+ Book on Statistics is a “Sugar-coated Civics Lesson”

other topics: click a “category” or use search box

 Barry Gewen, in the NY Times, reviews “The Numbers Game: The Commonsense Guide to Understanding Numbers in the News, in Politics, and in Life,” by Michel Blastland and Andrew Dilnot.  [Gotham Publishing]

“The Numbers Game” grew out of a popular BBC radio show called “More or Less.”  Their book appeared in Britain two years ago under a different title, and has been “extensively revised” for the American edition, although it apparently still has a somewhat British orientation.

It’s a fact (or is it?) that most people have more than the average number of feet.  Because of amputations, birth defects and the like, the average number of feet per person across the human population is slightly fewer than two.

What about the claim that Republicans enjoy sex more than Democrats do?  Well, the information that matters is that men vote Republican more than women, and men say they enjoy sex more than women say they do. 

The authors observe that most of us expect numbers to do too much.We like their precision.  We want to believe that statistics can tell us all we need to know about the world.

But they say precision comes at a price: before you can count something, you have to define what it is — and that’s not as simple as it sounds.

All those unemployment statistics?  Those conceal a host of decisions.  For example, how much can someone work and still be considered unemployed?  How hard does a person have to be looking for a job?

The Thatcher government changed the definition of “unemployed” either 23 or 27 times.

Another headache: “sampling.”  Most of the numbers needed involve numbers too big to count one by one.  But no matter how sophisticated the statistical techniques, they are still prone to error.

“Uncertainty is a fact of life,” write Blastland and Dilnot, even if it’s a given of human nature to look for meaning where there isn’t any.

They devote an entire chapter to chance in order to explain why the public sometimes sees patterns where there is no such thing.

They mention the villagers of Wishaw, England, who were convinced that a rise in the incidence of cancer in their area was caused by a nearby cellphone tower.  They pulled it down in the dead of night.

Blastland and Dilnot say the villagers didn’t know that “cancer clusters” occur naturally, just as a coin tossed 30 times will probably produce at least one sequence of four heads or four tails.  Throw some rice in the air and you’ll probably see patterns in the way it lands.   

A pattern doesn’t always mean a plan.

Statisticians even have a name for the phenomenon: “The Texas Sharpshooter Fallacy.”  Remember the legendary example of the gunslinger who fires away at the side of the barn and then draws his bull’s- eye around the holes that cluster.

Public misinformation and panic are fed by people who draw bull’s-eyes after the fact: activists with agendas, journalists flogging hot-sounding stories, uninformed “experts.”   There’s also sheer laziness — people who grab a cool statistic without doing the appropriate checking.

What advice do the authors have for us?  They advise us to beware of weaselly phrases like “could be as high as;”  broad comparisons, especially international ones, that aren’t really measuring like against like;  and false correlations.   It may be true that people with small hands read less well than people with big hands.  It’s also true that as we grow up, our hands get bigger.

The authors don’t want us to go so far that we end up distrusting statistics altogether.  Skepticism too easily curdles into cynicism — and that makes us worse citizens, not better.

They always conclude their discussions with the mantra that statistics are an invaluable tool when used properly.  They say that numbers are all we’ve got, often, and ignoring them is  “a dire alternative.”

sole source: Barry Gewen’s article in the NY Times on 2/3/09.  www.nytimes.com

tutoring in Columbus OH:   Adrienne Edwards   614-579-6021  or email   aedwardstutor@columbus.rr.com

Advertisements

Comments are closed.