The Corner

On Data Samples

From a reader:



I am not scientifically inclined, but the thing that has always bothered me the most about the whole global warming / climate change debate is the notion that any scientist is willing to make predictions based on such an infinitesimally small data set.  Even if “the record goes back a thousand years” (which itself is a dubious claim), the planet is estimated to be 4.5 BILLION years old (give or take a few hundred million years).   I am not mathematically inclined either, but doesn’t 1,000 / 4.5 BILLION = a statistically insignificant sample?

I think it’s a good question. Though I am not scientifically inclined either (which, according to Derb, means I should wear a pain collar and do menial labor at the command of my lab-frocked overlords. I kid, I kid). Still, in fairness, it seems to me we can probably trim down the 4.5 billion year denominator a bit.

4.5 billion years ago, Earth was still pretty much a ball of hot magma and therefore much warmer than even Washington, D.C., in August (and yes, half the point of this post is just to give me an excuse to say “hot magma” in my Dr. Evil voice). Life begins a billion years after that. But still: lots of magma. Instead we could start with the dawn of mammals, about 200 millon years ago. Or the dawn of homo sapiens, which starts (someone check my math!) 200,000 to 500,000 years ago. So yeah, even with the adjustments, 1,000 years is a pretty small sample.

But what do I know about science?


The Latest