No sacred cows

This junk study proves nothing about helicopter parenting

23 June 2018

9:00 AM

23 June 2018

9:00 AM

An academic paper by a group of child psychologists caused a stir earlier this week. ‘Helicopter parenting is bad for children,’ was how the Times reported it, and other news outlets summarised it in the same way. Here was proof, apparently, that wrapping your children in cotton wool and limiting their exposure to risk is bad for their emotional development and can lead to problems at school, as well as difficulties in later life.

A few years ago, when I was in the first flush of fatherhood, I would have leapt on this study as confirmation that my laissez-faire attitude to parenting was more effective than the more hands-on approach of my peers. Indeed, I have written columns in the past praising parents who leave children to their own devices and criticising schools for protecting them from failure. I’m a big fan of The Dangerous Book for Boys by Conn Iggulden and am constantly hurling my children up rock faces and telling them not to be so wet when they get stuck.

But I’m now more scientifically literate than I was and, having given this study a careful read, I’ve concluded that it’s an example of what Richard Feynman called ‘cargo cult science’. That is, it gives the appearance of being a robust piece of research, and uses lots of impressive-sounding technical lingo like ‘indexed’, ‘subscale’ and ‘covariates’, but it doesn’t actually tell us anything. In particular, it doesn’t tell us whether over-controlling parenting has a negative or positive effect on children’s ability to regulate their emotions. In Feynman’s words, the planes don’t land.

Why do I say this? Because the researchers overlook the fact that their findings could be confounded by genetics. This is a shortcoming of much academic research in developmental psychology. A typical study examines whether there’s a correlation between, say, children’s reading level at the age of five and how often they’re read to by their parents. If there is, the researchers conclude that there’s a causal relationship between the two. But we know that children’s reading levels are genetically influenced, so if the study doesn’t adjust for the fact that children share their parents’ genes the conclusion is worthless. To paraphrase Feynman, it’s junk science because the researchers haven’t considered other causes that could explain the result of their experiment.

In the case of the ‘helicopter parenting’ study, the researchers looked at how 422 different two-year-olds interacted with their mothers when they were asked to play together for four minutes and then tidy up after themselves. The mothers were then given a score of between 1 and 4 for each part of the experiment according to how ‘over-controlling’ they were, e.g., whether they constantly guided their child, created a structured environment, repeated commands, and so on. The researchers then examined the children again at the ages of five and ten to see if there was a correlation between how neurotic their mothers had been when they were toddlers and how good they were at regulating their emotions.

But the researchers made no attempt to adjust for the fact that children share their parents’ genes. That’s an unfortunate omission, given that we know that nearly all individual differences in measurable psychological traits are genetically influenced. One of the reasons we know this is thanks to the work of Thomas Bouchard, a psychologist at the University of Minnesota, who based his findings on studies of twins reared apart. Coincidentally, the lead author of the ‘helicopter parenting’ paper, Nicole Perry, is also a psychologist at Minnesota, which makes this oversight even more unforgivable. To cap it all, many of the findings of Bouchard and others have been confirmed by genome-wide association studies, so Dr Perry and her colleagues really don’t have an excuse.

No, I’m afraid this is cargo cult science. We know from the work of Judith Rich Harris and the behavioural geneticist Robert Plomin, among others, that the vast majority of parents have next to no influence on how children differ from each other. (Obviously, if a parent is extremely negligent or abusive then that does have an effect.) The individual personalities of your children are largely determined by a combination of their genes and their ‘non-shared environment’, i.e. pure happenstance. So ‘helicopter parenting’ won’t have any impact. Turns out my laissez-faire approach is fine after all, but not for the reasons set out in this junk study.

Got something to add? Join the discussion and comment below.

You might disagree with half of it, but you’ll enjoy reading all of it. Try your first 10 weeks for just $10

Show comments