Tuesday, September 29, 2009

Do the devout tend to be conservative in all religions? Using GSS data (N = 43,542), I calculated the association between frequency of attendance of religious services and level of political conservatism. The correlations are all positive for Abrahamic religions, but what about other religions? Look:



The relationship is negative, and the magnitude is noteworthy for all except Buddhism.

7 comments:

Jokah Macpherson said...

I am guessing the devout Muslims tend conservative in the same way that blacks do because I don't remember ever seeing Muslim groups out marching to show their support for George W. Bush.

I wonder if sin has anything to do with the conservative tendencies of Western religion practitioners. My knowledge of Eastern religions is limited but I get the sense that the concept of personal sin in the material world has a less central place in their theologies.

John said...

I always knew the Great Manitou was a lefty

bgc said...

Fascinating - more like this please!

Anonymous said...

Jokah Macpherson said...
I am guessing the devout Muslims tend conservative in the same way that blacks do because I don't remember ever seeing Muslim groups out marching to show their support for George W. Bush.

George W. Bush won the Muslim vote in the 2000 Presidential elections. It was the War on Terror that turned Muslims voters away from the Republican Party.

See: How Did Muslims Vote in 2000? by Alexander Rose.

B.B.

Jason Malloy said...

I don't like this. The sample size for the religious attendance question (RELACTIV) is 24 for Hindu, 5 for Other Eastern, and 5 for Native American.

For RELPERSN ("do you consider yourself a religious person" 1-4). The relationship between religiousness and conservatism is .17 for Christians (N=4,730), .25 for Jews (112), and .24 for Hindus (17). Ns are inappropriate for the other two.

A few more points:

1) Sample sizes are not high for exotic religions in the GSS, and a lot of these people are just educated white liberals high in openness to experience, not representative members of these religions.

2) Minority ethnic status affects the association, since "traditional" people are more religious as well as more ethnically engaged. Being ethnically engaged in America means being more liberal for everyone but white people.

So the association between religiousness and self-identified conservatism is .29 for whites, .07 for blacks, and .13 for "other race".

3) While there are probably other exceptions due to special circumstances, authoritarianism, conservatism, and religiousness appear to form a higher order factor Thomas Bouchard calls Traditionalism.

Jason Malloy said...

So the association between religiousness and self-identified conservatism is .29 for whites, .07 for blacks, and .13 for "other race".

In other words we call it "liberal" when nonwhites are ethnocentric and "conservative" when whites are ethnocentric. Otherwise these relationships are higher for nonwhites when you focus more on specific conservative beliefs rather than identity: e.g. abortion, sex, gender roles, homosexuality, etc.

Also, as Razib has discussed, social conservatism and religiousness don't seem to have a relationship in China, where religious belief is low.

Ron Guhname said...

Jason: My mistake not putting up the variable and the n's. I used ATTEND so the sample sizes are a bit better: Buddhist 83, Hindu 37, Other Eastern 11, and American Indian 12. But your points are well-taken.