Americans and Religion

[Maitre]

Here is the article the newspaper is referencing, for reference. Link

The salient point they make is that;I think this summary rings true from what I know of most Americans and their perceptions of American society and people responsibilities within it.

Just flipped through the article a bit and lighted on this particular bit of text:

The research includes a nationally representative random-digit dial (RDD) telephone survey (N = 2081) conducted during the summer of 2003. In addition, in-depth interviews and fieldwork were conducted in four U.S. cities (Los Angeles, Minneapolis-St. Paul, Atlanta, and Boston) by a team of graduate students in the summer of 2004.

-'Edgell P, Gerteis J, Hartmann D. Atheists As “Other”: Moral Boundaries and Cultural Membership in American Society. American Sociological Review. 2006;71(4):211-34.

So this is where the "In-depth" sample comes from. Not sure that I feel represented by folks in any of those cities, though I've never been to LA or the twin cities, so I'm not quite as sure about those two. Secondly, though the research that I do is quite a bit different from the methods used here, a few general pricipals still apply: if you frame a question in a deliberate fashion, you can get any answer you want. Lets review the options for "Agree with vision of America?" ---> (from Table 2a, pg 219)
a) Almost completely agree
b) Mostly agree
c) Somewhat agree
d) Not at all agree

Though the phrasing of the questions (stated on pg 217) was not as biasing as I had hoped after reviewing table 1 (Oh I was warming up to slam them if that was a forced choice list or rank order item), there still seems to be a pretty wide gulf between c and d, don't ya think?

Also from that table 2a, which if I wrote it would have included the N's for each of the groups, you can see that the 75% of the subgroups "Non-church attenders" and "Non-religious" (which seems to subsume non-church attenders according to the foot notes (still no N's) rated atheists as having some level of agreement with their vision of America, so what are we really talking about here? Looking at the data for church-goers indicates that they don't like the atheists, but even the non-conservative protestants are mostly tolerant (61% would not disapprove of marriage of a child to an atheist, only 32% feel that atheists disagree with vision for america).

So this study says to me that people like people who are like them. The churched don't like the non-churched, but the non-churched have no hang-ups about the non-churched. This seems like a verification of the classic bias that every intro psych or research methods course covers in the first two weeks: survey respondents tend to rate in-group individuals as better/higher than others as a reflection of their rating of themselves ("they're like me, and I'm good at this so they must be good at this")

The crushing result that Americans don't like atheists comes based on the extrapolation from the sample to the national population, but this relies on the representativeness of the sample. If this is based on ~140 folks who responded to the detailed survey, then we're talking about 140/~280,000,000 or a 0.00005% sample. I think the possibility that subgroup-representation may differ between the sample and the overall population in a sample this small.

Again, my research occupation is quite a bit more quantitative than this sort of thing, so there may be standards in qualitative research that have been met by this study that I'm missing, but personally, I wouldn't put too much stock in this particular result. I doesn't strike me as something new and different.

/2cents