Customer Reviews

More Damned Lies and Statistics: How Numbers Confuse Public Issues

5 star:		(4)
4 star:		(1)
3 star:		(0)
2 star:		(1)
1 star:		(0)

 

 

I didn't read the first Damned Lies, but the author says that he is making the same points but organizing the information differently for this follow-up. This is one of those books that has the potential to radically alter how we look at numbers. Best shows the reader how a data set can be manipulated to give a desired result.

Best is careful not to single out one side of the political spectrum when making his points. Says the author: "I don't believe that any particular group, faction, or ideology holds a monopoly on poor statistical reasoning." Rather than wallowing in this often-debated territory, the author turns to spheres of academia and the sciences, where radical-sounding results lead to more and more publications and grant dollars. This is a world not seen by most pundits and commentators.

When the issue of school shootings was sensationalized during the late 1990s and early 2000s, accounts in the popular media left out statistics that showed school crime had actually fallen over the past decade. The author calls this omission "missing numbers." Given what looked like a spike in shootings from around 1997 to 2001, few would believe, without seeing those numbers, that there was a clear, growing problem in our schools.

In his chapter, "Confusing Numbers," Best shows how figures can be reported, sometimes in a disingenuous manner, to make them sound better than they are. A good example of this is cited when the author turns to the Bush tax cuts of 2001. The administration claimed that their package would reduce the average family's taxes by over $1,000. Opponents shot back that half of all families would see less $100 of relief. Clearly, this is a case in which averaging wildly lopsided numbers doesn't tell the whole story.

The subject here isn't an exciting one, but given the author's ability to use highly relevant examples and his penchant for fair-mindedness, I was able to work through most of this one. Recommended for those interested in research, public policy, or statistics.

It's always refreshing to read a book in which the author strips away the wrapping around statistical figures to expose what those figures really could mean and how to question their credibility. In this book, as in its predecessor (Damned Lies and Statistics, 2001), the author warns against believing as facts the statistical figures that are always presented to us from various sources - both authoritative and otherwise. The solution is to be critical and to ask questions such as: Who produced those numbers? Exactly what was counted? What are those numbers really saying? Is there a way to present the information in a clearer more objective way? At the end, the author strongly argues in favor of the development of some system in society that would impart, what he calls, statistical literacy in the population at large. The book is clear and well written; it should be widely read.

Few "informational" books read as well as this one. The author disects commonally accepted data and shows how misleading statistics can be. It was an easy read and still the gray matter was stimulated and I put it down smarter than when I had picked it up. This would be a book journalists, kinesiologists, or sociologists would find useful to accompany their already inquisitive mind. I had heard the author on the radio and decided to pick up this book. I was pleased with the purchase and experience.

If you want to realize how you are manipulated by number merchants, then study this book, take notes, copiously underline sections. If however, you want to believe what is spoon fed you from both the right and definitely the left, stay away.

Excellent book, highly recommended. I reference this in my college courses quite often.

Most of us accept data and statistics at face value when we encouinter them. Whether they are presented in a broadcast, print journalism or an academic journal, we generally assume they are valid and accurate. Joel Best explains to us in this highly readable book, written in plain language for the layperson, why this may not necessarily be so. Foremost, he makes the point that all data and statistics are human constructs, elements that exist only because they are created by humans thinking about something, and therefore vulnerable to all the biases and error of which humans are capable. We don't always know exactly why someone chose to measure a particular item and not another, or what biases they bring to the table. And we ourselves bring our biases into play, believing more in numbers that supoport our beliefs or predisposition.

Best guides the reader on how to identify those numbers that are particularly problematic - "magical" numbers; "missing" numbers - for example, so that we are better prepared to accept or reject them. Reading this will put you in a stronger position to evaluate data and information, thereby becoming a more critical consumer of data and statistics-filled information.

Based on the reviews on the cover of this book, and the reviews of others, I bought it expecting something new and original. Boy was I disappointed. Indeed, in the preface the author owns up: "I should confess that, in writing this book, I have done little original research. I have borrowed most of my examples from works by other analysts, mostly social scientists and journalists." Boy howdy, did he ever "borrow" them. So don't look here for anything new. I suspect this was just a quickly dashed together "sequel" to his first book, which I haven't read. So my advice is to skip this one.