Naked Statistics: Stripping the Dread from the Data

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 StatisticsGrowing 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 anythingStatistics can be powerful for establishing an argumentHoweverStatistics can also be very easy to misuseThe 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 StatisticsDescription and comparisonInferenceAssessing risk and probabilityIdentifying 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 AccuracyThe 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 -> 11: 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 directionProbability:“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 deterministicThere 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.PollingThe 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 togetherNow 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 anythingStatistics can be powerful for establishing an argumentHoweverStatistics can also be very easy to misuse

Needed for my machine learning study. The concepts are explained in a simple way. Very fun read. I would recommend

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.

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.

Don't but this unless you are fsmiliar with American football or baseball. The book explain statistics concept using American baseball and football. As such I couldn't understand the examples becuase they assume you know the games. Try explaining the concept of 'means' using the number of hatrics scored in the champions league to someone that has no idea what a hatric is or what the champions league is. Good for an American readership but not British

Rather a laboured style, but an easy read on (rather basic) statistical theory. A misery for non-american readers due to the uniform choice of US-specific sports, etc as examples. Quite wearing - he misses the opportunity to use more obvious and more generally understandable examples such as the retail price index. If you are not a US sophomore (qv) then skip this book. There are better.

It's a good book and is written in a very accessible way. As the author mentions, it's a spiritual successor to the 1950s book, How to Lie with Statistics Personally, I found it far too focused on the US. Sports stats were used as examples throughout and I neither know or care about any of the sports mentioned.

This has been sitting on my wishlist for a while and bit the bullet and bought it when it was recommended by a couple of Professors for a "friendly refresher" of the basics of statistics, to accompany more advanced reading.I wasn't expecting much, but Wheelan blew me away with the awesomeness which he squeezed into this book. If it isn't already, this should be on the reading list of every undergraduate at University. Heck, it should be on EVERYONES' reading list! It starts with an introduction to the basics and progresses into more advanced material. All along the way, he explains all of the concepts extremely well and uses examples to get the point across. It's a shame a lot of his examples are Americanized, but you still get the point he's trying to make.As a graduate student, I kept up with this and found it a useful refresher and something which I can revisit to clarify topics in the future on a couple of the more advanced points. But, being honest, this is approachable for anybody with an interest in statistics and its usefulness in everyday life. If you have a really basic knowledge, you might have to re-read some of the concluding chapters once or twice, but Wheelan writes very well so I wouldn't expect this to be much of a chore.Just to conclude, it's a shame that more statistics book aren't like this. He makes a terrifying subject A LOT more approachable and something (SHOCK!) that we can find fun in! Wouldn't hesitate to recommend to anyone.

Naked Statistics is a good way to remind yourself what statistics is about, or if new to the subject, get a solid grasp of the basics. It is a fine complement to a dry textbook, in that it covers the groundwork in a clear, approachable and entertaining way that is not overly mathematically demanding. Appendices delve deeper into theory and can be read or ignored as the reader wishes.The first two thirds of the book is particularly good, breezing competently through key statistical concepts up to and including the Central Limit Theorem.Many people may be drawn to the book because of the growing importance of 'big data'. Wheelan takes this topic on board with a focus on regression analysis, and is not afraid to discuss the pitfalls as well as the benefits of the more abstract 'darker' arts of statistics. However, given the choice between a candid acknowledgements of the fundamental limitations of statistics and an uncomplicated view that 'as long as its done well all will be fine', Wheelan goes in the simpler, more positive direction, even when cheerfully supporting claims that over half of the top-flight peer reviewed scientific papers that draw conclusions from the techniques he proposes are likely to be wrong.Instead, Wheelan argues that brilliant statistical research simply requires brilliant researchers (guess who?) - and that brilliance is not about being good at the maths, but about a having a creative and intuitive grasp of what works. There are two problems with this. One is that observant readers may well spot flaws in the exemplars Wheelan presents as brilliant. The second (and more important) is that the power of statistics is meant to be its ability to reveal insights that are drawn entirely objectively, yet it is clear that many mistakes in statistical research are due to failings in the researchers' subjective and interpretive skills - in other words, the maths disappears - advanced stats is a matter of judgement (so why not rely on judgement and abandon the somewhat bogus claim of objectivity?).Consequently (and slightly disappointingly), Wheelan's concluding chapter is all about the amazing contribution statistics will continue to make to solving the world's most pressing problems, rather than a more reflective assessment of its strengths and weaknesses.All this said, this is a likeable and workmanlike book that treats a potentially dry subject with significant flair.