Field of Science

Why women are more religious. Part 2: It's not about risk at all

I've been spending some time rooting around in the research that's been done on religious differences between men and women. The previous post ruled out the idea that women are simply less willing to take risks about getting to heaven. The evidence is pretty convincing on that score.

But that doesn't rule out the possibility that it's something to do with risk. Religious people are more risk averse than non-religious, and women are more risk averse than men. It's very tempting to say that the two are linked. Maybe it's about risk in this life, rather than the next. After all, many people believe that their god will reward their good behaviour in some material way. They might also pray before doing something risky or dangerous, in the hope that that influence the odds in their favour.

Unfortunately, it doesn't look like that's the answer either. Jeremy Freese, at the University of Wisconsin-Madison, has reviewed survey data from the USA and Italy. These surveys were general-purpose, but they did include a measure of risk aversion: they asked respondents whether it is better in general to be cautious or bold. After controlling for age and income, Freese found that risk aversion didn't help explain the gender difference in going to religious services or in the importance of religion.

In fact, perhaps what looks like a solid relationship between risk attitudes and religiosity is not quite what it seems. There's an intriguing undergrad study looking at the relationship between risk aversion and religiosity, and it found that the answer you get depends on the order in which you ask the questions. If you ask about religion and gender first, and then risk aversion, you get the expected result: women and more religious people report that they are more risk averse. But if you ask about risk aversion first, and only then ask about gender/religion, the relationship disappears. In other words, if you remind people of their social status, then they will give you a response that matches the social stereotype. But you have to remind them first!

Well if it's not risk aversion, what is it? Take another look at what the sociologist Rodney Stark said in his 2002 paper:
The upshot is that some men are shortsighted and don't think ahead, Stark said, and so "going to prison or going to hell just doesn't matter to these men."
The evidence he used to support this claim put forward some evidence about risk aversion, but if you think about it this isn't anything to do with risk at all. It's about delayed gratification - your ability to resist temptation for a longer term goal. But just because Stark doesn't have any evidence, doesn't mean that it's wrong. Maybe this is why women are more religious.

However, the differences between men and women on this score are actually much smaller than most people think. In 2002 Irwin Silverman, a psychologist at Bowling Green State University in Ohio, analysed over 30 studies into sex differences in delayed gratification. He found that the gender gap was so small as to be almost undetectable.

What's more, Stark's argument doesn't make sense if you think about it. Delaying gratification is not necessarily a good thing. Ever heard the proverb 'A bird in the hand is worth two in the bush'? The psychological term for this is discounting - meaning that people put higher value on things they have now, rather than things they might have in the future. That makes sense, because the future is uncertain (you might die first, for example). The further away it is in time, the less chance you have of actually getting it. People who go for immediate gratification do so because they have a high discount rate. Arguably, men have a higher discount rate because, in our evolutionary past, men faced a high risk of being killed in early adulthood.

But that's irrelevant for religious rewards. One thing you can be certain of is that you will eventually meet your maker. The fact that it might be near or far doesn't affect the odds of that. The discount rate doesn't apply, and there's every reason to delay gratification.

So it seems that, whichever way you cut it, Rodney Starks' argument fails the test. Unfortunately, that still doesn't help explain why women are more religious than men. I have some ideas about this from the research I've read, and I'm going to address them in my next (and final, promise!) post on this topic.

But to finish off, here's a teaser. Women are not just more religious than men, they are also more superstitious. The best evidence for this comes from a 2007 paper by Benno Torgler, an economist now at at UC Berkley. He used data from the 1998 International Social Survey on religion, and found that whether he looked at astrology, fortune tellers, or good luck charms, there was always a strong effect of gender. What's more, superstitious people were less likely to be churchgoers, but more likely to consider themselves to be religious people.

Now, superstition has nothing to do with the prospect of rewards in the afterlife, but it is about helping maximise outcomes in this life. So perhaps there are some clues here?


ResearchBlogging.orgFreese, Jeremy (2004). Risk preferences and gender differences in religiosity: evidence from the World Values Survey Review of Religious Research, 46 (1), 88-91

Silverman, I. (2003). Gender Differences in Delay of Gratification: A Meta-Analysis. Sex Roles, 49 (9/10), 451-463 DOI: 10.1023/A:1025872421115

Torgler, B. (2007). Determinants of superstition. Journal of Socio-Economics, 36 (5), 713-733 DOI: 10.1016/j.socec.2007.01.007


  1. About forty years ago I read the hypothesis that women tend to be more superstitious because their social position, monetary success, and personal happiness depended more than men's on a whom they married, and that there was no way to know reliably how their choice would turn out.

  2. Interesting theory, Mona. That might mean that in general, superstition correlates inversely with the level of attained power (minus the kings of the hills, maybe, who tip over the edge into paranoia and need superstition back). Easily testable.

  3. Hi Mona, I think there's good evidence that that (or something like it) is part of it - inasmuch as people tend to believe in superstition if they sense that they are not in control of their lives.


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