Percentage of atheists in federal prison is even lower than we had thought.

I’ll start off by saying that Hemant Mehta is the straight up shit.  Why do I say that on this occasion?

For starters, the “atheists are 0.2% of the prison population” statistic is unreliable.  Even though it gets used everywhere, Hemant caught wise:

The 0.2% number has also been cited in book after book after book — including Victor Stenger‘s New York Times bestseller — and op-ed pieces, all referencing the exact same data.

If the statistic is wrong, we must stop using it. But can we really confirm or deny this information?

Yes we can — and I finally have some definitive information to back it up.

And back it up he did.  Hemant filed a Freedom of Information request with the Federal Bureau of Prisons and they provided the information.  Hemant has the full breakdown over at his site, but the short story is that the percentage of self-identified atheists in the federal prison system is 0.07%.

This is peculiar if atheists are the ones running around stealing, killing, and otherwise pillaging.

Enormous salute to Hemant for putting in the effort to secure this information.  Someone get that man a trophy.

ACTIVISM & GODLESSNESS: Purpose-driven life and secularism.
POLITICS & GAY RIGHTS: Arkansas about to pass an anti-gay law just like Indiana’s.
About JT Eberhard

When not defending the planet from inevitable apocalypse at the rotting hands of the undead, JT is a writer and public speaker about atheism, gay rights, and more. He spent two and a half years with the Secular Student Alliance as their first high school organizer. During that time he built the SSA’s high school program and oversaw the development of groups nationwide. JT is also the co-founder of the popular Skepticon conference and served as the events lead organizer during its first three years.

  • RhubarbTheBear

    I have some admittedly anecdotal evidence (from an ex-con friend) to suggest that the option of self-identifying as atheist in our correctional system is usually off the table. While I’d like to see change in that area, I doubt it would budge the statistics much.

    • Silent Service

      I to have a friend that recently left the federal (military) prison system. According to him participation in religious activities, while not mandatory, are highly encouraged as good for the rehabilitation of inmates. Self-identifying as atheist or agnostic was perceived as a problem as you would be treated as not fully engaged in your own rehabilitation. Inmates that do not wish to sabotage their own parole chances avoid anything that might identify them as not fully engaged. That could include identifying as atheist.

      • RhubarbTheBear

        Thanks for elaborating on this point. For those who have no ties to our correctional system, it may seem very well to use it as a cautionary tale at best. But many of us don’t have that luxury; if we are not incarcerated ourselves, we may have family members and loved ones in the system. If we really are going to use these stats to bolster our arguments, it would be a bit hypocritical to not also support the rights of current inmates to identify as atheists.

        Wow, something I’m sort of passionate about. I bet this is what JT feels like most of the time. :)

  • Myrick

    This doesn’t mean we’re better people. There have been studies showing a correlation between non-belief and education/intelligence. So, perhaps we’re simply better at getting away with crimes.

    • John H

      Strong privilege dynamics bias the outcome despite atheists being a marginalized group in many circumstances. Given the statistical overrepresentation of well-off White men in the set of self-identified atheists, we are less likely to find ourselves in circumstances that motivate lawbreaking, and we’re more likely to get away with crimes we do commit (not even that we’re better at covering our tracks or ‘smarter’ about crime, just that we’re less likely to be arrested, tried, and convicted).

      We’re not necessarily better people (to tie that claim to incarceration rates at all, one would have to demonstrate that our laws are ethically perfect and applied perfectly as well, and then attempt to correct for factors confounding incarceration rates), but this does rather refute the idea that we’re categorically unethical/immoral.

  • Epinephrine

    It’s not a very useful statistic though – the comment section there is full of people pointing out that there are many reasons to claim Christianity (or simply to not state that one is atheist) in prison.
    Another point is that the data isn’t comparable to anything. Before my current position, I worked doing statistical analysis of health trends using national datasets – and we have trouble comparing results obtained from the SAME survey in different years, thanks to things like changes in the % of respondents contacted by phone vs. in person, use of computers to record data rather than paper, changes in the way the question is phrased or the ordering of the answers, etc. Making comparisons between different surveys asking the same question of a similar population is likewise tricky – is the CCHS question comparable to the NPHS question?
    Here you have a population that can be distinguished from the population of other surveys by so many factors – even assuming that they had the same question and options, you’d at least need to correct for all the confounds (income, education, race, etc…) to make any meaningful comparison, and that’s assuming that they have every reason to be giving the same answers as those outside of prison (which they may not). There can be factors that affect reporting – BMI for example is biased in different ways among different populations, and even decoding the non-response rates is a task, since non-response rates seem to vary with other predictors of BMI.

    tl:dr – Nice to have the data, but it’s of very limited usefulness, and quoting it without caveats as to its comparability is not a good idea.

    • Artor

      JT abbreviated quite a bit in repeating this info. Hemant goes to great lengths explaining why the data is not to be relied upon; either the .2% or the .07% statistics. But in any case, it’s clear there’s not a lot of atheists in prison. While the data might not be entirely clear, the trend is obvious.

      • articulett

        Exactly… if religions helped people behave more morally as religionists imagine it does, then we’d expect to see a higher percentage of atheists in prison than in the general population. Instead we see a lower amount. Whatever moral edge people imagine that religion gives them, it doesn’t seem to bear out in any measurable statistics. Like their gods, their moral superiority seems to be largely in the believer’s minds.

        • Epinephrine

          if religions helped people behave more morally as religionists imagine it does, then we’d expect to see a higher percentage of atheists in prison than in the general population. Instead we see a lower amount.

          No, that’s exactly what you can’t read from it. You can’t compare between the two as the methods, questions, and populations are so different. As I was trying to point out, you’d have to adjust for income, education, race, etc. to even have a chance of making a valid comparison – but on top of that, you have the question of what pressures there are on atheists not to self-identify in prison. If you look at the “nones” there are about 17%, which is surprisingly similar to the rate of “nones” in the general population.
          It’s really not a statistic you can use to make any useful comparisons – you could track it to see if it changes over the years, which might indicate something, but you can’t make any reasonable claim by comparing it to the rates reported in surveys of the general public.

          • articulett

            No– you are incorrect. You can’t make the claim that atheism makes you moral– however you can use this data point to say that atheism doesn’t lead to the immorality that theists seem to expect it would. If religion enhanced morality, se should expect to see a measurable different in some area having to do with moralit; for example, if religion made people less likely to steal– we should see evidence of such. We don’t. Religiosity is more closely associated with societal dysfunction than it is to anything positive. This is not evidence that religion leads to dysfunction (correlation is not causation), however it IS evidence that religion is not essential for morality– and societies can be healthy and function very well without it.

          • Epinephrine

            No, you need a bunch of assumptions that don’t hold for your method to work. For example, you can’t randomly assign people to the groups “atheist” and “religious”, and you can’t tell whether the rate of professing atheism are the same in prison populations and in non-prison populations.
            You state that you should see a measurable difference in something like rates of theft – but what you see in prisons isn’t rate of theft by religion, it’s rates of being convicted of theft by religion professed in prison – and while rate of theft should be proportional to rate of conviction of theft, it’s affected by factors like race and socioeconomic status that also are related to atheism (and I really don’t have a grasp on the rate of atheism vs. professed atheism, but at least anecdotally the two may be quite distinct). If the covariates of atheism protect against conviction you could see the same effect despite equal rates of theft. Or if convicted atheists claim Christianity for in-prison benefit you would likewise see a similar pattern.

  • Gehennah

    To be fair, there is probably slightly higher percentage there, but from what I hear, being an Atheist in jail is a bad thing and going to church and stuff makes it seem as if you are building character, for early parole.

    But I doubt that the percentage in jail is actually the same percentage out of jail.

  • islandbrewer

    Ah, my take on it is that American theists ARE more moral and helpful, according to this data:

    The United States is number 1 (We’re number one!!!1!!! Woooo!) in percentage of our adult population who are incarcerated. Religious people, wanting the US to maintain its lead, voluntarily go into prison at much higher rates than us horrible atheists, who are obviously unpatriotic and don’t care about our standing in per capita adult incarceration rates.