Thursday, July 05, 2007
Unbelievable
In a continuation, the Chron is doubling down on its idiotic conclusion. Here's the only nod to the gaping hole in the study:
There are some potential drawbacks to the study, namely that because it used only university students, it might not apply perfectly to men and women of all age groups and education levels. But [Matthias Mehl] said if there were important biological differences between men and women's verbosity, they would have registered at least somewhat in the study. Not if they don't impact folks until later, nor if we're talking about social differences between men and women. I'm astounded that the Chron could actually keep reporting this with a straight face (take a look at some of the idiotic anecdotal comments surrounding it), even going so far as to claim:
That dispels the popular myth that women talk way more than men. What a bunch of dolts. This is front page material, too.
By the way, the reason I'm harping on this has nothing to do with the study or the result. The claim that the Chron wants to make may very well be true, and it wouldn't surprise me, though I can't say that I particularly care about average word counts among sexes. The problem is the contribution to scientific illiteracy. Studies and conclusions like this are part of the reason I never believe someone who claims "studies say X." When you draw a conclusion from a limited data set, you have to state your assumptions. Here, those assumptions would be something like:
Men and women talk about the same amount, if we assume men of all ages act like male college students, and women of all ages act like female college students. Of course, if they wrote that, that conclusion would be laughed out of the water, which is exactly what should happen.
. . .
|
. . .
|
