The growth of atheism and the increase in the numbers of atheists in America is repeated in every survey done on the subject. There is no doubt that atheists are growing in numbers in America; the only questions involve how much they are growing and why. If atheists are to have any hope to keep the momentum going, they need to understand how best to encourage increased skepticism towards theism, religion, the supernatural, and the paranormal.
A Gallup survey conducted conducted May 5-8, 2011 (margin of error: +/- 4%) shows that there are still few Americans who are or who at least admit to being atheists, but those numbers in 2011 are quite a bit higher than in past years. Rates of theism remain very high, but they are gradually going down.
Do you believe in God? (Yes — No — No Opinion)
- 2011 May 05-08: 92% — 7% — 1%
- 1967 Aug 24-29: 98% — 1% — *
- 1965 Nov ??-??: 98% — 2% — 1%
- 1954 Nov 11-16: 98% — 1% — 1%
- 1953 3/28-4/02: 98% — 1% — *
- 1947 Nov 07-12: 94% — 3% — 2%
- 1944 Nov 17-22: 96% — 1% — 2%
There is a huge gap in time between 1967 and 2011 because Gallup started asking a slightly different question. Instead of asking "do you believe in God," Gallup started asking "Do you believe in God or a universal spirit?" Answers to that question turned out to be pretty similar and in 2011 they decided to ask both, thus providing these numbers:
Do you believe in God or a Universal Spirit? (Yes — No — No Opinion)
- 2011 5/05-08: 91% — 8% — 1%
- 1994 12/16-18: 96% — 3% — 1%
- 1988 12/21-22: 95% — 5% — 1%
- 1983 ??/??: 95% — 3% — 2%
- 1978 10/10-13: 94% — 4% — 2%
- 1976 ??/??: 94% — 3% — 2%
It's interesting that when the question asks about belief in "God or a Universal Spirit," more people answer "no" than when the question simply asks about belief in "God". Why would that be? At first glance I'd expect the opposite, since asking about belief in "God" is more narrow and there are surely people who believe in a "Universal Spirit" that they wouldn't call "God". I'd expect the broader question to allow for more people answering "yes," but here we seem to see the opposite.
Perhaps by opening up the nature of the question, the survey is in a sense giving people permission to think outside the limited boundaries of traditional definitions of what "God" is, thus perhaps also giving them permission to answer "no" at all. Remember that surveys like these always have a lot of people who answer the way they think the questioner might want and/or the way they think is more socially respectable and moral.
This means that we should expect that the numbers of those answering "yes" are probably higher than those who actually do believe and those answering "no" are probably lower than those who actually don't believe. Interviewers can sometimes get more accurate data by simply by changing the order of questions, thus giving people the idea that certain answers are socially and morally acceptable.
Inclusion of "Universal Spirit" gives the impression that it's acceptable to not believe in something society calls "God". Once that switch is flipped, it may be that more people were willing to fill in "no" as an acceptable option, not simply "Universal Spirit." I wonder if the numbers of those willing to admit to atheism would be increased if the question were worded even more broadly — for example, by dropping the narrow reference to "God" and instead asking "do you believe in any sort of god or gods?"
Getting More Accurate Surveys of Atheists
Getting accurate surveys about atheists and even just the numbers of atheists is difficult given how much animus there is towards atheists and atheism. Short of a radical shift in these attitudes, how might research firms obtain better, more accurate information about atheists? Would rewording the questions in ways that move the bias away from the traditional Christian form of monotheism help or is there some other way of generating more accurate surveys?