In my post yesterday where I compared Catholics & Protestants in New England with Southerners in the McCain Belt, I was struck on the evolution question that in New England Protestants exhibited much more variance than Catholics. More Protestants rejected evolution or definitely believed it was true than Roman Catholics, who tended to agree that it was probably true. To me, this indicates the fissiparous tendencies of Protestantism, whereby new sects emerge from schisms within denominations, in contrast to the "broad church" philosophy of Roman Catholicism as well as the due deference to clerical elites. Though acceptance of some sort of evolutionary theory is not demanded by the Roman Catholic church, there is a general acceptance among the clerical caste as to the validity of general evolutionary processes. In contrast, liberal Protestants have arguably taken much more enthusiastically to evolutionary theory (e.g., Barack Obama, a member of the United Church of Christ, admits to believing in evolution with more certitude than angels), while conservative Protestants make its rejection a touchstone of their distinctiveness.
So I decided to go into the GSS, and see how the SCITEST4 variable relates to WORDSUM.
SCITEST4 asks: In your opinion, how true is this? " Human beings developed from earlier species of animals." The answers are:
DEFINITELY TRUE
PROBABLY TRUE
PROBABLY NOT TRUE
DEFINITELY NOT TRUE
These answers also have numeric values 1-4, and can be interpreted ordinally, which comes in handy.
WORDSUM is simply a vocabulary test, from 0 to 10 correct. One presumes those who have a higher score are more intelligent or educated.
Below are charts which show the difference categories of response in relation to evolution in terms of how the respondents distribute over the vocabulary test spectrum. Y = proportion, and X = WORDSUM correct score.
Converting the categorical responses to their numeric values (1-4), the correlation between WORDSUM and SCITEST4 for the groups is as follows:
All Americans = -0.19
Protestants = -0.13
Catholics = -0.16
Nones = -0.33
The negative value implies that those with higher WORDSUM scores selected the responses with "lower" values (e.g., DEFINITELY TRUE = 1). Note the "Nones." I'll leave speculation for the comments except to say that those with no religion in this sample are a diverse bunch. From the GSS:
Confidence in the existence of God for "Nones"
Don't believe - 12.1%
No way to find out - 21.7%
Some higher power - 26.9%
Believe sometimes - 6.6%
Believe but doubts - 15%
Know god exists - 17.7%
- Log in to post comments
Is there any way to rescale the graphs so that the y-axis is "proportion of the N-words-correct group" rather than proportion of the population as a whole? It's a little hard to get a sense of how the low ends compare because the absolute percentages are so small on this scale.
Is there any way to rescale the graphs so that the y-axis is "proportion of the N-words-correct group" rather than proportion of the population as a whole? It's a little hard to get a sense of how the low ends compare because the absolute percentages are so small on this scale.
the N's here are really small...5 isn't the median. but yeah, let me see. will post update.
ok, could you clarify what "N-words correct group" means? don't to add charts which are off base, and i'm a little confused.
I think the problem with the graphs is that you cannot see the proportions at any one point. For instance it clearly cannot be the case that only about 22% of religious nones who get 10 out of 10 in the wordsum believe evolution to be definitely true (it must be far higher, surely).
Why not a blog all about the McCain belt. You could call it Run When You Hear The Banjos