Statistics

That Study Claiming Religious Kids Are Less Altruistic That I Said Stinks Was Retracted

Long-time readers will recall the original criticisms of the paper “The Negative Association between Religiousness and Children’s Altruism across the World” by Jean Decety, Jason M. Cowell, Kang Lee, Susan Malcolm-Smith, Bilge Selcuk and Xinyue Zhou in Current Biology. This post garnered, and still garners, huge numbers of readers and comments.

Thanks to alert reader Luke Breuer, we learn that the paper has been withdrawn. But only partially for one of the reason’s we identified. Here’s what they say:

In our paper, we reported cross-cultural differences in how the religious environment of a child negatively impacted their sharing, their judgments of the actions of others, and how their parents evaluated them. An error in this article, our incorrect inclusion of country of origin as a covariate in many analyses, was pointed out in a correspondence from Shariff, Willard, Muthukrishna, Kramer, and Henrich (https://doi.org/10.1016/j.cub.2016.06.031). When we reanalyzed these data to correct this error, we found that country of origin, rather than religious affiliation, is the primary predictor of several of the outcomes. While our title finding that increased household religiousness predicts less sharing in children remains significant, we feel it necessary to explicitly correct the scientific record, and we are therefore retracting the article. We apologize to the scientific community for any inconvenience caused.

Here’s the relevant original criticism:

You know what happened next. Wee p-values through the terrible abuse of regression on the pseudo-quantified answers. A picture showing one of these is at the top. Notice the wee p-values? That makes the findings scientific.

All those dots are the answers to the pseudo-quantifications for each kid. The flat surface is the regression (expressing this and nothing else: the change in the central parameter of a normal distribution representing uncertainty in “altruism”; did you think it was something different?). Notice almost none of the dots are near this flat surface? That means this model has no real predictive value.

The picture you can see above, or in the original article.

It’s good they “retracted” the paper, which means, and has always meant, repudiated, since there is no way to retrieve all old copies or prevent people from reading the paper on the eternal internet. But they repudiated it for the weakest of reasons, the form of the regression. Not even for using p-values!

They more so should have repudiated it for saying they well measured altruism and religiosity. Which they most surely did not.

If you believe anybody can measure altruism or religiosity, you have a simple faith. That science can measure whatever it puts its mind to. Which it most certainly cannot.

We’ve done this before, but let’s do it again. Moby Thesaurus for altruism:

Benthamism, Christian charity, Christian love, agape,
beneficence, benevolence, benevolent disposition, benevolentness,
bigheartedness, brotherly love, caritas, charitableness, charity,
commitment, consecration, dedication, devotion, disinterest,
disinterestedness, do-goodism, flower power, generosity, giving,
goodwill, grace, greatheartedness, humaneness, humanitarianism,
humanity, humility, largeheartedness, love, love of mankind,
modesty, philanthropism, philanthropy, sacrifice, self-abasement,
self-abnegation, self-denial, self-devotion, self-effacement,
self-forgetfulness, self-immolation, self-neglect,
self-neglectfulness, self-renouncement, self-sacrifice,
self-subjection, selflessness, unacquisitiveness, unpossessiveness,
unselfishness, utilitarianism, welfarism, well-disposedness

Put that into one whole number on a scale from 1 to 5. And then do the same for religiosity. Then correlate the numbers, via statistical manipulation like regression, and then claim a causal force between the two. It’s not clear which causes which, but that doesn’t matter. No one will think to question you. Say greater altruism causes lesser religiosity, or greater religiosity causes lesser altruism. If your p is wee, you can get away with either.

Yes, it is possible to be more or less altruistic, or religious. This is what convinces people that numbers can be put to emotional (or intellectual) states. Higher is more, and surely 5 is more than 4. But is 5 – 4 the same distance from 4 – 3 and so on? For everybody? In every culture? For all time?

Are there no shades of meaning of the words, as the non-scientific thesaurus appears to indicate there are? Does every single person everywhere for all time hear the word “altruism” and respond in identical fashion?

And so on and on. It’s not that numbers cannot be put to things, it’s that putting numbers to things badly over-estimates certainty. As is spoken of in greater depth in this internationally available, award-eligible book.

Here’s Why That Study Claiming Religious Kids Are Less Altruistic Stinks.

Categories: Statistics

5 replies »

  1. Shock, shock. The wrong variable was credited. Oh well, it looked scientific.

    Love the thesaurus listing!

  2. In my unscientific observations of the friends of my teen daughters, the ones who are most “sharing” tend to be those who carry debit cards given to them by their parents. Most come from two income white collar families. With a virtually unlimited balance on their cards, it is easy for them to be generous. I am a single dad and we must stretch to make ends meet. The only money my daughters have is earned from babysitting. While they do share, they can’t do so to the extent their friends do. It has nothing to do with religion.

  3. I remember that study… the conclusions didn’t seem to fit the evidence. And, the evidence seemed conveniently arbitrary. Glad to hear it has been debunked “by their peers”.

  4. Another thing that you forgot to mention, and that further refutes said study is that in the same retraction of the authors it says that that of “punishments a little harsher” there was only a slight correlation in Muslims, and not in Christians, In addition to the 1170 and so many children in the 1/3 study, their parents were (not religious) or moderate religious “that is, in that sense, many were not necessarily atheists or agnostics, only (they did not go to church so much), or they had not had so many “spiritual experiences” although that is more ambiguous because not even the majority of religious people have had them, and another important fact that confirms why this study is useless
    THE ONLY THING I FOUND IN THIS STUDY IS THAT CHILDREN FROM MORE RELIGIOUS FAMILIES ARE ONLY 6 OR 7% LESS LIKELY TO SHARE THEIR STICKERS WITH STRANGE SCIENTISTS than children from “moderately religious or non-religious” families
    Although the authors still insist that this is “altruism” as you said and others that 7 or 6% can be attributed to statistical chance, or if they still insist that there is a slight “correlation” it can also be positive that the More religious kids are at least slightly less influential in giving their stickers to strange scientists, either because they want to give them to their siblings, or they want to share them with someone else, or they just don’t try to impress adults, but like I said that assumption 7% could perfectly be statistical chance
    https://www.cell.com/current-biology/fulltext/S0960-9822(16)30670-4?_returnURL=https%3A%2F%2Flinkinghub.elsevier.com%2Fretrieve%2Fpii%2FS0960982216306704%3Fshowall%3Dtrue

Leave a Reply

Your email address will not be published. Required fields are marked *