Saturday, July 05, 2008

Damned Statistics

"The mathematics of the collection, organization, and interpretation of numerical data, especially the analysis of population characteristics by inference from sampling." - American Heritage Dictionary

There it is in black and white. "...analysis of population characteristics by inference from sampling." A mathematician knows this. A statistician knows this. The problem is that much of the public don't seem to understand this.

Here are a few interesting statistics to discuss before your next bike ride...
  • The average person eats 8 spiders while asleep in their lifetime.
  • Women over 35 have a .5% chance of getting married.
  • About 11 percent of young citizens of the U.S. couldn't even locate the U.S. on a map.
  • One adult American in five thinks the Sun revolves around the Earth.
Not only are some of these statistics worthless, they are specious at best. A statistic is only as good as the sample it is based upon. What group of people has this nocturnal arachnid appetite? Only a .5% chance of getting married if you are a woman and older than 35? How young are these young citizens and how were they asked? One in five Americans think the earth goes around the sun?

These are all bad or misrepresented statistics. Here is a good motorcycle statistic for you: "Since 1997 motorcycle rider fatalities have increased by 89 percent from 2,116 to 4,008 in 2004." Sobering, yes. But for myself, it begs the question why? Is it lack of training? More squids on the road? Inattentive car and truck drivers? Increase in speed limits? Larger moto engine sizes? You can read for yourself here: if you like.

But really, statistics can be used or misused. When implemented properly, they can be used to bring to light real world problems. Unfortunately, there are those who would dissect these statistics and turn them against those trying to do good.

A good example are the statistics publicized by the National Coalition Against Domestic Violence. You see, this morning I started to do some research on using Google or Yahoo groups to open an e-mail discussion group for my little October ride. What did I find? I found some good links and a few groups, but predominantly what I saw turned my stomach. It seems there are those that would rather pick apart statements and statistics in an effort to discredit the work of a valid organization.

Questioning statistics is one thing. Questioning an organization's intentions is valid as well. Picking apart and criticizing valid statistics that follow the scientific method in an effort to belittle a real life problem is abhorrent.

Whether you eat 20 spiders or 2, the statistic is still valid.

Whether you use one definition of abuse or another, it does not minimize the problem. Pardon my language, but why don't these people quit their bitching and do something to solve the problem. Tweeking a number may make things look better but does nothing to help the woman or man down the street that was just hit in the eye by an intimate partner.

1 comment:

Stephen said...

Statistics are fun to read, but it is kind of hard to take any of them to heart. Thanks for the spider stat, rest assured, I will sleep well tonight!