The big scientific news today seems to be around a paper that was just published in the Journal of Current Biology, which shows how children from religious families are more selfish and mean compared to those from non-religious families.
Here’s the article: http://www.cell.com/current-biology/pdf/S0960-9822(15)01167-7.pdf
I see this news being paraded around by some like a trophy. “A win for the non-religious!” they say. “Religious people are immoral!” they chant. Calm down people…this is science not an inquisition.
This study, just like any other, is part of a process of empirical inquiry. That means that it is vulnerable to falsification and that the data must be carefully considered before generalizations are made. So let’s be critical.
- I’ve conducted studies using the Dictator Game, which is the method used by these researchers to measure generosity in children. The Dictator Game can be time consuming to set up and to deliver. I’ve also gone through the process of contacting schools to recruit children as participants in research – a major headache. Combining the Dictator Game with children recruited from schools presents a major challenge to this study: Selection bias. Selection bias is where the process of selecting participants for a study is non-random, which poses a problem for validity. In this case, I’m mostly concerned with sampling bias due to attrition or nonresponse, which would mean that the sample is not representative of the population being studied: “religious people.” More science.
- The models presented in this article only explained 18-20% of the variability in sharing behaviour. Although typical of studies attempting to predict human behaviours, this also means that 80% of the variability in sharing behaviour could not be predicted by the models they reported. That’s a lot of explainin’ that needs to be done! More science.
- The effect sizes reported were small to moderate. In the social sciences I feel like we tend to dismiss the magnitude of effects, as long as effect sizes are reported (IMHO). In clinical sciences, low effect sizes are often the result of poorly or inconsistently delivered treatments. In this case, the “treatment” was “religiousness.” I suspect that, given this was an inter-cultural study, the delivery of “religion” was dependent on cultural norms. Therefore, there’s going to be lots of inconsistency between cultures and less between religions within cultures. How do you measure the influence of culture? More science.
- In terms of measures, the terms “religiosity” and “non-religious” or “atheistic” are thrown around a lot in the social sciences, as are the tools used to measure them. My trouble with these terms is that they often don’t tend to capture the nuances of religious belief and practice. One thing to consider is that the tools used to measure these constructs are largely developed and tested in North America, and reflect a North American bias regarding religiosity. We need a broader debate regarding the validity of these measures before we start evaluating research based on their use. More science.
- This study was severely limited by its definition of religious identity. Only two groups had adequate sample sizes to be included in the final analysis (Muslims and Christians), and these two groups share a lot in common when it comes to measuring “religiosity.” Personally, I think they could be lumped into one sample for a study like this. Also, using “non-religious” as a comparison group introduces its own issues. Self-identifying as non-religious can mean many different thing. Are these individuals predominantly Atheist? In my view, self-identifying as an atheist is as belief-driven as calling yourself religious. Do self-identified atheists train their children in morality more systematically in spite of Christians and Muslims? Are the Christians and Muslims in this study those who may simply believe that mere adherence guarantees salvation? I suggest further inquiry. More science.
I haven’t been as thorough as I would like to be, but this should give you a sense of how complex research is, and the type of questions we could be posing as critical consumers of information.
Science is a means of generating knowledge and understanding our reality, let’s not use it to to further divide us.