Robin Hanson wrote:
> The June 2000 American Economic Review has an article by
> Blomberg & Harrington, "A Theory of Rigid Extremists and
> Flexible Moderates with an Application to the U.S. Congress".
> They offer data showing a correlation between political
> extremism and inflexibility in changing one's opinions.
> They refer to a 1950 army survey with a correlation
> between extreme views and holding those views strongly.
> And they present new data on US congressfolk explaining
> variance in ADA scores in terms of longer tenure and
> being closer to the median view.
> Their explanation is that opinion extremism and confidence
> are initially independent, but those who are less confident
> are more responsive to information, and thus move toward
> some common value, while those who are more confident stay
> put. Harrington has a larger research program on the
> evolution of flexibility and inflexibility
> (see: http://www.econ.jhu.edu/People/Harrington/)
> Since some people on this list might be considered extremists,
> what do you think of this explanation? Are you extreme
> because you are inflexible?
> It bothers me that his results seem so symmetric - it should
> be unusual for the ex post information to be smack in the
> middle of the prior distribution.
I suspect that most people regard the labels "extremist" and
"inflexible" to be perjorative. You're unlikely therefore to elicit
many responses, since most people don't identify themselves as such.
Someone you might label an "inflexible extremist" is likely to consider
themselves "consistent" and "principled."
Regarding the issue of flexibility, one explanation is that someone who
has spent a lot of time thinking about his/her belief system, and
believes it to be logically consistent, is likely to believe it more
strongly, and be more resistant to change than someone who hasn't given
it much thought.