Marginal Utility

by Drek

As regular readers of my blog know, I am not what you'd refer to as a religious person. I am, to the contrary, rather anti-religious and can be classed as a materialist atheist. Having been an atheist for some time now (somewhere in the vicinity of fifteen years if you count from when I "came out" about it) I have long had the impression that we are not the most liked minority group in the world. However, despite my hunch, I had no idea of the true scope of the problem. At least, not until I learned of a recent survey conducted by Sociologists at the University of Minnesota. The results of this survey are rather striking (from the original press release):

American’s increasing acceptance of religious diversity doesn’t extend to those who don’t believe in a god, according to a national survey by researchers in the University of Minnesota’s department of sociology.

From a telephone sampling of more than 2,000 households, university researchers found that Americans rate atheists below Muslims, recent immigrants, gays and lesbians and other minority groups in “sharing their vision of American society.” Atheists are also the minority group most Americans are least willing to allow their children to marry.

Well, as I've said before it's good to be loved, and this survey does indeed give me a warm fuzzy feeling inside. I can't say I'm really all that surprised- there's been plenty of anecdotal evidence in support of the idea that atheists aren't liked. Take, for instance, President George Bush (the first) who denied that atheists should be considered U.S. Citizens (which is a little terrifying given the policies of his son towards non-citizens). Then there's the preference given to religious parents over irreligious parents in custody disputes, which is pretty horrifying given that the United States is purported to be a secular democracy. Finally, there's the otherwise-innocent curiosity from people who should probably know better. So, the Minnesota study might seem like the straw that should break this camel's back.

Well, have no fear gentle readers, for I am made of sterner stuff than that! I have read Chick tracts and emerged with my mind unscarred! I have spent time perusing Answers in Genesis and emerged with my faculties untainted! Perhaps more importantly, I am a skeptical guy, and want to know more. Now, the survey in question won't see publication until the next issue of the American Sociological Review, thus thwarting the simplest approach to gaining more information. So, lacking a time machine (unlike, I think, my dissertation advisor, whom I am convinced is from the future) I must find another source of insight- namely, the General Social Survey.

The General Social Survey is a representative survey of the non-institutionalized U.S. population. It's been conducted every year or two for decades now, and includes data on a wide variety of issues from demographic factors to public opinion. Of particular use to us, within this survey there happen to be three questions pertaining to those who don't believe in god. These questions are: whether an atheist should be prevented from speaking in public, whether an atheist should be permitted to teach, and whether an atheist book should be removed from the library. So, you know, basic civil liberties. Presumably, the more people dislike atheists, the more likely they are to attempt to restrict their civil liberties. Thus, despite the lack of comparable questions, we should be able to get at least a rough handle on how disliked atheists are. Moreover, the same questions are asked for homosexuals and communists as well, so we can get a sense of perspective.

So what do we find when we examine the GSS? Well, believe it or not, using the 2004 GSS data* it appears that fully 85.9% of the population wouldn't curtail any of these civil liberties. The first lesson we take from this is that if atheists are hated, a vast majority of the population still seems to be relatively okay with us. The proportion of people indicating that between zero and three of the above civil liberties should be curtailed are:

Atheists:

0: 85.9%
1: 6.6%
2: 3.8%
3: 3.6%

Mean: 0.252

By comparison, take a look at the same breakdown for homosexuals and communists:

Homosexuals:

0: 89.7%
1: 4.9%
2: 2.2%
3: 3.2%

Mean: 0.189

Communists:

0: 85.9%
1: 5.4%
2: 3.5%
3: 5.2%

Mean: 0.281

So, based on these figures, it does look like atheists are arguably as hated, or more hated, than any of these three groups. Granted, we're running neck-in-neck with the communists, but hell, a lot of folks think that atheist is synonymous with communist anyway, so that makes a certain amount of sense. Well, that and the general U.S. disdain for communists that emerged during the cold war. Uncle Joe would be proud. In any case, it does appear that the survey from Minnesota is correct- atheists are among the most hated groups and are less liked than homosexuals. It's also apparent, however, that most people probably don't actually hate us. If these figures are any indication, most people are pretty okay with atheists, and I find that reassuring.**

What about the people who do hate atheists? What are they like? Well, let's find out! For the below, "atheist haters" are those who would restrict all three liberties, "non-haters" are all those who would restrict two or fewer, and "pro-atheists" are those who would restrict none.

Education: On average atheist haters have twelve years of education, which is roughly equivalent to a high school degree. This is supported by the GSS "degree" variable, which shows that the modal degree obtained by atheist haters is, indeed, a high school diploma. By contrast, non-haters have an average of 13.8 years of education and a higher proportion have obtained educational credentials beyond the high school diploma, though most still only have diplomas. Finally, pro-atheists have a mean education of 13.9 years, and have yet a higher proportion of post-high school degrees.

Age: Atheist haters have a mean age of 46.3 years, while non-haters and pro-atheists have ages of 44.6 and 44.5 respectively.

Race: 71.8% of atheist haters are white, 14.3% are black, and 13.8% belong to "other" ethnicities. Among non-haters, the proportions are 79% white, 12.1% black, and 8.7% "other." Finally, for pro-atheists the proportions are 79.8% white, 11.5% black, and 8.6% "other."

Sex: Of atheist haters, 38.2% are male and 61.7% are female. Among non-haters, 46.5% are male and 53.4% are female. Finally, for pro-atheists, 46.8% are male, and 53.1% are female.

Income: Modal household income for all three groups is above $25,000 a year, but only 53.6% of atheist haters reach this level, while for non-haters and pro-atheists, the proportions are respectively 71% and 72.1%.

Religion: Atheist haters are predominantly composed of Protestants (65%) and Catholics (26%). Moslems, "other Christian," "No religion," and "Inter-denominational" make up the remaining, but none of these groups account for more than 4% of all atheist haters. For non-haters, Protestants account for 50% while Catholics account for 25.1%. The remaining 25% is distributed among a variety of other faiths, with those professing no religion accounting for 14.7% of the total. Finally, pro-atheists are 49% Protestant, 25% Catholic and 15% "None." The remaining faiths do not individually account for more than 3% or so of the total. It's probably worth noting here that those who say they have no religion are not necessarily atheists, but rather may not subscribe to any particular religious doctrine.

Religious Attendance: Atheist haters attend an average of 36.9 religious services per year. Non-haters attend 22.7 times, and pro-atheists attend 22.2 times.

So, based on all of this, we can conclude the following about atheist haters: they are relatively poorly-educated, older, disproportionality unlikely to be white, and female with relatively low household incomes. They are mostly Christian, especially Protestant, and attend church much more frequently than average.

If we throw all of this into a logistic regression model using atheist hating as our dependent variable, though, we find that only three things are significantly related to the likelihood of hating thy neighbor. These are education level (which reduces the likelihood and is significant beyond the .001 level), Protestant faith (which increases the likelihood and is significant beyond the .05 level), and frequency of religious attendance (which increases the likelihood and is significant beyond the .001 level).

What all this means is simple: Protestants, and especially those who attend church frequently, are much more likely to hate atheists than other people. This is, of course, just as Jesus intended. Those who are more educated, however, are less likely to hate atheists. The remaining variables, including race, sex, and age are spurious and only appear associated with atheist hating because they're associated with education and/or religious devotion. Well, at least as an atheist I now have a pretty good idea who it is I should be worried about. Oddly, I pretty much had a sense I should watch out for the god-fearing before this, but it's nice to have independent confirmation.

And, in sum, I consider this to be encouraging. Atheists may be relatively more disliked than other groups, but we still seem to be largely accepted. While there is certainly a core of individuals who really don't care much for us, that core is largely made up of the ultra-religious. It should probably not come as a shock that this group finds atheists to be particularly threatening- especially when we live moral, worthy lives. Perhaps the study coming out next month will show this to be in error, but for the moment things look okay.

Maybe we atheists aren't feelin' the love, but at the same time, we also aren't feelin much hate. That's enough.

For a start.

* Okay, for those who are curious, I'm using the weighted GSS data from 2004. This is necessary because the 2004 data includes an African American oversample and has certain adjustments to account for non-responses. So, if you want to try to replicate what I'm doing, you'll need to use the "wt2004nr" weight variable. Please also note that I dashed these off in my spare time, so there may well be slight rounding errors and whatnot. Crap, for that matter there may be huge glaring faults. What can I say? I try to reserve the high-quality analysis for the paying gigs.

** As you might guess, I'm not exactly thrilled that fully 14% of the population wants to curtail at least one basic civil liberty. As an academic, I need to speak publicly, teach, and write books, so this is hardly a wonderful situation. Worse yet, by the time you're willing to restrict a group's civil liberties, you've probably already reached the point where you don't want to live next to them, let them babysit your children, or marry into your family. Still, it's a start, and I'm honestly surprised that the figures aren't any worse. And boy isn't that, in and of itself, depressing? I'm actually optimistic because fewer people than I expected hate me. Gosh, I love my life.