Uniform probability in what scale, though? Performing a Bayesian analysis with a uniform prior will generally give different results than, say, using a log scale on the dependent variable(s) and choosing a uniform prior on *that* scale. The Jeffreys prior provides a method of computing a non-informative prior that is invariant under re-parameterization, but is generally difficult to work with, and is never a uniform prior.
So yeah, the concept of an uninformative prior is more complicated than "just use a uniform distribution", and analysts need to be especially careful with priors used with small sample sizes!
Sure they do, within the bounds of non-infringement upon the rights of others. Tolerance doesn't mean you get to do whatever the hell you want. It means you get to do whatever you want until it affects someone else negatively, and then compromises need be drawn up, based on the severity of the negative affect, and the necessity of what you want to do.
I'm pretty sure none of that is hypocrisy. Some of those liberal assertions might turn out to be wrong (or at least wrong in certain circumstances), but then they'd just be wrong, not hypocritical.
I'd really like it if everyone's rights and freedoms were preserved, not just yours!
Maybe we should be considering the psychological and social benefits to allowing men to choose to throw rocks at baboons all day as they are biologically programmed to do?
Exactly! An economist from Yale, Keith Chen, made this interpretation in 2008, and experimental redesigns and rebuttals have raged since. A (dated) review of the debate is available at http://tierneylab.blogs.nytimes.com/2010/01/27/monkeys-candy-and-cognitive-dissonance/
I had my students at a high school summer math camp perform the M&M experiment with students and instructors in the lunch room. If my memory serves, after 200 trials our rejection rate was ~67%, pretty much perfectly in line with the pre-existing preference interpretation!
Slashdot Mirror
Uniform probability in what scale, though? Performing a Bayesian analysis with a uniform prior will generally give different results than, say, using a log scale on the dependent variable(s) and choosing a uniform prior on *that* scale. The Jeffreys prior provides a method of computing a non-informative prior that is invariant under re-parameterization, but is generally difficult to work with, and is never a uniform prior. So yeah, the concept of an uninformative prior is more complicated than "just use a uniform distribution", and analysts need to be especially careful with priors used with small sample sizes!
Sure they do, within the bounds of non-infringement upon the rights of others. Tolerance doesn't mean you get to do whatever the hell you want. It means you get to do whatever you want until it affects someone else negatively, and then compromises need be drawn up, based on the severity of the negative affect, and the necessity of what you want to do.
You're welcome to your hate-filled morality, but it doesn't give you the right to legislate discrimination.
I'm pretty sure none of that is hypocrisy. Some of those liberal assertions might turn out to be wrong (or at least wrong in certain circumstances), but then they'd just be wrong, not hypocritical. I'd really like it if everyone's rights and freedoms were preserved, not just yours!
Self-moderation is one of those things that we've always had more of 20 or 30 years ago.
I don't think you're using "parse" correctly... Fucking Anonymousites!
Maybe we should be considering the psychological and social benefits to allowing men to choose to throw rocks at baboons all day as they are biologically programmed to do?
Exactly! An economist from Yale, Keith Chen, made this interpretation in 2008, and experimental redesigns and rebuttals have raged since. A (dated) review of the debate is available at http://tierneylab.blogs.nytimes.com/2010/01/27/monkeys-candy-and-cognitive-dissonance/ I had my students at a high school summer math camp perform the M&M experiment with students and instructors in the lunch room. If my memory serves, after 200 trials our rejection rate was ~67%, pretty much perfectly in line with the pre-existing preference interpretation!