UCare generously supports MinnPost’s Second Opinion coverage; learn why.

Secular children show more altruism than religious ones, study finds

Children raised in secular households tend to be more generous and less punitive toward others than children raised in religious ones, according to new research published this month in the journal Current Biology.

“Our findings contradict the common-sense and popular assumption that children from religious households are more altruistic and kind toward others,” said Jean Decety, the study’s lead author and a neuroscientist at the University of Chicago, in a released statement. “In our study, kids from atheist and non-religious families were, in fact, more generous.”

The study involved almost 1,200 children between the ages 5 and 12 living in seven cities around the world: Chicago; Toronto; Amman, Jordan; Izmir and Istanbul, Turkey; Cape Town, South Africa; and Guangzhou, China.

The parents of these children filled out questionnaires about their religious beliefs and practices — and about how empathetic their children were toward others. Most of the parents described themselves as Muslim (43 percent), Christian (24 percent) or non-religious (28 percent). Other religions (or non-religions) represented included Jewish (1.6 percent), Buddhist (0.4 percent), agnostic (0.2 percent) and “other” (0.5 percent). 

Two experiments

In a variation of the “Dictator Game,” which is often used in behavioral economics research, the children were shown 30 stickers and instructed to select their 10 favorites. They were then told that there was not enough time for all the children in their school to receive stickers, but they could give some of their 10 stickers to their classmates by putting them in a special envelope. They were also given an envelope for any stickers they wished to keep for themselves. The researchers leading the experiment then turned away from the children, telling them to let them know when they were finished filling their envelopes.

The researchers assessed the children’s altruism by the average number of stickers shared.

In a second experiment, the children were shown short animation videos in which a character pushes or bumps against another, either accidentally or purposefully. After watching the video, the children were asked to rate (on a seven-point, child-friendly scale) the “meanness” of the behavior and the amount of punishment that should be administered.

Altruism findings

In their analysis of the data from the two experiments, Decety and his colleagues focused on differences between children living in Muslim, Christian and non-religious households. Children from the other religions were not numerous enough in this study to reach a statistically significant result.

The researchers found that the altruistic impulses of Muslim and Christian children in the first experiment were not much different. Both groups gave away, on average, slightly more than three out of their 10 stickers.

But that was significantly less than the children from the non-religious households. They gave away, on average, more than four stickers.

Furthermore, as the age of the children in the religious households rose, their comparative lack of generosity became even more pronounced.

Decety and his colleagues note that the children were asked to share their stickers with classmates from their own school and who, therefore, had a similar ethnic background.

“Therefore,” write the researchers, “the result cannot be simply explained by in-group versus out-group biases that are known to change children’s cooperative behaviors from an early age, nor by the known fact that religious people tend to be more altruistic toward individuals from their in-group.”

As background information in the study points out, some past research has suggested that religious adults give more to charity than their secular peers. Those studies, however, have tended to be based on self-reports of hypothetical giving.

“In fact,” write Decety and his colleagues, “a careful meta-examination of the studies measuring actual behavior shows that there is little evidence for such a positive relation.” 

Findings from second experiment

The second experiment revealed differences along religious/non-religious lines that were similar to those in the first experiment. Children in Muslim and Christian households tended to judge the pushing and shoving they saw in the videos as being “meaner” than did their peers from non-religious households. Both the Muslim and Christian groups of children also wanted, on average, harsher punishments for the characters in the videos doing the pushing and shoving.

Those results support other research that has found that religious adults tend to have more punitive attitudes toward people who have committed crimes, no matter what the level of the crime.

“For instance, within Christianity, fundamentalists tend to be more punitive and advocate harsher corrections than non-fundamentalists,” Decety and his colleagues write. “Moreover, Christians are also argued to view the moral wrongness of an action as a dichotomy and are less likely to discriminate between gradients of wrongness, yielding equal ratings for a variety of transgressions.”

A contradiction of parental views

Interestingly, the questionnaires filled out by the parents of the children in this study revealed that the religious parents were more likely than the secular ones to identify their child as being above average in terms of empathy and sensitivity to the plight of others

But that view appears to be contradicted by this study’s findings. It was the secular children who were, overall, more altruistic toward others and less judgmental and punitive about perceived infractions in social behavior.

The findings, therefore, “call into question whether religion is vital for moral development,” Decety and his colleagues write.

In fact, say the researchers, their findings support “the idea that the secularization of moral discourse will not reduce human kindness — in fact, it will do just the opposite.”

The study can be accessed and read in full at the Current Biology website.

You can also learn about all our free newsletter options.

Comments (6)

  1. Submitted by Dennis Tester on 11/09/2015 - 10:33 am.

    Here’s a little test

    Go down to Second Harvest sometime and count heads. How many kids are there representing their church and how many are there representing their secular family or public school?

    More food shelves and homeless shelters are helped by people of faith and their kids than atheists and their kids.

    Likewise, I always thought it was interesting when I’ve volunteered at Second Harvest that there’s more groups there helping out from evil corporate American than from the compassionate “non-profit community.” More valid than meaningless games played by the study’s authors.

  2. Submitted by Greg Kapphahn on 11/09/2015 - 01:05 pm.

    This Study is So FULL of Methodological Holes

    especially it’s failure to make even the slightest attempt to separate out what stripe of “religion” (conservative, moderate, or liberal forms of the various faiths) these families were involved in,…

    and it’s authors so clearly biased in the conclusions they draw from it’s very questionable statistics,…

    conclusions in which they paint ALL religious expression with a brush the darkness of which arises not from their “statistics,”…

    but from the preexisting dark attitudes toward “faith” which exist within their OWN hearts, minds, souls, and imaginations,…

    as to render their conclusions laughable if they weren’t so pathetically, transparently representative,…

    of the approach to “study” which begins with a presupposition, then seeks to find, create or manipulate evidence and statistics to prove yourself correct.

  3. Submitted by Beth-Ann Bloom on 11/09/2015 - 02:16 pm.

    Jumping to conclusions

    Likely in this age group you are seeing the impact of rule-bound behavior. Devout 5-12 year old Muslims are likely to be very good rule followers from observant families-this is the opposite behavior than the one tested for.

    It is unlikely that any conclusions can be reached from this “study.”

  4. Submitted by Ron Gotzman on 11/09/2015 - 03:39 pm.

    More testing please….

    I cannot wait until next week’s article that challenges the assumptions made in this study.

  5. Submitted by Jodie Castellani on 11/09/2015 - 05:53 pm.

    Flawed Study can’t produce Reliable Results

    Any first-year doctoral student can clearly see many confounds and problems with this study yet the sensationalistic media is taking this study viral. Why? Because it’s anti-religion of course. I also love that this study is (allegedly) about altruism yet some of the media are instead posting inflammatory headlines such as religionists are mean!” Mean? Being mean and selfish are too different constructs and altruism is something else altogether. They need to decide what the study is really about. First, Statistics 101 tells us that correlation does not imply causation. We also need to know the operational definitions of all of these constructs that are being thrown about. Christian? How did they decide who is representing Christians? Nominal, self-identified Christians who never attend church? Legalists who are there 7 times a week? Protestants? Catholics? Denominations of Protestants? These definitions alone are mind-numbing. Also, these are children. How do we know they fully ascribe to their parents belief systems? Do the 5 year olds have the cognitive development to even understand what that means? Then, we have to think about the confounds because they used samples from multiple nations with multiple culture and from different religions. These religions have very different theologies and value systems so how can they be compared wholesale. How do we know those results generalize anywhere outside their own culture? Then, what about the outcome measure? What about the strategy – those are very complex instructions for young children. Also, developmental psychology tells us that children’s moral development occurs in stages and yet they choose very young children to make this assessment–children who haven’t even achieved certain stages of moral development? They’re also lumping together 5 year olds and 12 year olds whose cognitive capacities and educational opportunities are inherently at odds. But this is just the start….there’s more but I’ll let the other academics tackle it.

    • Submitted by Joseph Hoover on 11/11/2015 - 08:49 am.

      Sensationalized by media but study still solid

      You start out stating there are many confounds and problems with the study but then go on attack the media and the way it is representing the study. Liberal or conservative the media loves to sell news and will often adjust news accordingly.
      Your issue with the study appears that it yet has to be replicated by others, and that there are many more avenues that need to be explored on the subject. However, it appears there isn’t an issue with the study itself.

Leave a Reply