Lies, damn lies and (my favorite) statistics

We have been thinking about statistics a bit recently (indeed it was the root cause of the letter, as was expounded in yesterday’s post.) As we pondered the topic, our focus drifted (can one focus and drift?) to the famous quote that inspired the title of this post.

The inevitable Wikipedia plunge revealed, to no great surprise, that the phrase’s authorship, variously bantered as belonging to Disraeli, or Twain or Dilke, or even an unnamed judge who once reportedly said that there were 3 kinds of unreliable witness: ‘simple liars, damn liars and experts.’

On the witness angle, we feel we must point to an experiment done by Daniel Simons (U of Illinois) and Christopher Chabris (Harvard). Few will value ‘eyewitness’ accounts as highly after they’ve considered this result.

The Wikipedia article also pointed, serendipitously, to Stephen Jay Gould’s article “The Median Isn’t the Message” which he wrote in 1984 after being diagnosed with abdominal mesothelemia, a serious ‘terminal’ cancer in 1982. Confronting the statiscally-significant median survival rate of his cancer – then 8 months – he wrote:

The problem may be briefly stated: What does “median mortality of eight months” signify in our vernacular? I suspect that most people, without training in statistics, would read such a statement as “I will probably be dead in eight months” – the very conclusion that must be avoided, since it isn’t so, and since attitude matters so much. …

If the median is the reality and variation around the median just a device for its calculation, the “I will probably be dead in eight months” may pass as a reasonable interpretation. …

But all evolutionary biologists know that variation itself is nature’s only irreducible essence. Variation is the hard reality, not a set of imperfect measures for a central tendency. Means and medians are the abstractions.

Statisticians, in my experience, concur with Gould and will tell you that statistics only apply to populations – they are meaningless when applied to the individual. The average American male may be 5 feet, 10 inches tall, but, if you happen to be Shaq, the statistic is all but meaningless.

In any case, Gould lived for 20 years after receiving his 8-month prognosis. I mention this, of course, because my cancer, glioma, comes with a 4-year median survival rate. According to this datum, I have a 50% chance of being alive this time next year (I was diagnosed in October, 2006), and have indeed, used that as a yardstick for planning events. Two years ago I started putting ‘things in order,’ which has resulted in me being better organized I’ve ever been.

Thus I had planned, in the countdown to my presumed expiration date, to inform the old school that I would be relinquishing a (visitor’s) board seat and possibly other responsibilities and proclivities as I focused on the important things, like my not-quite-2-year-old granddaughter and good Pinot Noir.

Gould’s tale has given me welcome pause to rethink the letter, which still sits, unsent, on my Mac’s right-hand monitor, the one reserved for serious projects. On the one hand, a recent, rigorously-conducted study tends to discount Gould’s contention that ‘attitude matters:’ on the other hand, every serious MD/PhD I know, people who deal daily with the terminally ill, believe, almost to a soul, that Gould is right. I hope these aren’t the judge’s ‘experts’…

Advertisements

About Chris Gulker

Chris Gulker, a self-described Infuential Blogger, lived in Menlo Park, California with spouse Linda. He passed away in late October 2010.
This entry was posted in All, My Brain, New Life. Bookmark the permalink.

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