brief attempt for those of you bothering to read the review: At its core, statistics is about organizing data in a meaningful way such that we can draw accurate conclusions about our environment...in other words, statistics can help us make sense of the world around us in a very meaningful and practical way. Statistics is really just a mathematically formalized form of what our brains do intuitively, and without that core concept of filtering and making sense of the information we're exposed to we would all be paralyzed - unable to make associations, draw conclusions, or derive any meaning from our environment. Learning would be impossible, life would be meaningless, and existence and survival itself would be impossible. Statistics takes that core function and focuses it in very particular ways, which if done correctly, can yield amazing results that add worthwhile knowledge and positivity to life. To quote the book, "Statistics is the most powerful tool we have for using information to some meaningful end." Further, if that description doesn't at least somewhat tempt you how about this: Even if you don't care about statistics, those that have enormous influence on your life and happiness (e.g. businesses, politicians, government organizations, media, and everybody else that has a vested interested in influencing your mindset, perspectives, and behaviors) very much care about statistics and use them, for both good and ill, to "help" you see and do things a certain way. Knowledge of statistics can help transform you from "propagandized" to "informed". This is one of my favorite aspects of the book, that is, the author acknowledges and recognizes the danger of inappropriately used statistics; he even conceived the book as an homage to How to Lie with Statistics: 1st (First) Edition , and as he very correctly points out right in the introduction of the book, "Statistics is like a high-caliber weapon: helpful when used correctly and potentially disastrous in the wrong hands." Amazingly, this book is a fun and enjoyable read. No seriously. Mr. Wheelan exposes himself as a fantastic teacher, bringing an enthusiasm and energy to the subject coupled with a gift of explaining ideas and concepts in a way that most (not just "stat people") can understand and appreciate. Although this isn't a "textbook" per se, I think it remains relevant for the whole spectrum of people from total "newbies" all the way to hard-core statistics researchers. Why? Fundamentals. The author explains the fundamentals in a clear concise way using practical, relevant, and familiar examples from our everyday lives. For those new to the subject you'll get an "intro to statistics" class from a very gifted Ivy-league professor that will give you a solid well-rooted realistic foundation that you can both apply to your everyday life as well as prepare you for higher level statistics courses if you so choose. For those already well-versed in statistics or professionals, it will reacquaint you with why what you do matters as well as provide you a wonderful way to explain to others why what you do matters. Personally, I'm somewhere in the middle in that I've had a reasonably high level statistics background (relative to the average college graduate, anyway) and as I read the book I kept thinking to myself, "Man, if only I had had this before taking my first stats course!" Outstanding book.

How many times have you wondered how two different people looking at the same report managed to draw totally opposite conclusions? How many times have you looked at a set of data yourself and decided that the data must be wrong because it doesn't show what you "know" to be true? Well, I suspect that you aren't alone. I suspect that you also have an aversion to "doing the math" necessary to convince yourself that the data is actually correct. Wheelan claims that "Statistics is like a high-caliber weapon: helpful when used correctly and potentially disastrous in the wrong hands." This is an incredibly important topic as we move forward into our increasingly connected world where we are creating data at an accelerating rate. The data is only raw knowledge, how we analyze that data is what allows us to derive useful information to create actionable knowledge. The tools we use to analyze the data is, like it or not, statistics. Wheelan gives us an easy to follow and understand overview of statistics. Even if you didn't like math in school, you will find this book useful as well as insightful. Here's the rub. No matter how hard we try, when using descriptive statistics, we lose the detail in the data. We've "summarized" it and just like the summary of a good book, we will miss a lot of the nuance in the story. Neither this review nor the summary of data known as descriptive statistics is an exception to that rule. And because of that loss of fidelity, we will find as Wheelan puts its, "Smart and honest people will often disagree about what the data are trying to tell us." Wheelan includes many examples of how statistics is enlightening and infuriating. The examples are from everyday life and you are likely to recognize them; things like "Money Ball" and the "Wall Street failure." You will learn about correlation, basic probability, the Central Limit Theorem, Inferences, Polling, Regression Analysis and more. All without tears! So why should you be interested? Because the big questions will be answered by statistical analysis and you will want to know at least enough to not be "taken in" by those who would purposely try and confuse you or convert you to their views using statistics and data. Wheelan believes that there are five "large" questions that statistics will help us answer in the future. 1. What is the future of football? Will it die of concussions? 2. What (if anything) is causing the dramatic rise in the incidence of autism? 3. How can we identify and reward good teachers and schools? 4. What are the best tools for fighting global poverty? 5. Who gets to know what about you? Wheelan doesn't answer these questions. The data is still being collected and analyzed. What he has done, however, is set the reader on the right path to understanding the complexity of those questions and why there aren't any easy answers. He has given the reader at least a working knowledge of statistics so that we can decide for ourselves who is doing the best job of presenting the information and knowledge buried in ever expanding universe of dots for us to connect. The sub title of this excellent book is "Stripping the dread from the data." What I found is that now I know how to make sure that I strip the dread from the statistics which applies to the data. The dread is when the proper use of statistics on that data shows my cherished beliefs to be ill conceived!

The rise of social media we have massive amounts of data at our fingertips. What we do with all this information is up to us. Have you ever have logged on to face the only to discover the friend who has a post with one single data point and uses that as the basis of their argument? It’s perfectly acceptable to take a stance on a hot topic but you’ll come out looking foolish if you don’t understand the statistics and data points behind your argument. Don’t be that person. In comes Naked Statistics Growing up, I was never fond of mathematics. I struggled in high school with a solid 65 average (that is borderline failing in the United States), and a long believed that now just wasn’t for me. It was until recently ago that I discovered I really do love numbers and specifically statistics. It’s around this all the time whether through social media or my favorite thing, gaming. Looking back, I think educators were not ever explaining to me in a way that grabbed my attention that made me want to further my math education. The blame can’t lie solely on the educators I grew up with, so I recently took it into my own hands to further my own understanding of math. I stumbled across the audio book Naked Statistics by Charles Wheelan, a professor at Dartmouth college. I found the narration to the enticing and upon finishing the audio book actually bought the physical copy and marked it up heavily with notes of my own. Throughout the book the author gives great detail about how statistics can truly improve our lives while the same time helping readers understand the power that comes with it and how the power can be used and negative ways as well. Now there are a few key points I want to highlight about the book: Statistics rarely offer a single “right” way of doing anything Statistics can be powerful for establishing an argument However Statistics can also be very easy to misuse The book opens with an explanation of the classic Monty Hall problem, which is a brain teaser in the form of a probability puzzle loosely based on the 1980’s television show “Let’s Make a Deal”. I’d never quite comprehend it the full detail of this problem until this book. I encourage you to look into unknown because not only to find to be interesting but also provided a great starting point for understanding for help powerful statistics can truly be. Uses of Statistics Description and comparison Inference Assessing risk and probability Identifying important relationships (correlation) Descriptive Statistics “Who was the best baseball player of all time?” Descriptive statistics can be technically accurate yet pretty misleading. The author explains that “The good news is that these descriptive statistics gives us a manageable and meaningful summary of the underlying phenomenon,” and continues with “the bad news is that any simplification invites abuse.” Wheelan highlights the Gini Index, which is a measure of statistical dispersion intended to represent the income distribution of the nation’s residents and most commonly used to measure inequality. The problem however is that this is a mere useless stat. It simply takes the per capita income of a nation and divides it by the number of people. But who was actually earning how much of that income? That is the key missing piece. Precision vs Accuracy The author goes on to describe the differences between precision and accuracy. Precision: “The exactitude with which we can express something” Accuracy: “Measure of whether a figure is broadly consistent with the truth” Correlation “Measures the degree to which two phenomena are related to one another. Correlation coefficient: Expressed on a scale of -1 -> 1 1: every change in one variable is associated with an equivalent change in the other variable in the same direction -1: every change in one variable is associated with an equivalent change in the other variable in the opposite direction Probability: “Study of events in outcomes including elements of uncertainty.” There were a few key takeaways from this chapter: Don’t play Lotto (Pg. 80) You should invest in the stock market if you have a long-term investment horizon (because stocks typically have the best long-term returns) Probability is not deterministic There are several problems with probability, however. The 2008 Wall Street crash and financial crisis largely had to do with computers misusing probability. The VaR model, or Value at Risk, which was called “potentially catastrophic”. Value at risk is a measure of risk in investments, and estimates how much a set of investments might lose given normal market conditions and a set time period, such as any day. This model offered 99% reassurance about 1% is the difference between success and failure. The importance of data (Garbage in and out) When working with numbers you assume that you retrieve and good data, but no amount of fancy analysis can make up for fundamentally flawed data. Selection bias: “How have you chosen the samples you are evaluating?” Self-selection bias: “Will arise whenever individuals volunteer to be any treatment group.” Publication bias: “positive findings are more likely to be published” This distorts the research that we see, or more evidently, do not see. This pious comes from not directly his studies themselves but from the skewed information that reaches the public. Typically, medical journals now require that any study be registered at the beginning of the project to help combat this problem of only publishing information that researchers want to see. Inference “The process by which data speaks to us enabling us to draw meaningful conclusions.” The typical tool that researchers will use is a hypothesis, which is a proposed explanation made on the basis of limited evidence as a starting point for further investigation. While statistical inference offers many benefits it also comes with some significant pitfalls. Notably, statistical inference is based on a probability, And not some sort of absolute certainty. Polling The power polling comes also comes from the central limit theorem, similar to other forms of sampling. Where it differs however, is that we don’t care about the mean, but we actually care about the percentage, or proportion relative to our results. But again, this can be powerful for example the poll of 1000 individuals can offer meaningful insight into the attitudes of an entire nation. Regression Analysis “Allows us to quantify the relationship between a particular variable and how come we care about, while controlling for other factors.” It’s important to note there are many pitfalls of regression analysis, and is the hydrogen bomb of the statistical arsenal! Most of the studies that you read about in the newspaper are based on regression analysis. Bringing it all together Now that you’re more informed, consider these aspects of statistics the next time you stumble across an infographic on social media which illustrates only one point, or discover an article which takes a stand on a topic. I never realized how large of a role statistics plays in my life, but the author’s enthusiasm and clear instruction helped me to not only appreciate it, but has also further sparked my interest. If you have other books you’d like to recommend regarding math, business, or computer science, I’d love to hear about them. Key Takeaways: Statistics rarely offer a single “right” way of doing anything Statistics can be powerful for establishing an argument However Statistics can also be very easy to misuse

For lay people interested in better understand the world of data and statistics, this book and Tim Harford's "The Data Dectective" are the two books I most highly recommend. I first read this while in undergrad. Now with a few years of professional work, and a nearly completed grad program, I decided to re-read Naked Statistics. This book was one of the things that sparked my interest in statistics and data in undergrad. Now with more experience and expertise, I realize just how thorough yet succinct the text is. Wheelan writes in a very informal way that I think is approachable for a wide audience. It's certainly not a dry read like all my beloved textbooks. I particularly love how he focuses on intuition and non-technical explanations first then dives into more nitty-gritty details after he's shown how non-intimidating and intuitive the statistical idea really is. I think this approach will prevent readers from getting turned off when the mathematical details are explained (though those are also simplified significantly for lay readers). I plan on teaching data and statistics soon at my local technical college, and will use this book as my outline for an intro to stats class and make it required reading. The combination of detail and approachable content is perfect to educate without turning students off the subject. Of course more advanced texts are needed to learn how to practice statistics in the real world, but this is a good addition to those more rigorous texts.

Naked Statistics is very well done. This book gives you an entry look into the world of statistics. If you have ever been curious about how statistics work, and never took a class, this book would be good for you. Charles Wheelan talks about what statistics does and what it looks at, but doesn’t get in to how it works. He really simplifies the topic to make it easy for anyone to understand. The book is written from a perspective that the audience hates statistics and has never wanted to look into it at all. Wheelan does a great job in addressing that, and he tries to make the process less painful. His examples are off the wall and make reading the book entertaining. This book doesn’t really teach you statistics, but tries to peak your curiosity, so you do continue on your learning journey. He doesn’t go into all of the pesky details of why the math works, but he does have appendixes and sources of where he gathers all of his sources for each section of statistics he goes over. I really liked how Wheelan explained how different groups can mess with data depending on how it is interpreted or how people can skew the methodology to produce certain results. He has a great explanation of how people lie with statistics or distort confidence in results.

I have been reading this book and I like it so far, but there are 2 things that bothers me a lot: 1) A lot of sport metaphors which is fine in general, but the problem is they are based on american football and baseball terms. That is absolutely OK for a reader born in USA, but let me say that a lot of people across the ocean do not know much about it! I am sure that there are more broadly respected sports, i.e. soccer. Just a little bit more of case about international readers... 2) When it comes down to inference, I found that the following statement in chapter 9: ==== You may recall from the last chapter that we can create a standard error for each of our samples: S/sqrt(N) where S is the standard deviation of the sample and N is the number of observations. ==== is very confusing because chapter 8 says: ==== If you’ve followed the logic this far, then the formula for the standard error follows naturally: SE = S/sqrt(N) where S is the standard deviation of the population from which the sample is drawn, and N is the size of the sample. ==== How did we replace the standard deviation of the *population* with the standard deviation of the *sample* so easily? I didn't not find any explanation of that..I guess that it's because the sample's standard deviation is "similar" to the population it was drawn from, but does is it enough to change the formula so drastically?

Statistics can be really interesting, and most of it isn't that difficult, according to Charles Wheelan in this book. Statistics can be used to explain everything from DNA testing to the idiocy of playing the lottery; statistics can help us identify the factors associated with diseases like cancer and heart disease; it can even help us spot cheating on standardised tests. The book goes on to discuss a range of basic statistical concepts including correlation, basic probability, standard deviation, the central limit theorem, sampling errors, and regression analysis. Each concept is introduced and explained with interesting and amusing stories having very little technical mathematical content, and several chapters have appendices containing slightly more difficult explanations. If made-up examples reflect the maker's mind patterns and prejudices, perhaps the author has something to answer for. Busloads of International Festival of Sausage participants and international marathon runners seem frequently to be involved in crashes, and all of them are either concussed or unable to speak English, so that we have to use mean weights and standard deviations to determine whether a busload contains sausage-eaters or runners. The is a witty and enjoyable introduction to basic statistics. Most readers will already have at least some knowledge of probability and statistics, but the author's entertaining style keeps the reader engaged while the really simple stuff is explained. Those who were hoping for a text book with plenty of Greek symbols will be largely disappointed; this book is firmly aimed at non-mathematicians with almost no previous knowledge of statistics.

I read this book because it was added to the FSOT recommended reading list this year and I found it delightful. This is the third book I have read on statistics this year (which probably says something about me) and I believe it is the best. The author makes good use of humor, which usually hits, to introduce the basic concepts of stats and does all the way through linear regression and T-tests. This book will even book good for college students in introductory statistics courses who need a break from the regular textbook. After reading this, I am tempted to ready the author’s similar book on economics