The Numbers Guy in the WSJ

One of the regular features in the WSJ is one written Carl Bialik, call the Numbers Guy. It attempts to clear up confusion in stories in the media which use numbers. Bialik has a great way of explaining complex issues and he did it again in a story about the Discovery Channel's hyped up documentary about the supposed discovery of the tombs of Jesus and his family. As noted in an earlier post, the documentary's film-makers make a great deal of a statistics professor's discussion of probabilities - without offering the very logical alternative to the trail of numbers. Bialik commented in this morning's column -

"Had the professor assumed the inscription could be for any Mary, a very common name then, it would be far less likely that Christ's family is in the tomb. The mathematical finding would become "statistically not significant," Prof. Feuerverger tells me. Similarly, the name "Yose" -- as one of Jesus' four brothers was called in the Gospel of Mark -- is a derivative of Yosef, another common name. There, too, the finding would be less conclusive if the professor had considered "Yose" applicable to any Yosef.

Even if there was consensus on the interpretation of the names, there are no comprehensive records showing how frequently they occurred in the population at that time. Prof. Feuerverger relied on modern books about ossuaries and ancient texts to tally the occurrence of certain names in the area then. That falls far short of a complete census.

"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.)"

Last night as I was driving back from a dinner I listened to a guy on the radio named John Ziegler interview one of the other "experts" on the documentary - a professor from UNC Charlotte. ZIegler went on about how sure the data was - yet the key story is that the real numbers suggest a very shaky case. Ziegler was clearly trying to reinforce his own "agnostic" views rather than to discuss the implications of such a documentary. In the Bialik article UCLA assistant professor of statistics commented "I wouldn't be comfortable coming up with a number like this, becuse the general audience will not understand that it is very, very subjective," And indeed Professor Feuergiver commented that he was uncomfortable in the way that his calculations, which were calculated with the normal caveats of statistics, were used. What bothers me about this program was not the potential conclusion but rather the very sloppy way that the producers used the data. But then why should I be surprised about that with the MSM?

BIalik does what I think a journalist should do, on a consistent basis, he tries to explain complex stories (that use some form of numbers) in a way that clarifies rather than sensationalizes. He contributes a lot to making the world more understandable.