More Mush about Wrongful Convictions from the NYTimes

In a NYTimes article dated August 19, 2007, the reporter wrote:

Why Would Someone Falsely Confess?

Because most suspects who confess to a crime are in fact guilty, it is not surprising that most of us have a hard time accepting the idea that someone would falsely confess. ...

That [innocent people don't confess to crimes they didn't commit] is certainly the conventional wisdom. [Tillers: Oh yeah? Sez who?] ... According to the Innocence Project, 49 people whose convictions relied on false confessions have been proved innocent and released from prison based on DNA evidence.

Another math quiz: If 25% of the wrongful convictions overturned by the Innocence Project [Tillers: assume circa 49/207] are the "result of" [Tillers: What does this mean?] false confessions, does it follow that circa 25% of all confessions are "false"? (Hint: "No, not necessarily." [By now you know the spiel here, right?])

Well now, let me get this straight. The suggestion is that 75% of wrongful convictions are the result of mistaken identifications and 25% are the result of false confessions. Therefore: There are no other causes of wrongful convictions. Can that be? Oh wait; I see now that the Innocence Project (reportedly) asserted that in 25% of the wrongful convictions it overturned there was "reliance" on a false confession. Well, there's mush for you. We want to know the extent to which false confessions (and mistaken identifications etc.) are responsible for wrongful convictions. That's because we want to know how much the frequency of wrongful convictions would be reduced if the frequency of false confessions (or mistaken identifications etc.) were reduced. The mushy numbers we are given here don't give us answers to such questions.

Numbers, numbers, ... ooooh those %$&% numbers!