Friday, March 09, 2007

The Numbers Guy on the Talpiot Tomb Calculations

There is an article in this morning's Wall Street Journal Online on the "Jesus' Family Tomb" documentary (thanks to Stephen Goranson for pointing it out):

Odds of 'Lost Tomb' Being Jesus' Family Rest on Assumptions
Carl Bialik
. . . "As you pile on more assumptions, you're building a house of cards," says Keith Devlin, a Stanford mathematician and NPR's "Math Guy." (Scientific American also challenged the calculation on its Web site.) . . . .

. . . . "I wouldn't be comfortable coming up with a number like this, because the general audience will not understand that it is very, very subjective," says Ivo Dinov, assistant professor of statistics at the University of California, Los Angeles . . . .

. . . . He still hasn't provided full documentation of his calculation, saying he'd wait until his paper, not yet completed, is accepted by a journal. "There is a mismatch between how the media works and how academia works," Prof. Feuerverger says. "Obviously it would have been a whole lot better if I had completed the paper" before the documentary aired . . . .

. . . . . "When I was doing the calculation, I was naively unaware of the extent to which the filmmakers might be depending on the ultimate result of it," he says. "I did carry out the calculation in every good faith. I hoped it would be interpreted in that light."
The article doesn't tell us much more than we already knew, but it does help to underline the point that I have been making since my post last week on The Statistical Case for the Identity of the "Jesus Family Tomb", viz. that any statistical calculation is only as good as the data fed to the statistician; the numbers are only as good as the assumptions they are based on, and here there are major problems with those assumptions.

No comments: