Friday, March 09, 2007

Smart, white, conservative men to the rescue again: I just can't think of very many political values that I cherish more than free speech. So I decided to run a multiple regression with General Social Survey data to identify folks who are its weakest supporters. For my dependent variable, I chose whether or not you favored allowing a racist to speak in public (1 for yes, 0 for no).

Here are the standardized coefficients, from strongest to weakest:

Standardized Coefficients

IQ .15
Years of education .12
Frequency of church attendance -.11
Being a woman -.07
Politically liberal -.05
Being black -.04
Income .03

So, a free speecher is likely to be smart, educated, non-churchgoing, male, conservative, white, and wealthy. The first three are especially important. Generally, I like churchy people, but I suppose they do have an authoritarian streak. But look how it's the libs who want to shut people up. And I'm not surprised that the mommies want to wash our mouths out with soap for saying bad words.

"The very aim and end of our institutions is just this: that we may think what we like and say what we think."
~ Oliver Wendell Holmes


  1. Anonymous3:27 PM

    Bryan Caplan points out that paternalism should be called maternalism because women are more in favor of it.

    What other free-speech questions are in the survey?

  2. Anonymous3:42 PM

    Personality should be a better predictor than IQ.

    Eysenck found a correlation between authoritarianism and personality. Moreover, he recognized similiarities in personality among fascists and communists: 'authoritarianism-of-the-left' (1954, 'The Psychology of Politics', London : Routledge, New York : Praeger)

  3. tggp: I'm not sure about the full list, but I do know that they ask about racists, anti-religionists, and I think communists speaking in public and being a teacher.

  4. I am curious how the correlations change with different questions. It seems logical that a group particularly aggrieved by an activity would be more illiberal about it. Thus I'd suggest that the strength of the blackness correlate that you found is an artifact of asking about racism.

    Are liberals less illiberal about communists and anti-religionists? I would guess so.

    Using only the racism question is likely skew the correlations against liberals as a group, and also against liberal subpopulations (blacks and women).


What's the profile of a person who believes astrology is scientific? (Answer: the exact opposite of me)

Clearly, anyone who believes that astrology is scientific doesn't understand what science is, but how common is this belief, and what...