Customer Reviews

Damned Lies and Statistics: Untangling Numbers from the Media, Politicians, and Activists

5 star:		(16)
4 star:		(8)
3 star:		(2)
2 star:		(6)
1 star:		(0)

 

 

Are ten percent of Americans gay? Is the white male in the work force rapidly becoming a minority? Are 150,000 young American women dying each year from anorexia?

Joel Best clearly answers "no" to each of these three questions and, more importantly, shows why many people would say "yes". His point is that descriptive statistics are the product of a social activity, not just a representation of society. Social advocacy causes people to collect the data that they feel will best support their preconceived notions: They talk to unrepresentative groups. They start to collect new measures and then wonder why the "statistics" have grown since ten years earlier (when they weren't much -- if at all -- measured). They multiply erroneous assumptions. They mutate data. And the press and other publications carry the mutations forward.

This book offers plenty of illustrations of intentions run amok. Many of the reports provide useful information for a classroom lecture on the need to discern if a person is "speaking rot", as Harold Macmillan once said was the primary purpose of an education.

A good, crisp 171 pages in length, it is absent discussion of the more difficult inferential statistics and, as a result, it is easy to understand by the lay person.

This is a book about reading and understanding statistics. It is not a book on research methods. As a book that helps to analyze and think critically about statistics, however, it is a book on methodology: the critical comparison of method issues.

Best’s point is a central issue in modern industrial democracy. If we are going to make effective policy choices as citizens and voters, we must understand the issues on which we make decisions. The same holds true for the decisions we make in business life and in research. Many of the choices we make are based on statistical evidence. To make informed choices, therefore, we must be able to think about statistics.

A quick summary of the issues and topics in this book offers a good overview of clear thinking on statistical issues. Chapter 1, “the importance of social statistics,” explains where statistics come from, how we use them, and why they are important. Chapter 2, “soft facts,” discusses sources of bad statistics. Guessing, poor definitions, poor measures, and bad samples are the primary sources of based statistics. Good statistics require good data; clear, reasonable definitions; clear, reasonable measures; and appropriate samples.

Chapter 3 catalogues “mutant statistics,” the methods for mangling numbers. Most of these arise from violating the four requirements of good statistics, but a new problem arises here. Where is relatively easy to spot bad statistics, mutant statistics require a second level of understanding. As statistics mutate, they take on a history, and it becomes necessary to unravel the history to understand just how - and why - they are mutant. Transformation, confusion, and compound errors create chains of based statistics that become difficult to trace and categorize.

Chapter 4, “apples and oranges,” describes the dangers of inappropriate comparison. Dangers arise when comparisons over time involve changing and unchanging measures, and projections. Comparison among places and groups lead to problems not merely in the data measured, but in the ways that data may be gathered and collated. Comparison among social problems also creates unique difficulties. Best offers logic of comparison to help the reader understand how to make sense of good comparison and bad.

Chapter 5, “stat wars,” describes the problems that arise when advocates use questionable numbers to make a case. Chapter 6, “thinking about social statistics,” sums up Best’s advice on understanding statistics. Don’t be awestruck in the face of numbers, and don’t be cynical about them, he suggests. Be critical and thoughtful.

This book is recommended for every non-statistical researcher who is required to make some use of statistical results in his or her work. It will be especially helpful for those designers who belong to the 2% of the population that one study identifies as victims of UFO abduction. ....

Joel Best writes a book about statistics for everyman. He does it without using confusing bell curves, without charts of T1 -T2 - T3 standard distributions, and without the i.e. "greater than-lesser than" hieroglyphics one finds in statistical textbooks. This is a book that should accompany a short course given every year of schooling from 7th grade through the end of high school. Think of how difficult this would make it for manipulators of every kind i.e. social researchers, social activists, big corporations e.g. tobacco companies, government agencies, the major media, charities, big labor unions, congress, the White house, who all share something in common; an agenda with which they seek to manipulate the common man.

A democracy only works if it has an educated citizenry, and the understanding of the manipulation of statistics, in a society such as ours, probably the most complex in the history of mankind, is essential. The author doesn't try to overwhelm the reader with the many nuances of statistical research and evaluation, he instead implores him or her to rise above being awestruck, naive or cynical about the numbers. He implores us all to engage in critical analysis, in critical thinking. He uses many examples of statistics that are obviously incorrect and tells us how to look behind their numbers and their subsequent conclusions.

It doesn't take long to read, and it should be required reading for all of us who vote in our myriad elections. "If it were to be the case the world would be a better place", and from a practical standpoint the Florida post election fiasco would have been resolved sooner. And, social security would be reformed to the benefit of the bottom half of income tax payers.

Learning how statistics are compiled, manipulated, and used is crucial to keeping any society on an even keel. Let's hear it for Joel Best and let's hope he writes his next work as a metaphor. Maybe Disney will pick it up so the message will reach, and reinforce the children

How many times have we heard that 25% of women on American college campuses have been raped? Or, that 130,000 young women die each year from anorexia? Or, that most medical research has customarily been performed on behalf of men?

Joel Best, a professor at the University of Delaware, has written a highly readable treatise on statistics, and how we can become better consumers of the statistical information that permeates the environment in which we live. Not only does he share some egregious examples of the misuse of statistics (as those described in the above paragraph), but he also explains how to become more discerning about statistics as they are (ab)used by partisans of various causes.

This book is especially timely in the wake of the furor that erupted at a University of California campus recently when the Independent Women's Forum took out an ad in the student newspaper declaring "Take Back the Campus." The ad was critical of several statistics that are used by advocacy groups to distort the facts about male college students in their relations with women. One is the "25% of women ... raped" statistic cited above. Rape is a serious crime, that is universally abhorred, but a review of the study used to establish the 25% figure showed that the figure was misleading at best. Professor Best gives a thorough evaluation of this situation, leading to a conclusion that the actual figure is most likely less than 3%.

You will not find a better book on how to read statistics and understand their implications. I strongly those who want to know how to discover the truth about important issues of our day to read this book.

This is a short book which could be shorter without compromising its message. The author explains in very easily understandable terms the perils of misinterpreting public presentations of statistical data. Perhaps he was trying to keep it simple, but I felt that he repeated himself too often. I would have appreciated more examples of actual cases of flawed statistical reporting. It would have helped his case to provide some examples of good uses of statistics. With the author, I wish the media, as the willing or unwitting conveyors of misleading statistics, were more mindful of the flawed information they present to the public.

If you have no backround in statistics and you tend to believe the numbers you see in the newspapers or on TV, then this book is a must for you. It offers a wealth of examples and explanations. (I would give it 5 stars for such readers.) On the other hand if you are already skeptical and have some knowledge of statistical methods you may find this book slow reading. Still it may have some value for you because it tells the story behind some of the frequently cited statistics that is interesting to read even if you know the statistics cannot be right. (3 stars.)

Statistics are not simple to grasp. Joel Best (a sociologist, past editor of the journal Social Problems) has done a great service by providing the most basic introduction to the topic, indicating how to proceed in trying to make sense of the statistics we encounter every day. I will use it as a supplemental text for my Sociological Research course in the Fall of 2002, and I'm already using examples in my Sociology 1010 (intro) course.

Best frames the common views of statistics with a 3-part typology -- Naive, Cynical and Critical. Very similar to the public's reception of mass media news in general it seems to me, Best sees the Naive and Cynical views as only superficially opposite, while in reality more likely to overlap and reinforce one another. (Many people will say they don't believe what they hear on TV, but if they lack any other sources of information, and don't critically evaluate what they do hear, they are likely to absorb and believe much of it regardless.) Best's goal is to educate citizens in the capacity to be Critical of statistics, that is, to start them up a learning curve of questioning, analyzing and judging statistics and claims made on the basis of statistics.

How is it, I wonder, that so many first-year college students are seemingly clueless when it comes to statistics? A basic social statistics course would be far more valuable than much of whatever else it is that is taught in high schools these days!

This is a topic that has a lot of promise and one can do so much with it. However, if you're looking for anything more than just the basics (with a few good examples from social statistics), don't bother with this book. This book starts out by identifying the reasons why statistics get distorted, but then after every 20 pages or so this book reads like the script of groundhog day - the same three points (innumeracy, people with a vested interest and apples and oranges comparisons) are referenced again and again. The topic is fascinating but there is very little substance to keep you occupied for more than 30 minutes.

This book delves into a few of the difficulties an individual would face if he or she tried to get an accurate count of and ultimately make conclusions about a phenomenon. Best gives the reader plenty of examples from the social sciences which demonstrate some of the problems encountered when trying to describe, measure, and ultimately make conclusions about various issues.

Best adequately picks up the subject matter just before Darrell Huff's timeless text, How to Lie with Statistics, begins. Like Huff, Best argues that people have incentives to put numbers in front of us (Huff refers to this as axe-grinding), and it behooves us to know who is putting the statistic in front of us, why they chose to put this statistic before us, and most importantly, just how they derived this statistic.

Because proponents (and opponents) of an issue, whom Best describes as 'Advocates', can control the way a statistic is generated and presented, we must closely scrutinize the numbers before us so that we can 'untangle the few facts from the various fictions'. Toward this end, the book gives the reader some very helpful questions he or she can ask when attempting to interpret a descriptive (summary) statistic.

The practical utility of Best's text is not solely limited to issues in the social sciences. One could easily apply Best's argument to, for example, the ongoing environmental debate. As such, this book is a vital component in developing critical thinking skills that can be applied in all areas both personal and professional.

Finally, although this text focuses exclusively on descriptive statistics, limiting itself to contentious and controversial topics in the social sciences, readers should take note that Mr. Best is not presenting a truly original and comprehensive treatment of the subject matter. For example, David S. Moore, author of the text Statistics, Concepts and Controversies, provides a concise, yet rigorous, entertaining and accessible treatment of the same subject matter- including a wider range of examples culled from the fields of education, social and medical sciences, all in the first fifty pages of his text.

"There are lies, damned lies, and statistics," everyone knows Mark Twain said. But maybe it isn't really that clear; in his autobiography, Mark Twain includes the quip, but attributes it to Disraeli. In fact, there is a lot about statistics that is not clear, and Joel Best, a professor of sociology and criminal justice at the University of Delaware, wants to clarify it for us. In _Damned Lies and Statistics: Untangling Numbers from the Media, Politicians, and Activists_ (University of California Press), he addresses the statistically impaired, which all of us are, more or less. We can judge statistics better, and he has written clearly and enthusiastically about how to do so.

The problem is that we are ready to treat statistics as facts. The media like to report them, for they are regarded as little nuggets of truth. The public tends to be receptive of numerical claims, and indeed, some statistics are true, if limited, and useful. But plenty of statistics are bad statistics. It is reasonable, when seeing a statistical claim, to ask who made it and why. An activist group may select a statistic that highlights a social problem, while a government agency may select one that shows there is no need to worry. No matter the origin of the statistic, how it was created needs to be critically evaluated.

Best's examples are great fun. Thirty years ago, no one had heard of the crime of stalking, which loomed into importance in the 1990s. There was not a Stalker Information Bureau compiling statistics on the issue, so when an initial news report said that "researchers suggest that up to 200,000 people exhibit a stalker's traits," it was not really an arguable number; but it was a number, a nugget, a "fact." Then a television talk show declared that "there are an estimated 200,000 stalkers in the United States, and those are only the ones we have track of." Then _Cosmopolitan_ modified it upwards to "Some two hundred thousand people in the U.S. pursue the famous. No one knows how many people stalk the rest of us, but the figure is probably higher." A statistic has a life of its own, and mutates, just like any story passed by word of mouth, or nowadays on the internet. Activists calling attention to the problem of anorexia estimated that 150,000 American women were anorexic, and noted that anorexia could lead to death. Sure enough, reports based on this began to say that 150,000 women per year were dying of anorexia (the correct figure is about 70).

_Damned Lies and Statistics_ will inevitably invite comparison to the classic _How to Lie with Statistics_, but it concentrates more on the social value and advocative use of statistics, as well as how to start to tell the good ones from the bad ones. It is an accessible and amusing book which counsels us to be neither naïve about statistics nor cynical, and helps us steer through the numbers so many groups ceaselessly throw at us.