Wed, 15 Oct 2003

More fun with statistics

More fun with statistics ... this stuff really pisses me off! I find it simply incredible that our vaunted news media, with such a nose for bogus claims by politicians, businessmen and celebs, will take nearly any data presented as "scientific" as absolute, indisputable fact.

Note: I'm not arguing the effects of secondhand smoke here, only the public policy effects of bad or skewed science. I certainly have a position myself on the issue, but it's pretty much a low priority for me, as I don't eat out much and hardly travel at all anymore...

Helena, Montana passed an 'Indoor Clean Air' act that became effective in June 0f 2002. When local voters later (successfully) tried to rescind the ordinance, two of the "activist" doctors involved "discovered" that the law had dramatically reduced heart attacks in helena in just three months.

Here's the NY Times editorial today that set me off:

The Secondhand Smoking Gun

Here's a link to a Power Point presentation (in PDF format) of the actual study:

Helena Study

Like I said, the ban went into effect June 1, and was suspended December 1. Notice the pattern of the number of heart attacks (starting in January of 2002):
January: 8
February: 6
March: 2
April: 8
May: 6
June: 2
July: 1
August: 3
September: 7
October: 5
November: 4
December: 6

The number of heart attacks in June was 2 - a dramatic drop indeed! Unfortunately, that's the same number of heart attacks recorded for March, well before the ban was in place!

Note the spike in heart attacks in September - might one surmise that elderly farmers are cutting wheat in September? Note also the generally higher number of attacks in the winter months as opposed to the summer months. Chest colds and flu have long been factors in the onset of other problems, especially for a senior population.

Note the very low number of attacks overall - this is a tiny study! If these results were reproducible, wouldn't it be logical to assume that California, with it's overall ban, would have seen such similar dramatic results?

Ach - I've been beaten to the punch (thank you, Google). While just now looking for more data on this, I noticed Jacob Sullum's fine debunking of this study in Reason Magazine.

Let's wrap this with my favorite quote from Mark Twain:

There are three kinds of lies: lies, damn lies and statistics!

/Home | 0 writebacks | permanent link

comment...

 
Notes: If you put a <mailto:> link in the URL field your address will not be mangled: this could be a bad idea as your email address could be easily harvested by bots designed for SPAM. The comments field should now format correctly for line feeds and carriage returns: when you hit the 'Enter' or 'Return' keys in your comment it should break to a new line. The text should wrap cleanly. Please let me know if it doesn't. No HTML tags will pass through - entering links seems to be the main cause of comment SPAM. Also, please be sure that Javascript is enabled in your browser before attempting to post a writeback. Sorry for any inconvenience, but this really helps cut down on the amount of comment SPAM I have to deal with.
 
 Name:
 URL:(optional)
 Title: (optional)
 Comments:  
Save my Name and URL/Email for next time