September 12

# Great Description of Crash Statistics

By Mike Spack

September 12, 2012

There was an article in the Star Tribune a few days ago titled "Pedestrian Deaths Rise Sharply in Minnesota."  Below is a fantastic letter to the editor regarding the article and the use of statistics.  I'm really impressed how Michael Hoy could illuminate the situation with a couple of paragraphs.  He has a talent for applying stats to this real world situation.

The Sept. 10 story "Pedestrian deaths rise sharply" stated that "experts worry there is no stopping the trend." I am a retired actuary, and I believe I know enough statistics to explain that there is nothing abnormal about this sudden "rise."

The statistic for total pedestrian deaths is a binomial distribution. You cross the street, and you either get killed or you don't.

Over the last 11 years, Minnesota has averaged about 39 pedestrian deaths per year. That is the mean. And, in a binomial distribution, with a low incidence of deaths to crossings, the standard deviation is the square root of the mean.

In this case, with the mean being 39, the standard deviation is about six. The statistical rule is this: About two-thirds of the time, the statistic will lie within one standard deviation of the mean. This is what happened in the last 11 years. Two-thirds of the years had a death count between 33 and 45.

Also, about 95 percent of the time the statistic will lie within two standard deviations of the mean. Again, this is just about what happened.

We still need to look both ways when we cross the street. Attention to safety is very important. However, the "trends" of the last 11 years are readily explained statistically as nothing but chance fluctuations.

News is always more interesting when there is a fright element, but there is no reason to believe that walking is any worse now than before.

MICHAEL D. HOY, EXCELSIOR

• Cherian Eapen says:

Michael Hoy’s take on this is excellent. Thank you for posting.