A recent study, headed up by Jean Decety, has been used to suggest that secular individuals are more generous and moral than religious individuals. In fact, Decety argues that his research shows morality to basically be a part of our biology and that religion not only fails to add to our morality but even appears to detract from it. Thus he argues that a secular worldview is best for us. Finally, he states that evangelical Christians will not accept his work because they “don’t want science.” But there are serious methodological problems with Decety’s study, problems a trained social scientist should recognize, never mind his or her worldview.

The difficulties starts with Decety and his co-authors using a measure known as “the dictator game” to determine the level of altruism among their subjects. However, it is highly questionable that this game can assess altruism. This tool instead may measure compliance to the instructor of the game.

The authors also wanted to look at issues of morality by examining children’s punitive nature when confronted with mean actions. The authors clearly interpret more mercy for the guilty as a higher level of morality. But that is an ethical, and not a scientific, question. Is it really more moral to avoid punishment no matter what the circumstances? Indeed, there is a balance between punishment and mercy that all of us must consider. Some tend to err on the side of punishment and some on the side of mercy, but almost all of us will punish and almost all of us will have mercy under the right conditions. Is it moral to punish a mass murderer with a $100 fine? Most of us would consider that an injustice. Is it moral to punish a traffic ticket with a ten-year jail sentence? Most of us would consider that unmerciful.

We all draw that line between mercy and justice somewhere. It seems that the authors have determined that the proper place to draw the line is closer to the mercy side. They judge the kids closest to where they will draw the line as more moral. Since at least the lead author has identified himself as secular, it is not surprising that secular children are closer to that line than religious children.

The authors are welcome to their opinion as to what the balance of mercy and justice should be, but since the question is an ethical, and not a scientific, question, this study is inadequate for assessing whether secular kids have higher morality.

I was also concerned about the lack of certain control variables in the research. For example, it seems to me that the education level of the parent would be important. In many countries secular individuals tend to have higher levels of education than religious individuals. Their kids likely went to different schools, which may socialize children to respond differently to the dictator game. (Notice I did not say “to be more altruistic” since I am not convinced that the dictator game measures altruism.) It may be that the educational level of the parents, rather than secularity, shaped the results of this research.

I would also like to have seen controls for religious attendance, gender and ethnicity as well. A solid research project also would have included qualitative measures such as interviews and focus groups to see if secular kids understand morality differently than religious kids do. Then the researchers would have been in a position to interpret the meaning of any quantitative differences between the two groups. This current study is devoid of such a qualitative assessment, so it is unwise to make the strong assertions that Decety made.

Rarely do social scientists decide a research question on the basis of a single study; but if they were to do so, they certainly would not do so with research using such questionable measurements.

Moreover, to understand any research one should explore what others have found about the topic. There is other research looking at the relationship of generosity and religion. Religion has been shown to correlate to the willingness of individuals to volunteer, to give money to charity and even to be nice.

That research focused on adults instead of children, but this produces important questions. If religious children are so selfish and immoral, then how do they become generous, friendly adults? Conversely, how do the altruistic secular children grow up to be, on average, less generous and kind as adults than their religious counterparts? The authors do not seem to have grappled with the greater research literature on this topic. If they had, then they would have addressed those questions and been a little more circumspect about making pronouncements about secular superiority in morality.

I am an evangelical Christian and I do not accept Decety’s assertions about generosity and religiosity. But Decety is wrong in claiming that my rejection is due to the fact that I “don’t want science.” Given the weaknesses in his study that I just outlined, it is more accurate to say that I don’t want bad science.