Field of Science

The happiness smile

Earlier this year I blogged about a study done by Luke Galen, a psychologist at Grand Valley State University. Well, he's now written an account of it in Free Inquiry - you can download it free here.

The study is a survey of the labels non-believers apply to themselves, as well as a survey of their personality types and other demographics.

One fascinating finding is some evidence that contradicts the widely held assumption that religious people are happiest. What he points out is that, because studies on happiness usually include few atheists, they are often lumped in with the 'undecideds'.

But what his study suggests is that it's precisely these fence-sitters who are the unhappiest. People who are firm in their convictions - either firmly religious or firmly non-religious - tend to be happier. Here's how Galen puts it:

When we distinguished strong varieties of nonbelief, such as atheism, from weaker nonbelief, a curvilinear relationship emerged (see Shaver, Lenauer, and Sadd). Those nonbelievers most confident in their nonbelief tended to be the most emotionally healthy, relative to the “fence sitters” who reported more negative emotions. Similarly, life satisfaction was lower among the spirituals relative to the other three belief labels. Therefore, having uncertainty regarding one’s religious views appears to be associated with relatively greater emotional instability.
Now, the problem with Galen's study is that it's not a random sample of nonbelievers. He recruited members of the Center for Inquiry, who by their nature are going to be 'joiners' rather than loners.

So I took a look at the World Values Survey. This survey took in over 50 countries, and is a random sample of people in those countries. Among a range of questions they asked how happy people were, and how important religion was in their lives.

I lumped together all the responses from all the countries to get this chart of happiness by 'importance of religion'.

Take a look first at the people who are 'very happy'. Sure enough, people for whom religion is not at all important are more likely to be very happy than people in the middle (although people who are religious devotees come out top).

So it looks like Galen is right!

But look at the 'Quite happy' line. It's the mirror image! So being firm in your convictions has the effect of shifting you from being quite happy to very happy. But it doesn't seem to have much effect on people who are unhappy.

Now, this is a very top-line analysis. No doubt there are all sorts of demographic reasons that might explain these patterns (they're probably linked to gender and social status, for a start).

But I think the important take home is that the naive belief that more religion = more happiness needs to be challenged.

By the way, the study website is here, but still under construction!

Edit: I added a new figure. Basically the same, but divides people according to whether they are religious, not religious, or a confirmed atheist.

Creative Commons LicenseThis work by Tom Rees is licensed under a Creative Commons Attribution-Share Alike 2.0 UK: England & Wales License.


  1. I agree with your take home message that more religion = more happiness needs to be challenged.
    All theists may not say that more religion = more happiness.
    Religious beliefs/behaviors can make life more stressful.A tribal in India may be concerned about how to finance the witch doctor for the goat that needs to be sacrificed on the new moon day and so be unhappy due to the anxiety he experiences.
    A Christian, who refuses to pay a bribe to a Government official (because of his values) to get a driving license might have to keep going back to his office for months together to get was was due to him.This could lead to unhappiness.
    The goal of religion may not be just "making people happy".It is for overall beneficence of individuals and communities,even if it is painful and difficult at times.

  2. The graph can be interpreted as
    "Among people who are experiencing extremes of emotions,greater proportion of them have stronger convictions regarding religion.
    Those who are experiencing emotions moderately,greater proportion of them have moderate convictions regarding religion."

    But from this given data,we cannot say if emotions have led to convictions or convictions have led to emotions.It could be either way.

    Though my religious nature could be more stable than emotional states,my own interpretation of important religion is to me could depend on my emotional state at the time of inquiry,creating possibility for biased results.

  3. The problem with the values survey regarding my findings on the curvilinear relationship between belief and life satisfaction and emotional stability is that the lowest category is: religion is "not important". This conflates those who are indifferent to religion and those who have strong disbelief. What my study found was that those with strong disbelief are happier than those with unsure beliefs. Most of the previous research tends to use measures that lump those categories together or uses some measure like church attendence that does the same lumping. But there are some previous studies that found the same curve effect:
    Ross, C.E. (1990). Religion and psychological distress. Journal for the Scientific Study of Religion, 29, 236-245.
    Riley, J., Best, S., & Charlton, B.G. (2005). Religious believers and strong atheists may both be less depressed than existentially-uncertain people. Quarterly Journal of Medicine, 98, 840.
    Shaver, P., Lenauer, M., & Sadd, S. (1980). Religiousness, conversion, and subjective well-being: The "healthy-minded" religion of modern American women. Journal of Psychiatry, 137, 1563-1568.
    Buggle, F., Bister, D., Nohe, G., Schneider, W., & Uhmann, K. (2000). Are atheists more depressed than religious people? Free Inquiry, 20, 50-55.
    Diener, E., & Clilfton, D. (2002). Life satisfaction and religiosity in broad probability samples. Psychological Inquiry, 13, 206-209.
    Eliassen, A.H., Taylor, J., & Lloyd, D.A. (2005). Subjective religiosity and depression in the transition to adulthood. Journal for the Scientific Study of Religion, 44, 187-199.
    Maselko, J., & Buka, S. (2008). Religious activity and lifetime prevalence of psychiatric disorder. Social Psychiatry and Psychiatric Epidemiology, 43, 18-24.

  4. Also, (this is Luke by the way),
    I controlled for age, sex, income, marital status, and education, and the effect was still significant. In the large CFI sample, the groups were all equally educated. So the curve effect is not attributable to those demographic differences.

  5. Dheeraj, I think that doing things because of your principles will make you happy, rather than unhappy, even if they result in financial or other loss. I suspect that the problem comes with people who do things because they are under social pressure, but they don't do it from conviction.

  6. Luke, the 5th wave WVS also included a question on whether the respondent was a 'convinced atheist'. I've edited the post to include a figure showing those. There's a hint of a curve effect.

    I also wonder whether the WVS in the original graph is an artefact. It may be that their might be a personality type that is more comfortable with responding at the extreme of the range. In other words, it might simply be that people who answer at the top or bottom of the range on one scale are more likely to answer top or bottom on others.

    Interesting that your result is after controlling for those demographics. Did it weaken the effect, though? It would be interesting to know who the categories interact with the demographics.

  7. Tom,
    I'm trying different types of statistical analyses currently. Using the demographics (age, sex, education, income, married) only eliminated the overall sigificance once, for the church vs. local CFI groups for life satisfaction as the D.V. Otherwise, the overall sig holds using those controls. The other main culprit that i'm exploring is social support. Because in previous research those with greater belief conviction tend to be involved more and receive more social support. On that variable, the church groups really outperform the secular people.

    i'm assembling other research on moral behaviors and the curvilinear effect to see if i can maybe write a review paper. There are several moral categories where the highly religious and the atheists look very similar morally with the fence sitters intermediate. For example Bock and Warren found that the extremes were more disobedient (humane) on the Milgram shock paradigm. Alcohol consumption also often looks curvilinear with religion. Oliner's data with helpers of Jews during WWII also looks curved. So there may be behaviors in which those with general conviction, regardless of the relgious content, is higher.

  8. Luke, will you send me an email if you do get a review out? It's a very interesting topic!

  9. In his NYT review of Terry Eagleton's book, Stanley Fish declared that, "[t]he religions I know are about nothing but doubt and dissent, and the struggles of faith, the dark night of the soul, feelings of unworthiness, serial backsliding, the abyss of despair."

    Apparently, the joyous believer is insufficiently devout.


Markup Key:
- <b>bold</b> = bold
- <i>italic</i> = italic
- <a href="">FoS</a> = FoS