Sunday, March 11, 2007

Maybe if the researchers faked a Jamaican accent

The study of ESP (or other forms of parapsychology) has a fascinating history. I'm a bit sad to see the lab close. I have no belief in paraPSY, but I appreciate the approach to science outlined here:

The culture of science, at its purest, is one of freedom in which any idea can be tested regardless of how far-fetched it might seem. “I don’t believe in anything Bob is doing, but I support his right to do it,” said Will Happer, a professor of physics at Princeton.

But as someone pretty well-versed in statistics, I'm not aware of any statistical procedure that could separate randomness or simple error probability from the following claim:

Analyzing data from such trials, the PEAR team concluded that people could alter the behavior of these machines very slightly, changing about 2 or 3 flips out of 10,000.

Though it's a bit more complicated than this from a statistical perspective, and you have to differentiate "statistical significance" and "practical significance": If you flipped a coin 10,000 times, how shocked would you be to get 5002 heads and 4998 tails?

First posted at Psyche Killer qu'est que c'est on 2/13/07

No comments: