There Are More Atheists Than We Thought?

In a 2017 study, psychologist Dr. Will Gervais determined that polling organizations like Pew or Gallup are underestimating the share of atheists in the US population. How many Americans are atheists? How plausible is Dr. Gervais' study?

Twitter: @andrewmarkhenry

Bibliography:

[Orlymusicboy]

Living at home at about age 20, I was horrified to learn that my father had put our entire non-religious family down as Anglican. He didn’t even have a reason - he just filled that in as if it was obvious to lie.

[badatpseudoscience]

It would be interesting to test the actual phrase "I am an atheist" against phrases like "I do not believe in a higher power".

[corrda]

Many people I know have trouble answering a question like that. Many people don't understand what Thiest, Athiest and Agnostic even mean, yet alone know where they fit in. That 3-11% therefore is "people who are Athiest and know what Athiesm is". Which doesn't include a lot of Athiests who know very little about their own beliefs or lackthereof.

[pchemist]

Another confounding issue: I'm a Buddhist, so also an atheist, but often those are counted as separate categories.

[MusicIan]

All I know is anecdotal, but I meet atheists pretty regularly, and agnostics very regularly. I live right in the "Bible Belt" too.

[rosemaryoverton]

Most teenagers seem to be atheists at my son's school. I am a Christian, but I'm a firm believer in knowing what I believe and also knowing about other religions as well. Your channel is very informative and easy to understand. Thanks.

[RobinPillage.]

This is definitely a very complicated subject. Most people don't want to have to deal with the backlash from family, profession etc. There's a lot of closet atheists out there that don't want others to know and even more that have yet to go ahead and just admit it to themselves. It's easier for these people to just play along knowing really how they feel or just pushing it out of their mind and falling just short of even admitting it to themselves. Going from a believer to a non believer is a major process for most people. So I don't know if the number for atheists is too low or the number for believers is too high.

[riflemanracing]

Aren't most of the surveys done over land lines? I'm in my mid 30s and I have never had a land line ever. I think traditional polls miss a lot.

[nielsdaemen]

*I can't wait for day the first atheist US president is elected!*

[BigKevSexyMan]

Could be true. We would need more surveys with differing methodologies before we can get a truly accurate number.

[museofchartreuse]

"I do not believe in God" might be problematic to answer for polytheists and those who believe in the Goddess. It 
should be rephrased to "I do not believe in any deities" .

[StefanMilo]

I can't really speak on his methodology but in my own personal opinion 3-11% seems like a very low result.

I imagine many feel uncomfortable saying they are an atheist as they think it has some sort of stigma.

[timothymclean]

It wouldn't surprise me if this were true. The wide range of estimates in other results (3-11%, with one estimate being almost _four times_ as high as the other) suggests that _something_ other than peoples' beliefs is heavily influencing the results.

[TheWizardYeof]

As a recent high school graduate, I can honestly say most people I know aren’t religious/don’t believe in deities. It just really isn’t a thing many people care about these days, and I’m very happy to see it happening.

[nelsongalvan]

I think it is true. A lot of people associate God with God of the Bible and in order to believe in "God", you have to believe that 100% of the Bible is literally true.
Also people associate "atheism" with being a naturalist/materialist and thoygh many people don't believe in the God of the Bible, they may believe in the supernatural, the afterlife, and all sorts of superstitions.

[_lrr_]

One thing that you didn't mention in the video and I think is important is this: the study referenced is still in pre-print, which means that the version currently available has not been approved by peer-review. Note that the pre-print has been available (in some version) since January 2017, indicating a long review process possibly (although not necessarily) due to weaknesses in the methods, data, and/or interpretations. Maybe even involving one or more rejections and re-submissions to other journals along the way. Also, the last edit of the pre-print is from July 2018, indicating that the authors are actively working on this. All this doesn't mean the conclusions are wrong, or the manuscript is of lesser quality, it just means that we have to be very careful with the way we treat their results and read their conclusions with an even more critical eye.

The technique is actually using a very simple but very cleaver trick. As you said, you divide (at random) your population in two: test and control. First you measure the number of questions that are true for morally innocuous statements in your control group. Lets say you get that, on average, 3.5 out of 5 statements are true for people. Then you do the same for the test group, but adding the morally charged question (I believe in God), and assuming that the base rate for the 5 innocuous questions is the same, you can calculate the rate for the morally charged question. For example, if this time you get that, on average, 3.7 out of 6 questions are true for people, you can infer that (on average) 3.5 corresponds to the first 5 questions, while the other 0.2 corresponds to the extra question, therefore the statement is true for 20% of the people.

[brettd]

I know this is an older video, but I'm reminded of how my family would insist on lying and putting us all down as "Baptists" on everything because they thought that, living in the Bible Belt, we'd get better medical care if that was our stated affiliation.

[febodelenda]

In Italy polling 'shyness' has been repeatdly pointed at exolaining the failure of polls not predicting the electoral success of Berlusconi.

[jeremykientz]

I was nervous about calling myself an atheist for years... then I grew a pair lol

[solentbum]

Some 30 years ago I, along with others, was required to answer two questionnaires as part of my application for a sensitive position at work. After finishing the two papers we, as a group, were informed that of the nearly 300 questions only four were actually related to the job. The majority of the questions were designed to settle the subjects into a frame of mind to give truthful answers of the type that could be considered 'difficult', such as stealing, equality and similar.
The methodology suggested by Dr. Gervais seems to be based on a well tried technique to find the truth.