If he’s a statistician, why is the numerical reasoning in this press release so bad?

Half of this post is based on comments forwarded to me by Carl Phillips after I alerted him to this typical yet still outrageous press release from the inimitable yet wholly dispensible John Banzhaf (and I in turn have James Dunworth to thank for bringing it to my attention). I do have active tobacco alerts but it could be that even the alerts were wary of sending through such spam like material. (My apologies to both Carl and James for placing their names in such proximity to Banzhaf’s.)

John Banzhaf

2010-02-27 20:41:13 – When President Barack Obama undergoes his first medical exam as President on Sunday, he should receive some bad news: his smoking increases his odds of dying of lung cancer during the next 10 years about 1000%, and his overall odds of dying during the next 10 years by about 160%, says mathematician, statistician, and public interest law professor John Banzhaf, Executive Director of Action on Smoking and Health (ASH).

Once again playing fast and loose with both the numbers and his usual self aggrandizement (it is somewhat misleading of him to describe himself as a mathematician and statistician, and clearly a falsehood to fashion himself as working for the public interest when each of his press releases makes it more plain that the man sees community, democracy and autonomy as impediments to his goals.

Let us consider the numbers.

One needs to start by observing that the chance of the President dying in the next ten years from disease are extremely small, so those multipliers (if they were right) are clearly a misleading use of statistics.

Second, those figures usually represent regular heavy smokers as compared to abstainers, but Obama is not a regular heavy smoker. Moreover, much of that risk is due to being “a smoker” (the average type of person who smokes), which he is not, rather than from “smoking” which he does.

Since his diet, weight, medical care (he has medical people on call 24-7 so, e.g., would be much less likely to die from a cardiac event — which is the most likely disease death for someone his age with no preexisting bad diagnoses — than the rest of us), etc. are near-perfect, his risks are much lower than those for a “smoker”.

Finally, as a controversial president in troubled times, his risk of dying violently is, unfortunately, probably ten times his very low risk of dying from disease, maybe 100, so doubling his risk of dying from disease would not nearly double his chances of dying in the next ten years. So Banzhaf is clearly demonstrating that he has a typical guy-on-the-next-barstool understanding of statistics, just enough to make really dumb mistakes.

“These are evidence of his fundamental ignorance about the matters on which he claims expertise. It is certainly the case that he and many of his fellow travelers make a habit of lying to people, but in this case it actually seems to be a matter of getting in over his head — knowing so little about the topic that he does not even realize he does not understand it — rather than dishonesty (not that the former is a whole lot better).” –CVP

A somewhat recognizable fellow

Ironically, there are rumors that President Obama chair-smoked during the luncheon break in the health care reform summit he chaired on Thursday. If so, it appears that his doctors may have a harder time convincing him to quit, especially if his efforts directed as the health care crisis continue to create the heavy stresses which may be contributing to his smoking behavior.

Though Banzhaf does point out the fact that health care reform can lead to smoking, he does not sufficiently account for the increased risk of “chair-smoking” which is certain to be much more health averse than regular tobacco smoking. Not only do chairs contain plastics and other synthetics which do not occur in tobacco, but given that they are not as yet covered under the FDA umbrella are neither proscribed for ingestion nor are the contents adequately listed. They also are easily accessible to children and are deliberately manufactured in colours known to appeal to children and young adults.

-PLB

Advertisements

One thought on “If he’s a statistician, why is the numerical reasoning in this press release so bad?

  1. 34. A person essentially help to make seriously posts I would state. This is the very first time I frequented your web page and thus far? I surprised with the research you made to make this particular publish extraordinary. Wonderful job!

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s