Welcome!

Thanks for stopping by! If you like what you read, tell your friends! If you don't like what you read, tell your enemies! Either way, please post a comment, even if it's just to tell us how much we suck! (We're really needy!) You can even follow us @JasonBerner! Or don't! See if we care!







Thursday, January 6, 2011

More Damn Lies



Just sayin'.

***************************************************
Quite a bit about statistics in the news today. First, the New York Police Department has commissioned a panel to determine whether commanders are cooking the books. Specifically, they worry that precinct commanders may downgrade felonies into misdemeanors or fail to submit reports in order to make it seem as if the rate of serious crime is going down. Disturbing if true, but an understandable response to the reliance on statistical performance measures.

The NYPD depends on a program called CompStat, which looks at rates of different crimes in different areas and then places the onus on precinct commanders to reduce those rates. If sufficient progress is not achieved, these commanders' careers may suffer. So commanders have a clear incentive to minimize felony rates.

On the other hand, programs like CompStat guarantee their own ultimate failure. Initial progress may be relatively easy--reducing the number of armed robberies from 100 to, say, 75 in a given area: a nice 25% drop. But as numbers decrease, it becomes ever harder to achieve further reductions. And in a city like New York, where the crime rate is unlikely ever to hit zero, the incentive to fudge one's numbers to achieve continued "success" may become overwhelming.

This is not to excuse precinct commanders who deliberately brush serious crimes under the rug. But those who live by statistics must adapt those statistics to measure results once "success" has been achieved.

Or one could just make up numbers, as it appears "Doctor" Andrew Wakefield did several years ago when he trumpeted results of a study that linked vaccines to autism. His report has finally been revealed to be a hoax, which may or may not convince vaccine-averse parents to protect their kids against nasty diseases. Numerous studies undertaken in the years since Wakefield's hoax that showed no link between vaccines and autism didn't convince these folks, so we are less than optimistic.

On the bright side, trillions of measles viruses that would have died needless deaths due to vaccinations did get to live out their lives happily in children's bloodstreams since 1998, when the original study came out.

Finally, a prominent psychology journal will publish an article offering compelling proof of the existence of ESP. Professor Daryl J. Bem of Cornell University bases his conclusions upon (among other things) an experiment wherein volunteers were asked to predict which side of a covered computer monitor contained an image. When the image was erotic in nature, participants were correctly able to guess--sorry, "predict"--which side contained the image 53% of the time (as opposed to 50%, which would be attributable to "chance"). Interestingly, the participants did not do better than chance for non-erotic pictures.

In other words, smut increases one's psychic ability by approximately 3%. So, before you make any major investments in the stock market, you might do well to flip through some back issues of Hustler.

Solipsistography:
"Taking Control, G.O.P. Overhauls Rules in House"
"New York City to Examine Reliability of Its Crime Reports"
"Study Linking Vaccine to Autism Was Fraud, Journal Reports"
"Journal's Paper on ESP Expected to Prompt Outrage"
Image of Dr. Evil from Moviefone.co.uk.

No comments:

Post a Comment