What bias?

By Megan McArdle

William Saletan has a piece on the conservative v. liberal study in Slate, titled "Rigging a study to make conservatives look stupid". I'm not sure "rigging" is the right word. But Saletan does make an interesting point, which is that even if the study really did identify receptiveness to change, that would not be the same as identifying who was more likely to be right:

Frank Sulloway, a Berkeley professor who co-authored a damning psychological analysis of conservatism four years ago, illustrates the problem. Appearing in the Times as a researcher "not connected to the study"—despite having co-written his similar 2003 analysis with one of its authors—Sulloway endorsed the study and pointed out, "There is ample data from the history of science showing that social and political liberals indeed do tend to support major revolutions in science." That's true: When new ideas turn out to be right, liberals are vindicated. But when new ideas turn out to be wrong, they cease to be "revolutions in science," so it's hard to keep score of liberalism's net results. And that's in science, where errors, being relatively factual, are easiest to prove and correct. In culture and politics, errors can be unrecoverable.

This article available online at:

http://www.theatlantic.com/business/archive/2007/09/what-bias/1927/