New Issue of Evolution: Education and Outreach
T. Ryan Gregory flags the availability of the newest issue of "Evolution: Education and Outreach" a few days ago. The main emphasis is on 'tree thinking' and there are some very good resources available there. However, of interest to me was Mazur's article on the relationship of religiosity, political conservatism, education, and acceptance of several scientific propositions about evolution, plate tectonics, the Big Bang, and heliocentrism.
9 Comments
Just Bob · 22 November 2010
Uh oh, we're in for some "conservatism doesn't make you creationist" from JK.
RBH · 22 November 2010
Just Bob · 22 November 2010
And that, among other things, is what we love you for!
A true gentleman and scholar!
Robert Byers · 24 November 2010
Fre thinking? From the evolution censors?!
Is this a trick?
yes AMEN to free thinking! Why not start in the house of thinking. Schools. In order to think about origin ideas or origin contentions one must have full access to all ideas in order to freely think.
In short the censorship of public institutions must be ended.
I suspect, as I always note, that these people just want the prestige of claiming to be free thinkers but are not in reality.
Establishments , in the wrong, never welcome free thinking.
Not in Canada anyways,.
Roger · 24 November 2010
Stanton · 24 November 2010
Rich Blinne · 24 November 2010
John Kwok · 25 November 2010
Solely on statistical grounds, Mazur's analysis is flawed. His regression coefficient of 0.64 shows that there is a lot of variability in the data. What he should have done - and he does not mention that he did this - was to use some kind of data transformation (e. g. logarithmic (base 10), natural logarithmic (base e), square root) that would have transformed the polling data into one more closely approximating a normal distribution. Data transformations of the kinds I have described are standard statistical practice and, in fact, were stressed to me when I took introductory biostatistic courses in graduate school.
Of course this also begs the question as to whether ordinary least squares linear regression analysis was the best statistical technique used; even Mazur acknowledges that probit analysis would have been more accurate.
Since Mazur's statistical analysis is flawed, I don't think it is necessary for me to comment on the predictors he found (Betas) or how he has interpreted them.
John Kwok · 25 November 2010