Thursday, 5 July 2007

Is DNA testing over-hyped?

A couple of days ago I blogged on prominent genealogist Michael John Neill's skepticism on DNA and genealogy. He now has a follow-up posting.

He gives an example where, in trying to establish a relationship between two people several generations back, using descendants of those two as surrogates to provide "their" DNA, one has to be aware that there could be a non-paternity event in an intervening generation.

The problem here is not with the DNA test, which shows the lack of a relationship between the two people tested, but rather the assignment of those people as surrogates for another supposed ancestor.

There's a lot of scope for hidden hanky-panky in comparing distant cousins. Suppose the chance of a non-paternity event is 5% for any given child. Then the chance of at least one such event in 10 generations, five up from one person tested to the common ancestor and five down to the second person tested, is 1- (0.95^10) = about 40%. In this case conventional paper records would also be leading one of the two astray in following their ancestry back.

In ending the posting Michael John Neill makes the statement "The problem I have with DNA for genealogy is that it tends to be hyped more than necessary and pitched as a cure-all."

How much hype is necessary? Everyone is entitled to their view and it depends on the definition, "advertising or promotional ploy," "something deliberately misleading; a deception" or something in between. I'd point to genealogical examples where the hype is much greater than that for DNA. Geni.com and Footnote.com are examples, for different reasons.

As for DNA being pitched as a cure all, I don't recall any examples by companies involved in genetic genealogy. Perhaps Michael John Neill would oblige with some cases.

No comments: