Why the Latest Study Showing 'Religious' Children are Less Moral is Just Bad Science

11/11/2015 10:58 | Updated 07 November 2016

The Hindenburg Disaster - Bad science never ends well

Word is out. Belief in God will make your children less moral people, so say researchers from the University of Chicago. It's hit the news too, and looking at the comments sections, boy, those un-judgmental atheists are really showing how humble they can be.

1,151 children aged 5-12 participated, from USA, Canada, South Africa, Turkey, Jordan and China, with 23.9% from Christian families, 43% Muslim and 27.6% non-religious. Altruism was tested by asking children to choose ten stickers (of thirty), to keep and to see how many they would be willing to give away, after watching a video of children being physically bullied. How judgmental the children were was measured by how "mean" they viewed the video and how severely they thought the culprits should be punished.

Religiosity of the parents was also determined by asking parents if they identify as religious, noting frequency of attendance to religious functions and asking if home was a spiritual place or not. Socioeconomic status (as measured by maternal education) and the child's disposition were also assessed.

The results? Non-religious children were more altruistic, less judgmental and less severe on bad behaviour. Thus, the more religious your upbringing, the less moral you are likely to be! Right?

Wrong. Here's why.

Firstly, the conclusion is totally unsupported by the evidence. The study shows a correlation (not causation!) of -0.173 between religiosity and altruism. Correlation is measured on a scale of -1 to +1 with 0 meaning no correlation. To draw the authors' conclusion from this meagre result is laughable. This small correlation indicates that other unaccounted factors are at work. What could they be?

The first is that the authors did not publish which children were from which country. Taking all the Muslims from Jordan and Turkey, all the Christians from Chicago and all the atheists from Toronto or China will result in very significant cultural biases along religious lines. Take China - the most atheistic nation on Earth. Might the tendencies of Chinese children to share perhaps have less to do with their atheism than their communist upbringing?

Here are some more.

"Religiosity" was also biased by culture: Muslims pray five times a day. How many Christians go to church that often in a week? Frequency of mosque/church attendance is therefore a biased tool. Religious self-identification is also biased as the Middle East regards religion more positively than places like China, where you could be arrested. Additionally, the authors didn't look at whether being more atheist makes you less altruistic. Why only tell half the story?

The political outlook of families was not at all accounted for either. Turkey recently voted in Erdoğan with a conservative majority while Canada just voted in Trudeau with a liberal majority. Conservatives have been shown to use physical discipline with their children more often than their politically liberal counterparts. Is it not possible that such differences in parenting techniques may also affect a child's opinion as to how severely bad behaviour should be dealt with, independent of religiosity?

Socio-economic status was also poorly assessed. In religiously conservative families, the father's level of education is more reflective of socio-economic status than the mother's - he more often, being the breadwinner. In 2007 in Turkey, 24% of unemployed women had tertiary level education, as compared to 46% in Jordan. In 2013 in Jordan, female unemployment rate was 11% as compared to 3.8% in China. How can measuring maternal education be a valid marker of socio-economic status, with such vast differences?

Interpretation of the study was similarly biased. That religious children had greater negative reactions to bad behaviour could be interpreted as greater abhorrence of injustice, yet instead, the authors framed this negatively, as judgmental tendencies. That religiously educated children may punish bad behavior severer than the non-religious could merely indicate an immature but developing moral compass. The leniency of non-religious children could be interpreted as a lack of a moral compass in the first place and an inability to recognise wrongful actions.

In short, analysis of this study reveals a greater interest in grabbing headlines rather than serious science. Humility is essential for honest research, and especially so when drawing conclusions about the moral tendencies of all Christians and Muslims - 3.8 billion people, constituting 48% of the world's population - especially when your correlation values are as weak as -0.173. Given the moralistic concern of the study, it is ironic that humility is the one key ingredient profoundly lacking in its authorship.